CN112519765A - Vehicle control method, apparatus, device, and medium - Google Patents

Abstract

The embodiments of the present application disclose a vehicle control method, device, device and medium, which relate to automatic driving technology in the field of computer technology. Wherein, the vehicle control method includes: acquiring the motion trajectory information of the dynamic obstacles identified during the driving process of the vehicle; inputting the motion trajectory information into the trained machine learning model to predict the motion intention of the dynamic obstacles; The predicted motion intention controls the route planning of the vehicle or the driving state of the vehicle. In the embodiment of the present application, the motion trajectory of the dynamic obstacle is identified by the machine learning model, so as to determine the motion intention of the dynamic obstacle, and then the vehicle is controlled according to the motion intention, thereby improving the accuracy of the vehicle control.

Description

Vehicle control method, device, device and medium

technical field

The embodiments of the present application relate to computer technologies, in particular to automatic driving technologies, and in particular to a vehicle control method, apparatus, device, and medium.

Background technique

In autonomous driving technology, the autonomous driving system of the vehicle will perform path planning and control the driving state of the vehicle according to the surrounding conditions of the vehicle.

When the vehicle is driving in the lane, it needs to avoid the dynamic obstacles that appear in the lane. The dynamic obstacle may be a vehicle switching lanes, an overtaking vehicle, a jammed vehicle, a pedestrian crossing the road, or other situations. The motion of these dynamic obstacles is small and may have tentative reciprocating motion. Therefore, it is difficult for the automatic driving system to determine the motion intention of dynamic obstacles through the identification of long and stable motion trajectories, which causes difficulties in the path planning and driving state control of the vehicle.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a vehicle control method, device, device, and medium, which improve the accuracy of vehicle control.

In a first aspect, an embodiment of the present application provides a vehicle control method, the method comprising:

acquiring the motion trajectory information of the dynamic obstacles identified during the driving of the vehicle;

inputting the motion trajectory information into a trained machine learning model to predict the motion intent of the dynamic obstacle;

According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle is controlled.

In the embodiment of the present application, the motion trajectory information of the dynamic obstacle identified during the driving of the vehicle is obtained, and the motion trajectory information of the dynamic obstacle is input into the trained machine learning model, so as to predict the motion intention of the dynamic obstacle, and then According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle are controlled. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of dynamic obstacles through long and stable motion trajectories, which causes difficulties in the path planning and driving state control of vehicles, and realizes the recognition of the motion trajectories of dynamic obstacles through machine learning models. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

In addition, the vehicle control method according to the above embodiment of the present application may also have the following additional technical features:

Optionally, the motion trajectory sample set required in the training process of the machine learning model is a sample set marked with motion intent, and the acquisition process of the motion trajectory sample set specifically includes:

Obtain the motion trajectory information of the dynamic obstacles collected during the vehicle driving in the lane;

According to the set intention phenomenon rule, determine the movement intention from the movement track information of the dynamic obstacle;

Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample.

An embodiment in the above application has the following advantages or beneficial effects: by determining the motion intention of the dynamic obstacle according to the set intention phenomenon rules, and forming the motion intention sample from the binary set of the motion trajectory information and the motion intention, the improved performance is improved. Analyzing the speed of the motion trajectories of dynamic obstacles helps to obtain motion trajectories samples.

Optionally, according to the set intention phenomenon rule, determining the movement intention from the movement trajectory information of the dynamic obstacle includes at least one of the following:

If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as crossing the lane, determining that the motion is intended to be a pedestrian crossing the lane;

If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as going straight along the same edge of the lane, determining that the motion intention is that the pedestrian goes straight along the lane;

If the dynamic obstacle is identified as a vehicle and the motion trajectory information is identified as entering the current lane from an adjacent lane and is in front of the current vehicle, then it is determined that the motion is intended to switch lanes or overtake.

An embodiment in the above application has the following advantages or beneficial effects: by determining the motion intent of the dynamic obstacle according to the signs indicating the motion intent appearing in the motion trajectory of the dynamic obstacle, the complexity of determining the motion intent of the dynamic obstacle is simplified.

Optionally, the acquisition process of the required motion intent sample set in the training process of the machine learning model specifically includes:

Obtain the motion trajectory information of the dynamic obstacles collected during the vehicle driving in the lane;

obtaining the stable driving control instruction of the vehicle with respect to the motion trajectory information of the dynamic obstacle, as the motion intention;

Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample.

An embodiment in the above application has the following advantages or beneficial effects: by taking the stable driving control instruction of the vehicle for the motion trajectory information of the dynamic obstacle as the motion intent, and forming the motion intent sample from the binary set of the motion trajectory information and the motion intent , which provides favorable conditions for obtaining motion intent sample sets.

Optionally, obtaining the stable driving control instruction of the vehicle with respect to the motion trajectory information of the dynamic obstacle, as the motion intention includes:

obtaining at least one driving control command for the vehicle in response to the dynamic obstacle entering the driving influence range of the vehicle;

In response to the dynamic obstacle being out of the driving influence range of the vehicle, selecting a last driving control command from the at least one driving control command as the stable driving control command;

The motion intention is acquired based on the stable driving control instruction.

An embodiment in the above application has the following advantages or beneficial effects: by acquiring the stable driving control instruction from at least one driving control instruction, and obtaining the motion intention based on the stable driving control instruction, it can provide reference meaning for the subsequent dynamic obstacle motion intention .

Optionally, inputting the motion trajectory information into a trained machine learning model to predict the motion intention of the dynamic obstacle includes:

Inputting the motion trajectory information into the machine learning model to identify motion intent;

If the identified motion intent result is uncertain according to the probability value of the motion intent recognition result, return to the collection operation of motion trajectory information, and superimpose the newly collected motion trajectory information into the historically collected motion trajectory information for input the machine learning model until the motion intent of the dynamic obstacle is determined.

An embodiment in the above application has the following advantages or beneficial effects: when the motion intention of the dynamic obstacle cannot be accurately determined according to the collected motion trajectory information, new motion trajectory information is acquired, and the newly collected motion trajectory information and history are combined. The motion trajectory information is used for re-motion intention recognition to improve the motion intention recognition accuracy of dynamic obstacles.

Optionally, the motion intent is obtained by labeling the motion trajectory samples after clustering in the motion trajectory sample set that does not contain the motion intent.

An embodiment in the above application has the following advantages or beneficial effects: by clustering the motion trajectory samples, and labeling each category after the clustering, the motion intention of the motion trajectory samples is obtained.

Optionally, the data format of the motion track information is to use at least one track point recorded sequentially to represent the motion track information, and each track point records the speed of the dynamic obstacle moving along the lane line and the direction perpendicular to the lane line. A 2-tuple of the velocity of the movement.

An embodiment in the above application has the following advantages or beneficial effects: recording the moving speed of the dynamic obstacle along the direction of the lane line and the speed of moving perpendicular to the direction of the lane line through each track point, providing reliable information for subsequent determination of the moving intention of the dynamic obstacle. condition.

In a second aspect, an embodiment of the present application further provides a vehicle control device, the device comprising:

an acquisition motion trajectory module for acquiring motion trajectory information of dynamic obstacles identified during the driving process of the vehicle;

A motion intention prediction module for inputting the motion trajectory information into the trained machine learning model to predict the motion intention of the dynamic obstacle;

A vehicle control module, configured to control the route planning of the vehicle or the driving state of the vehicle according to the predicted motion intention.

In the embodiment of the present application, the motion trajectory information of the dynamic obstacle identified during the driving of the vehicle is obtained, and the motion trajectory information of the dynamic obstacle is input into the trained machine learning model, so as to predict the motion intention of the dynamic obstacle, and then According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle are controlled. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of dynamic obstacles through long and stable motion trajectories, which causes difficulties in the path planning and driving state control of vehicles, and realizes the recognition of the motion trajectories of dynamic obstacles through machine learning models. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

In a third aspect, an embodiment of the present application also provides an electronic device, the 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 vehicle control method of any embodiment.

In the embodiment of the present application, the motion trajectory information of the dynamic obstacle identified during the driving of the vehicle is obtained, and the motion trajectory information of the dynamic obstacle is input into the trained machine learning model, so as to predict the motion intention of the dynamic obstacle, and then According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle are controlled. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of the dynamic obstacle through a long and stable motion trajectory, which causes difficulties in the path planning and driving state control of the vehicle, and realizes the recognition of the motion trajectory of the dynamic obstacle through the machine learning model. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

In a fourth aspect, the embodiments of the present application further provide a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to cause the computer to execute the vehicle control method described in any of the embodiments.

In the embodiment of the present application, the motion trajectory information of the dynamic obstacle identified during the driving of the vehicle is obtained, and the motion trajectory information of the dynamic obstacle is input into the trained machine learning model, so as to predict the motion intention of the dynamic obstacle, and then According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle are controlled. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of dynamic obstacles through long and stable motion trajectories, which causes difficulties in the path planning and driving state control of vehicles, and realizes the recognition of the motion trajectories of dynamic obstacles through machine learning models. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

Other effects of the above-mentioned optional manners of the present application will be described below with reference to specific embodiments.

Description of drawings

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

1 is a schematic flowchart of a vehicle control method provided in Embodiment 1 of the present application;

2 is a schematic flowchart of a machine learning model training process provided in Embodiment 2 of the present application;

3 is a schematic flowchart of another machine learning model training process provided in Embodiment 2 of the present application;

4 is a schematic flowchart of another machine learning model training process provided in Embodiment 2 of the present application;

5 is a schematic flowchart of another vehicle control method provided in Embodiment 3 of the present application;

6 is a schematic structural diagram of a vehicle control device provided in Embodiment 4 of the present application;

FIG. 7 is a schematic structural diagram of an electronic device according to Embodiment 5 of the present application.

Detailed ways

Exemplary embodiments of the present application are described below with reference to the accompanying drawings, which include various details of the embodiments of the present application 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 application. Also, descriptions of well-known functions and constructions are omitted from the following description for clarity and conciseness.

Aiming at the problem in the related art that it is difficult for an automatic driving system to determine the motion intention of a dynamic obstacle through the identification of a long and stable motion trajectory, which causes difficulties in the path planning and driving state control of the vehicle, a vehicle control system is proposed. method.

In the embodiment of the present application, by acquiring the motion trajectory information of the dynamic obstacle identified during the driving of the vehicle, and inputting the motion trajectory information of the dynamic obstacle into the trained machine learning model, the motion intention of the dynamic obstacle is predicted, Then, according to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle are controlled. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of the dynamic obstacle through a long and stable motion trajectory, which causes difficulties in the path planning and driving state control of the vehicle, and realizes the recognition of the motion trajectory of the dynamic obstacle through the machine learning model. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

The following describes a vehicle control method, device, device, and medium according to the embodiments of the present application with reference to the accompanying drawings for detailed description.

Example 1

FIG. 1 is a schematic flowchart of a vehicle control method provided in Embodiment 1 of the present application. The embodiment of the present application can be applied to a scenario in which a vehicle is controlled based on the motion intention of a dynamic obstacle identified on the way of the vehicle. It is executed by the vehicle control device to realize the control of the path planning or the driving state of the vehicle, and the device can be realized by software and/hardware, and can be integrated into the interior of the electronic device. In this embodiment, the electronic device may be any hardware device with a data processing function, such as an on-board computer, an intelligent pilot and the like. The method specifically includes the following steps:

S101 , acquiring motion track information of a dynamic obstacle identified during the driving of the vehicle.

Wherein, in this embodiment, the dynamic obstacles may include vehicles, pedestrians and others. Others can be animals such as cats and dogs, or objects such as carriages.

Usually, the dynamic obstacle is in a state of motion, that is, the position of the dynamic obstacle is different at different times. In this regard, in this embodiment, a camera set on the current vehicle can collect continuous multiple frames of images including dynamic obstacles, and analyze the position of the dynamic obstacles in each frame of the above-mentioned multiple frames of images, so as to obtain the dynamic obstacles. Movement track information. Alternatively, the position of dynamic obstacles can be obtained by collecting laser point cloud data through lidar, and the motion trajectory information can be obtained by collecting multiple consecutive frames. This embodiment of the present application does not limit the acquisition method of the motion trajectory information.

Wherein, the setting position of the camera can be adaptively set according to actual application needs, which is not specifically limited here. For example, a camera may be provided at the front and rear of the current vehicle, respectively; or, a camera may also be provided on the roof of the current vehicle, and so on.

It should be noted that, if a camera is set on the roof of the current vehicle, the camera can be set as a 360-degree panoramic camera in this embodiment, so as to obtain motion trajectory information of dynamic obstacles in different directions.

Optionally, the motion track information of the dynamic obstacle may be formed by a large number of track points. The data format of the motion track information of the dynamic obstacle in this embodiment is that the motion track information is represented by at least one track point recorded sequentially, and each track A 2-tuple that records the velocity of the dynamic obstacle in the direction of the lane line and the velocity in the direction perpendicular to the lane line.

Wherein, the trajectory point may be the position point of the dynamic obstacle at different times.

Since the movement of the dynamic obstacle has a time attribute, in this embodiment, multiple position points of the dynamic obstacle can be recorded sequentially according to the time attribute, and the movement track information of the dynamic obstacle can be formed by the recorded multiple position points.

Correspondingly, the velocity of the dynamic obstacle moving along the lane line and the velocity perpendicular to the lane line recorded by each track point can be calculated and obtained according to the position information of the dynamic obstacle in the collection time period.

S102: Input the motion trajectory information into a trained machine learning model to predict the motion intention of the dynamic obstacle.

In this embodiment, the motion intention of the dynamic obstacle may be determined according to the type of the dynamic obstacle.

For example, if the dynamic obstacle is a vehicle, the motion intention of the vehicle may include: changing lanes, accelerating, decelerating, overtaking, jamming, and emergency stop. If the dynamic obstacle is a pedestrian, the motion intention of the pedestrian may include: going straight, cutting into the lane, crossing the lane, starting, accelerating, and stopping suddenly.

Optionally, in this embodiment, the motion trajectory information of the dynamic obstacle may be used as input data into a trained machine learning model, and the motion trajectory of the dynamic obstacle may be intentionally identified by the machine learning model to predict the dynamic obstacle. The motion intent of the obstacle.

It should be noted that the specific process of training the machine learning model in this embodiment will be described in detail in the following example, and will not be described in detail here.

S103, according to the predicted motion intention, control the route planning of the vehicle or the driving state of the vehicle.

Wherein, controlling the route planning of the vehicle refers to re-planning the traveling route of the vehicle; controlling the traveling state of the vehicle may refer to reducing the traveling speed of the vehicle, etc., which are not specifically limited here.

In this embodiment, if according to the motion intention of the dynamic obstacle, it is determined that the dynamic obstacle will appear within the safe range of the current vehicle's trajectory, then the current vehicle is subjected to path planning or driving speed control;

If, according to the motion intention of the dynamic obstacle, it is determined that the dynamic obstacle will not appear within the safe range of the current vehicle running track, the current driving state of the current vehicle is maintained.

Among them, the so-called safety range is the range that may affect the safe driving of the vehicle. Within the safety range, the vehicle needs to avoid obstacles, detour, monitor and other processing. The safe range can be determined according to the current vehicle speed. For example, if the current vehicle speed is 80 kilometers per hour (km/h), the safe range of the current vehicle trajectory can be 80 meters (m); for another example, if the current vehicle speed is 100km/h, the current The safe range of the vehicle's trajectory is 100m. That is, the faster the vehicle travels, the larger the required safety margin, and vice versa.

That is to say, in the embodiment of the present application, by determining the motion intention of the dynamic obstacle and comparing it with the safety range of the current vehicle's trajectory, if the dynamic obstacle appears within the safety range of the current vehicle's trajectory, it means that the current vehicle is in accordance with Continuing to drive along the current driving path in the current driving state may collide with dynamic obstacles. At this time, the current driving path of the vehicle needs to be re-planned to achieve deceleration and/or detour to avoid collision with dynamic obstacles; if dynamic obstacles do not Appears within the safe range of the current vehicle's running track, which means that the current vehicle continues to drive according to the current driving state and will not collide with dynamic obstacles. At this time, there is no need to adjust the current driving state of the vehicle, so as to realize the intelligent driving of the vehicle. Controls to avoid situations where the vehicle snakes around dynamic obstacles.

The vehicle control method provided by the embodiment of the present application predicts the dynamic obstacle by acquiring the motion trajectory information of the dynamic obstacle identified during the driving process of the vehicle, and inputting the motion trajectory information of the dynamic obstacle into the trained machine learning model. The motion intention of the object is then controlled according to the predicted motion intention to control the route planning of the vehicle or the driving state of the vehicle. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of dynamic obstacles through long and stable motion trajectories, which causes difficulties in the path planning and driving state control of vehicles, and realizes the recognition of the motion trajectories of dynamic obstacles through machine learning models. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved. In the automatic driving system, it is necessary to obtain the conditions of the surrounding environment in real time, and adjust the driving path and speed of the vehicle accordingly, so as to achieve safe driving. However, the motion trajectories of some dynamic obstacles can quickly indicate the intention, that is, the motion intention of the dynamic obstacles can be identified within a few frames, so as to make corresponding vehicle control reflections. However, the motion trajectory of some dynamic obstacles is not stable, and they are in a tentative state for a relatively long period of time, making it difficult to identify the motion intention. Typically, other vehicles and pedestrians are also observing the surrounding environment to test whether it is possible to switch lanes or cross lanes. In this case, it is difficult for an autonomous vehicle to make a correct judgment and plan its own path and speed in a short and tentative motion trajectory. If conservative and safe processing methods are adopted at this time, such as reducing the speed, driving safety can be ensured, but it will greatly interfere with the normal driving speed of the vehicle, and may frequently require waiting for other vehicles and pedestrians. Therefore, the above-mentioned technical solutions are proposed in the embodiments of the present application, which can learn a large number of exploratory and unclear motion trajectory sample data with the help of a machine learning model, so that the motion intention can be predicted by using as few motion trajectory points as possible, so as to make out the response control of the vehicle.

Embodiment 2

The following describes in detail the process of training the machine learning model in the embodiments of the present application with reference to FIGS. 2 to 4 .

It should be noted that, in the embodiment of the present application, the motion trajectory samples required in the training process of the machine learning model can be divided into: a sample set marked with motion intention and a sample set without motion intention marked.

First, with reference to FIGS. 2 to 3 , when the motion trajectory sample set required in the training process of the machine learning model in the embodiment of the present application is the sample set marked with motion intention, according to the sample set marked with motion intention, the machine learning model The training process is described in detail.

FIG. 2 is a schematic flowchart of a training process of a machine learning model according to Embodiment 2 of the present application. The machine learning model training process in the embodiment of the present application specifically includes the following steps:

S201 , acquiring the motion trajectory information of the dynamic obstacles collected during the vehicle running in the lane.

Wherein, the implementation process and principle of S201 in this embodiment are similar to the acquisition of the motion track information of the dynamic obstacle in the first embodiment, which will not be repeated here. For the specific process, refer to S101 in the first embodiment.

S202, according to the set intention phenomenon rule, determine the movement intention from the movement track information of the dynamic obstacle.

Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample.

In this embodiment of the present application, the setting intention phenomenon rule may be set according to actual application needs, which is not specifically limited here.

The rule of setting intention phenomenon is that there are signs of motion intention in the motion trajectory. When acquiring training samples, the motion intention can be waited for by collecting the motion trajectory for a long time. From the massively collected motion trajectory data, samples that show clear motion intentions can be screened out.

Exemplarily, in this embodiment, according to the set intention phenomenon rule, determining the movement intention from the movement track information of the dynamic obstacle includes at least one of the following:

If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as crossing the lane, determining that the motion is intended to be a pedestrian crossing the lane;

If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as going straight along the same edge of the lane, determining that the motion intention is that the pedestrian goes straight along the lane;

If the dynamic obstacle is identified as a vehicle and the motion trajectory information is identified as entering the current lane from an adjacent lane and being in front of the current vehicle, it is determined that the motion is intended to be lane switching, overtaking or jamming.

If the dynamic obstacle is identified as a pedestrian and it is identified that the motion trajectory information suddenly ends at any position in the lane, determining that the motion intention is a pedestrian emergency stop;

If it is recognized that the dynamic obstacle is a pedestrian and that the motion trajectory information changes suddenly after a period of time, determining that the motion intention is a pedestrian starting;

If it is recognized that the dynamic obstacle is a pedestrian and the time from the previous position to the current position of the motion trajectory information is recognized to be shortened, it is determined that the motion intention is pedestrian acceleration;

If the dynamic obstacle is identified as a vehicle and it is identified that the time from the last position to the current position of the motion trajectory information is shortened, determining that the motion intention is vehicle acceleration;

If the dynamic obstacle is identified as a vehicle and the time from the last position to the current position of the motion trajectory information is identified to be longer, determining that the motion intention is vehicle deceleration;

If the dynamic obstacle is identified as a vehicle and it is identified that the motion trajectory information ends abruptly at any position, it is determined that the motion intention is an emergency stop of the vehicle.

Further, after the motion intention is determined from the motion trajectory information of the dynamic obstacle, in this embodiment of the present application, a binary combination of the motion trajectory information of the dynamic obstacle and the determined motion intention can be combined into a motion intention sample, so as to obtain a motion intention sample marked with motion. A sample set of intents.

Wherein, in this embodiment, the motion trajectory information is formed by a plurality of trajectory point sequences, and the motion intention can be determined according to the type of dynamic obstacles. For example, if the dynamic obstacle is a pedestrian, the motion intention may include the pedestrian crossing the lane and the pedestrian going straight along the lane; for another example, if the dynamic obstacle is a vehicle, the motion intention may include vehicle acceleration, vehicle overtaking, etc.

For example, if the dynamic obstacle is a pedestrian, and the obtained motion trajectory information is to move from point A to point B, and from point B to point C, and the movement intention is to cross the lane, you can pass [A-B-C, cross Road] to form a sample of pedestrian motion intent.

S203, using the acquired motion trajectory sample set marked with motion intention to train the machine learning model.

That is to say, in this embodiment of the present application, the motion trajectory samples are input into the machine learning model, and the motion intention corresponding to the above motion trajectory samples is used as the training result, and the machine learning model is repeatedly trained until the motion trajectory samples are input. Corresponds to the motion intention. At this point, this model is taken as the final machine learning model.

FIG. 3 is a schematic flowchart of another machine learning model training process provided in Embodiment 2 of the present application. The machine learning model training process in the embodiment of the present application specifically includes the following steps:

S301 , acquiring the motion trajectory information of the dynamic obstacle collected during the vehicle running in the lane.

Wherein, the implementation process and principle of S301 are similar to the acquisition of the motion trajectory information of the dynamic obstacle in the first embodiment, which will not be repeated here. For the specific process, refer to S101 in the first embodiment.

S302 , obtaining a stable driving control instruction of the vehicle with respect to the motion trajectory information of the dynamic obstacle, as the motion intention.

Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample.

In the embodiment of the present application, the stable driving control instruction refers to an instruction for causing the dynamic obstacle to escape the driving influence range of the current vehicle. The driving influence range of the vehicle can be the same or different from the safety range. The driving influence range of the vehicle may refer to the range that may affect the safe driving of the vehicle.

Optionally, in this embodiment, the stable driving control instruction for obtaining the motion trajectory information of the vehicle for the dynamic obstacle is obtained as the motion intention including:

obtaining at least one driving control command for the vehicle in response to the dynamic obstacle entering within the driving influence range of the vehicle;

In response to the dynamic obstacle being out of the driving influence range of the vehicle, selecting a last driving control command from the at least one driving control command as the stable driving control command;

The motion intention is acquired based on the stable driving control instruction.

In this embodiment, the driving control instruction is an instruction that is adaptively taken based on the motion trajectory information of the dynamic obstacle. Since the motion trajectory information may be tentative, the control instruction may also be a series of instructions. That is, the control command can reflect the motion intention of the dynamic obstacle. For example, if the dynamic obstacle is a pedestrian, and the pedestrian wants to wait until there are few vehicles to cross the lane, he may approach the lane at the side of the lane for a while, and then look away from the lane for a while, and a tentative motion trajectory will appear. When the autonomous vehicle recognizes that the pedestrian is approaching the lane, it should generally give a control command to reduce the speed, but it can accelerate when the pedestrian is away from the lane. This series is in response to the pedestrian's tentative driving command. Until the pedestrian actually crosses the lane, the vehicle makes the last control command to slow down or stop; or, if the pedestrian has not crossed the lane and is overtaken by the vehicle, then the last control command of the vehicle is to accelerate or drive at normal speed. The last control command can be used as a stable driving control command in response to the pedestrian trajectory information, which has reference significance and can be used as a training sample. In the subsequent prediction of the motion intention, if the pedestrian has a similar motion trajectory again, the corresponding stable driving control command can be used to respond to the motion intention of the pedestrian.

S303, using the obtained motion trajectory sample set marked with motion intention to train the machine learning model.

The implementation process and principle of S303 are the same as or similar to those of S203 in the second embodiment, and the specific process may refer to S203, which will not be repeated here.

In another implementation scenario of the present application, in this embodiment, a sample set of manually annotating motion trajectories of dynamic obstacles can also be directly obtained from a database. Then, according to the above-mentioned motion trajectory sample set, the machine learning model is trained.

Further, with reference to FIG. 4 , when the motion trajectory sample set required in the training process of the machine learning model in the embodiment of the present application is a sample set without motion intent, the machine learning model is analyzed according to the sample set without motion intent. The training process is described in detail.

FIG. 4 is a schematic flowchart of a training process of another machine learning model provided in Embodiment 2 of the present application. The training process of the machine learning model in the embodiment of the present application specifically includes the following steps:

S401 , acquiring the motion trajectory information of the dynamic obstacle collected during the vehicle running in the lane.

Wherein, the implementation process and principle of S401 are similar to the acquisition of the motion trajectory information of the dynamic obstacle in the first embodiment, which will not be repeated here. For the specific process, refer to S101 in the first embodiment.

S402: After clustering the motion trajectory information of the dynamic obstacles, perform intention labeling to obtain a motion trajectory sample set marked with motion intentions.

In this embodiment, different categories can be obtained by clustering the motion trajectories of the dynamic obstacles. Then, the different categories obtained by clustering are labeled with motion intentions by manual labeling, and the motion trajectory sample set is obtained.

S403, using the acquired motion trajectory sample set marked with motion intention to train the machine learning model.

The implementation process and principle of S403 are the same as or similar to those of S203 in the second embodiment, and the specific process may refer to S203, which will not be repeated here.

That is to say, in the embodiment of the present application, the different motion trajectory categories obtained by clustering are labeled with motion intentions by manual labeling, so as to obtain a motion trajectory sample set, so as to train a machine learning model according to the motion trajectory sample set.

Embodiment 3

It can be seen from the above analysis that in the embodiment of the present application, the motion trajectory information of the dynamic obstacle is input into the trained machine learning model to predict the motion intention of the dynamic obstacle, and according to the predicted motion intention, the vehicle's route planning or the vehicle's movement intention are controlled. driving status.

In another implementation scenario of the present application, when the motion trajectory information is input into the trained machine learning model for motion intent recognition, if the motion intent result cannot be recognized according to the probability value of the motion intent recognition result, in this embodiment, the The new motion trajectory information and the newly collected motion trajectory information are superimposed on the historically collected motion trajectory information for input into the machine learning model until the motion intention of the dynamic obstacle is determined. The above situation of the vehicle control method according to the embodiment of the present application will be described in detail below with reference to FIG. 5 .

FIG. 5 is a schematic flowchart of another vehicle control method provided in Embodiment 3 of the present application. As shown in Figure 5, the vehicle control method includes the following steps:

S501 , acquiring motion trajectory information of dynamic obstacles identified during the driving of the vehicle.

S502: Input the motion trajectory information into the machine learning model to identify motion intentions.

S503, if according to the probability value of the motion intention recognition result, it is recognized that the motion intention result is uncertain, then return to perform the collection operation of motion trajectory information, and superimpose the newly collected motion trajectory information into the historically collected motion trajectory information for use in S503. The machine learning model is input until the motion intent of the dynamic obstacle is determined.

In this embodiment, during the training process, various motion intentions can be marked for each motion trajectory information as samples, and the machine learning model can be trained by using the motion trajectory sample set marked with motion intentions, so that the machine learning model can recognize the motion trajectory information in the training process. When , the probability values of the motion trajectory information corresponding to various motion intentions are also obtained accordingly. If any probability value of multiple motion intentions exceeds the preset threshold, the motion intention exceeding the preset threshold is determined as the motion intention of the motion trajectory information; if the probability values of multiple motion intentions do not exceed the preset threshold, then It shows that the motion intention determined according to the motion trajectory information of the dynamic obstacle has a large error and cannot be trusted, that is, the motion intention result is uncertain.

For example, the motion intention of the dynamic obstacle identified by the machine learning model includes: crossing the lane and going straight along the lane, and the probability value of crossing the lane is 30%, and the probability value of going straight along the lane is 50%, if the preset threshold is 70%, then it can be determined that neither of the above two intention probability values exceeds 70%, and the motion intention of the dynamic obstacle cannot be determined at this time.

Further, in this embodiment, when it is determined that the trajectory points in the motion trajectory information are less than the preset number value, or the speed of the trajectory points is lower than the preset speed value, it means that the data for predicting the motion intention is not representative, and the It can be recognized that the motion intent result is indeterminate.

The preset quantity value may be adaptively set according to actual application needs, and is not specifically limited here. For example, the preset number value is set to 1 or 2.

The preset speed thresholds are set according to the dynamic obstacle type. That is to say, when the dynamic obstacles are different, the corresponding preset speed thresholds are also different.

Exemplarily, when the acquired motion trajectory information of the dynamic obstacle is input into the trained machine learning model to recognize the motion intention, if the result of the motion intention is identified as uncertain, it means that according to the collected motion trajectory information at this time. The motion intent of dynamic obstacles cannot be accurately determined. To this end, this embodiment can return to the motion trajectory information collection operation, and accumulate the motion trajectory information newly collected in the next frame with the historically collected multi-frame motion trajectory information, and then input the accumulated motion trajectory information into the trained machine. In the learning model, the recognition of motion intention is carried out. As the motion trajectories are further increased, it is more likely to recognize motion intent. If the motion intention of the motion track is recognized, end; otherwise, continue to return to the motion track information collection operation and the motion track information superposition operation, and input the superimposed motion track information into the trained machine learning model until the obtained Motion intent for dynamic obstacles.

S504, according to the predicted motion intention, control the route planning of the vehicle or the driving state of the vehicle.

That is to say, when the motion intention of the dynamic obstacle cannot be accurately determined according to the collected motion track information in this embodiment, the motion intention recognition is performed again by acquiring new motion track information and combining the newly collected motion track information with the historical motion track information. , which can improve the accuracy of motion intention recognition of dynamic obstacles.

Embodiment 4

In order to achieve the above purpose, a fourth embodiment of the present application proposes a vehicle control device. FIG. 6 is a schematic structural diagram of a vehicle control device provided in Embodiment 4 of the present application.

As shown in FIG. 6 , the vehicle control apparatus according to the embodiment of the present application includes: a movement trajectory acquisition module 610 , a movement intention prediction module 620 , and a vehicle control module 630 .

Wherein, the obtaining motion track module 610 is used to obtain the motion track information of the dynamic obstacles identified during the driving process of the vehicle;

The motion intention prediction module 620 is configured to input the motion trajectory information into the trained machine learning model to predict the motion intention of the dynamic obstacle;

The vehicle control module 630 is configured to control the route planning of the vehicle or the driving state of the vehicle according to the predicted motion intention.

As an optional implementation manner of the embodiment of the present application, if the motion trajectory sample set required in the training process of the machine learning model is a sample set marked with motion intention, the vehicle control device further includes: an acquisition module and a first determination module.

Wherein, the first acquisition module is used to acquire the motion trajectory information of the dynamic obstacles collected during the vehicle driving in the lane;

a first determination module, configured to determine the motion intention from the motion trajectory information of the dynamic obstacle according to the set intention phenomenon rule;

Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample.

As an optional implementation manner of the embodiment of the present application, the first determination module is specifically used for:

If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as crossing the lane, determining that the motion is intended to be a pedestrian crossing the lane;

If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as going straight along the same edge of the lane, determining that the motion intention is that the pedestrian goes straight along the lane;

If the dynamic obstacle is identified as a vehicle and the motion trajectory information is identified as entering the current lane from an adjacent lane and is in front of the current vehicle, then it is determined that the motion is intended to switch lanes or overtake.

As an optional implementation manner of the embodiment of the present application, the vehicle control device further includes: a second acquisition module.

The first acquisition module is used to acquire the motion trajectory information of the dynamic obstacles collected during the vehicle running in the lane;

a second acquiring module, configured to acquire the stable driving control instruction of the vehicle for the motion trajectory information of the dynamic obstacle, as the motion intention;

Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample.

As an optional implementation manner of the embodiment of the present application, the second acquisition module is specifically used for:

obtaining at least one driving control command for the vehicle in response to the dynamic obstacle entering within the driving influence range of the vehicle;

In response to the dynamic obstacle being out of the driving influence range of the vehicle, selecting a last driving control command from the at least one driving control command as the stable driving control command;

The motion intention is acquired based on the stable driving control instruction.

As an optional implementation manner of this embodiment of the present application, the motion-intent prediction module 620 is specifically configured to:

Inputting the motion trajectory information into the machine learning model to identify motion intent;

If, according to the probability value of the motion intent recognition result, the motion intent result is identified as uncertain, return to the collection operation of motion trajectory information, and superimpose the newly collected motion trajectory into the historically collected motion trajectory for input to the machine. The model is learned until the motion intent of the dynamic obstacle is determined.

As an optional implementation manner of the embodiment of the present application, the motion intent is obtained by labeling the motion trajectory samples in the motion trajectory sample set that do not contain the motion intent after clustering.

As an optional implementation manner of the embodiment of the present application, the data format of the motion track information is to use at least one track point recorded sequentially to represent the motion track, and each track point records the direction of the dynamic obstacle along the lane line A 2-tuple of the speed of the movement and the speed of the movement perpendicular to the lane line

It should be noted that the foregoing explanations of the vehicle control method embodiment are also applicable to the vehicle control device of this embodiment, and the implementation principles thereof are similar, which will not be repeated here.

In the vehicle control device provided by the embodiment of the present application, the embodiment of the present application obtains the motion trajectory information of the dynamic obstacle identified during the driving process of the vehicle, and inputs the motion trajectory information of the dynamic obstacle into the trained machine learning model, To predict the motion intention of dynamic obstacles, and then control the route planning of the vehicle or the driving state of the vehicle according to the predicted motion intention. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of the dynamic obstacle through a long and stable motion trajectory, which causes difficulties in the path planning and driving state control of the vehicle, and realizes the recognition of the motion trajectory of the dynamic obstacle through the machine learning model. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

Embodiment 5

Referring to FIG. 7 , an embodiment of the present application provides an electronic device 700, which includes: one or more processors 720; a memory 710 that is communicatively connected to the at least one processor 720; Instructions executed by the at least one processor 720, the instructions are executed by the at least one processor 720, so that the at least one processor 720 can execute the vehicle control method according to any embodiment of the present application, the methods, including:

acquiring the motion trajectory information of the dynamic obstacles identified during the driving of the vehicle;

inputting the motion trajectory information into a trained machine learning model to predict the motion intent of the dynamic obstacle;

According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle is controlled.

Of course, those skilled in the art can understand that the processor 720 can also implement the technical solution of the vehicle control method provided by any embodiment of the present application.

The electronic device 700 shown in FIG. 7 is only an example, and should not impose any limitations on the functions and scope of use of the embodiments of the present application.

As shown in FIG. 7, electronic device 700 takes the form of a general-purpose computing device. Components of electronic device 700 may include, but are not limited to, one or more processors 720, memory 710, and a bus 750 connecting various system components including memory 710 and processor 720.

Bus 750 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of a variety of bus structures. By way of example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect ( PCI) bus.

Electronic device 700 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 700, including volatile and non-volatile media, removable and non-removable media.

Memory 710 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 711 and/or cache memory 712 . Electronic device 700 may further include other removable/non-removable, volatile/non-volatile computer system storage media. For example only, storage system 713 may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 7, commonly referred to as a "hard drive"). Although not shown in Figure 7, a disk drive for reading and writing to removable non-volatile magnetic disks (eg "floppy disks") and removable non-volatile optical disks (eg CD-ROM, DVD-ROM) may be provided or other optical media) to read and write optical drives. In these cases, each drive may be connected to bus 750 through one or more data media interfaces. Memory 710 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present application.

A program/utility 714 having a set (at least one) of program modules 715, which may be stored, for example, in memory 710, such program modules 715 including, but not limited to, an operating system, one or more application programs, other program modules, and program data , each or some combination of these examples may include an implementation of a network environment. Program modules 715 generally perform the functions and/or methods of any of the embodiments described herein.

The electronic device 700 may also communicate with one or more external devices 760 (eg, keyboard, pointing device, display 770, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 700, and/or communicate with the electronic device 700. Any device (eg, network card, modem, etc.) that enables the electronic device 700 to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interface 730 . Also, the electronic device 700 may communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network such as the Internet) through a network adapter 740 . As shown in FIG. 7 , the network adapter 740 communicates with other modules of the electronic device 700 through the bus 750 . It should be appreciated that, although not shown, other hardware and/or software modules may be used in conjunction with electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives and data backup storage systems.

The processor 720 executes various functional applications and data processing by running the programs stored in the memory 710, for example, to implement the vehicle control method provided by the embodiments of the present application.

It should be noted that the foregoing explanations of the vehicle control method embodiment are also applicable to the electronic device of this embodiment, and the implementation principles thereof are similar, which will not be repeated here.

For the electronic device provided by the embodiment of the present application, the embodiment of the present application obtains the motion trajectory information of the dynamic obstacle identified during the driving of the vehicle, and inputs the motion trajectory information of the dynamic obstacle into the trained machine learning model to Predict the motion intention of dynamic obstacles, and then control the route planning of the vehicle or the driving state of the vehicle according to the predicted motion intention. It solves the problem that it is difficult for the automatic driving system to determine the motion intention of the dynamic obstacle through a long and stable motion trajectory, which causes difficulties in the path planning and driving state control of the vehicle, and realizes the recognition of the motion trajectory of the dynamic obstacle through the machine learning model. In order to determine the motion intention of the dynamic obstacle, and then control the vehicle according to the motion intention, the accuracy of the vehicle control is improved.

Embodiment 6

This embodiment provides a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to cause the computer to execute the vehicle control method described in any of the embodiments of this application, and the method includes:

acquiring the motion trajectory information of the dynamic obstacles identified during the driving of the vehicle;

inputting the motion trajectory information into a trained machine learning model to predict the motion intent of the dynamic obstacle;

According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle is controlled.

Of course, a non-transitory computer-readable storage medium storing computer instructions provided by the embodiments of the present application, the computer instructions of which are used to enable the computer to execute the instructions are not limited to the above-mentioned method operations, and can also execute any embodiment of the present application. Relevant operations in the provided vehicle control method.

The computer-readable storage medium of the embodiments of the present application may adopt any combination of one or more computer-readable mediums. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples (a non-exhaustive list) of computer readable storage media include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer-readable signal medium may include a propagated data signal in baseband or as part of a carrier wave, with computer-readable program code embodied thereon. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device .

Program code embodied on a computer readable medium may be transmitted using any suitable medium, including - but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for performing the operations of the present application may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, but also conventional Procedural programming language - such as the "C" language or similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through Internet connection).

Note that the above are only preferred embodiments of the present application and applied technical principles. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and can also include more other equivalent embodiments without departing from the concept of the present application. The scope is determined by the scope of the appended claims.

Claims (11)

1. a vehicle control method, is characterized in that, comprises: acquiring the motion trajectory information of the dynamic obstacles identified during the driving of the vehicle; inputting the motion trajectory information into a trained machine learning model to predict the motion intent of the dynamic obstacle; According to the predicted motion intention, the route planning of the vehicle or the driving state of the vehicle is controlled. 2. The method according to claim 1, wherein the required motion trajectory sample set in the training process of the machine learning model is a sample set marked with motion intention, and the acquisition process of the motion trajectory sample set is concrete include: Obtain the motion trajectory information of the dynamic obstacles collected during the vehicle driving in the lane; According to the set intention phenomenon rule, determine the movement intention from the movement track information of the dynamic obstacle; Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample. 3. The method according to claim 2, wherein, according to the set intention phenomenon rule, determining the movement intention from the movement track information of the dynamic obstacle comprises at least one of the following: If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as crossing the lane, determining that the motion is intended to be a pedestrian crossing the lane; If the dynamic obstacle is identified as a pedestrian and the motion trajectory information is identified as going straight along the same edge of the lane, determining that the motion intention is that the pedestrian goes straight along the lane; If the dynamic obstacle is identified as a vehicle and the motion trajectory information is identified as entering the current lane from an adjacent lane and is in front of the current vehicle, then it is determined that the motion is intended to switch lanes or overtake. 4. The method according to claim 1, wherein the acquisition process of the required motion intent sample set in the training process of the machine learning model specifically comprises: Obtain the motion trajectory information of the dynamic obstacles collected during the vehicle driving in the lane; obtaining the stable driving control instruction of the vehicle with respect to the motion trajectory information of the dynamic obstacle, as the motion intention; Wherein, a binary set of the motion trajectory information and motion intent constitutes the motion intent sample. 5 . The method according to claim 4 , wherein obtaining a stable driving control instruction of the vehicle with respect to the motion trajectory information of the dynamic obstacle, as the motion intention, comprises: 6 . obtaining at least one driving control command for the vehicle in response to the dynamic obstacle entering the driving influence range of the vehicle; In response to the dynamic obstacle being out of the driving influence range of the vehicle, selecting a last driving control command from the at least one driving control command as the stable driving control command; The motion intention is acquired based on the stable driving control instruction. 6. The method according to claim 1, wherein inputting the motion trajectory information into a trained machine learning model to predict the motion intention of the dynamic obstacle comprises: Inputting the motion trajectory information into the machine learning model to identify motion intent; If, according to the probability value of the motion intention recognition result, the motion intention result is identified as uncertain, then return to the collection operation of motion trajectory information, and superimpose the newly collected motion trajectory information into the historically collected motion trajectory information for inputting all the motion trajectory information. The machine learning model is used until the motion intention of the dynamic obstacle is determined. 7 . The method according to claim 1 , wherein the motion intent is obtained by labeling the motion trajectory samples in the motion trajectory sample set that do not contain the motion intent after clustering. 8 . 8. The method according to claim 1, wherein the data format of the motion track information is to use at least one track point recorded sequentially to represent the motion track information, and each track point records the dynamic obstacle along the lane A 2-tuple of the speed of the line direction movement and the speed of the movement perpendicular to the lane line direction. 9. A vehicle control device, comprising: an acquisition motion trajectory module for acquiring motion trajectory information of dynamic obstacles identified during the driving process of the vehicle; A motion intention prediction module for inputting the motion trajectory information into the trained machine learning model to predict the motion intention of the dynamic obstacle; A vehicle control module, configured to control the route planning of the vehicle or the driving state of the vehicle according to the predicted motion intention. 10. 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 of claims 1-8 vehicle control method. 11. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the vehicle control method according to any one of claims 1-8.

Images