CN115520118B - Advanced driving assistance system based on CAN-TSN gateway - Google Patents
Advanced driving assistance system based on CAN-TSN gateway Download PDFInfo
- Publication number
- CN115520118B CN115520118B CN202211000155.6A CN202211000155A CN115520118B CN 115520118 B CN115520118 B CN 115520118B CN 202211000155 A CN202211000155 A CN 202211000155A CN 115520118 B CN115520118 B CN 115520118B
- Authority
- CN
- China
- Prior art keywords
- tsn
- automobile
- data
- gateway
- subsystem
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/023—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for transmission of signals between vehicle parts or subsystems
- B60R16/0231—Circuits relating to the driving or the functioning of the vehicle
- B60R16/0232—Circuits relating to the driving or the functioning of the vehicle for measuring vehicle parameters and indicating critical, abnormal or dangerous conditions
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Mechanical Engineering (AREA)
- Small-Scale Networks (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of automatic driving, and discloses an advanced auxiliary driving system based on a CAN-TSN gateway, which comprises an auxiliary driving subsystem, an automobile control subsystem and a central gateway; the auxiliary driving subsystem detects road condition information and driver behaviors; the automobile control subsystem detects the automobile state and detects obstacles; the central gateway is connected with the auxiliary driving subsystem and the automobile control subsystem, adopts a CAN-TSN protocol conversion algorithm to carry out rate matching, comprises a CAN-TSN conversion strategy and a TSN-CAN conversion strategy, wherein the CAN-TSN conversion strategy packs according to the message period and the load, and the TSN-CAN conversion strategy divides the domain to which the TSN frame is sent according to the requirement. According to the system provided by the invention, the vehicle state data acquired by the OBD is transmitted to the auxiliary driving subsystem through the CAN-TSN gateway, more data support is provided for the decision of the auxiliary driving subsystem, multi-perception fusion is realized, the decision is fed back to the automobile control subsystem, and the safety and reliability of the automatic driving automobile are improved.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to an advanced auxiliary driving system based on a CAN-TSN gateway.
Background
With the development of intelligent networking of automobiles, an automobile electronic and electric architecture is turned into a domain centralized electronic and electric architecture by a distributed automobile electronic and electric architecture based on an ECU. Such as BOSCH corporation, divides automobiles into power domain, body domain, chassis domain, cabin domain, and advanced auxiliary driving domain. The existing advanced auxiliary driving system is mainly used for collecting traffic environment data through external sensors such as millimeter wave radar, laser radar, vehicle-mounted cameras, inertial navigation and the like, accurately identifying various traffic elements and providing support for an automatic driving automobile decision system. The existing advanced auxiliary driving system mainly takes the road condition environment outside the vehicle as the main part, has less perception on the information such as the state of the driver in the automobile, the running state of the automobile and the like, and lacks perception on the information such as the speed of the automobile, the rotating speed of the tire, the temperature of the engine, whether the driver is tired to drive, whether bad driving behaviors exist or not.
Problems with existing advanced driver assistance systems include: 1. the automatic driving decision is mainly carried out by collecting road condition information, the data is single, the automobile state detection and the driver information collection and decision are lacked, and the data cannot be transmitted to an automobile control system. 2. The advanced auxiliary driving system and the automobile control system have the problems of incompatible communication protocols, incompatible rates and the like, and interconnection and intercommunication are needed to be realized through a gateway. 3. The existing CAN-TSN gateway mainly focuses on theory, and a used verification platform lacks vehicle-standard authentication, does not have practical landing feasibility and lacks a practical scheme.
Disclosure of Invention
The invention provides an advanced auxiliary driving system based on a CAN-TSN gateway, which transmits vehicle state data acquired by OBD to an auxiliary driving subsystem through the CAN-TSN gateway, provides more data support for decision making of the auxiliary driving subsystem, realizes multi-perception fusion, feeds back decision making to an automobile control subsystem, and improves the safety and reliability of an automatic driving automobile.
The invention provides an advanced auxiliary driving system based on a CAN-TSN gateway, which comprises an auxiliary driving subsystem, an automobile control subsystem and a central gateway;
The auxiliary driving subsystem is used for detecting road condition information and driver behaviors; the automobile control subsystem is used for detecting the state of an automobile and detecting obstacles by adopting a millimeter wave radar; the central gateway is used for connecting a first domain controller of the auxiliary driving subsystem and a second domain controller of the automobile control subsystem, is a CAN-TSN gateway, adopts a CAN-TSN protocol conversion algorithm to perform rate matching, distributes IP addresses for the first domain controller and the second domain controller, adopts Socket transmission of a TCP/IP network layer and Ethernet frame/TSN frame transmission of a data link layer;
The CAN-TSN protocol conversion algorithm comprises a CAN-TSN conversion strategy and a TSN-CAN conversion strategy, the CAN-TSN conversion strategy is packaged according to the message period and the load, and the TSN-CAN conversion strategy divides the TSN frame into domains which are required to be sent.
Further, the CAN-TSN conversion strategy converts the CAN message of the automobile control subsystem into a TSN frame and forwards the TSN frame to the auxiliary driving subsystem, and the steps include:
S11, loading the CAN message into a TSN frame;
s12, judging whether the TSN frame is overloaded or not;
s13, when the TSN frame is overloaded, creating the TSN frame, and returning to the step S11;
s14, judging whether real-time scheduling is met or not when the TSN frame is not overloaded;
s15, when real-time scheduling is not met, repackaging the message, and returning to the step S11;
S16, forwarding is carried out when the real-time scheduling is met.
Further, the TSN-CAN conversion strategy sends the decision of the auxiliary driving subsystem to the automobile control subsystem, which comprises the steps of:
s21, splitting a TSN frame;
s22, filling the data into CAN information;
S23, judging whether filling is possible;
s24, when the filling is impossible, a new CAN message is created, and the step S22 is returned;
s25, judging whether real-time scheduling is met or not when filling is possible;
s26, returning to the step S21 when real-time scheduling is not met;
and S27, forwarding when the real-time scheduling is met.
Further, the auxiliary driving subsystem comprises a first domain controller, an on-vehicle external camera, an on-vehicle internal camera and an on-vehicle embedded GPU, and is communicated through a TSN network;
the vehicle-mounted external camera collects road condition information outwards of the vehicle and carries out real-time target detection through a first setting algorithm; the detection content comprises a trolley, a truck, a bus, pedestrians, bicycles and traffic signs;
The vehicle-mounted internal camera faces the vehicle to collect driver behavior information, and fatigue driving detection and bad behavior detection are carried out through a second setting algorithm;
The vehicle-mounted embedded GPU is used for receiving the road condition information and the driver behavior information, performing data processing, and transmitting a processing result to the first domain controller;
And the first domain controller is used for performing perception fusion on the data transmitted by the vehicle-mounted embedded GPU by adopting a multi-perception fusion algorithm to form the decision, and sending the decision to the central gateway.
Further, the step of the multi-perception fusion algorithm includes:
s31, acquiring data of an observation target by a plurality of sensors of different types;
S32, carrying out feature extraction and transformation on output data of a plurality of sensors of different types, and carrying out pattern recognition processing on a feature vector Yi to obtain description data of each sensor about a target;
S33, sensor synchronization including time synchronization and space synchronization is performed;
S34, grouping the description data of the sensors about the target according to the same target, and synthesizing the sensor data of the target by utilizing a fusion algorithm to obtain consistency interpretation and description of the observed target.
Further, the time synchronization and the space synchronization include IEEE 802.1AS protocol and sensor calibration.
Further, the first setting algorithm is YOLOv algorithm, and the second setting algorithm is lightweight MobileNet.
Further, the automobile control subsystem comprises a second domain controller, an automobile power simulator, a millimeter wave radar and an automobile OBD instrument panel, and is communicated through an on-board CAN bus;
The second domain controller is used for receiving and analyzing CAN bus data, packaging the data into an Ethernet/TSN frame format and transmitting the Ethernet/TSN frame format to the central gateway through a Socket;
the automobile power simulator sends an automobile state signal in a first set period; the automobile state signals comprise automobile speed, tire rotating speed, cooling liquid temperature, air inlet manifold pressure, throttle position, ignition advance angle, engine load, residual oil quantity, air-fuel ratio and turbine pressure;
the millimeter wave radar collects barrier information in a second set period and calculates the angle, distance and speed of the barrier; wherein the obstacle information comprises a transverse distance, a longitudinal distance, a transverse speed and a longitudinal speed;
and the automobile OBD instrument panel reads the data of the automobile power simulator through an OBD interface, and displays the data in real time and gives an abnormal alarm.
The beneficial effects of the invention are as follows:
1. The invention can expand the functions of the existing advanced auxiliary driving system, provide more information for the decision of the automatic driving algorithm, such as the running condition of the automobile, the information of the automobile body and the like, and perform multi-perception fusion, thereby comprehensively improving the safety and the accuracy of the automatic driving algorithm in various scenes.
2. According to the invention, a part of low-bandwidth sensors (such as millimeter wave radar) are placed in a vehicle body control domain, so that the load pressure of the existing automatic driving domain is reduced, and the real-time performance of the low-bandwidth sensors is improved.
3. The invention is compatible with various OBD interfaces and millimeter wave radars, CAN dynamically adjust a real-time scheduling algorithm of CAN-TSN protocol conversion according to data characteristics, and provides a gateway design scheme capable of falling to the ground based on a vehicle-specification-level development kit.
Drawings
Fig. 1 is a functional schematic diagram of an advanced driving support system based on a CAN-TSN gateway according to the present invention.
Fig. 2 is a schematic diagram of an advanced driving support system architecture based on a CAN-TSN gateway according to the present invention.
Fig. 3 is a schematic diagram of an internal architecture of the advanced driving support system based on the CAN-TSN gateway of the present invention.
Fig. 4 is a schematic diagram of a multi-perception fusion algorithm according to the present invention.
Fig. 5 is a flow chart of a CAN-TSN forwarding strategy in the CAN-TSN protocol conversion algorithm of the present invention.
Fig. 6 is a flowchart of a TSN-CAN forwarding strategy in the CAN-TSN protocol conversion algorithm of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
There is a need in advanced driving assistance systems to use high-speed on-board ethernet technology such as Time Sensitive Networks (TSNs) to meet the requirements of the autopilot domain for high bandwidth and low latency. While conventional automotive control systems, such as power domain, base domain, and body domain systems, still use Controller Area Network (CAN) as a backbone communication network. Therefore, in order to realize interconnection of all domains in the domain centralized electronic and electric architecture, the invention provides the CAN-TSN gateway to send the vehicle state data in the automobile control domain to the auxiliary driving subsystem, solves the problems of rate matching and protocol conversion among different networks, and achieves the auxiliary decision-making function of the auxiliary driving subsystem.
As shown in fig. 1-3, the present invention provides an advanced driving assistance system based on a CAN-TSN gateway, which includes an driving assistance subsystem, an automobile control subsystem and a central gateway, wherein the driving assistance subsystem (advanced driving assistance domain) and the automobile control subsystem (automobile power control domain) establish a connection through the vehicle gateway. The auxiliary driving subsystem is used for realizing road condition information detection and driver behavior detection, the automobile control subsystem is used for realizing automobile state detection and millimeter wave radar obstacle detection, and the central gateway is used for connecting the domain controllers of the two domains to realize direct protocol conversion and rate matching of different domains.
The auxiliary driving subsystem (advanced auxiliary driving domain) is mainly composed of Ethernet/TSN, the behavior of a driver is detected in real time through an in-vehicle camera facing the vehicle, road condition information is detected in real time through an in-vehicle camera facing the outside of the vehicle, and data is sent to the exchanger through the domain controller. The automobile control subsystem (automobile power control domain) is mainly composed of a CAN bus, automobile state display is achieved through an automobile OBD instrument panel, power simulation is achieved through an automobile ECU simulator, and obstacle distance and speed detection is achieved through a millimeter wave radar.
Auxiliary driving subsystem: the method is used for detecting road condition information and driver behaviors.
The auxiliary driving subsystem comprises a first domain controller, a vehicle-mounted external camera, a vehicle-mounted internal camera and a vehicle-mounted embedded GPU, and is communicated through a TSN network;
the vehicle-mounted external camera collects road condition information outwards of the vehicle and carries out real-time target detection through a first setting algorithm; the detection content comprises a trolley, a truck, a bus, pedestrians, bicycles and traffic signs, and the first setting algorithm is YOLOv algorithm.
And the vehicle-mounted internal camera is used for collecting driver behavior information facing the vehicle, and fatigue driving detection and bad behavior detection are performed through a second setting algorithm, wherein the second setting algorithm is lightweight MobileNet.
The vehicle-mounted embedded GPU is used for receiving the road condition information and the driver behavior information, performing data processing, and transmitting a processing result to the first domain controller;
And the first domain controller is used for performing perception fusion on the data transmitted by the vehicle-mounted embedded GPU by adopting a multi-perception fusion algorithm to form the decision, and sending the decision to the central gateway.
As described above, the auxiliary driving subsystem takes the embedded GPU as a computing unit, acquires information through the vehicle-mounted camera, and communicates through the TSN network. The road condition information is collected towards the outside of the vehicle, real-time target detection is carried out through YOLOv algorithm, and the main detection content comprises road condition information such as trolley, truck, bus, pedestrian, bicycle and traffic sign. Driver information is collected in the vehicle, and fatigue driving detection and bad behavior detection, such as call making, smoking and the like, are realized through lightweight MobileNet. And (3) carrying out data processing through the embedded GPU, summarizing the processing result to the first domain controller for perception fusion, forming a decision through a multi-perception fusion algorithm, and finally sending the decision to the central gateway through the first domain controller.
As shown in fig. 4, the steps of the multi-perception fusion algorithm include:
s31, acquiring data of an observation target by a plurality of sensors of different types;
S32, carrying out feature extraction and transformation on output data of a plurality of sensors of different types, and carrying out pattern recognition processing on a feature vector Yi to obtain description data of each sensor about a target;
S33, sensor synchronization including time synchronization and space synchronization is performed; the time synchronization and the space synchronization comprise IEEE802.1AS protocol and sensor calibration.
S34, grouping the description data of the sensors about the target according to the same target, and synthesizing the sensor data of the target by utilizing a fusion algorithm to obtain consistency interpretation and description of the observed target.
As described above, the principle of the multi-perception fusion algorithm is: firstly, data of an observation target are collected through a plurality of sensors of different types, and the output data of the sensors are subjected to feature extraction and transformation. The characteristic vector Yi is subjected to pattern recognition processing, and the description of each sensor about the target is completed. Secondly, sensor synchronization is carried out, including time synchronization and space synchronization, and design techniques include IEEE802.1AS, sensor calibration and the like. And finally, grouping the description data of each sensor about the target according to the same target, and synthesizing the sensor data of the target by utilizing a fusion algorithm to obtain consistency interpretation and description of the target.
Automobile control subsystem: the method is used for detecting the state of the automobile and detecting the obstacle by adopting a millimeter wave radar.
The automobile control subsystem comprises a second domain controller, an automobile power simulator, a millimeter wave radar and an automobile OBD instrument panel, and is communicated through an on-board CAN bus;
The second domain controller is used for receiving and analyzing CAN bus data, packaging the data into an Ethernet/TSN frame format and transmitting the Ethernet/TSN frame format to the central gateway through a Socket;
the automobile power simulator sends an automobile state signal in a first set period; the automobile state signals comprise automobile speed, tire rotating speed, cooling liquid temperature, air inlet manifold pressure, throttle position, ignition advance angle, engine load, residual oil quantity, air-fuel ratio and turbine pressure;
the millimeter wave radar collects barrier information in a second set period and calculates the angle, distance and speed of the barrier; wherein the obstacle information comprises a transverse distance, a longitudinal distance, a transverse speed and a longitudinal speed;
and the automobile OBD instrument panel reads the data of the automobile power simulator through an OBD interface, and displays the data in real time and gives an abnormal alarm.
As described above, the automobile control subsystem is composed of the second domain controller, the automobile OBD instrument panel, the automobile power simulator and the millimeter wave radar, and communicates through the vehicle-mounted CAN bus. The second domain controller realizes CAN bus data receiving and analyzing, packages the data into an Ethernet/TSN frame format, and transmits the Ethernet/TSN frame format to the central gateway through a Socket. The vehicle power simulator sends vehicle status signals at 50ms cycles including vehicle speed, tire speed, coolant temperature, intake air temperature, intake manifold pressure, throttle position, spark advance angle, engine load, residual oil, air-fuel ratio, turbine pressure, etc. The millimeter wave radar collects obstacle information including a transverse distance, a longitudinal distance, a transverse speed, a longitudinal speed and the like in a period of 30ms, and calculates the angle, the distance and the speed of the obstacle through the existing algorithm. The automobile OBD instrument panel can read the data of the power simulator through the OBD interface, and real-time display and abnormal alarm of the data are realized.
The central gateway:
The central gateway is used for connecting a first domain controller of the auxiliary driving subsystem and a second domain controller of the automobile control subsystem, is a CAN-TSN gateway, adopts a CAN-TSN protocol conversion algorithm to perform rate matching, distributes IP addresses for the first domain controller and the second domain controller, adopts Socket transmission of a TCP/IP network layer and Ethernet frame/TSN frame transmission of a data link layer;
The CAN-TSN protocol conversion algorithm comprises a CAN-TSN conversion strategy and a TSN-CAN conversion strategy, the CAN-TSN conversion strategy is packaged according to the message period and the load, and the TSN-CAN conversion strategy divides the TSN frame into domains which are required to be sent.
As described above, with the CAN-TSN gateway as the central gateway, the main functions include network routing, protocol conversion, and rate matching. The central gateway distributes IP addresses for each controller and supports Socket transmission of a TCP/IP network layer and Ethernet frame/TSN frame transmission of a data link layer; since the ethernet/TSN supports 100M or even 1000M networks, the CAN bus on board only supports 1M/s. Therefore, the gateway cannot directly forward and a CAN-TSN protocol conversion algorithm is required to perform rate matching. The CAN-TSN protocol conversion algorithm is divided into a CAN-TSN conversion strategy and a TSN-CAN conversion strategy, the CAN-to-TSN conversion strategy is packaged according to the message period and the load, and the bandwidth utilization rate is improved as much as possible on the premise of meeting the real-time performance and the safety. The TSN-to-CAN conversion strategy is to divide the fields to which the TSN frames are sent according to the needs, so that the number of CAN message creation is reduced as much as possible while the CAN message communication matrix is met.
As shown in fig. 5, the CAN-TSN conversion strategy converts the CAN message of the automobile control subsystem into a TSN frame and forwards the TSN frame to the auxiliary driving subsystem, and the steps include:
S11, loading the CAN message into a TSN frame;
s12, judging whether the TSN frame is overloaded or not;
s13, when the TSN frame is overloaded, creating the TSN frame, and returning to the step S11;
s14, judging whether real-time scheduling is met or not when the TSN frame is not overloaded;
s15, when real-time scheduling is not met, repackaging the message, and returning to the step S11;
S16, forwarding is carried out when the real-time scheduling is met.
As shown in fig. 6, the TSN-CAN conversion strategy sends the decision of the auxiliary driving subsystem to the automobile control subsystem, which includes the steps of:
s21, splitting a TSN frame;
s22, filling the data into CAN information;
S23, judging whether filling is possible;
s24, when the filling is impossible, a new CAN message is created, and the step S22 is returned;
s25, judging whether real-time scheduling is met or not when filling is possible;
s26, returning to the step S21 when real-time scheduling is not met;
and S27, forwarding when the real-time scheduling is met.
As described above, the CAN-TSN protocol conversion algorithm is divided into a CAN-TSN conversion strategy and a TSN-CAN conversion strategy, and the CAN-to-TSN conversion strategy is packaged according to the message period and the load, so that the bandwidth utilization rate is improved as much as possible on the premise of meeting the real-time performance and the safety. The TSN-to-CAN conversion strategy is to divide the fields to which the TSN frames are sent according to the needs, so that the number of CAN message creation is reduced as much as possible while the CAN message communication matrix is met.
As shown in fig. 1 and 2, the system of the present invention can be functionally divided into an outside detection, an inside detection, and an automobile state detection, and the outside detection: road condition detection, common road sign recognition and obstacle detection; and (3) in-vehicle detection: fatigue driving detection and bad driving behavior detection of a driver; and (3) detecting the state of the automobile: and (5) detecting and displaying the state and simulating the automobile braking. In fig. 2, EPS (Electric Power Steering, electric power steering system). The environment sensing sensor commonly used by ADAS mainly comprises a camera, a millimeter wave radar and an ultrasonic radar. Navi refers to a vehicle self-contained voice electronic navigation system (NAVI), and an on-board computer integrating multimedia functions and capable of simultaneously executing multiple tasks. V2X (Vehicle to Everything) is communicated with surrounding vehicles, equipment and base stations by taking the vehicle as the center, so that a series of traffic information such as real-time road conditions, road information, pedestrian information and the like is obtained, driving safety is improved, congestion is reduced, traffic efficiency is improved, vehicle-mounted entertainment information is provided and the like.
The invention adopts the CAN-TSN gateway to send the vehicle state data in the automobile control domain to the auxiliary driving subsystem, solves the problem of rate matching and protocol conversion among different networks, and achieves the auxiliary decision-making function of the auxiliary driving subsystem. Vehicle state monitoring and data acquisition are carried out through an automobile OBD interface, vehicle state data acquired by OBD are transmitted to an auxiliary driving subsystem through a CAN-TSN gateway, more data support is provided for decision making of the auxiliary driving subsystem, and multi-perception fusion is realized. And the decision is fed back to the automobile control subsystem to realize an auxiliary driving system integrating sensing, decision making and executing, so that the safety and reliability of the automatic driving automobile are improved.
The device model related in the invention comprises: the number of the NXP vehicle-mounted Ethernet switches is 1: the model is SJA1105Q-EVB; NXP car specification level microcontroller 2: the main model is S32K148EVB; NXP car specification level on-chip devices 2: the main model is I.MX6; NXP car specification PHY devices 2: the main model is TJA1101; 2 vehicle-mounted Ethernet transparent transmission modules; 4 vehicle millimeter wave radars: the main model is NA Lei Keji SR73; 1 automobile ECU power simulators; vehicle-mounted embedded GPU 2: the main model is Yingweida Jetson Nano; 2 vehicle-mounted cameras;
The invention adopts an automobile ECU simulator supporting an OBD interface to simulate a power domain data source, adopts a millimeter wave radar as a vehicle body domain data source, adopts a microcontroller as a node in a CAN domain, adopts a vehicle-mounted Ethernet switch as a central gateway, and adopts on-chip equipment as a domain controller. Acquiring the running state and road condition information of the automobile from an OBD interface and a millimeter wave radar, carrying out data analysis on a CAN node, converting the CAN message into a TSN frame through a central gateway and transmitting the TSN frame to an auxiliary driving subsystem; the method comprises the steps of collecting internal and external information of an automobile through a camera, deploying an AI detection algorithm on an embedded GPU, sending data to a domain controller for sensing fusion to form a decision, and sending the decision to an automobile control subsystem through a central gateway for decision making.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that comprises the element.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes using the descriptions and drawings of the present invention or directly or indirectly applied to other related technical fields are included in the scope of the invention.
Claims (5)
1. The advanced auxiliary driving system based on the CAN-TSN gateway is characterized by comprising an auxiliary driving subsystem, an automobile control subsystem and a central gateway;
The auxiliary driving subsystem is used for detecting road condition information and driver behaviors; the automobile control subsystem is used for detecting the state of an automobile and detecting obstacles by adopting a millimeter wave radar; the central gateway is used for connecting a first domain controller of the auxiliary driving subsystem and a second domain controller of the automobile control subsystem, is a CAN-TSN gateway, adopts a CAN-TSN protocol conversion algorithm to perform rate matching, distributes IP addresses for the first domain controller and the second domain controller, adopts Socket transmission of a TCP/IP network layer and Ethernet frame/TSN frame transmission of a data link layer;
The CAN-TSN protocol conversion algorithm comprises a CAN-TSN conversion strategy and a TSN-CAN conversion strategy, wherein the CAN-TSN conversion strategy is packaged according to a message period and a load, and the TSN-CAN conversion strategy divides a TSN frame into domains according to the need;
The CAN-TSN conversion strategy converts the CAN message of the automobile control subsystem into a TSN frame and forwards the TSN frame to the auxiliary driving subsystem, and the method comprises the following steps of:
S11, loading the CAN message into a TSN frame;
s12, judging whether the TSN frame is overloaded or not;
s13, when the TSN frame is overloaded, creating the TSN frame, and returning to the step S11;
s14, judging whether real-time scheduling is met or not when the TSN frame is not overloaded;
s15, when real-time scheduling is not met, repackaging the message, and returning to the step S11;
S16, forwarding when real-time scheduling is met;
the TSN-CAN conversion strategy sends the decision of the auxiliary driving subsystem to the automobile control subsystem, and the method comprises the following steps of:
s21, splitting a TSN frame;
s22, filling the data into CAN information;
S23, judging whether filling is possible;
s24, when the filling is impossible, a new CAN message is created, and the step S22 is returned;
s25, judging whether real-time scheduling is met or not when filling is possible;
s26, returning to the step S21 when real-time scheduling is not met;
s27, forwarding when real-time scheduling is met;
the auxiliary driving subsystem comprises a first domain controller, a vehicle-mounted external camera, a vehicle-mounted internal camera and a vehicle-mounted embedded GPU, and is communicated through a TSN network;
the vehicle-mounted external camera collects road condition information outwards of the vehicle and carries out real-time target detection through a first setting algorithm; the detection content comprises a trolley, a truck, a bus, pedestrians, bicycles and traffic signs;
The vehicle-mounted internal camera faces the vehicle to collect driver behavior information, and fatigue driving detection and bad behavior detection are carried out through a second setting algorithm;
The vehicle-mounted embedded GPU is used for receiving the road condition information and the driver behavior information, performing data processing, and transmitting a processing result to the first domain controller;
And the first domain controller is used for performing perception fusion on the data transmitted by the vehicle-mounted embedded GPU by adopting a multi-perception fusion algorithm to form the decision, and sending the decision to the central gateway.
2. The advanced driver assistance system based on a CAN-TSN gateway of claim 1, wherein the step of the multi-awareness fusion algorithm comprises:
s31, acquiring data of an observation target by a plurality of sensors of different types;
S32, carrying out feature extraction and transformation on output data of a plurality of sensors of different types, and carrying out pattern recognition processing on a feature vector Yi to obtain description data of each sensor about a target;
S33, sensor synchronization including time synchronization and space synchronization is performed;
S34, grouping the description data of the sensors about the target according to the same target, and synthesizing the sensor data of the target by utilizing a fusion algorithm to obtain consistency interpretation and description of the observed target.
3. Advanced driver assistance system based on CAN-TSN gateway according to claim 2, characterized in that the time and spatial synchronization comprises IEEE 802.1AS protocol and sensor calibration.
4. The CAN-TSN gateway-based advanced driver assistance system of claim 3, wherein the first set-up algorithm is YOLOv algorithm and the second set-up algorithm is lightweight MobileNet.
5. The CAN-TSN gateway-based advanced driver assistance system of claim 1, wherein the automotive control subsystem comprises a second domain controller, an automotive power simulator, a millimeter wave radar, and an automotive OBD dashboard, and communicates over an onboard CAN bus;
The second domain controller is used for receiving and analyzing CAN bus data, packaging the data into an Ethernet/TSN frame format and transmitting the Ethernet/TSN frame format to the central gateway through a Socket;
the automobile power simulator sends an automobile state signal in a first set period; the automobile state signals comprise automobile speed, tire rotating speed, cooling liquid temperature, air inlet manifold pressure, throttle position, ignition advance angle, engine load, residual oil quantity, air-fuel ratio and turbine pressure;
the millimeter wave radar collects barrier information in a second set period and calculates the angle, distance and speed of the barrier; wherein the obstacle information comprises a transverse distance, a longitudinal distance, a transverse speed and a longitudinal speed;
and the automobile OBD instrument panel reads the data of the automobile power simulator through an OBD interface, and displays the data in real time and gives an abnormal alarm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211000155.6A CN115520118B (en) | 2022-08-19 | 2022-08-19 | Advanced driving assistance system based on CAN-TSN gateway |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211000155.6A CN115520118B (en) | 2022-08-19 | 2022-08-19 | Advanced driving assistance system based on CAN-TSN gateway |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115520118A CN115520118A (en) | 2022-12-27 |
| CN115520118B true CN115520118B (en) | 2024-09-20 |
Family
ID=84696223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211000155.6A Active CN115520118B (en) | 2022-08-19 | 2022-08-19 | Advanced driving assistance system based on CAN-TSN gateway |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115520118B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117580095A (en) * | 2023-11-28 | 2024-02-20 | 北京理工大学深圳汽车研究院(电动车辆国家工程实验室深圳研究院) | Downlink transmission scheduling method and system for CAN-TSN gateway |
| CN118042012B (en) * | 2023-12-26 | 2025-03-28 | 北京国科天迅科技股份有限公司 | Bus data control method and device based on time-sensitive network |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109917765A (en) * | 2019-03-28 | 2019-06-21 | 东软睿驰汽车技术（沈阳）有限公司 | A kind of collecting and distributing type domain controller system of the network architecture based on automated driving system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112543153A (en) * | 2019-09-20 | 2021-03-23 | 华为技术有限公司 | Message forwarding method, device, system, equipment and storage medium |
| CN111626905A (en) * | 2020-05-26 | 2020-09-04 | 深圳技术大学 | Passenger safety monitoring method and device and computer readable storage medium |
| CN112835037B (en) * | 2020-12-29 | 2021-12-07 | 清华大学 | All-weather target detection method based on fusion of vision and millimeter waves |
| CN113923077A (en) * | 2021-07-28 | 2022-01-11 | 北京大学 | MODBUS-TSN gateway chip |
-
2022
- 2022-08-19 CN CN202211000155.6A patent/CN115520118B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109917765A (en) * | 2019-03-28 | 2019-06-21 | 东软睿驰汽车技术（沈阳）有限公司 | A kind of collecting and distributing type domain controller system of the network architecture based on automated driving system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115520118A (en) | 2022-12-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109421630B (en) | Controller architecture for monitoring health of autonomous vehicles | |
| CN115520118B (en) | Advanced driving assistance system based on CAN-TSN gateway | |
| CN110798658A (en) | Remote vehicle moving control method, mobile terminal and vehicle | |
| CN115334110A (en) | System architecture, communication method, vehicle, medium and chip for vehicle control | |
| CN110971567A (en) | Vehicle, cloud server, vehicle equipment, media device and data integration method | |
| CN112188440B (en) | Vehicle-road cooperative parallel simulation test method and system | |
| CN110576808B (en) | Vehicle, vehicle machine equipment and scene information pushing method based on artificial intelligence | |
| CN108833589B (en) | Vehicle, vehicle equipment and vehicle equipment navigation data synchronization method thereof | |
| CN115314526B (en) | System architecture, transmission method, vehicle, medium and chip for self-vehicle position recognition | |
| CN111311939A (en) | Vehicle, vehicle equipment and traffic light information map display method thereof | |
| CN110793537A (en) | Navigation path recommendation method, vehicle machine and vehicle | |
| WO2023028858A1 (en) | Test method and system | |
| KR20250029103A (en) | Vehicle data collection and transmission apparatus and method | |
| Prasad et al. | Development of a prototype ev autonomous vehicle for systematic research | |
| CN103021187B (en) | A kind of car alarming method and car-mounted terminal based on car networking technology | |
| CN115348657A (en) | System architecture and method for vehicle time synchronization and vehicle | |
| CN112017459A (en) | Vehicle, vehicle equipment and driving assistance method for signal lamp recognition thereof | |
| CN111210541A (en) | Vehicle, vehicle equipment and intelligent speed limit information prompting method | |
| Zhao et al. | Virtual traffic simulator for connected and automated vehicles | |
| CN110853387A (en) | Road condition panoramic video construction display method, system, server and vehicle | |
| CN115145246B (en) | Method and device for testing controller, vehicle, storage medium and chip | |
| KR20180092227A (en) | Integrated vehicle dynamics analysis apparatus | |
| CN115334109B (en) | System architecture, transmission method, vehicle, medium and chip for traffic signal recognition | |
| CN110689713A (en) | Vehicle, vehicle equipment and vehicle association control method thereof | |
| CN110843797A (en) | Vehicle, vehicle equipment and screen display prompting method for vehicle running state |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |