CN114771534A - Control method, training method, vehicle, equipment and medium for autonomous vehicle - Google Patents

Abstract

The disclosure provides a control method of an automatic driving vehicle, a training method and device of a deep learning model, the automatic driving vehicle, electronic equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence, in particular to the technical fields of automatic driving, intelligent transportation, high-precision maps, cloud services, car networking and the like. The specific implementation scheme is as follows: in response to receiving the lane change instruction, determining a first target lane change entry from a plurality of lane change entries based on target scene data, the target scene data including data related to the plurality of lane change entries; determining a lane change planning path based on the first target lane change convergence port; and controlling the vehicle to run according to the lane-changing planned route.

Description

Control method, training method, vehicle, equipment and medium for autonomous vehicle

technical field

The present disclosure relates to the technical field of artificial intelligence, and in particular, to the technical fields of autonomous driving, intelligent transportation, high-precision maps, cloud services, and Internet of Vehicles. A control method for an autonomous vehicle, a training method for a deep learning model, an apparatus, an autonomous vehicle, an electronic device, a storage medium, and a program product.

Background technique

Vehicles operating in autonomous mode can free occupants, especially drivers, from some driving-related responsibilities. When operating in autonomous driving mode, the vehicle can navigate to various locations using onboard sensors, allowing the vehicle to travel with minimal human interaction or in some situations without any passengers.

Lane-changing driving is generally driving to change lanes to a corresponding turning lane in response to an instruction for steering driving, or lane-changing driving in response to a command for bypassing a construction road section. However, how to determine a reasonable lane-changing planning path from a complex lane-changing driving scenario and control the vehicle to automatically change lanes is an important manifestation of the autonomous driving capability.

SUMMARY OF THE INVENTION

The present disclosure provides a control method for an automatic driving vehicle, a training method for a deep learning model, an apparatus, an automatic driving vehicle, an electronic device, a storage medium, and a program product.

According to an aspect of the present disclosure, there is provided a control method for an automatic driving vehicle, comprising: in response to receiving a lane change instruction, determining a first target lane change entrance from a plurality of lane change entrances based on target scene data, The target scene data includes data related to a plurality of lane-change entrances; a planned lane-change path is determined based on the first target lane-change entrance; and the vehicle is controlled to change lanes according to the lane-change planned path.

According to another aspect of the present disclosure, a method for training a deep learning model is provided, comprising: determining a training sample, wherein the training sample includes sample scene data and a label, and the sample scene data includes a Data, the label includes a positive sample label and a negative sample label, the positive sample label is used to indicate the first target sample change lane entry, and the first target sample lane change entry includes the successfully imported samples from the multiple sample lane change entrances The lane change entrance, the negative sample label is used to indicate the second target sample lane change entrance, and the second target sample lane change entrance includes samples other than the first target sample lane change entrance among the multiple sample lane change entrances lane-changing entrance; and using the training samples to train a deep learning model to obtain a trained deep learning model.

According to another aspect of the present disclosure, there is provided a control device for an automatic driving vehicle, comprising: a first determination module configured to, in response to receiving a lane change instruction, determine from a plurality of lane change entrances based on target scene data a first target lane change entrance, wherein the target scene data includes data related to a plurality of lane change entrances; a second determination module for determining a planned lane change path based on the first target lane change entrance; and a driving module , which is used to control the vehicle to change lanes according to the planned lane change path.

According to another aspect of the present disclosure, there is provided a training device for a deep learning model, comprising: a sample determination module for acquiring training samples, wherein the training samples include sample scene data and labels, and the sample scene data includes a plurality of samples The data related to the lane change entrance, the label includes a positive sample label and a negative sample label, the positive sample label is used to indicate the first target sample lane change entrance, and the first target sample lane change entrance includes multiple sample lane change entrances. The successfully imported sample lane change entry, the negative sample label is used to indicate the second target sample lane change entry, and the second target sample lane change entry includes a plurality of sample lane change entries except the first target sample entry. a sample change channel entrance other than the channel entrance; and a training module for training a deep learning model by using the training samples to obtain a trained deep learning model.

According to another aspect of the present disclosure, there is provided an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor. The at least one processor executes to enable the at least one processor to perform a method as disclosed.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform a method as disclosed.

According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program that, when executed by a processor, implements a method as disclosed herein.

According to another aspect of the present disclosure, there is provided an autonomous vehicle including the electronic device as described in the present disclosure.

It should be understood that what is described in this section is not intended to identify key or critical features of embodiments of the disclosure, nor is it intended to limit the scope of the disclosure. Other features of the present disclosure will become readily understood from the following description.

Description of drawings

The accompanying drawings are used for better understanding of the present solution, and do not constitute a limitation to the present disclosure. in:

FIG. 1 schematically shows an exemplary system architecture to which a control method and apparatus for an automatic driving vehicle can be applied according to an embodiment of the present disclosure;

FIG. 2 schematically shows an application scenario diagram of a method for controlling an autonomous driving vehicle according to an embodiment of the present disclosure;

FIG. 3 schematically shows a flowchart of a control method for an automatic driving vehicle according to an embodiment of the present disclosure;

FIG. 4 schematically shows a scene diagram of a control method for an automatic driving vehicle according to an embodiment of the present disclosure;

FIG. 5 schematically shows a schematic flowchart of a control method for an automatic driving vehicle according to an embodiment of the present disclosure;

FIG. 6 schematically shows a flowchart of a training method for a deep learning model according to an embodiment of the present disclosure;

FIG. 7 schematically shows a block diagram of a control device of an autonomous driving vehicle according to an embodiment of the present disclosure;

FIG. 8 schematically shows a block diagram of an apparatus for training a deep learning model according to an embodiment of the present disclosure; and

FIG. 9 schematically shows a block diagram of an electronic device suitable for implementing a control method of an autonomous driving vehicle according to an embodiment of the present disclosure.

Detailed ways

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding and should be considered as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted from the following description for clarity and conciseness.

The present disclosure provides a control method for an automatic driving vehicle, a training method for a deep learning model, an apparatus, an automatic driving vehicle, an electronic device, a storage medium, and a program product.

According to an embodiment of the present disclosure, there is provided a control method for an automatic driving vehicle, comprising: in response to receiving a lane change instruction, determining a first target lane change entrance from a plurality of lane change entrances based on target scene data, The target scene data includes data related to a plurality of lane change entrances; based on the first target lane change entrance, a planned lane change path is determined; and the vehicle is controlled to travel according to the lane change planned path.

In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision, disclosure and application of the user's personal information involved are all in compliance with the relevant laws and regulations, and necessary confidentiality measures have been taken, and do not violate the Public order and good customs.

In the technical solution of the present disclosure, the authorization or consent of the user is obtained before the user's personal information is obtained or collected.

FIG. 1 schematically shows an exemplary system architecture to which a control method and apparatus for an autonomous driving vehicle can be applied according to an embodiment of the present disclosure.

It should be noted that FIG. 1 is only an example of a system architecture to which the embodiments of the present disclosure can be applied, so as to help those skilled in the art to understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used for other A device, system, environment or scene. For example, in another embodiment, an exemplary system architecture to which the control method and device for an autonomous vehicle can be applied may include an in-vehicle terminal of an autonomous vehicle, but the in-vehicle terminal may implement the embodiments of the present disclosure without interacting with a server Provided are a control method and device for an automatic driving vehicle.

As shown in FIG. 1 , the system architecture 100 system according to this embodiment may include an autonomous vehicle 101 , a network 102 and a server 103 . Autonomous vehicle 101 may be communicatively coupled to one or more servers 103 via network 102 . The network 102 may be any type of network, eg, a wired or wireless local area network (LAN), a wide area network (WAN) such as the Internet, a cellular network, a satellite network, or a combination thereof. The server 103 may be any type of server or server cluster, eg, a web or cloud server, an application server, a backend server, or a combination thereof. The server may be a data analysis server, a content server, a traffic information server, a map and point of interest (MPOI) server, or a location server, or the like.

Autonomous vehicle 101 may refer to a vehicle configured to operate in an autonomous driving mode. But it is not limited to this. Autonomous vehicles can also operate in manual mode, in fully autonomous driving mode, or in partially autonomous driving mode.

The autonomous driving vehicle 101 may include an on-board terminal, a vehicle control module, a wireless communication module, a user interface module, and a sensing module. The autonomous vehicle 101 may also include common components included in ordinary vehicles, such as: an engine, wheels, steering wheel, transmission, and the like. Common components can be controlled by the vehicle terminal and the vehicle control module using various communication commands, such as acceleration commands, deceleration commands, steering commands, and braking commands.

The various modules in the autonomous vehicle 101 may be communicatively coupled to each other via interconnects, buses, networks, or a combination thereof. For example, they may be communicatively coupled to each other via a controller area network (CAN) bus. CAN bus is a vehicle bus standard designed to allow microcontrollers and devices to communicate with each other in applications without a host.

Sensing modules may include, but are not limited to, one or more cameras, global positioning system (GPS) units, inertial measurement units (IMUs), radar units, and light detection and ranging (LIDAR) units. The GPS unit may include a transceiver operable to provide information about the location of the autonomous vehicle. The IMU unit can sense position and orientation changes of the autonomous vehicle based on inertial acceleration. A radar unit may represent a system that utilizes radio signals to sense obstacles within the surrounding environment of an autonomous vehicle. In addition to sensing the obstacle, the radar unit may additionally sense the speed and/or heading of the obstacle. LIDAR units can use lasers to sense obstacles in the environment of the autonomous vehicle. The LIDAR unit may include, among other components, one or more laser sources, laser scanners, and one or more detectors. The cameras may include one or more devices used to capture images of the environment surrounding the autonomous vehicle. The cameras can be still cameras and/or video cameras. The camera may be mechanically movable, eg by mounting the camera on a rotating or tilting platform.

The sensing module may also include other sensors such as: sonar sensors, infrared sensors, steering sensors, accelerator sensors, brake sensors, and audio sensors (eg, microphones). Audio sensors may be configured to collect sound from the environment surrounding the autonomous vehicle. The steering sensor may be configured to sense the steering angle of the steering wheel, the wheels of the autonomous vehicle, or a combination thereof. The accelerator sensor and brake sensor sense the accelerator position and brake position of the autonomous vehicle, respectively. In some cases, the accelerator sensor and brake sensor may be integrated into an integrated accelerator/brake sensor.

Vehicle control modules may include, but are not limited to, steering units, accelerator units (also referred to as acceleration units), and braking units. Steering units are used to adjust the direction or heading of an autonomous vehicle. The throttle unit is used to control the speed of the electric motor or engine, which in turn controls the speed and acceleration of the autonomous vehicle. The braking unit slows the self-driving vehicle by providing friction to slow the wheels or tires of the self-driving vehicle.

The wireless communication module allows communication between the autonomous vehicle and external modules such as devices, sensors, other vehicles, and the like. For example, the wireless communication module may communicate wirelessly with one or more devices directly or via a communication network, eg, with a server over a network. The wireless communication module may use any cellular communication network or wireless local area network (WLAN), eg, WiFi, to communicate with another component or module. User interface modules may be part of peripheral devices implemented within the autonomous vehicle, including, for example, keyboards, touch screen displays, microphones, speakers, and the like.

Some or all of the functions of the autonomous vehicle 101 may be controlled or managed by the onboard terminal, especially when operating in an autonomous driving mode. The in-vehicle terminal includes the necessary hardware (eg, processor, memory, storage device) and software (eg, operating system, planning and routing programs) to connect from the sensing module, control module, wireless communication module, and/or user interface module Receive information, process the received information, and generate instructions for controlling the autonomous vehicle. Alternatively, the vehicle-mounted terminal may be integrated with the control module.

For example, a user as a passenger may specify the start location and destination of the trip, eg, via a user interface module. The in-vehicle terminal obtains trip-related data. For example, the in-vehicle terminal may obtain the location and the drivable route from the MPOI server, which may be a part of the server. The location server provides location services, and the MPOI server provides map services. Alternatively, such locations and maps may be cached locally in persistent storage in the vehicle terminal.

When the autonomous vehicle moves along the drivable path, the in-vehicle terminal can also obtain real-time traffic information from the traffic information system or server. The server may be operated by a third party entity. The function of the server can be integrated with the vehicle terminal. Based on real-time traffic information, and location information, and real-time local environmental data detected or sensed by the sensing module, the in-vehicle terminal can plan an optimal path and control the autonomous vehicle according to the planned optimal path, eg, via a control module, to safely and Efficiently reach the designated destination.

It should be understood that the numbers of autonomous vehicles, networks and servers in FIG. 1 are merely illustrative. There can be any number of autonomous vehicles, networks, and servers depending on the implementation needs.

It should be noted that the sequence numbers of the respective operations in the following methods are only used as representations of the operations for the convenience of description, and should not be regarded as representing the execution order of the respective operations. The methods need not be performed in the exact order shown unless explicitly stated.

FIG. 2 schematically shows an application scenario diagram of a method for controlling an autonomous driving vehicle according to an embodiment of the present disclosure.

As shown in FIG. 2 , when the vehicle ADC201 (ie, the autonomous driving vehicle, hereinafter referred to as the vehicle) needs to turn, such as turning left, the vehicle-mounted terminal mounted on the vehicle ADC201 or the server in communication with the vehicle ADC201 generates a lane change instruction, such as Generates instructions for changing lanes from the straight lane to the left turn lane. It may be that the vehicle-mounted terminal receives the target scene data collected from the sensing module in response to receiving the lane change instruction. The control method for an automatic driving vehicle provided by the embodiment of the present disclosure is executed based on the target scene data. But it is not limited to this. It is also possible that the server receives target scene data collected from the sensing module in response to receiving the lane change instruction, and executes the method for controlling an autonomous driving vehicle provided by the embodiments of the present disclosure based on the target scene data. The following embodiments will take the in-vehicle terminal as an execution subject for illustration, and will not be repeated here.

As shown in FIG. 2 , the target scene data may include obstacle data about obstacles OBS201 , OBS202 , OBS203 , and OBS204 on the left-turn lane. The target scene data may also include data related to multiple lane change entrances, such as data related to lane change entrances GAP201, GAP202, and GAP203, and may also include vehicle data, environment data around the vehicle, and road traffic rule data. The first target lane change entrance can be determined from a plurality of lane change entrances by using the control method for an automatic driving vehicle provided by the embodiment of the present disclosure. Based on the first target lane change entrance, a planned lane change path is determined. And control the vehicle ADC201 to change lanes according to the lane change planning path.

Using the control method for an automatic driving vehicle provided by the embodiment of the present disclosure, according to the state of the vehicle and the state of surrounding obstacles, the optional range of lane change is changed from the lane change entrance adjacent to the vehicle on the target lane to the multiple lanes on the target lane. There is a lane-changing entrance, which expands the optional range of lane-changing, reduces the restriction of lane-changing driving, and improves the intelligence and flexibility of lane-changing driving.

FIG. 3 schematically shows a flowchart of a control method for an automatic driving vehicle according to an embodiment of the present disclosure.

As shown in FIG. 3, the method includes operations S310-S330.

In operation S310, in response to receiving the lane change instruction, a first target lane change entrance is determined from the plurality of lane change entrances based on the target scene data. The target scene data includes data related to a plurality of lane change entrances.

In operation S320, a planned lane change path is determined based on the first target lane change entrance.

In operation S330, the vehicle is controlled to change lanes according to the planned lane change route.

According to embodiments of the present disclosure, in response to receiving a lane change instruction, the lane change task may be accomplished through a plurality of lane change phases. The plurality of lane change phases may include: a lane change planning phase, and a lane change execution phase.

According to an embodiment of the present disclosure, the lane change planning stage may include a stage of generating a lane change intention or generating a lane change task in response to receiving a lane change instruction. The lane change execution stage may include: a stage of controlling the vehicle to change lanes according to the planned lane change path.

For example, when the vehicle is going straight, in response to receiving a lane change instruction, the on-board terminal can enter the lane change planning stage from the straight drive stage, and will determine the first target lane change entrance from the target lane according to the target scene data, based on the first target lane change entry. A target lane-change entrance, and a lane-change planning path is generated. In the lane change execution stage, the planned path will be based on the lane change at the first target lane change entrance, and the vehicle will be controlled to perform the lane change and merge operation.

According to an embodiment of the present disclosure, the lane change entrance may refer to a travel space on the target lane where the vehicle is allowed to perform a lane change task. This travel space is larger than the vehicle's movement space during lane change tasks.

According to other embodiments of the present disclosure, in the lane change planning stage, a lane change entrance adjacent to the vehicle on the target lane may be used as the first target lane change entrance according to a predetermined lane change rule. The lane-change entrance adjacent to the vehicle on the target lane can be understood as: a lane-change entrance that satisfies a predetermined lane change condition. Satisfying the predetermined lane change condition may refer to: controlling the vehicle to change lanes according to the planned lane change path, and to be able to complete the lane change task within a predetermined period of time. The lane-change planning path may include not only the lane-change planning trajectory, but also a safe driving speed. The predetermined duration may be 8s, but is not limited to this, as long as it is a duration that can ensure safe lane change. When it is determined that the target scene data does not meet the predetermined lane change conditions, for example, if it is determined that there is no lane change entrance at the adjacent position of the vehicle, it will stay in the lane change planning stage and wait for the lane change entrance that meets the predetermined lane change conditions. From the lane change planning stage to the lane change execution stage, the lane change task cannot be performed.

According to an embodiment of the present disclosure, in the lane change planning stage, the first target lane change entrance may be determined from a plurality of lane change entrances based on the target scene data. For example, in a traffic scene where obstacles on the target lane form a traffic flow, not only the lane change entrance at the position adjacent to the vehicle on the target lane, but also the forward lane change of the vehicle on the target lane can be taken into consideration. Merge entrances and rear-facing lane change entrances for vehicles are taken into account to increase lane change opportunities. For example, in a case where it is determined that there is no lane change adjacent to the vehicle on the target lane among the plurality of lane change entrances, the forward lane change entrance and the rearward lane change of the plurality of lane change entrances may be selected. A first target lane-change junction is determined in the junction. This expands the selection range of variable road routes, expands the timing of changing lanes, reduces the restrictions on changing lanes, and improves the intelligence and flexibility of changing lanes.

According to other embodiments of the present disclosure, the planned lane change path may be adjusted according to the first target lane change entrance. For example, it is possible to adjust the predetermined lane change conditions specified by the lane change, to adjust the safe driving speed, or to adjust the predetermined time period. For example, the safe driving speed is adjusted from 60km/h to 80km/h, or the predetermined duration is adjusted from 8s to 10s, so that a planned lane change path that satisfies the predetermined conditions can be generated according to the first target lane change entrance, and then the vehicle can be controlled Change lanes according to the planned lane change route, and complete the lane change task when the adjusted preset lane change conditions are met.

According to an embodiment of the present disclosure, when the first target lane change entrance is determined, a lane change adjustment phase (prepare change lane task) may be added after the lane change planning phase. For example, the multiple lane change phases are changed from including a lane change planning phase and a lane change execution phase to include a lane change planning phase, a lane change adjustment phase, and a lane change execution phase.

According to an embodiment of the present disclosure, the lane change adjustment phase may include a phase of adjusting the traveling speed of the vehicle. For example, in a case where it is determined that the target scene data does not meet the predetermined lane change condition, the driving speed is adjusted so that the adjusted target scene data meets the predetermined lane change condition.

According to the embodiments of the present disclosure, by adjusting the driving speed of the vehicle, the relative speed and relative position between the vehicle and the obstacles on the target lane can be adjusted, so that the adjusted target scene data, for example, converges with the first target lane change. The adjusted target scene data related to the entrance satisfies the predetermined lane change condition, and can transition from the lane change adjustment stage to the lane change generation stage.

FIG. 4 schematically shows a scene diagram of a control method for an automatic driving vehicle according to an embodiment of the present disclosure.

As shown in FIG. 4 , in response to receiving a lane change instruction to change lanes from the straight lane to the left turn lane, the lane change planning stage is entered, and the target scene data is determined. The target scene data can be used to characterize: in the target lane, such as the left-turn lane, there are obstacles OBS401, OBS402, OBS403, in front of the obstacle OBS401, there is a first lane change entrance, and there is a first lane between the obstacle OBS402 and the obstacle OBS403. The second lane change entrance, behind the obstacle OBS403, has the third lane change entrance.

The in-vehicle terminal determines the second lane change entrance as the first target lane change entrance from the first lane change entrance, the second lane change entrance, and the third lane change entrance based on the target scene data.

As shown in Figure 4, the vehicle-mounted terminal can generate an initial lane change planning path based on the target scene data, and determine the required lane change duration. In the case of driving at a safe driving speed according to the initial lane change planning path and the lane change duration is longer than the predetermined time duration, it can be determined that the target scene data does not meet the predetermined lane change conditions. In this case, the vehicle terminal can control the vehicle ADC401 to enter the lane change adjustment stage. The magnitude of the traveling speed of the vehicle ADC401 in the traveling direction of the straight lane is adjusted so that the lane-change entrance GAP402 becomes the lane-change entrance adjacent to the vehicle ADC401 in the left-turn lane. Further, when it is determined that the adjusted target scene data satisfies the predetermined lane change condition, a planned lane change path is generated. Entering the lane change execution stage, the vehicle ADC401 is controlled to change lanes according to the lane change planning path, and finally realize the completion of the lane change task.

Using the control method for an autonomous driving vehicle provided by the embodiment of the present disclosure, when it is determined that the target scene data related to the current driving position and the first target lane change entrance does not meet the predetermined lane change condition, a lane change adjustment stage can be set, Make the adjusted target scene data meet the predetermined lane change conditions, and then by setting the lane change adjustment stage, actively create the lane change entry opportunity, effectively improve the lane change entry capability, and improve the automatic lane change process while ensuring the safety of the lane change process. Intelligence and flexibility in driving mode.

According to an embodiment of the present disclosure, in a case where it is determined that the target scene data does not satisfy the predetermined lane change condition, the target acceleration for adjusting the traveling speed is determined. When it is determined that the target acceleration satisfies a predetermined adjustment speed condition, the travel speed is adjusted according to the target acceleration. When it is determined that the target acceleration does not satisfy the adjustment speed condition, the operation of adjusting the traveling speed is stopped.

According to an embodiment of the present disclosure, the predetermined adjustment speed condition may include a safe driving speed condition, but is not limited thereto, and may also include a somatosensory condition.

For example, in the case of adjusting the traveling speed by adjusting the target acceleration, the vehicle accelerates according to the target acceleration, and the adjusted traveling speed cannot be greater than the safe traveling speed. Alternatively, deceleration is performed according to the target acceleration, and the vehicle cannot decelerate too much to satisfy the somatosensory conditions, such as emergency braking.

By using the control method for an automatic driving vehicle provided by the embodiments of the present disclosure, the safety and comfort of the lane changing process can be improved while the lane changing capability is improved.

According to the embodiments of the present disclosure, in the lane change adjustment stage, in the case of adjusting the driving speed, the current scene data can be collected in real time by the sensing module, and the vehicle terminal receives the current scene data from the sensing module. The current scene data may include target scene data in the process of adjusting the driving speed. The in-vehicle terminal determines that the current scene data satisfies the predetermined lane change cancellation condition, and can cancel the operation of adjusting the driving speed and cancel the lane change if the current scene data meets the predetermined lane change cancellation condition.

According to an embodiment of the present disclosure, satisfying the predetermined cancellation lane change condition may include not satisfying the predetermined lane change condition. For example, the current scene data indicates that the space of the first target lane change entrance is smaller than the movement space of the vehicle during the lane change process, and there are no other lane change entrances on the target lane.

According to the embodiments of the present disclosure, in the lane change adjustment stage, in the case of adjusting the driving speed, the current scene data can be collected in real time by the sensing module, and the vehicle terminal receives the current scene data from the sensing module. The in-vehicle terminal determines the second target lane change entrance from the plurality of lane change entrances based on the current scene data. Based on the first target lane change entrance, a first lane change cost value is determined. The first lane change cost value is used to represent: the cost value of changing lanes according to the first target lane change entrance. Based on the second target lane change entry, a second lane change cost is determined. The second lane-change cost is used to represent: the cost of changing lanes at the entrance of the second target lane-change. Based on the first lane change cost value and the second lane change cost value, a new target lane change entrance is determined from the first target lane change entrance and the second target lane change entrance. Based on the new target lane change entrance, the vehicle terminal determines the updated lane change planning path. Control the vehicle to change lanes according to the updated lane change planning path.

According to an embodiment of the present disclosure, the first lane change cost value or the second lane change cost value may include at least one of the following: a somatosensory value, an aging value, a safety value, and the like.

According to an embodiment of the present disclosure, the somatosensory value may be used to characterize the comfort level felt by the body. For example, when the vehicle is driving smoothly, the comfort level of the occupants is high, and the somatosensory value is high; on the contrary, when the vehicle brakes suddenly, the comfort level of the passengers is low, and the somatosensory value is low. The aging value can be used to characterize driving efficiency. For example, for the same distance, the shorter the travel time, the higher the aging value. Safety values can be used to characterize driving safety. For example, when the vehicle is driving, if the risk of collision between the vehicle and surrounding obstacles is low, the safety value is high. On the contrary, if the risk of collision between the vehicle and surrounding obstacles is high, the safety value is low.

According to an embodiment of the present disclosure, a first planned lane change path is determined based on the first target lane change entrance. Based on the first lane change planning path, the first lane change cost value is determined. For example, the weighted summation obtains the first lane-change cost value based on the first somatosensory value, the first aging value, and the first safety value obtained from the first lane-change planned route. Similarly, based on the second target lane change entrance, a second lane change planned path is determined. Based on the second lane change planning path, the second lane change cost value is determined.

According to the embodiments of the present disclosure, the first lane change cost value and the second lane change cost value may be compared, the larger value is used as the target cost value, and the lane change entry corresponding to the target cost value is used as the new target lane change Import entrance.

By using the control method for an automatic driving vehicle provided by the embodiments of the present disclosure, by determining the cost value, comprehensive performances such as comfort, safety, and lane-changing efficiency during the lane changing process can be improved.

FIG. 5 schematically shows a flow chart of a control method for an automatic driving vehicle according to another embodiment of the present disclosure.

As shown in FIG. 5 , the first target lane change junction 530 may be determined from the plurality of lane change junctions based on the target scene data 520 using the lane change gap selection model N520 . For example, the target scene data 520 related to a plurality of lane change entrances is input into the lane change gap selection model N520, and a classification result for indicating the first target lane change entrance is output.

As shown in FIG. 5 , the feature extraction model N510 and the path planning model N530 can also be used to assist in the completion of the control method. Feature data satisfying predetermined extraction conditions can be extracted from the traffic scene data 510 by using the feature extraction model N510 as the target scene data 520 . For example, the traffic scene data 510 may be input into the feature extraction model N510 to obtain the target scene data 520 . The lane change planning path 540 may be determined using the path planning model N530. For example, data such as the current vehicle position 550 and the first target lane change entrance 530 may be input into the route planning model N530 to obtain the lane change planned route 540 . Finally, the vehicle is controlled to change lanes according to the planned lane change path 540 .

According to an embodiment of the present disclosure, the lane change gap selection model may include a deep learning model such as an SVM (Support Vector Machine, support vector machine), a neural network, and the like.

According to an embodiment of the present disclosure, the feature extraction model may be a screening rule, but is not limited thereto, and may also include, for example, CNN (Convolutional Neural Network, convolutional neural network), RNN (Recurrent Neural Network, cyclic neural network) for feature extraction neural network) and other deep learning models.

According to an embodiment of the present disclosure, the path planning model may include a graph search method, a rapid search random tree (RRT) algorithm, and the like.

According to an embodiment of the present disclosure, the traffic scene data may include: time data, vehicle data, data of obstacles related to obstacles, environment data, road traffic rule data, and the like. The feature data satisfying the predetermined extraction condition may include at least one of the following: data of obstacles related to lane change, vehicle data, environment data, and road traffic rule data.

According to an embodiment of the present disclosure, the data of the obstacle related to the lane change may include: the size of the obstacle, the driving speed of the obstacle, the driving direction of the obstacle, and the driving acceleration of the obstacle and other state data and attribute data. The vehicle data may include: the size of the vehicle, the driving speed of the vehicle, the driving direction of the vehicle, the driving acceleration of the vehicle and other state data and attribute data. The environmental data may include: weather, visibility, road muddyness, road congestion, road construction and other objective driving environment data. The road traffic rule data may include subjective driving rule data such as speed limit rules, irreversible driving rules, and rules not to change lanes across solid lines.

According to an embodiment of the present disclosure, taking vehicle data as an example, the vehicle data in the traffic scene data may include the license plate of the vehicle, the annual inspection data of the vehicle, the size of the vehicle, and the status data of the vehicle. After being processed by the feature extraction model, the vehicle data in the target scene data may include the state data of the vehicle and the size of the vehicle.

Using the method for controlling an autonomous vehicle provided by the embodiment of the present disclosure, target scene data can be extracted from a large amount of traffic scene data by means of feature extraction, so that the target scene data is simplified data, thereby reducing the amount of data processing and improving Data processing efficiency.

FIG. 6 schematically shows a flowchart of a training method of a deep learning model according to an embodiment of the present disclosure.

As shown in FIG. 6 , the method includes operations S610-S620.

In operation S610, training samples are determined. The training samples include sample scene data and labels, the sample scene data includes data related to multiple sample lane change entrances, the labels include positive sample labels and negative sample labels, and the positive sample labels are used to indicate the first target sample lane change entrances, The first target sample lane change entry includes a successfully imported sample lane change entry among the multiple sample lane change entrances, and the negative sample label is used to indicate the second target sample lane change entry, the second target sample lane change entry. The inlets include sample re-laning inlets other than the first target sample re-laning inlet among the plurality of sample re-laning inlets.

In operation S620, a deep learning model is trained using the training samples to obtain a trained deep learning model.

According to the embodiment of the present disclosure, the trained deep learning model can be used as a lane change gap selection model, and the determination of the first target lane change entrance can be regarded as a classification problem. Add the first target sample lane change entrance and the second target sample lane change entrance to the training sample, so that the deep learning model not only learns the feature data of the successfully imported sample lane change entrance during the training process, but also learns To the feature data of the unselected sample lane change entrance, the trained deep learning model has high accuracy and strong robustness in the selection of lane change entrance.

According to other embodiments of the present disclosure, a target sample lane change entrance can be determined from a plurality of sample lane change entrances according to a corresponding lane change entrance selection rule, and the effect can be verified through simulation.

According to the embodiments of the present disclosure, compared with the method of using the selection rule to determine the first target lane change entrance, using the intelligent determination method of the lane change gap selection model to determine the first target lane change entrance can make the determination process shorter , high efficiency, and high intelligence.

According to an embodiment of the present disclosure, for operation S610, determining a training sample may include: acquiring initial sample scene data. Feature data satisfying predetermined extraction conditions is extracted from the initial sample scene data as sample scene data.

According to an embodiment of the present disclosure, the initial sample scene data may include: closed-loop data, such as the data of the successful lane change of the vehicle; open-loop data, such as the data of the successful lane change of the vehicle controlled by the human driver; and the successful lane change of other vehicles collected by the vehicle The data.

According to an embodiment of the present disclosure, the feature data satisfying the predetermined extraction condition includes at least one of the following: data of obstacles related to lane change, vehicle data, environment data, and road traffic rule data.

Using the training method of the deep learning model provided by the embodiment of the present disclosure, the sample scene data can be extracted from a large amount of initial sample scene data by means of feature extraction, so that the sample scene data is simplified data, thereby reducing the amount of data processing , to improve the training efficiency of deep learning models.

According to the embodiments of the present disclosure, the process operation code of the control method of the autonomous driving vehicle can call the process operation code related to the training method of the deep learning model. For example, extracting feature data satisfying predetermined extraction conditions from initial sample scene data as an operation code for sample scene data operations, and performing extraction of feature data satisfying predetermined extraction conditions from traffic scene data as target scene data operations The code is the same. The above method can ensure the consistency of the training process and the application process of the lane-changing gap selection model.

FIG. 7 schematically shows a block diagram of a control apparatus of an autonomous driving vehicle according to an embodiment of the present disclosure.

As shown in FIG. 7 , the control device 700 of the automatic driving vehicle includes: a first determination module 710 , a second determination module 720 , and a driving module 730 .

The first determination module 710 is configured to, in response to receiving the lane change instruction, determine the first target lane change entrance from the plurality of lane change entrances based on the target scene data, wherein the target scene data includes a plurality of lane change entrances. Entry-related data.

The second determination module 720 is configured to determine a planned lane change path based on the first target lane change entrance.

The driving module 730 is configured to control the vehicle to change lanes according to the planned lane change route.

According to an embodiment of the present disclosure, the control apparatus for an automatic driving vehicle further includes, before the first determination module: an extraction module.

The extraction module is used for extracting feature data that satisfies predetermined extraction conditions from the traffic scene data as target scene data.

According to an embodiment of the present disclosure, the control apparatus for an automatic driving vehicle further includes, before the second determination module: an adjustment module.

The adjustment module is configured to adjust the driving speed when it is determined that the target scene data does not meet the predetermined lane change condition, so that the adjusted lane change scene data meets the predetermined lane change condition.

According to an embodiment of the present disclosure, the control device for an automatic driving vehicle further includes, in the case of adjusting the driving speed when it is determined that the target scene data does not meet the predetermined lane change condition: a third determination module and a cancellation module.

The third determination module is used to determine the current scene data. The current scene data includes target scene data in the process of adjusting the travel speed.

The cancellation module is configured to cancel the operation of adjusting the driving speed and cancel the lane change when it is determined that the current scene data satisfies the predetermined conditions for canceling the lane change.

According to an embodiment of the present disclosure, the control device for an automatic driving vehicle further includes, in the case of adjusting the driving speed in the case where it is determined that the target scene data does not meet the predetermined lane change condition: a fourth determination module, a fifth determination module, a fourth determination module, a fifth determination module, a fourth determination module, and a fifth determination module Six determine the module, update the module.

The fourth determination module is configured to determine the second target lane change entrance from the plurality of lane change entrances based on the current scene data.

The fifth determination module is configured to determine the first lane change cost based on the first target lane change entrance. The first lane change cost value is used to represent: the cost value of changing lanes according to the first target lane change entrance.

The sixth determination module is configured to determine the second lane change cost based on the second target lane change entrance. The second lane-change cost is used to represent: the cost of changing lanes at the entrance of the second target lane-change.

The update module is used for determining a new target lane change entrance from the first target lane change entrance and the second target lane change entrance based on the first lane change cost value and the second lane change cost value, so as to change the target lane based on the new target lane change entrance. Roadway entrance, to determine the updated lane change planning path.

According to an embodiment of the present disclosure, the predetermined lane change condition includes: a condition for completing the lane change within a predetermined period of time at a safe driving speed.

According to an embodiment of the present disclosure, the feature data satisfying the predetermined extraction condition includes at least one of the following: data of obstacles related to lane change, vehicle data, environment data, and road traffic rule data.

According to an embodiment of the present disclosure, the adjustment module includes an acceleration adjustment unit and a speed adjustment unit.

The acceleration adjustment unit is configured to determine a target acceleration for adjusting the traveling speed when it is determined that the target scene data does not meet the predetermined lane change condition.

The speed adjustment unit is configured to adjust the traveling speed according to the target driving degree when it is determined that the target acceleration satisfies a predetermined adjustment speed condition.

FIG. 8 schematically shows a block diagram of an apparatus for training a deep learning model according to an embodiment of the present disclosure.

As shown in FIG. 8 , the training device 800 of the deep learning model includes: a sample determination module 810 and a training module 820 .

The sample determination module 810 is configured to obtain training samples, wherein the training samples include sample scene data and labels, the sample scene data includes data related to multiple sample lane change entrances, and the labels include positive sample labels and negative sample labels, and the positive samples The label is used to indicate the first target sample re-route entry, the first target sample re-entry entry includes a successfully imported sample re-route entry among the multiple sample re-entry entries, and the negative sample label is used to indicate the second target The sample lane change entrance, the second target sample lane change entrance includes a sample lane change entrance other than the first target sample lane change entrance among the plurality of sample lane change entrances.

The training module 820 is used for training the deep learning model by using the training samples to obtain the trained deep learning model.

According to an embodiment of the present disclosure, a sample determination module: a data acquisition unit, a data extraction unit.

The data acquisition unit is used to acquire initial sample scene data.

A data extraction unit, configured to extract feature data that meets predetermined extraction conditions from the initial sample scene data, as sample scene data.

According to an embodiment of the present disclosure, the feature data satisfying the predetermined extraction condition includes at least one of the following: obstacle data, vehicle data, environment data, and road traffic rule data related to lane change.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, an autonomous driving vehicle, and a computer program product.

According to an embodiment of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are processed by the at least one processor The processor executes to enable at least one processor to execute the method as an embodiment of the present disclosure.

According to an embodiment of the present disclosure, there is a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to perform a method according to an embodiment of the present disclosure.

According to an embodiment of the present disclosure, a computer program product includes a computer program that, when executed by a processor, implements a method as an embodiment of the present disclosure.

According to an embodiment of the present disclosure, an automatic driving vehicle configured with the above electronic device can implement the control method of the automatic driving vehicle described in the above embodiment when the configured electronic device is executed by its processor.

FIG. 9 shows a schematic block diagram of an example electronic device 900 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in FIG. 9 , the device 900 includes a computing unit 901 that can be executed according to a computer program stored in a read only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903 Various appropriate actions and handling. In the RAM 903, various programs and data necessary for the operation of the device 900 can also be stored. The computing unit 901 , the ROM 902 , and the RAM 903 are connected to each other through a bus 904 . An input/output (I/O) interface 905 is also connected to bus 904 .

Various components in the device 900 are connected to the I/O interface 905, including: an input unit 906, such as a keyboard, mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc. ; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 909 allows the device 900 to exchange information/data with other devices through a computer network such as the Internet and/or various telecommunication networks.

Computing unit 901 may be various general-purpose and/or special-purpose processing components with processing and computing capabilities. Some examples of computing units 901 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various specialized artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 executes the various methods and processes described above, such as a control method of an autonomous vehicle. For example, in some embodiments, a control method for an autonomous vehicle may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 908 . In some embodiments, part or all of the computer program may be loaded and/or installed on device 900 via ROM 902 and/or communication unit 909 . When the computer program is loaded into RAM 903 and executed by computing unit 901, one or more steps of the above-described control method for an autonomous vehicle may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the control method of the autonomous vehicle by any other suitable means (eg, by means of firmware).

Various implementations of the systems and techniques described herein above may be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips system (SOC), complex programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpretable on a programmable system including at least one programmable processor that The processor, which may be a special purpose or general-purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device an output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer or other programmable data processing apparatus, such that the program code, when executed by the processor or controller, performs the functions/functions specified in the flowcharts and/or block diagrams. Action is implemented. The program code may execute entirely on the machine, partly on the machine, partly on the machine and partly on a remote machine as a stand-alone software package or entirely on the remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in connection with the instruction execution system, apparatus or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), fiber optics, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device (eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user ); and a keyboard and pointing device (eg, a mouse or trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including acoustic input, voice input, or tactile input) to receive input from the user.

The systems and techniques described herein may be implemented on a computing system that includes back-end components (eg, as a data server), or a computing system that includes middleware components (eg, an application server), or a computing system that includes front-end components (eg, a user's computer having a graphical user interface or web browser through which a user may interact with implementations of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system may be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

A computer system can include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a distributed system server, or a server combined with blockchain.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, the steps described in the present disclosure can be executed in parallel, sequentially, or in different orders. As long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, there is no limitation herein.

The above-mentioned specific embodiments do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure should be included within the protection scope of the present disclosure.

Claims (26)

1. A control method for an autonomous vehicle, comprising: In response to receiving the lane change instruction, determining a first target lane change entry from a plurality of lane change entrances based on target scene data, wherein the target scene data includes data related to the plurality of lane change entrances ; determining a planned lane change path based on the first target lane change entrance; and The vehicle is controlled to change lanes according to the planned lane change route. 2. The method of claim 1, further comprising, before said determining a first target lane change entrance from a plurality of lane change entrances based on the target scene data: Feature data satisfying predetermined extraction conditions is extracted from the traffic scene data as the target scene data. 3. The method of claim 1, further comprising, before said determining a planned lane change path based on the first target lane change entrance: In the case that it is determined that the target scene data does not meet the predetermined lane change condition, the driving speed is adjusted so that the adjusted target scene data meets the predetermined lane change condition. 4. The method according to claim 3, further comprising, in the case of adjusting the driving speed in the case of determining that the target scene data does not meet the predetermined lane change condition: determining current scene data, wherein the current scene data includes target scene data in the process of adjusting the driving speed; and If it is determined that the current scene data satisfies a predetermined lane change cancellation condition, the execution of the operation of adjusting the driving speed is cancelled, and the lane change is cancelled. 5. The method according to claim 4, further comprising, in the case of adjusting the driving speed in the case of determining that the lane change scene data does not meet the predetermined lane change condition: determining a second target lane change entrance from the plurality of lane change entrances based on the current scene data; Based on the first target lane-change entrance, a first lane-change cost is determined, wherein the first lane-change cost is used to represent: the cost of changing lanes according to the first target lane-change entrance; Based on the second target lane-change entrance, a second lane-change cost is determined, wherein the second lane-change cost is used to represent: the cost of changing lanes according to the second target lane-change entrance; and Based on the first lane change cost value and the second lane change cost value, a new target lane change entrance is determined from the first target lane change entrance and the second target lane change entrance so as to be based on the Describe the new target lane change entrance, and determine the updated lane change planning path. 6 . The method of claim 3 , wherein the predetermined lane change condition comprises a condition for completing the lane change within a predetermined period of time at a safe driving speed. 7 . 7. The method according to claim 2, wherein the feature data satisfying a predetermined extraction condition comprises at least one of the following: Data of obstacles, vehicle data, environmental data, and road traffic rules data related to lane changes. 8 . The method according to claim 3 , wherein, in the case where it is determined that the target scene data does not meet a predetermined lane change condition, adjusting the driving speed comprises: 8 . determining the target acceleration for adjusting the driving speed in the case that it is determined that the target scene data does not meet the predetermined lane change condition; When it is determined that the target acceleration satisfies a predetermined adjustment speed condition, the travel speed is adjusted according to the target driving degree. 9. A training method for a deep learning model, comprising: Determine a training sample, wherein the training sample includes sample scene data and a label, the sample scene data includes data related to a plurality of sample lane change entrances, the label includes a positive sample label and a negative sample label, the positive sample label The sample label is used to indicate a first target sample change lane entry, the first target sample lane change entrance includes a successfully imported sample lane change entrance among the plurality of sample lane change entrances, the negative sample The label is used to indicate a second target sample lane change entry, and the second target sample lane change entry includes sample changes other than the first target sample lane change entry among the plurality of sample lane change entrances. Gateway entrance; and The deep learning model is trained by using the training samples to obtain the trained deep learning model. 10. The method of claim 9, wherein the determining training samples comprises: obtain initial sample scene data; and Feature data satisfying a predetermined extraction condition is extracted from the initial sample scene data as the sample scene data. 11. The method according to claim 10, wherein the feature data satisfying a predetermined extraction condition comprises at least one of the following: Obstacle data, vehicle data, environmental data, and road traffic rules data related to lane changes. 12. A control device for an autonomous vehicle, comprising: The first determination module is configured to, in response to receiving a lane change instruction, determine a first target lane change entrance from a plurality of lane change entrances based on target scene data, wherein the target scene data includes a combination of the plurality of lane change entrances. Data related to lane change entrances; a second determining module, configured to determine a planned lane change path based on the first target lane change entrance; and The driving module is used to control the vehicle to change lanes according to the planned lane change path. 13. The apparatus of claim 12, further comprising, before the first determination module: The extraction module is used for extracting feature data satisfying predetermined extraction conditions from the traffic scene data as the target scene data. 14. The apparatus of claim 12, further comprising, before the second determination module: An adjustment module, configured to adjust the driving speed when it is determined that the target scene data does not meet the predetermined lane change condition, so that the adjusted target scene data meets the predetermined lane change condition. 15. The apparatus according to claim 14, further comprising, in the case of adjusting the driving speed in the case of determining that the target scene data does not meet the predetermined lane change condition: a third determining module, configured to determine current scene data, wherein the current scene data includes target scene data in the process of adjusting the driving speed; and The canceling module is configured to cancel executing the operation of adjusting the driving speed and cancel the lane change when it is determined that the current scene data satisfies the predetermined canceling lane change condition. 16. The apparatus according to claim 15, further comprising, in the case of adjusting the travel speed in the case of determining that the target scene data does not meet the predetermined lane change condition: a fourth determination module, configured to determine a second target lane change entrance from a plurality of lane change entrances based on the current scene data; a fifth determination module, configured to determine a first lane-change cost based on the first target lane-change entry, wherein the first lane-change cost is used to represent: changing lanes according to the first target lane-change entry the cost of driving; A sixth determination module, configured to determine a second lane-change cost based on the second target lane-change entry, wherein the second lane-change cost is used to represent: changing lanes according to the second target lane-change entry the cost of driving; and an update module for determining a new target lane change entry from the first target lane change entry and the second target lane change entry based on the first lane change cost and the second lane change cost entry, so as to determine an updated planned lane change path based on the new target lane change entry entry. 17. The apparatus of claim 14, wherein the predetermined lane change condition includes a condition for completing the lane change within a predetermined period of time at a safe driving speed. 18. The apparatus according to claim 13, wherein the feature data satisfying a predetermined extraction condition comprises at least one of the following: Data of obstacles, vehicle data, environmental data, and road traffic rules data related to lane changes. 19. The apparatus of claim 14, wherein the adjustment module comprises: an acceleration adjustment unit, configured to determine a target acceleration for adjusting the driving speed when it is determined that the target scene data does not meet a predetermined lane change condition; A speed adjustment unit, configured to adjust the travel speed according to the target driving degree when it is determined that the target acceleration satisfies a predetermined speed adjustment condition. 20. A training device for a deep learning model, comprising: A sample determination module for acquiring training samples, wherein the training samples include sample scene data and labels, the sample scene data includes data related to a plurality of sample lane change entrances, and the labels include positive sample labels and negative sample labels a sample label, where the positive sample label is used to indicate a first target sample change lane entry, and the first target sample lane change entry includes a successfully imported sample lane change sink among the plurality of sample lane change entrances an entry, the negative sample label is used to indicate a second target sample lane change entry, and the second target sample lane change entry includes the first target sample lane change among the plurality of sample lane changes Sample re-route entrances other than sink entrances; and A training module is used to train a deep learning model by using the training samples to obtain a trained deep learning model. 21. The apparatus of claim 20, wherein the sample determination module: a data acquisition unit for acquiring initial sample scene data; and A data extraction unit, configured to extract feature data that satisfies a predetermined extraction condition from the initial sample scene data, as the sample scene data. 22. The apparatus according to claim 21, wherein the feature data satisfying a predetermined extraction condition comprises at least one of the following: Obstacle data, vehicle data, environmental data, and road traffic rules data related to lane changes. 23. An electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the execution of any one of claims 1 to 11 Methods. 24. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1 to 11. 25. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1 to 11. 26. An autonomous vehicle comprising: the electronic device of claim 23.

Images