CN118701084A - Vehicle, vehicle control method, storage medium, and electronic device - Google Patents

Abstract

The present invention provides a vehicle, a vehicle control method, a storage medium, and an electronic device, wherein the vehicle includes: a perception and computing component, which is used to obtain the perception information of the entire vehicle outside the vehicle, determine the decision relationship of the perception information of at least two different perception types using the basic parameters in the perception information, and generate a vehicle decision instruction corresponding to the perception information based on the decision relationship; a motion and control component, which is connected to the perception and computing component in communication, is used to receive the vehicle decision instruction issued by the perception and computing component, and generates the vehicle control instruction of the entire vehicle according to the fusion of the vehicle decision instruction and the local control information; a drive and execution component, which is connected to the motion and control component in communication, is used to receive the vehicle control instruction, and executes the vehicle control task locally based on the vehicle control instruction. Through the present invention, the waste of repeated use of processor resources of each controller is solved, and the drive and execution components are easy to form standardized products, reducing the cost of the entire vehicle.

Description

Vehicle, vehicle control method, storage medium, and electronic device

Technical Field

The present invention relates to the field of vehicle technology, and in particular to a vehicle, a vehicle control method, a storage medium, and an electronic device.

Background Art

Among the related technologies, automotive electronic and electrical architecture is the core of the development of automotive electrification and intelligence. It can run through the entire vehicle design and connect the entire vehicle architecture layout. The design level of the electronic and electrical architecture directly affects the vehicle's intelligence level, safety experience, and cost value.

The main body of transportation is gradually changing from "humans" to "smart machines", the living space inside the car is gradually evolving from "a single simple living space centered on the driver" to "multiple smart living spaces centered on families and social groups", and the in-car technical architecture is gradually evolving from the EE architecture oriented towards "humans" to the architecture oriented towards "machine intelligence".

In the related technologies, the automotive architecture schemes are very different, but the main forms are basically similar, mainly including the architecture forms of vehicle control, power & chassis, driving & cockpit, and regional control. In the related technologies, the automotive electronic and electrical architecture includes distributed electronic and electrical architecture and domain centralized electronic and electrical architecture. Distributed electronic and electrical architecture is an early form of vehicle electronic and electrical architecture. Under this architecture, the automotive electronic and electrical system consists of multiple independent control units, each of which controls a specific automotive function. For example, the engine electronic control unit (ECU) is responsible for controlling the working state of the engine, and the brake control unit is responsible for controlling the operation of the brake system. These control units exchange data through the control bus to achieve information sharing and collaborative work. Under this architecture, the coordination of vehicle resources is relatively low, and each ECU basically performs functional logic control in this field independently. However, with the continuous development of automotive electronic and electrical systems, higher-level architectures are gradually introduced to cope with increasingly complex automotive functions and performance requirements. The domain-centralized electrical and electronic architecture is a form of automotive electrical and electronic architecture. Compared with the distributed architecture, it is more centralized and efficient. Under this architecture, the automotive electrical and electronic system is divided into different functional domains, and each functional domain is uniformly controlled by a domain controller. This architecture combines modern automotive technologies such as intelligence, networking, and electrification to meet the requirements of intelligent connected vehicles for computing power, safety protection, etc. The typical division method of the domain-centralized electrical and electronic architecture is to divide the entire vehicle's electrical and electronic architecture into five domains: power domain, chassis domain, body domain, cockpit domain, and autonomous driving domain. Each domain is responsible for controlling a large functional module of the car and is uniformly controlled and managed through the domain controller. For example, the power domain controller mainly controls the vehicle's powertrain, optimizes power performance, and ensures power safety; the chassis domain controller is responsible for controlling the vehicle's driving behavior and driving posture, including functions such as brake system management and transmission system management. In the domain-centralized architecture, the domain controller is an independent controller for each domain, and a processor with strong core computing power is matched internally to meet the requirements of intelligent connected vehicles for computing power. In addition, the domain-centralized architecture also divides the vehicle into several independent functional modules according to the function and communication rate requirements, which improves the security and reliability of the system. If an attacker wants to attack the entire vehicle through a certain function, the domain controller where the function is located can monitor and eliminate hidden dangers in time without affecting other functional domains. The domain-centralized electronic and electrical architecture improves the integration, security and reliability of the automotive system by dividing different functional domains and using high-performance domain controllers.

The overall trend of the development of automotive electronic and electrical architecture is from distributed to domain control and then to domain centralization. The distributed electronic and electrical architecture has limited improvement on the intelligence level of the entire vehicle, and brings about an increase in wiring harness costs and vehicle weight. However, users' higher requirements for the intelligence level of the entire vehicle make the domain centralized electronic and electrical architecture slightly insufficient in future development.

With respect to the above-mentioned problems existing in the related technologies, no efficient and accurate solutions have been found yet.

Summary of the invention

The present invention provides a vehicle, a vehicle control method, a storage medium, and an electronic device to solve the technical problems in the related art.

According to one embodiment of the present invention, a vehicle is provided, comprising: a perception and computing component, for acquiring perception information of the entire vehicle outside the vehicle, the perception information comprising perception information of at least two different perception types, the perception information containing basic parameters, using the basic parameters in the perception information to determine a decision relationship between the at least two different perception types of perception information, and generating a vehicle decision instruction corresponding to the perception information based on the decision relationship; a motion and control component, communicatively connected to the perception and computing component, for receiving a vehicle decision instruction issued by the perception and computing component, and generating a vehicle control instruction for the entire vehicle based on the fusion of the vehicle decision instruction and local control information; a drive and execution component, communicatively connected to the motion and control component, for receiving the vehicle control instruction, and locally executing a vehicle control task based on the vehicle control instruction.

Optionally, the perception and calculation component includes:

An information acquisition unit, configured to acquire basic parameters and target parameters of the perception information corresponding to each perception type; the basic parameters include: one or more of a sensor identifier, a sensor position, a perception data value, a perception time, and a change trend of the perception data; the target parameters include: one or more of a vehicle controlled motion parameter associated with the perception information, an identifier of a controlled component of the vehicle, and a parameter of a controlled component of the vehicle;

A decision relationship determination unit is used to determine perception information of at least two different perception types with the same target parameter information as a perception information group, determine the association relationship between basic parameters of each perception information in the perception information group, the association relationship between the basic parameters includes: a constraint relationship, a priority relationship and a fusion relationship; determine the decision relationship of the perception information in the perception information group according to the association relationship between the basic parameters, and the decision relationship of the perception information in the perception information group varies according to the difference of the basic parameters.

Optionally, the perception and computing component includes: the perception and computing component includes: a first processing unit, used to obtain at least one perception information related to the target parameter and the cockpit, and generate a cockpit decision instruction based on the acquired perception information; a second processing unit, used to obtain at least one perception information related to the target parameter and the driving, and generate a self-driving decision instruction based on the acquired perception information; wherein the vehicle decision instruction includes the cockpit decision instruction and the self-driving decision instruction.

Optionally, the perception and computing component includes: a perception unit, used to identify the perception information of the entire vehicle outside the vehicle using a preset perception strategy to obtain a perception result; the information acquisition unit obtains the basic parameters and target parameters of the perception information corresponding to each perception type in the perception result, and the decision relationship determination unit determines the decision relationship of the perception information in the perception information group according to the correlation relationship between the basic parameters; a decision instruction generation unit, connected to the decision relationship determination unit, for calculating the vehicle decision instructions of the perception result according to the decision relationship and a preset computing strategy.

Optionally, the perception and computing component also includes: a vehicle control instruction analysis unit, which is used to receive the vehicle control instruction fed back by the motion and control component, and analyze the vehicle control instruction to obtain the corresponding target vehicle parameters; a strategy adjustment unit, which is used to add the perception information corresponding to the target vehicle parameters to the preset perception strategy when the perception information identified by the preset perception strategy does not contain the perception information corresponding to the target vehicle parameters, and, according to the target vehicle parameters; a feedback information acquisition unit, which is used to obtain the vehicle state parameters after executing the vehicle control task locally based on the vehicle control instruction, and the state parameters include: execution state parameters and vehicle state parameters, and adjust the preset calculation strategy according to the difference between the target vehicle parameters and the state parameters corresponding to the target vehicle parameters.

Optionally, the motion and control component includes: a motion domain controller, used to generate vehicle control instructions for the entire vehicle based on the vehicle decision instructions and local control information, wherein the vehicle control instructions are used to perform vehicle motion control and vehicle balance control on the vehicle.

Optionally, the motion domain controller includes: a main microprocessor MCU, used to search for local control information matching the vehicle decision instructions, update the local control information based on the vehicle decision instructions, and generate vehicle control instructions for the vehicle fusion system; a redundant MCU, used to perform safety verification on the vehicle control instructions and take over the main MCU when the main MCU fails; the vehicle fusion system, the vehicle fusion system includes: a vehicle control unit VCU, a vehicle thermal management unit TMS, a chassis steering unit, a braking unit, and a suspension unit; wherein the VCU is used to control the vehicle's power system and braking system, the TMS is used to control the vehicle's thermal management system, the chassis steering unit is used to control the vehicle's lateral motion, the braking unit is used to control the vehicle's longitudinal motion, and the suspension unit is used to control the vehicle's vertical motion.

Optionally, the drive and execution component includes: a plurality of regional controllers, which are used to receive vehicle control instructions sent by the motion and control component, generate drive commands according to the vehicle control instructions, and send the drive commands to corresponding target drivers; a plurality of drivers, each driver is connected to at least one of the regional controllers and execution components, and is used to receive the drive commands and drive the corresponding execution components based on the drive commands.

Optionally, the several regional controllers are also used to collect real physical signals of the execution components through sensors, convert the real physical signals into network communication data, and feed back the network communication data to the motion and control components.

Optionally, the actuator includes at least one of the following: a wire-controlled braking component, a wire-controlled steering component, a wire-controlled suspension component, and a power motor drive component.

Optionally, the sensor includes at least one of the following: a temperature sensor, a pressure sensor, a light sensor, and a rainfall sensor.

Optionally, the vehicle also includes: a power supply network for supplying power to the sensing and computing components, the motion and control components, and the driving and execution components.

Optionally, the vehicle further includes: an optical communication network, the communication network including a plurality of communication links, and the perception and computing component and the motion and control component are connected via the communication links.

Optionally, the vehicle further includes an external perception device, and the external perception device is connected to the perception and computing component via a communication link, and the communication link is used for transmitting external vehicle-wide perception information collected by the external perception device in real time.

Optionally, the external perception device includes at least one of the following: a camera, a radar.

According to another embodiment of the present invention, a vehicle control method is also provided, which is applied to the vehicle described in the above embodiment, including: receiving a vehicle control instruction sent by the motion and control component; generating a drive command according to the vehicle control instruction; searching for a target driver that matches the drive command; and sending the drive command to the target driver so that the target driver drives the corresponding execution component based on the drive command.

Optionally, after sending the drive command to the target drive, the method further includes: collecting real physical signals of the execution component through sensors; converting the real physical signals into network communication data; and feeding back the network communication data to the motion and control component.

According to another aspect of an embodiment of the present application, a storage medium is further provided, which includes a stored program, and the above steps are executed when the program is run.

According to another aspect of an embodiment of the present application, there is also provided an electronic device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; wherein: the memory is used to store computer programs; and the processor is used to execute the steps in the above method by running the program stored in the memory.

The embodiment of the present application also provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the steps in the above method.

Beneficial effects of the present invention:

A vehicle provided in an embodiment of the present application includes: a perception and computing component, used to obtain perception information of the entire vehicle outside the vehicle, the perception information includes perception information of at least two different perception types, the perception information contains basic parameters, the basic parameters in the perception information are used to determine the decision relationship between the at least two different perception types of perception information, and based on the decision relationship, a vehicle decision instruction corresponding to the perception information is generated; a motion and control component, which is communicatively connected to the perception and computing component, used to receive the vehicle decision instruction issued by the perception and computing component, and generate a vehicle control instruction for the entire vehicle according to the fusion of the vehicle decision instruction and local control information; a drive and execution component, which is communicatively connected to the motion and control component, used to receive the vehicle control instruction, and locally execute a vehicle control task based on the vehicle control instruction.

This solution no longer uses a distributed transmission solution, but integrates all sensor data into the perception and computing components, solving the technical problems of repeated use of processor resources of each controller, high transmission latency, and multiple and complex wiring harnesses. It effectively ensures vehicle safety, reduces computing power waste, integrates computing power at the controller level, reduces the number of controllers and chips, and thus reduces vehicle costs. In addition, before the motion and control components generate control instructions, this solution can determine the decision relationship, that is, the coordination relationship, between the perception information of different perception types in advance. Even if the perception information changes, the relationship between the perception information of different perception types can be clarified in advance, thereby quickly generating accurate vehicle decision instructions, which is convenient for the subsequent rapid generation of vehicle control instructions.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

FIG1 is a schematic structural diagram of a vehicle according to an embodiment of the present invention;

FIG2 is a schematic diagram of a scenario according to an embodiment of the present invention;

FIG3 is a schematic diagram of the sensing and computing components of the present invention;

FIG4 is a schematic diagram of a framework of a motion domain controller according to an embodiment of the present invention;

FIG5 is a schematic diagram of the implementation of the motion and control components in an embodiment of the present invention;

FIG6 is a power supply network diagram of a vehicle electrical and electronic architecture according to an embodiment of the present invention;

7 is a schematic diagram of a vehicle communication network in an embodiment of the present invention;

FIG8 is a control flow chart of a vehicle in an embodiment of the present invention;

FIG9 is a flow chart of a vehicle control method according to an embodiment of the present invention;

FIG10 is an overall structural diagram of a vehicle in an embodiment of the present invention;

11 is a structural block diagram of a vehicle control device according to an embodiment of the present invention;

FIG. 12 is a schematic diagram of another scenario according to an embodiment of the present invention.

DETAILED DESCRIPTION

In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only embodiments of a part of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

Example 1

In this embodiment, a vehicle is provided. FIG. 1 is a schematic structural diagram of a vehicle according to an embodiment of the present invention. As shown in FIG. 1 , the vehicle comprises:

A perception and calculation component 10 is used to obtain perception information of the entire vehicle outside the vehicle, determine a decision relationship between the perception information of the at least two different perception types using basic parameters in the perception information, and generate a vehicle decision instruction corresponding to the perception information based on the decision relationship;

The perception and computing components in the vehicle architecture of this embodiment are similar to the human brain and are responsible for overall calculation and decision-making, including receiving information, memorizing and learning, and issuing decision instructions after calculation.

Among them, the perception information includes perception information of at least two different perception types, where different perception types refer to different types of sensors, and different perception types refer to detection data of the vehicle obtained from different dimensions. The detection data of the vehicle obtained from different dimensions can be the same or different. For example: different sensors can obtain vehicle speed information. Specific examples: visual sensor data and radar sensor data belong to different perception types; radar speed sensors and wheel speed sensors belong to different perception types. The perception information contains basic parameters, and the acquisition parameters include: one or more of sensor identification, sensor position, perception data value, perception time and change trend of perception data. The correlation relationship in the basic parameters is used to determine the decision relationship between the perception information.

The motion and control component 12 is connected to the sensing and computing component for receiving the vehicle decision instruction issued by the sensing and computing component, and generating the vehicle control instruction of the whole vehicle according to the fusion of the vehicle decision instruction and the local control information;

The vehicle control instructions of the entire vehicle include all vehicle control instructions in the motion domain (the motion domain of this embodiment covers all control domains of the traditional automobile architecture, namely the power domain, chassis domain, body domain, cockpit domain and autonomous driving domain), which may be an instruction set including at least one vehicle control instruction.

The local control information of the motion and control component of this embodiment is pre-configured vehicle control information, such as input navigation routes, vehicle driving modes, user preferences, system preferences, etc.

The motion and control component is responsible for connecting the perception and computing components and the drive and execution components, processing information interaction. Compared with the traditional architecture, it connects the "horizontal, longitudinal and vertical" of the vehicle chassis and the vehicle power strategy fusion control, reduces transmission delay, realizes vehicle motion control and balance control, and improves user experience;

The driving and executing component 14 is connected to the motion and controlling component for receiving the vehicle control instruction and executing the vehicle control task locally based on the vehicle control instruction.

The drive and execution components are responsible for receiving control instructions. By moving the logical calculations of the traditional controller to the motion and control components, the local drive with strong implementation control is retained. Compared with the traditional architecture, standardized, serialized configurable components can be formed to solve cost issues.

Through the above scheme, a new vehicle electronic and electrical architecture is realized, including three components: a perception and computing component, which is used to obtain the perception information of the whole vehicle outside the vehicle, the perception information includes perception information of at least two different perception types, determine the decision relationship of the perception information of at least two different perception types, and generate a vehicle decision instruction corresponding to the perception information based on the decision relationship; a motion and control component, which is connected to the perception and computing component in communication, is used to receive the vehicle decision instruction issued by the perception and computing component, and generates the vehicle control instruction of the whole vehicle according to the fusion of the vehicle decision instruction and local control information; a driving and execution component, which is connected to the motion and control component in communication, is used to receive the vehicle control instruction, and executes the vehicle control task locally based on the vehicle control instruction. This scheme no longer adopts a distributed transmission scheme, and all sensor data are uniformly collected into the perception and computing component, which solves the technical problems of repeated use and waste of processor resources of each controller, high transmission delay, and more wiring harnesses and complex layout, effectively ensures vehicle safety, reduces computing power waste, integrates computing power at the controller level, reduces the number of controllers and chips, and thus reduces the cost of the whole vehicle. Moreover, before the motion and control components generate control instructions, the solution can determine in advance the decision-making relationship, that is, the coordination relationship, between the perception information of different perception types. Even if the perception information changes, the relationship between the perception information of different perception types can be clarified in advance, thereby quickly generating accurate vehicle decision instructions, which is convenient for the subsequent rapid generation of vehicle control instructions.

The solution provided in the embodiment of the present application has the following beneficial effects:

1. The relevant sensing end sensor information is unified to the sensing and computing components for processing, and decision output is made, while adapting to AI, big data self-learning and other capabilities; it solves the technical problems in the traditional vehicle architecture, such as poor real-time data transmission, the waste of processor resources due to the decentralized processing of each controller, high transmission delay, and many wiring harnesses with complex layout;

2. The motion and control component is responsible for the movement of the entire vehicle, integrates sensor data, integrates power, chassis functions, and movement, and uniformly controls steering, braking, and suspension adjustment, effectively ensuring vehicle safety, reducing computing power waste, and making it easier to achieve high-level vehicle functions, such as intelligent drifting, extreme cornering, and extreme climbing;

3. By designing a human-like architecture, the driving and execution components can be standardized, reducing repeated development and repeated investment;

4. A new vehicle architecture consisting of perception and computing components, motion and control components, and drive and execution components integrates computing power at the controller level, reducing the number of controllers, chips, and wiring harnesses, thereby reducing the cost of the entire vehicle.

This three-center vehicle architecture draws on the design of the human nervous system and embodies the end-to-end concept. The core of the architecture includes the first center (data processing and decision center), the second center (motion control center), and the third center (execution end). This architecture has significant advantages in the following aspects:

1. Data Processing and Decision-making Center (First Center): Similar to the human brain, this center is responsible for collecting and processing data from various vehicle sensors. Through advanced data big models and AI technology, this center can quickly and accurately analyze massive amounts of data and generate efficient and accurate decision-making instructions. This centralized data processing method ensures data consistency and efficient decision-making, avoiding delays and errors that may be caused by decentralized processing.

2. Motion Control Center (Second Center): Equivalent to the cerebellum of the human body, this center formulates the vehicle's motion control strategy based on the decision-making instructions of the First Center. The independence of the Motion Control Center enables it to focus on real-time vehicle dynamic control, ensuring rapid response and execution of instructions. This structure with clear division of labor makes the control process more professional and efficient.

3. Execution end (third center): As a specific execution agency, this center is responsible for implementing the motion control strategy of the second center. This includes controlling the engine, braking system, steering system, etc. The modular design of the execution end enables it to quickly respond to the instructions of the motion control center to ensure the smooth operation and safety of the vehicle.

Architecture Advantages

1. End-to-end integrated design: The entire system forms an end-to-end closed-loop process from data acquisition, decision generation, motion control to final execution. This design ensures that the process from data input to command output is highly coherent and consistent, avoiding information loss or processing delays in the intermediate links.

2. Efficient data processing and decision-making capabilities: The First Center integrates advanced data big models and AI technology to quickly process massive sensor data and generate accurate decision-making instructions. This centralized processing method greatly improves the decision-making efficiency and accuracy of the system.

3. Professional motion control: The second center focuses on the formulation of motion control strategies, making the vehicle more flexible and accurate in executing the instructions of the first center. The existence of the motion control center reduces the burden of the first center, allowing it to focus more on high-level decision-making analysis.

4. Rapid response and execution: The modular design and focused execution function of the third center ensure rapid response and accurate execution of instructions. The independence and efficiency of each execution unit improve the reliability and security of the entire system.

5. Scalability and adaptability: The modular design of this three-center architecture enables the system to have strong scalability and adaptability. New functions or new technologies can be flexibly integrated into the corresponding centers to enhance the overall performance and functionality of the system.

In short, this vehicle architecture significantly improves the system's processing capabilities, response speed, and overall efficiency through end-to-end design concepts and a highly integrated three-center structure, ensuring the safety, stability, and intelligence of vehicle operation.

In the prior art, all perception information is input into the processor for centralized processing. This method results in that different perception information can only be processed according to predetermined strategies. Here, the predetermined strategy is usually fixed for the perception information itself. However, in specific scenarios, the perception information is composed of multiple information. The correlation relationship between different information is different and changes dynamically. Therefore, processing the perception information as a whole is likely to ignore the changes brought about by multiple information within the perception information.

In the present application, the decision relationship between perception information of different perception types is determined. The decision relationship, that is, the coordination relationship, between the perception information of different perception types can be determined in advance. Even if the perception information changes, the relationship between the perception information of different perception types can be clarified in advance, which is convenient for the subsequent rapid generation of accurate vehicle decision instructions.

In addition, in the prior art, when the same perception information is compared with other perception information, the perception information is usually judged as a whole. For example, when the perception information has different priorities, the different priorities are usually considered to be pre-set. But in fact, the perception information contains a variety of information, and the relationship between different information is not just priority, but there are other relationships. Figure 2 is a scene diagram according to an embodiment of the present invention. In Figure 2, perception information 1 and perception information 2, according to the prior art, can only distinguish priority 1 and priority 2 using the arrow on the left, but the perception information can also include: basic parameters and target parameters, and the relationship between each basic parameter is also different.

To this end, in one implementation of this embodiment, the perception and calculation component includes: an information acquisition unit and a decision relationship determination unit, wherein:

The information acquisition unit is used to acquire basic parameters and target parameters of the perception information corresponding to each perception type.

In the embodiment of the present application, each perception information has a basic parameter indicating its source and a target parameter indicating its purpose, wherein the source refers to the information related to the collection of the perception information, for example: the basic parameters include: one or more of the sensor identification, sensor position, perception data value, perception time and the change trend of the perception data. The target parameters include: one or more of the motion parameters of the vehicle being controlled, the identification of the controlled components of the vehicle and the parameters of the controlled components of the vehicle associated with the perception information. The target parameter refers to the parameters of the vehicle that may be related to the perception information, for example: the wheel speed sensor, in the sensing information output by it, the corresponding basic parameters are: the ID, position, speed value, speed collection time, and speed change trend of the wheel speed sensor; the corresponding target parameter is: vehicle speed. Another example: the vehicle angle sensor, in the sensing information output by it, the corresponding basic parameters are: the ID, position, angle value, angle collection time, and angle change trend of the angle sensor; the corresponding target parameter is: vehicle tilt angle.

A decision relationship determination unit is used to determine the perception information of at least two different perception types with the same target parameter information as a perception information group, and determine the association relationship between the basic parameters of each perception information in the perception information group; determine the decision relationship of the perception information in the perception information group according to the association relationship between the basic parameters, and the decision relationship of the perception information in the perception information group varies according to the difference of the basic parameters.

In one embodiment of the present application, the perception information at least includes: first radar perception information, second radar perception information, image perception information, and gear perception information; wherein the basic information of the first radar perception information includes: first radar identification, vehicle front position, vehicle front obstacle distance, first detection time, and vehicle front obstacle distance change rate; the basic information of the second radar perception information includes: second radar identification, vehicle rear position, vehicle rear obstacle distance, second detection time, and vehicle rear obstacle distance change rate; the basic information of the image perception information includes: third identification, vehicle rear position, vehicle rear image, and vehicle rear image change; the basic data of the gear perception information includes: gear sensor identification, current gear, and gear change; the target parameters of the first radar perception information, the second radar perception information, the image perception information, and the gear perception information all include: vehicle speed;

The decision relationship unit is used to define the first radar perception information, the second radar perception information, the image perception information and the gear perception information as a perception information group; when the gear is in reverse gear or switched to reverse gear, the decision relationship unit determines that the priority of the obstacle distance behind the vehicle in the basic parameters of the first radar perception information and the second radar perception information is higher than the priority of the obstacle distance in front of the vehicle, that is, the association relationship is a priority relationship, and determines that the decision relationship of the perception information in the perception information is a fusion relationship; and, when the rate of change of the rear obstacle distance is that the obstacle distance behind the vehicle gradually decreases to a preset distance threshold, the decision relationship unit determines that the priority of the obstacle distance in front of the vehicle in the basic parameters of the first radar perception information and the second radar perception information is higher than the priority of the obstacle distance behind the vehicle, that is, a priority relationship, and determines that the decision relationship of the perception information in the perception information group is a priority relationship.

The aforementioned perception information includes at least: first radar perception information, second radar perception information, image perception information, and gear position perception information. In actual application scenarios, it can be a reversing scenario, wherein the first radar perception information, the second radar perception information, the image perception information, and the gear position perception information are collected when reversing. The target parameters of these perception information are all related to the vehicle speed.

When the vehicle shifts to reverse gear, at this time, the correlation relationship between the distance of the obstacle behind the vehicle and the distance of the obstacle in front of the vehicle in the basic parameters of the first radar perception information and the second radar perception information is a priority relationship, and the priority of the distance of the obstacle behind the vehicle is higher than the priority of the distance of the obstacle in front of the vehicle. At this time, the decision relationship in the perception information group is also a priority relationship. According to this decision relationship, the subsequent vehicle decision instruction can be a distance priority instruction for the obstacle behind the vehicle. At this time, when the distance of the obstacle behind the vehicle reaches the preset threshold, the vehicle can directly brake the vehicle according to the distance of the obstacle behind the vehicle, that is, stop the vehicle, and avoid collision between the vehicle and the vehicle behind.

When the rate of change of the distance to the rear obstacle is that the distance to the rear obstacle gradually decreases to the preset distance threshold, it means that even if braking measures are taken to stop the vehicle, the rear vehicle may still move forward, so that the distance between the two is further reduced. At this time, the braking measures are obviously unable to meet the safety control of the vehicle. Then, at this time, it is determined that the priority of the distance to the obstacle in front of the vehicle in the basic parameters of the first radar perception information and the second radar perception information is greater than the priority of the distance to the obstacle behind the vehicle, that is, the priority relationship, and the decision relationship of the perception information in the perception information group is determined to be a priority relationship. The perception information here is the priority relationship as above, but the specific information of the priority is different. When it is determined that the priority of the distance to the obstacle in front of the vehicle is greater than the priority of the distance to the obstacle behind the vehicle, the decision instruction of the vehicle is: the distance priority instruction of the obstacle in front of the vehicle, and when the distance to the obstacle in front of the vehicle is greater than the preset safety value (safety distance, for example: 2 meters), the vehicle is subsequently controlled to move forward to increase the distance to the obstacle behind the vehicle to avoid collision with the rear vehicle.

Through the above embodiments, it can be seen that in this application, by utilizing the basic information in the perception information, the decision relationship between the perception information can be dynamically adjusted, avoiding the disadvantages of the fixed relationship between the perception information, and being more adaptable to the scene.

When determining the relationship between the sensing information, if two sensing information of different sensing types, one is about the vehicle speed and the other is about the vehicle tilt angle, then there will be no conflict between the two sensing information. Therefore, in the embodiment of the present application, the main consideration is the relationship between the sensing information of different sensing types for the same target parameter information. For example: the target parameter information is the vehicle speed, but one sensing type is the wheel speed sensor and the other sensing type is the radar speed sensor, then further judgment is needed between the two sensing information obtained. For example: when the vehicle is sliding on the ice, the speed of the wheel speed sensor is 40km/h, and the speed calculated by the radar speed sensor is 60km/h, so it is necessary to distinguish the relationship between the sensing data of the same target parameter information.

In the embodiment of the present application, the association relationship between the basic parameters includes: a constraint relationship, a priority relationship and a fusion relationship, wherein the constraint relationship is non-A and B, that is, only the A sensing information is used to control the vehicle; the priority relationship means that there is a priority relationship between the two sensing information, and when controlling the vehicle, the sensing information with a higher priority is selected; the fusion relationship means that after the two sensing information are fused, for example: deviation compensation; the vehicle is controlled together; in addition, the association relationship may also include: a selection relationship, for example: when one condition is met, one sensing information is selected, and when another condition is met, another sensing information is selected. In this regard, in the embodiment of the present application, there is no limitation.

In specific applications, weights of basic parameters can be assigned. When the decision relationship is determined by using the association relationship of basic parameters, it can be calculated based on the weight assignment. For example, if the weight of the sensor position is 90%, and the association relationship of the sensor position is a fusion relationship, then the decision relationship of the overall perception parameter is a fusion relationship.

In the embodiment of the present application, the perception and computing component obtains the perception information of various different perception types such as the vehicle's camera, radar, ultrasonic wave, inertial measurement unit, GNSS (global positioning system), infrared sensor, speed sensor information and execution unit feedback information and status information (including suspension height, power torque, engine speed, etc.), and based on the perception information, combined with AI large model data, implements vehicle decision instructions in several specific ways. In one scenario, the perception information obtained by the corresponding sensors has large differences or environmental constraints in the basic information of the perception information of sensors of different categories. By judging the priority of the basic information, relying on key information (such as vehicle status, vehicle speed, lighting information, etc.), the basic information that needs to be restricted is eliminated, the priority is dynamically adjusted, and the decision instructions are generated in combination with the local large model learning results.

In another scenario, in a certain state, such as driving state, by integrating the perception information of different types of sensors and the basic information of the perception information, it is determined to perform fusion processing on the perception information of the sensors to implement decision instructions (for example, sensitive road information, engine torque, vehicle speed and steering state, and comprehensive acquisition of camera data, lidar data, GNSS data, etc., to issue decision instructions).

In other scenarios, we can also add confidence factors to the basic information of the sensor's perception information, combine the daily large model learning results (such as daily road sections, driving habits, etc.), integrate the information of the current sensor input, and improve and enhance the adaptive algorithm based on the confidence factors contained in the basic information in the perception information, and then generate decision instructions in specific scenarios. Alternatively, we can also use deep learning models to process and fuse multiple sensor data in a multi-mode mixed situation. For example, convolutional neural networks can be used for image processing collected by sensors, and recursive neural networks can be used for time series data analysis, so as to achieve better decision instructions.

In one implementation of the present embodiment, the perception and computing component includes: a first processing unit, used to obtain at least one perception information related to the target parameter and the cockpit, and generate a cockpit decision instruction based on the acquired perception information; a second processing unit, used to obtain at least one perception information related to the target parameter and the driving, and generate a self-driving decision instruction based on the acquired perception information; wherein the vehicle decision instruction includes the cockpit decision instruction and the self-driving decision instruction.

The core of the perception and computing components is the cockpit and autonomous driving functions, which corresponds to the supercomputing properties of the implementation computer. The vehicle architecture of this embodiment will have different physical forms for autonomous driving requirements of different capability levels. The core methods are one-chip (single SOC realizes cockpit and intelligent driving control sharing a chip example) and multi-chip (cockpit and intelligent driving are implemented on different chips, but still on the same controller).

In one example, the perception and computing component includes: a perception unit and a decision instruction generation unit, wherein the perception unit is used to identify the external full-vehicle perception information using a preset perception strategy to obtain a perception result; in addition, the information acquisition unit obtains the basic parameters and target parameters of the perception information corresponding to each perception type in the perception result, and the decision relationship determination unit determines the decision relationship of the perception information in the perception information group according to the correlation relationship between the basic parameters; the decision instruction generation unit is connected to the decision relationship determination unit, and is used to use a vehicle decision instruction to calculate the perception result based on the decision relationship and the preset strategy.

In this example, the perception unit obtains real-time environmental information around the vehicle through on-board sensors (such as radar, camera, lidar, etc.), including road conditions, other vehicle locations, pedestrians, traffic signals, etc. The computing unit processes and analyzes the perceived information, and uses various algorithms (such as image recognition, target tracking, obstacle detection, etc.) to identify and understand the environment, including strategies for advanced functions such as big data, cockpit functions, driving, and parking.

For example, the perception module: uses high-definition cameras, millimeter-wave radars, lidar and other sensors to perceive the environment, receive perception information, and perform information processing.

Computing module: Use deep learning, computer vision and other algorithms to process and analyze environmental information, identify targets such as roads, vehicles, pedestrians, and predict their movement trajectories.

Optionally, in the embodiment of the present application, the perception and calculation component further includes: a vehicle control instruction analysis unit, a strategy adjustment unit, and a feedback information acquisition unit, wherein:

The vehicle control instruction analysis unit is used to receive the vehicle control instruction fed back by the motion and control component, and analyze the vehicle control instruction to obtain corresponding target vehicle parameters.

In the embodiment of the present application, the target vehicle control parameter may refer to a specific control parameter of the vehicle, for example, the engine speed, throttle amount, light brightness, gear position, etc.;

A strategy adjustment unit is used to add the perception information corresponding to the target vehicle parameters to the preset perception strategy when the perception information identified by the preset perception strategy does not include the perception information corresponding to the target vehicle parameters, and according to the target vehicle parameters.

In the embodiment of the present application, when the preset perception strategy does not have the perception information of the light, and it is judged based on the perception information that the vehicle is about to enter the tunnel, and the vehicle queue instruction includes turning on the light, then at this time, the sensing information of the light corresponding to the instruction to turn on the light of the vehicle needs to be added to the preset perception strategy, and then the perception of the light is added later, so as to control the high beam, low beam, etc. later. In addition, if the current perception information contains perception information that is not related to the target vehicle parameters, it can also be eliminated later, or gradually eliminated to avoid redundancy of perception information.

The feedback information acquisition unit is used to acquire the state parameters of the vehicle after the vehicle control task is locally executed based on the vehicle control instruction, and the state parameters include: execution state parameters and whole vehicle state parameters.

In the present application, execution status refers to the difference between the parameters after the execution of the vehicle control instruction and the target vehicle parameters. For example, the target vehicle parameter corresponding to the vehicle control instruction is a speed as high as 2500 rpm; but the actual engine speed is adjusted to 2300 rpm, then 2300 rpm is the execution status parameter; the overall status parameter refers to the vehicle tilt angle, vehicle speed or other parameters after the vehicle control instruction is executed.

The strategy adjustment unit is further used to adjust the preset calculation strategy according to the difference between the target vehicle parameters and the state parameters corresponding to the target vehicle parameters.

In the embodiment of the present application, the difference between the target vehicle parameters and the state parameters corresponding to the target vehicle parameters reflects whether the adjustment is in place or exceeds the set value. For example, the vehicle control instruction is to increase the vehicle speed by 100km/h, which is specifically increased by controlling the engine to 2500 revolutions, but in fact the vehicle speed is only increased to 90km/h. In order to reach 120km/h in the subsequent vehicle control instructions, the engine speed needs to be increased to 2600 revolutions to finally reach a vehicle speed of 100km/h. Through feedback regulation, the strategy adjustment unit can adjust the preset perception strategy and the preset calculation strategy in a timely manner.

It is also used to receive vehicle control instructions fed back by the motion and control component, and use the vehicle control instructions to dynamically adjust the perception strategy and the calculation strategy.

After processing the environmental information, the perception and computation component passes the specific strategy (algorithm calculation results) to the motion and control component to provide it with a basis for decision-making; at the same time, it receives feedback from the motion and control component and adjusts the perception and computation strategies based on the feedback to improve the robustness and accuracy of the system.

Figure 3 is a schematic diagram of the perception and computing components of the present invention. In addition to the functions included in the traditional cockpit and autonomous driving controller, the perception and computing components integrate AI and big data learning, can process and fuse information according to the algorithm, and transmit it to the motion domain controller for execution after unified calculation. The traditional architecture completely separates the intelligent driving and cockpit functions, and there will be repeated acquisition of information. Under the traditional architecture, the implementation of autonomous driving is that the autonomous driving controller collects sensor data in the autonomous driving mode, and after limited processing, it controls multiple controllers such as steering, braking, and power respectively, which will have great limitations in improving the level of self-driving. Under the new architecture, by improving the SOC example level, integrating information processing, and unified transmission, it is easy to achieve a higher level of autonomous driving.

Gradually realize the continuous experience of all scenarios to achieve the purpose of liberating people. The algorithm evolves to "end-to-end integration of perception and regulation and control", enhances the perception ability, and strengthens the use of raw data from sensors. Gradually realize the smart living space of "integration of objects and numbers, integration of virtual and real" to achieve the purpose of pleasing people.

In one implementation of this embodiment, the motion and control component includes: a motion domain controller, which is used to generate vehicle control instructions for the entire vehicle based on the vehicle decision instructions and local control information, wherein the vehicle control instructions are used to perform vehicle motion control and vehicle balance control on the vehicle.

The task of the motion and control component is to realize vehicle motion and balance control, integrating the whole vehicle VCU (vehicle control), TMS (vehicle thermal management) and the chassis lateral (steering), longitudinal (braking) and vertical (suspension) control strategies. It can adopt heterogeneous SOC+MCU, or MCU+MCU, or an internally isolated single SOC to form a motion domain controller.

In one example, the motion domain controller includes: a main microprocessor MCU, used to find local control information matching the vehicle decision instruction, update the local control information based on the vehicle decision instruction, and generate a vehicle control instruction for the vehicle fusion system; a redundant MCU, used to perform safety verification on the vehicle control instruction and take over the main MCU when the main MCU fails; the vehicle fusion system, the vehicle fusion system includes: a vehicle control unit VCU, a vehicle thermal management unit TMS, a chassis steering unit, a braking unit, and a suspension unit; wherein the VCU is used to control the vehicle's power system and braking system, the TMS is used to control the vehicle's thermal management system, the chassis steering unit is used to control the vehicle's lateral motion, the braking unit is used to control the vehicle's longitudinal motion, and the suspension unit is used to control the vehicle's vertical motion.

In this example, the motion domain controller makes an execution decision based on the environmental information and calculation results (vehicle decision instructions) provided by the perception and computing components, determines the vehicle's driving trajectory, speed, and optimal route, and clarifies the control instructions based on the decision results to ensure that the vehicle drives according to the preset trajectory and speed. Receive environmental information from the perception and computing components, make decisions based on this information, send the decision results to the drive and execution components, and receive feedback from the drive and execution components to ensure the accurate execution of control instructions. For example, use planning algorithms for path planning to ensure that the vehicle finds the optimal driving path in a complex environment, and use preset algorithms to accurately control the vehicle to ensure that the vehicle drives according to the preset trajectory and speed.

4 is a schematic diagram of the framework of the motion domain controller in an embodiment of the present invention, including a main module (main MCU) and a redundant module (redundant MCU), a power module (TMS, VCU module, BMS main, EMS), a caliper module (right caliper drive, left caliper drive), a suspension module (ECDC drive, ECAS drive), a steering SBW/EPS, a braking EMB/IBCU, etc.

In one implementation of this embodiment, the drive and execution component includes: a plurality of regional controllers, which are used to receive vehicle control instructions sent by the motion and control component, generate drive commands according to the vehicle control instructions, and send the drive commands to the corresponding target drivers; a plurality of drivers (drive controllers), each driver is connected to at least one of the regional controllers and execution components (actuators), and is used to receive the drive commands and drive the corresponding execution components based on the drive commands.

In this implementation, the core of the drive and execution components is the regional controller (such as VIU1, VIU2) and the drivers of each execution component. The regional controller converts the real physical signal of the sensor (mainly analog quantity) into network communication data and transmits it to the motion domain control center. The motion and control components send drive commands to VIU and EMB (wire control brake), SBW (wire control steering), AHC (wire control suspension), power motor drive, etc., thereby realizing vehicle control. The drive and execution components drive the vehicle to perform corresponding actions, such as acceleration, deceleration, steering, etc. according to the control instructions of the motion and control components.

The perception and computing component feeds back the actual state of the vehicle (such as position, speed, acceleration, etc.) to the motion and control component for real-time adjustment. The drive and execution component receives the control instructions from the motion and control component, drives the vehicle to perform corresponding actions, and feeds back the actual state of the vehicle to the motion and control component for real-time decision-making and adjustment.

FIG5 is a schematic diagram of the implementation of the motion and control components in an embodiment of the present invention. The power supply system of the motion domain controller has redundant inputs, the external sensor signal has redundant inputs, and the main system receives the data of the perception and calculation components. After comprehensive decision-making, the motion domain controller implements TMS, power control, and horizontal, vertical and vertical three-way fusion control. The redundant system performs data and safety verification, and takes over the control when the main system fails unexpectedly to ensure driving safety. The motion domain controller hardware integrates the drive of the suspension, thermal management and VCU parts, and adopts the main system and redundant system. The functions related to vehicle safety mainly involve driving, braking and other aspects. Therefore, when designing, it is considered that the main and redundant systems realize information interaction and control with the chassis part (steering, braking) and the power part through two CANFD buses. The main and auxiliary systems of the motion domain controller are connected to two power supply systems and two sensor systems respectively, and perform data processing respectively, and synchronously receive the execution instructions issued by the perception and calculation components. The CANFD bus is used between the main system and the redundant system for data interaction and verification. The main system adopts powerful chips (such as heterogeneous SOC, MCU, MCU+MCU, SOC+MCU or two redundant chips integrated in one SOC chip architecture) with strong computing power foundation and multi-core architecture, which can easily realize integrated vehicle thermal management, XYZ three-way motion independent control and fusion control. The redundant system uses MCU with functional safety level ASIL D to realize redundant backup control and safety verification of driving motion.

The drive and execution components include hardware such as motors, steering mechanisms, and brake systems, which are used to drive the vehicle to perform actions such as acceleration, deceleration, and steering. At the same time, various sensors are used to obtain information such as the actual position, speed, and acceleration of the vehicle, and these information are sent to the motion and control components through the information transmission module.

Under the new vehicle architecture of this embodiment, the core of the existence of the drive and execution components is to control the standardization of the actuators, thereby reducing costs and avoiding repeated development caused by the huge differences between the current vehicle states.

Optionally, the execution component may be, but is not limited to: a wire-controlled braking component, a wire-controlled steering component, a wire-controlled suspension component, and a power motor drive component.

In one example, the several regional controllers are also used to collect real physical signals of the execution components through sensors, convert the real physical signals into network communication data, and feed back the network communication data to the motion and control components.

Optionally, the sensor may be, but is not limited to: a temperature sensor, a pressure sensor, a light sensor, and a rain sensor.

In this embodiment, the vehicle further includes: a power supply network for supplying power to the sensing and computing component, the motion and control component, and the driving and execution component.

Optionally, the power supply network may be a 48V power supply network. Of course, it may also be adapted to power supply networks of other voltages, such as 12V.

The vehicle of this embodiment also provides a 48V power supply network system. The power supply of the sensing and computing components, the motion and control components, the driving and execution components, and the drive controller and actuator are powered by 48V, which can effectively reduce the load current, wire diameter, wire loss, etc. At the same time, the motion and control components adopt dual redundant power supply lines to improve system safety.

The energy system of the power supply network is similar to the human heart and blood vessels, which transmit energy and execute actions. Compared with the traditional vehicle architecture, the energy layer introduces a 48V power supply system to solve the problems of increased vehicle wiring harness weight and cost due to the increased degree of intelligence, as well as the problem of insufficient 12V power supply.

FIG6 is a power supply network diagram of the vehicle electronic and electrical architecture of an embodiment of the present invention, which is powered by a high-voltage battery through a DCDC control, a voltage distribution controller, and then to each center of the architecture, using a 48V voltage. The motion and control components are related to driving safety because they integrate vehicle management and motion domain control, so they have a separate independent redundant power supply line to ensure driving safety in the event of abnormalities in the voltage distribution controller and wiring harness. The power supply of the perception and computing components, motion and control components, and drive and execution components are all powered by 48V, and high-power loads (such as wire-controlled steering drive, wire-controlled brake drive, and power motor drive) are powered by a 48V distribution network, and traditional low-power sensors and loads are still distributed by 5V/12V.

In this power distribution network, the power distribution controller is mainly used for 48V power distribution and protection. It has a built-in electronic fuse. When a short circuit occurs at any node in the electrical circuit, it can be quickly shut down and feedback the fault. The energy system is similar to the human heart and blood vessels, which transmit energy and execute actions. Compared with the traditional vehicle architecture, the energy layer introduces a 48V power supply system to solve the problems of increased weight and cost of the vehicle wiring harness due to the increased degree of intelligence, as well as the problem of insufficient 12V power supply.

In this embodiment, the vehicle further includes: an optical communication network, the communication network includes a plurality of communication links, and the perception and computing component and the motion and control component are connected via the communication links.

The communication network of this embodiment can be an optical communication network, or other forms of communication networks, or a hybrid communication network formed by a combination of multiple communication networks, such as passive optical communication + CAN/CANFD. The communication link can be a wired communication link (such as an optical communication line) or a wireless communication link.

In one example, the vehicle further includes an external perception device, and the external perception device is connected to the perception and computing component via a communication link, and the communication link is used to transmit external vehicle-wide perception information collected by the external perception device in real time.

Optionally, the external sensing device may be, but is not limited to, a camera or a radar.

The communication system of this embodiment is similar to the human nervous system, connecting various small systems for communication and command transmission. Corresponding to the whole vehicle system, passive optical communication + CAN/CANFD is adopted. Compared with the traditional architecture, the newly added optical communication method solves the problems of insufficient bandwidth, poor anti-interference and insufficient real-time performance of the traditional architecture. At the current stage, most parking lots use Ethernet technology based on twisted pair cables, which has the problems of high cost and limited transmission bandwidth, and is limited to the future development needs of intelligence.

The communication technology capabilities related to intelligent driving need to evolve towards "redundant transmission of execution and sensing data, backbone line bandwidth of 10 Gigabits, sensor line bandwidth of over 10 Gigabits, and deterministic transmission at the physical layer" to provide a reliable neural network for liberated people; the communication technology capabilities related to smart life need to evolve towards "multi-center data interaction, seamless dynamic networking" to provide a flexible neural network for happy people.

FIG7 is a schematic diagram of a vehicle communication network in an embodiment of the present invention, which proposes a hybrid communication network architecture based on optical communication + CAN/CANFD. On this architecture, the communication medium and communication method of communication transmission are reduced to achieve more stable, safe and reliable information transmission, and can effectively reduce wiring harnesses and reduce costs. It is particularly noted that in this architecture, optical communication technology only represents one way to implement this architecture, and represents the implementation method of the future vehicle architecture communication network. At this stage, the application of this architecture can still use the traditional Ethernet communication method.

Optical communication can be used between the perception and computing components and the motion and control components due to the high requirements for data transmission bandwidth (≥1000M) and reliability. The amount of data between the camera, radar and the perception and computing components is large, so optical communication is used. The motion and control components and the drive and execution components mainly realize data acquisition transmission and drive control. The data transmission bandwidth requirements are relatively low, and the reliability and real-time requirements are relatively high. Considering the comprehensive cost, CANFD can effectively realize low-cost and high-reliability communication.

In an implementation scenario of the optical communication network, the perception and computing component serves as the communication link terminal OLT, with four optical paths in front, back, left and right to achieve connection with the motion domain controller and the vehicle camera and radar, ensuring stable and reliable data transmission. The communication technology capabilities related to intelligent driving need to evolve to "redundant transmission of both execution and sensing data, 10 Gigabit bandwidth of the backbone line, more than 10 Gigabit bandwidth of the sensor line, and deterministic transmission at the physical layer" to provide a reliable neural network for liberated people; the communication technology capabilities related to smart life need to evolve to "multi-center data interaction, non-sensing dynamic networking" to provide a flexible neural network for happy people. This optical path arrangement method under the communication network architecture reduces the number of OLTs, increases the multiplexing rate, and solves the current problems of large number of CAN lines, complex layout, high cost and high bandwidth requirements.

FIG8 is a control flow chart of a vehicle in an embodiment of the present invention, including:

S80, the perception and computing component, receives external sensor information (radar, camera, display, etc.), performs calculations and decisions, completes in-cabin interactions, and issues driving decision instructions.

S81, the motion and control component receives instructions from the perception and computing component and integrates them with the control information it receives, realizes the motion control and drive of the whole vehicle and issues execution instructions to the components of the drive and execution components.

S82, the driving and executing component, after receiving the control instruction from the motion and control component, implements the corresponding action and completes the execution of the local task.

Take the application scenario of obstacle avoidance operation as an example. In this application scenario, the perception and computing component senses obstacles that suddenly appear in front of it, such as pedestrians, animals or other vehicles, through sensors. The computing center quickly analyzes the location, speed and other information of the obstacle and assesses the collision risk. The obstacle information and its collision risk assessment results are sent to the motion and control component.

The motion and control component immediately initiates the obstacle avoidance strategy upon receiving obstacle information. It replans the driving path based on the position and speed of the obstacle, and calculates the acceleration, deceleration and steering instructions required for obstacle avoidance through the control algorithm. It sends the obstacle avoidance instructions to the drive and execution component, and continuously receives feedback from the center to ensure that the vehicle can safely avoid obstacles.

The drive and execution components quickly adjust the vehicle's driving state according to the obstacle avoidance instructions, such as emergency braking, quick steering, etc., to avoid obstacles. They receive and execute obstacle avoidance instructions, and at the same time, they feed back the actual state of the vehicle to the motion and control components so that the control center can monitor and adjust the obstacle avoidance process in real time.

In this embodiment, a vehicle control method is provided. FIG. 9 is a flow chart of a vehicle control method according to an embodiment of the present invention, which is applied to the motion and control component of the vehicle in the above embodiment. As shown in FIG. 9 , the process includes the following steps:

Step S902, receiving a vehicle control instruction sent by the motion and control component;

Step S904, generating a driving command according to the vehicle control instruction;

Step S906, searching for a target driver matching the drive command;

Step S908: sending the driving command to the target driver, so that the target driver drives the corresponding execution component based on the driving command.

Optionally, after the drive command is sent to the target driver, the method further includes: collecting a real physical signal of the execution component through a sensor; converting the real physical signal into network communication data; and feeding back the network communication data to the motion and control component.

Figure 10 is an overall architecture diagram of a vehicle in an embodiment of the present invention. The development of the whole vehicle is like the development of a humanoid robot, providing an architecture similar to the human body structure. Through human-like comparison, each component has independent responsibilities, which runs through the whole vehicle action, is more centralized, rather than the traditional functional domain division, and establishes a deep connection between the control units of the whole vehicle. It includes perception and computing components, motion and control components, drive and execution components, sensor units, control units, execution units, and the communication network and power supply network necessary for the implementation of the architecture.

The perception and computing components are similar to the human brain and are responsible for overall computing and decision-making: receiving information, memorizing and learning, and issuing decision instructions after calculation; the motion and control components are responsible for information interaction, body motion control, and balance control; the drive and execution components are responsible for executing and processing some local tasks such as unconditional reflex control of the limbs and body; the communication system is similar to the human nervous system, connecting various places to transmit commands; it develops from CANFD→Ether→Optical Ethernet→Passive Optical Ethernet; the energy system is similar to the human heart and blood vessels, and is responsible for energy transmission.

Actuators (hands and feet) mainly refer to electrical components such as headlights, compressors, blowers, wiper motors, etc. in the vehicle electrical system. Sensors (skin) mainly refer to sensor components such as temperature, pressure, light, rainfall, etc. in the vehicle sensors.

In Figure 10, RCM is the rear headlight drive module, HCM is the front headlight drive module, AHC is the active suspension controller, EMB is the wire-controlled brake system, SBW is the wire-controlled steering system, PEBB is the high-voltage system integrated drive controller, VIU1 and VIU2 are regional controllers, BMS is the battery power management system, and TBOX is the intelligent remote network communication system; it should be particularly noted that the AHC, EMB, SBW, and PEBB mentioned in the present invention are phased products of the development of controllers, and the present invention particularly emphasizes controllers with corresponding functions, such as chassis vertical controllers, chassis longitudinal controllers, chassis lateral controllers, and power drive controllers.

In the vehicle based on the humanoid architecture of this embodiment, the sensor information of the relevant perception end is unified to the perception and computing components for processing, and the decision output is made, while adapting to AI, big data self-learning and other capabilities; in the traditional architecture, sensor data is transmitted to multiple controllers, and the multiple controllers process the information in a decentralized manner and then transmit it to the autonomous driving domain controller (or other functional domain controllers, each functional domain controller is implemented in the same way). In this mode, the real-time data transmission is not good, the decentralized processing of each controller results in the reuse and waste of processor resources, the transmission delay is high, and the wiring harnesses are numerous and complex in layout. This embodiment can solve the above problems.

The motion and control components realize the fusion of motion energy. In the traditional architecture, steering, braking, suspension, power and driving are completely separated, and information is transmitted through the bus, which can often only realize basic steering, braking, energy recovery and other functions; under the humanoid architecture, the motion and control components are responsible for the movement of the whole vehicle. Under the traditional form, if it is at high speed or on a slippery road, it is difficult to keep the vehicle stable by steering. There is always a delay in the actions of various components, which is prone to accidents such as rollover. Under the vehicle architecture of this embodiment, the sensor data is integrated, and the steering, braking and suspension adjustment are uniformly controlled to effectively ensure the safety of the vehicle.

The solution of this embodiment can also reduce the waste of computing power. By integrating power and chassis functions and unified motion control, it is easier to perform high-level driving functions, such as intelligent drifting, extreme cornering, extreme climbing, etc. The three components integrate computing power at the controller level, reducing the number of controllers and chips, and thus reducing the cost of the entire vehicle.

Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

Example 2

In this embodiment, a vehicle control is also provided, and the device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and/or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

FIG. 11 is a structural block diagram of a vehicle control device according to an embodiment of the present invention. As shown in FIG. 10 , the device includes:

A receiving module 110, for receiving a vehicle control instruction sent by the motion and control component;

A generating module 112, configured to generate a driving command according to the vehicle control command;

A search module 114, configured to search for a target drive that matches the drive command;

The sending module 116 is used to send the driving command to the target driver, so that the target driver drives the corresponding execution component based on the driving command.

Optionally, the device also includes: an acquisition module, which is used to collect the real physical signal of the execution component through a sensor after the sending module sends the drive command to the target driver; a conversion module, which is used to convert the real physical signal into network communication data; and a feedback module, which is used to feed back the network communication data to the motion and control component.

It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

Example 3

An embodiment of the present invention further provides a storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

S1, receiving a vehicle control instruction sent by the motion and control component;

S2, generating a driving command according to the vehicle control command;

S3, searching for a target drive that matches the drive command;

S4, sending the driving command to the target driver, so that the target driver drives the corresponding execution component based on the driving command.

Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

Optionally, the electronic device may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.

Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

S1, receiving a vehicle control instruction sent by the motion and control component;

S2, generating a driving command according to the vehicle control command;

S3, searching for a target drive that matches the drive command;

S4, sending the driving command to the target driver, so that the target driver drives the corresponding execution component based on the driving command.

The solutions of this patent are described in detail below in conjunction with specific scenarios.

1. Control logic of the brain

For vehicles, through the above settings, the inside of the vehicle (similar to the brain) consists of two centers, the perception center and the computing center. The perception center is specifically used for the perception and processing of external environmental information (including direct information acquisition such as pedestrians, vehicles, locations, distances, road conditions, traffic lights, and indirect information obtained from the cloud). The computing center is based on AI and large models similar to the inherent modules of the large brain, and realizes data fusion processing. Based on the data content, data reconstruction is performed, and decision instructions are generated through data decision-making, competition, judgment, fusion and other strategies based on the category, execution degree, priority, etc. of the data; inside the vehicle (similar to the cerebellum), the functional logic control of the whole vehicle motion control execution is integrated in the motion and control components, including the basic functional control of the chassis, power, and energy management, as well as the motion energy fusion control, and the corresponding actuators are driven to perform specific actions, liberating the demand for computing power for the execution components under the traditional architecture. The specific detailed structure is shown in Figure 12, which is another scenario schematic diagram according to an embodiment of the present invention.

2. In the perception and computing center, the perception unit and the computing decision unit can greatly improve the hardware performance and functional scalability through fine linkage and collaboration. The detailed solution analysis is as follows:

1) Perception unit module:

Environmental perception module: Use sensors (cameras, radars, lidars, etc.) to perceive the surrounding environment and obtain environmental information, torque, angle, speed, external status and other information.

On-board data acquisition module: obtains real-time status information inside the vehicle, such as wheel torque, steering angle, vehicle speed, body status, etc.

Cloud data interface module: Receives information from other vehicles and infrastructure transmitted from the cloud to realize vehicle-to-everything (V2X) functions.

2) Calculation and decision-making unit module:

Data fusion processing module: Fusion processing of multi-source data obtained by the perception unit, using filtering, competition, combination, judgment, supplementation and other methods to improve the accuracy and consistency of the data.

AI module and neural network management module: Analyze and make decisions in complex environments based on deep learning, large models and AI algorithms. Manage and optimize various neural network models, and support online learning and model updates.

Logical management and decision-making module: Combines data fusion results to perform high-level logical management and decision generation.

Target recognition module: Identify and understand targets and dynamic behaviors in the environment.

Safety redundancy module: Design system redundancy and fault management to ensure the safety and reliability of the system.

Model training and update module: Model training and update based on real-time data and cloud resources to improve the adaptability and intelligence of the system.

Specific implementation method:

1) Data fusion and environmental perception linkage:

The multi-source data collected by the environment perception module, the vehicle data acquisition module and the cloud data interface module are fused through the data fusion processing module. The filtering method is used to remove noise and interference, the competition and joint methods are used to integrate data from different sources, and the judgment and supplementation methods are used to make reasonable supplements and adjustments when the data is incomplete or conflicting.

This multi-level data fusion processing improves the accuracy and robustness of environmental perception, enabling the computing decision-making unit to perform analysis and make decisions based on more reliable data.

2) The big model is linked with the AI module and the neural network management module:

The large models and AI modules provide powerful computing power and intelligent decision-making in complex scenarios, while the neural network management module is responsible for managing and optimizing these models.

The neural network management module can perform online learning and model optimization based on real-time data and feedback to ensure the real-time and accuracy of AI decisions. The model training and update module continuously improves the performance and adaptability of the model through cloud resources.

3. Logic management and decision-making module and target recognition module linkage:

The logic management and decision-making module relies on the environment and target information provided by the target cognition module to perform high-level logical reasoning and decision-making.

The target recognition module analyzes the perception data in real time, identifies and understands key targets in the environment (such as pedestrians, vehicles, traffic signs, etc.), and passes this information to the logic management and decision-making module to ensure the accuracy and rationality of the decision.

4. The safety redundancy module is linked with each module:

The safety redundancy module is designed with redundant backup and fault management of the system to ensure that the system can continue to operate safely when any single point failure occurs.

Each functional module has redundant design for key tasks. The data fusion processing module can call redundant data sources when processing data. The logic management and decision-making module can make adjustments based on the feedback from the security redundant module when making decisions to ensure safety.

5. Model training and updating module is linked with other modules:

The model training and update module uses cloud resources to continuously train and update various neural networks and AI models, ensuring that the models have good adaptability when facing new environments and new scenarios.

The data and feedback information generated by each functional module during operation are passed to the model training and update module for offline analysis and model optimization. The optimized model is updated via OTA (Over The Air) and deployed to the computing decision unit to improve the overall performance of the system.

Achievable results:

①. Improve hardware performance:

The high efficiency and accuracy of data fusion processing reduce repeated calculations and error processing in the calculation and decision-making units, and improve the overall calculation efficiency.

The optimization and continuous learning of the neural network management module improve the operating efficiency of the AI model and reduce the consumption of hardware resources.

The design of the safety redundant module ensures that the system can still operate efficiently under various fault conditions, reducing the performance degradation caused by faults.

②. Function expansion:

The modular and linkage design allows the system to flexibly expand new functional modules. For example, the function of the perception unit can be expanded by adding new sensors and data interfaces, and the function of the calculation and decision-making unit can be expanded by introducing new AI algorithms and models.

The linkage between the data fusion processing module and the model training and updating module can support new data sources and new model types, enabling the system to adapt to more complex and changing environments.

③. Improved intelligence and adaptability:

The collaborative work of the big model with the AI module and the neural network management module enables the system to be highly intelligent and capable of autonomous learning, enabling it to make optimal decisions in different driving scenarios.

Real-time online learning and model update mechanisms enable the system to continuously adapt and optimize, and to flexibly adjust in the face of changing environments and demands.

3. Control logic of the cerebellum:

To realize the cerebellum's motor fusion control function and ensure efficient linkage between its modules, the specific linkage examples are as follows:

1) Linkage between horizontal and vertical control

Steering and speed coordination: When turning, the lateral control module adjusts the steering angle according to the brain's instructions, and the longitudinal control module synchronously adjusts the speed to ensure that the vehicle turns smoothly.

Lane Keeping and Adaptive Cruise Control: Lane Keeping Assist (LKA) and Adaptive Cruise Control (ACC) work together to ensure that the vehicle maintains a safe distance from the vehicle ahead while maintaining its lane.

2) Linkage between longitudinal control and vertical control

Speed and suspension coordination: At different speeds, the longitudinal control module adjusts the vehicle speed, and the vertical control module adjusts the suspension state according to road conditions and speed to improve driving comfort and stability.

Active suspension and brake control: During emergency braking, the vertical control module adjusts the suspension system to reduce vehicle pitch and improve braking stability.

3) Linkage between power drive control and energy management

Power distribution and energy recovery: The power drive control module optimizes power output according to driving conditions, and the energy management module recovers energy during deceleration and braking to improve energy utilization efficiency.

Four-wheel drive control and battery management: In four-wheel drive mode, the power drive control module optimizes the torque distribution between the front and rear axles, and the energy management module monitors the battery status to ensure efficient use of energy and safety.

4) Global linkage of status monitoring and feedback

Real-time status monitoring and adjustment: The status monitoring and feedback module monitors the status of each vehicle component in real time, and transmits feedback data to each control module to ensure the correct execution of each instruction.

Fault detection and safety redundancy: When the status monitoring and feedback module detects an abnormality or fault, it promptly notifies the relevant modules to make adjustments or switch to the redundant system to ensure driving safety.

Achievable beneficial effects:

1) Improve driving safety

Coordinated control: The coordinated work of the lateral, longitudinal and vertical control modules ensures the stability and safety of the vehicle in various driving situations (such as sharp turns, acceleration, braking).

Real-time feedback adjustment: The real-time monitoring and feedback mechanism of the status monitoring and feedback module ensures the accurate execution of various instructions and reduces potential risks.

2) Improve driving comfort

Suspension adjustment: The vertical control module optimizes the vehicle's suspension status in real time through suspension adjustment and active suspension system, reducing road bumps and improving ride comfort.

Smooth acceleration and braking: The adaptive cruise and regenerative braking functions of the longitudinal control module enable smooth acceleration and braking to enhance the riding experience.

3) Optimize energy utilization

Energy recovery: The energy management module improves energy utilization efficiency, extends driving range and reduces operating costs by optimizing the battery charging and discharging process and regenerative braking.

Power optimization: The power drive control module optimizes power distribution, improves fuel or electricity efficiency, and improves the economy and environmental protection of the vehicle.

4) Enhance system reliability

Redundant design: The power drive control module improves the efficiency of fuel or electricity use by optimizing power distribution, thereby improving the economy and environmental protection of the vehicle.

Fault detection and processing: Real-time fault detection and processing mechanism reduces the impact of system faults on vehicle operation and improves system reliability and stability.

Through the above detailed module design and linkage mechanism, the cerebellum can efficiently formulate and execute the vehicle's motion control strategy based on the brain's decision instructions, ensuring driving safety, comfort and energy efficiency. This modular and linkage design also provides a good foundation for future function expansion and performance improvement.

Specific improvements:

1) Comprehensive control and optimization:

Traditional architecture: Traditional architecture usually has each module working independently and lacks comprehensive coordination, which may lead to slow control response and low efficiency.

Motion and Control Center: Through the comprehensive optimization of lateral, longitudinal, vertical and power control, the coordinated work of various control modules is achieved to improve the overall performance of the vehicle.

2) Energy management and efficiency improvement:

Traditional architecture: Energy management is decentralized, and energy recovery and utilization efficiency is low.

Motion and Control Center: A unified energy management module optimizes energy utilization and recovery, improving energy efficiency and driving range.

3) Intelligence and adaptability:

Traditional architecture: low intelligence, lack of adaptive control and real-time adjustment capabilities.

Motion and control center: Through the high-level decision-making of the brain and the real-time control of the cerebellum, intelligent motion control and environmental adaptation are realized, improving the intelligence level and adaptability of the vehicle.

4) Safety and reliability:

Traditional architecture: high risk of single point failure and lack of effective redundancy design.

Motion and control center: The status monitoring and feedback module provides real-time monitoring and redundant design to ensure the reliability and safety of the system in various situations.

3. Architecture

In the humanoid architecture, the signal rules will change compared to the traditional automotive architecture. This change is mainly reflected in the signal type, data format, communication protocol and control strategy. The following is the specific scheme of signal transmission between the brain, cerebellum and execution end and the possible changes in signal rules.

The signal output from the cerebrum to the cerebellum. The cerebrum is mainly responsible for high-level decision-making and complex calculations. Its signal output to the cerebellum requires clear instructions and rich environmental information. The main signals include:

1) High-level decision-making instructions:

Path planning instructions: including target location, path point sequence, driving route, etc.

Behavioral decision instructions: instructions for operations such as changing lanes, overtaking, and parking.

2) Environmental perception information:

Target object information: location, speed, trajectory, etc. of detected vehicles, pedestrians, and obstacles.

Road condition information: road type, road surface conditions, traffic signs, etc.

3) Status information:

The overall status of the vehicle: such as current speed, acceleration, position, direction, etc.

4) Safety redundancy information:

Fault detection results: including fault information of any perception and computing modules.

Signal arrangement and processing of the cerebellum: After receiving the signal from the brain, the cerebellum needs to perform detailed processing and arrange control strategies to generate specific execution instructions. The main processing steps and signal arrangement include:

1) Signal analysis and classification:

Parse high-level decision instructions and convert them into specific motion control parameters (such as steering angle, acceleration, deceleration, etc.).

Environmental perception information processing: Combined with sensor data, further identify and track the target object and update the real-time environmental model.

2) Motion control strategy arrangement:

Lateral control strategy: Calculates steering angle and steering torque based on steering instructions and lane keeping requirements.

Longitudinal control strategy: Calculate acceleration, deceleration and target speed based on acceleration command and adaptive cruise control requirements.

Vertical control strategy: adjust suspension system parameters according to road condition information and suspension adjustment requirements.

3) Comprehensive control strategy generation:

Multi-control strategy integration: Integrate lateral, longitudinal and vertical control strategies to generate coordinated control instructions to ensure smooth operation and response speed of the vehicle.

4) Real-time status monitoring and feedback adjustment:

Monitor feedback from the actuator: monitor the status of each actuator (such as steering system, power system, suspension system, etc.) in real time and make feedback adjustments.

Fault handling and safety redundancy: When a fault is detected, switch to the safety redundancy strategy to ensure stable operation of the system.

The cerebellum outputs signals to the drive and execution components. The cerebellum generates specific execution instructions based on the comprehensive control strategy and outputs them to each execution end. The main signals include:

1) Steering system instructions:

Steering angle command: specific steering angle and steering torque.

Lane Keeping Assist Command: Automatically adjusts steering to maintain lane.

2) Power system instructions:

Acceleration/deceleration instructions: target acceleration, deceleration and speed values.

Power distribution command: torque distribution between the front and rear axles, output power of the engine or electric motor.

3) Suspension system instructions:

Suspension adjustment instructions: adjust the stiffness and damping of the suspension system to adapt to different road conditions.

Active suspension control instructions: adjust suspension status in real time to improve vehicle stability and comfort.

4) Energy management instructions:

Energy management and recovery instructions: Control the battery charging and discharging process to optimize energy utilization.

Regenerative Braking Command: Recovers energy during deceleration and braking.

Specific changes to signal rules

1) Signal type and data format:

Standardized interface: Use unified data formats and communication protocols (such as CAN, Ethernet, etc.) to ensure efficient communication between the brain, cerebellum and execution end.

High-bandwidth communication: supports high-bandwidth data transmission to meet the needs of real-time environmental perception and high-frequency control instructions.

2) Real-time and priority management:

Real-time guarantee: Ensure that high-priority control instructions and emergency signals can be transmitted and processed in a timely manner, reducing communication delays.

Priority management: Set the priority of different types of signals to ensure the transmission reliability of key control instructions.

3) Redundancy and fault detection mechanism:

Bidirectional feedback: Bidirectional feedback mechanism between the cerebrum and cerebellum, and between the cerebellum and the executive end, ensures the reliability and accuracy of signal transmission.

Redundant design: Redundant design is used for key signals and control instructions to ensure normal operation in case of signal loss or failure.

4)Flexibility and scalability:

Modular design: The signal rules and communication protocol design are flexible and extensible, supporting the integration of new functional modules and sensors.

Dynamic adjustment: Dynamically adjust control strategies and signal transmission methods according to real-time needs to improve the adaptability and intelligence level of the system.

Possible results

By changing and designing the above signal rules, the following beneficial effects can be achieved:

1) Improve vehicle control accuracy: Efficient communication and signal transmission between the cerebrum and cerebellum ensure that high-level decision-making instructions can be accurately converted into specific motion control instructions, improving the accuracy and response speed of vehicle control.

2) Enhance system reliability: Bidirectional feedback mechanism and redundant design ensure the reliability of signal transmission and system stability, and reduce the risks caused by signal loss or failure.

3) Optimize energy utilization efficiency: The energy management module works in coordination with the power control system to ensure efficient use of energy and improve the vehicle's endurance and economy.

4) Improve driving comfort: Dynamic adjustment of the suspension system and comprehensive optimization of motion control strategies ensure smooth operation of the vehicle under different road conditions and improve driving and riding comfort.

5) Realize intelligent expansion: Modular design and flexible signal rules support the expansion and upgrade of system functions, improve the intelligence level and adaptability of the system, and meet the development needs of future autonomous driving and intelligent transportation.

The new brain, cerebellum, and execution-end architecture have made significant improvements in signal rules, ensuring that the system can operate efficiently, accurately, and reliably, and has good scalability and intelligence capabilities.

Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM/RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

It should be understood that the terms used in the text are only for the purpose of describing specific example embodiments, and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used in the text may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and/or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and/or combinations thereof. The method steps, processes, and operations described in the text are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims (19)

1. A vehicle, characterized by comprising:

The sensing and calculating component is used for acquiring sensing information of the whole vehicle outside the vehicle, the sensing information comprises at least two sensing information types, the sensing information comprises basic parameters, the basic parameters in the sensing information are utilized to determine decision relations of the sensing information of the at least two sensing information types, and vehicle decision instructions corresponding to the sensing information are generated based on the decision relations;

The motion and control assembly is in communication connection with the perception and calculation assembly and is used for receiving a vehicle decision instruction sent by the perception and calculation assembly and generating a vehicle control instruction of the whole vehicle according to the vehicle decision instruction and the local control information;

And the driving and executing assembly is in communication connection with the motion and control assembly and is used for receiving the vehicle control command and locally executing the vehicle control task based on the vehicle control command.

2. The vehicle of claim 1, wherein the sensing and computing component comprises:

the information acquisition unit is used for acquiring basic parameters and target parameters of the perception information corresponding to each perception type; the base parameters include: one or more of sensor identification, sensor location, perceived data value, perceived time, and perceived data trend; the target parameters include: one or more of a motion parameter of the vehicle being controlled, an identification of a controlled component of the vehicle, and a parameter of the controlled component of the vehicle associated with the perception information;

The decision relation determining unit is configured to determine, as a sensing information group, sensing information of at least two different sensing types, which are the same as the target parameter information, and determine an association relation between basic parameters of each sensing information in the sensing information group, where the association relation between the basic parameters includes: a constraint relationship, a priority relationship and a fusion relationship; and determining the decision relation of the perception information in the perception information group according to the association relation between the basic parameters, wherein the decision relation of the perception information in the perception information group is changed according to the different basic parameters.

3. The vehicle of claim 2, wherein the sensing and computing component comprises:

the first processing unit is used for acquiring at least one piece of perception information of the target parameter related to the cabin and generating a cabin decision instruction according to the acquired perception information;

The second processing unit is used for acquiring at least one piece of perception information related to driving of the target parameter and generating a self-driving decision instruction according to the acquired perception information;

the vehicle decision instruction comprises the cabin decision instruction and the self-driving decision instruction.

4. The vehicle of claim 2, wherein the sensing and computing component comprises:

the sensing unit is used for recognizing sensing information of the whole vehicle outside the vehicle by adopting a preset sensing strategy to obtain a sensing result;

The information acquisition unit acquires basic parameters and target parameters of the perception information corresponding to each perception type in the perception result, and the decision relation determination unit determines the decision relation of the perception information in the perception information group according to the association relation between the basic parameters;

And the decision instruction generation unit is connected with the decision relation determination unit and is used for calculating the vehicle decision instruction of the perception result according to the decision relation and a preset calculation strategy.

5. The vehicle of claim 4, wherein the sensing and computing component further comprises:

the vehicle control instruction analysis unit is used for receiving the vehicle control instruction fed back by the motion and control assembly and analyzing the vehicle control instruction to obtain corresponding target vehicle parameters;

The strategy adjustment unit is used for adding the perception information corresponding to the target vehicle parameter into the preset perception strategy when the perception information identified by the preset perception strategy does not contain the perception information corresponding to the target vehicle parameter, and according to the target vehicle parameter;

the feedback information acquisition unit is used for acquiring state parameters of the vehicle after the vehicle control task is locally executed based on the vehicle control instruction, and the state parameters comprise: executing the state parameters and the whole vehicle state parameters, and adjusting the preset calculation strategy according to the target vehicle parameters and the difference of the state parameters corresponding to the target vehicle parameters.

6. The vehicle of claim 1, wherein the motion and control assembly comprises:

And the motion domain controller is used for generating a vehicle control instruction of the whole vehicle according to the vehicle decision instruction and the local control information, wherein the vehicle control instruction is used for controlling the motion of the whole vehicle and controlling the balance of the whole vehicle.

7. The vehicle of claim 6, wherein the motion domain controller comprises:

The main microprocessor MCU is used for searching local control information matched with the vehicle decision instruction, updating the local control information based on the vehicle decision instruction and generating a vehicle control instruction for a vehicle fusion system;

The redundant MCU is used for carrying out safety check on the vehicle control instruction and taking over the main MCU when the main MCU fails;

The vehicle fusion system is described in the context of a vehicle, the vehicle fusion system includes: the system comprises a whole vehicle control unit VCU, a whole vehicle thermal management unit TMS, a chassis steering unit, a braking unit and a suspension unit; the VCU is used for controlling a power system and a braking system of a vehicle, the TMS is used for controlling a thermal management system of the vehicle, the chassis steering unit is used for controlling transverse motion of the vehicle, the braking unit is used for controlling longitudinal motion of the vehicle, and the suspension unit is used for controlling vertical motion of the vehicle.

8. The vehicle of claim 1, wherein the drive and implement assembly comprises:

The plurality of regional controllers are used for receiving the vehicle control command sent by the motion and control assembly, generating a driving command according to the vehicle control command and sending the driving command to a corresponding target driver;

And each driver is connected with at least one regional controller and the execution component, and is used for receiving the driving command and driving the corresponding execution component based on the driving command.

9. The vehicle of claim 8, wherein the plurality of zone controllers are further configured to collect real physical signals of the implement by sensors, convert the real physical signals into network communication data, and feed the network communication data back to the motion and control assembly.

10. The vehicle of claim 8, characterized in that the execution means comprise at least one of: a drive-by-wire component, a drive-by-wire steering component, a drive-by-wire suspension component, and a power motor driving component.

11. The vehicle of claim 9, wherein the sensor comprises at least one of: temperature sensor, pressure sensor, light sensor, rainfall sensor.

12. The vehicle of claim 1, characterized in that the vehicle further comprises: a power supply network for powering the sensing and computing components, the motion and control components, and the drive and execution components.

13. The vehicle of claim 1, characterized in that the vehicle further comprises: a communication network comprising a plurality of communication links, the sensing and computing components and the motion and control components being connected by communication links.

14. The vehicle of claim 13, further comprising an external awareness device coupled to the awareness and computing assembly via a communication link for transmitting in real-time external all-vehicle awareness information collected by the external awareness device.

15. The vehicle of claim 14, characterized in that the external awareness device comprises at least one of: camera, radar.

16. A vehicle control method, characterized by being applied to the vehicle according to any one of claims 1 to 15, comprising:

Receiving a vehicle control instruction sent by the motion and control assembly;

Generating a driving command according to the vehicle control command;

searching a target driver matched with the driving command;

And issuing the driving command to the target driver so that the target driver drives the corresponding execution component based on the driving command.

17. The method of claim 16, wherein after issuing the drive command to the target drive, the method further comprises:

acquiring a real physical signal of the execution part through a sensor;

converting the actual physical signal into network communication data;

The network communication data is fed back to the motion and control component.

18. A storage medium having a computer program stored therein, wherein the computer program is arranged to execute the vehicle control method of any one of claims 16 to 17 when run.

19. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program for the vehicle control method as claimed in any of the claims 16 to 17.