Vehicle control method and vehicle controller

Abstract

The invention relates to the field of automatic driving, in particular to a vehicle control method and a vehicle controller. The method is used for controlling the vehicle to run safely. The method comprises the steps that first sensing information shared by other vehicles is received, the other vehicles comprise a second vehicle, the second vehicle comprises a tractor and a trailer, and the first sensing information of the second vehicle comprises driving state information of the second vehicle and position information of the trailer; according to the first sensing information of the other vehicles, determining a predicted moving track of a tractor of the second vehicle and a predicted moving track of a trailer of the second vehicle; and controlling the first vehicle to run according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving, and in particular to a vehicle control method and a vehicle controller. Background Art

[0002] At present, vehicles with trailers mounted on them are a common means of transportation and are widely used in various logistics and transportation operation scenarios.

[0003] Compared with conventional vehicles, the movement trajectory of vehicles with trailers is more complex, which increases the risk of traffic accidents. Summary of the Invention

[0004] In order to solve the above technical problems, the present disclosure provides a vehicle control method and a vehicle controller.

[0005] In a first aspect, the present disclosure provides a vehicle control method. The method includes: receiving first perception information shared by other vehicles, the other vehicles including a second vehicle, the second vehicle including a tractor and a trailer, the first perception information of the second vehicle including driving status information of the second vehicle and position information of the trailer; determining a predicted movement trajectory of the tractor of the second vehicle and a predicted movement trajectory of the trailer of the second vehicle based on the first perception information of the other vehicles; and controlling the movement of the first vehicle based on the predicted movement trajectory of the tractor of the second vehicle and the predicted movement trajectory of the trailer of the second vehicle.

[0006] In some implementations, the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle are determined based on the first perception information of other vehicles, including: extracting feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle from the first perception information of other vehicles and the second perception information detected by the on-board sensor of the first vehicle; inputting the feature information into a trajectory prediction model to obtain the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle output by the trajectory prediction model.

[0007] In some implementations, feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle is extracted from the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle, including: constructing a spatiotemporal relationship dynamic graph between the first vehicle and the other vehicles based on the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle; the spatiotemporal relationship dynamic graph includes nodes corresponding one-to-one to the tractor of the second vehicle, the trailer of the second vehicle and the first vehicle; and extracting feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle in the spatiotemporal relationship dynamic graph.

[0008] In some implementations, the control of the first vehicle to travel according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle comprises: calculating a first probability of the first vehicle colliding with the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle; and controlling the first vehicle to perform a risk-avoiding operation when the first probability is greater than a probability threshold.

[0009] In some implementations, the calculation of the first probability of the first vehicle colliding with the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle comprises: mapping the predicted moving trajectory of the tractor of the second vehicle, the predicted moving trajectory of the trailer of the second vehicle, and the predicted moving trajectory of the first vehicle to a grid space to obtain trajectory densities corresponding to each grid in the grid space; the grid space comprises a plurality of grids obtained by gridding a space-time in which the first vehicle is located, each grid corresponding to a time point and a space region; and calculating the first probability of the first vehicle colliding with the second vehicle according to the trajectory densities corresponding to each grid in the grid space.

[0010] In some implementations, the calculation of the first probability of the first vehicle colliding with the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle comprises: mapping the predicted moving trajectory of the tractor of the second vehicle, the predicted moving trajectory of the trailer of the second vehicle, and the predicted moving trajectory of the first vehicle to a grid space to obtain trajectory densities corresponding to each grid in the grid space; the grid space comprises a plurality of grids obtained by gridding a space-time in which the first vehicle is located, each grid corresponding to a time point and a space region; calculating a second probability of the first vehicle colliding with the second vehicle according to the trajectory densities corresponding to each grid in the grid space; calculating a risk distribution of each region in an environment in which the first vehicle is located by using a deep potential network according to the first perception information of the other vehicles and second perception information detected by the on-board sensor of the first vehicle; and adjusting the second probability according to the risk distribution of each region in the environment in which the first vehicle is located to obtain the first probability of the first vehicle colliding with the second vehicle.

[0011] In some implementations, the control of the first vehicle to travel according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle comprises: determining a planned travel trajectory of the first vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle; and controlling the first vehicle to travel according to the planned travel trajectory; and / or the method further comprises: controlling the first vehicle to perform a risk-avoiding operation when the first perception information of any of the other vehicles includes a collision signal predicting that any of the other vehicles will collide with the first vehicle soon.

[0012] In some implementations, the first vehicle includes a tractor and a trailer, and the method further includes: obtaining driving state information of the first vehicle and position information of the trailer of the first vehicle; if the first vehicle is in a stationary state, sending third perception information outward at a preset transmission power; the third perception information includes the driving state information of the first vehicle and the position information of the trailer of the first vehicle; if the first vehicle is in a moving state, determining a signal transmission power according to a moving speed of the first vehicle, and sending the third perception information outward according to the signal transmission power; the third perception information includes the driving state information of the first vehicle and the position information of the trailer of the first vehicle.

[0013] In some implementations, if a first probability of a collision between the first vehicle and the second vehicle calculated according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle is greater than a probability threshold, the third perception information carries a collision signal that the first vehicle and the second vehicle are about to collide and an identifier of the second vehicle.

[0014] In a second aspect, a vehicle controller is provided, applied to a first vehicle, and includes: an obtaining unit configured to receive first perception information shared by other vehicles, the other vehicles including a second vehicle, the second vehicle including a tractor and a trailer, and the first perception information of the second vehicle including driving state information of the second vehicle and position information of the trailer; a processing unit configured to determine, according to the first perception information of the other vehicles, a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle; and the processing unit is further configured to control the first vehicle to drive according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle.

[0015] In some implementations, the processing unit is configured to determine, according to the first perception information of the other vehicles, the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, including: the processing unit is configured to extract feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle from the first perception information of the other vehicles and second perception information detected by a vehicle-mounted sensor of the first vehicle; and the processing unit is configured to input the feature information into a trajectory prediction model to obtain the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle output by the trajectory prediction model.

[0016] In some implementations, the processing unit is configured to extract feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle from the first perception information of the other vehicles and the second perception information detected by the on-board sensors of the first vehicle, including: the processing unit is configured to construct a spatio-temporal relationship dynamic graph of the first vehicle and the other vehicles according to the first perception information of the other vehicles and the second perception information detected by the on-board sensors of the first vehicle; the spatio-temporal relationship dynamic graph includes nodes corresponding to the tractor of the second vehicle, the trailer of the second vehicle, and the first vehicle one by one; and the processing unit is configured to extract feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle in the spatio-temporal relationship dynamic graph.

[0017] In some implementations, the processing unit is configured to control the first vehicle to travel according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, including: the processing unit is configured to calculate a first probability of a collision between the first vehicle and the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle; and the processing unit is configured to control the first vehicle to perform a risk-avoiding operation if the first probability is greater than a probability threshold.

[0018] In some implementations, the processing unit is configured to calculate a first probability of a collision between the first vehicle and the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, including: the processing unit is configured to map the predicted moving trajectory of the tractor of the second vehicle, the predicted moving trajectory of the trailer of the second vehicle, and the predicted moving trajectory of the first vehicle to a grid space to obtain trajectory densities corresponding to each grid in the grid space; the grid space includes a plurality of grids obtained by gridding a spatio-temporal space in which the first vehicle is located, each grid corresponding to a time point and a spatial region; and the processing unit is configured to calculate the first probability of the collision between the first vehicle and the second vehicle according to the trajectory densities corresponding to each grid in the grid space.

[0019] In some embodiments, the processing unit, for calculating the first probability of the first vehicle colliding with the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, comprises: the processing unit, for mapping the predicted moving trajectory of the tractor of the second vehicle, the predicted moving trajectory of the trailer of the second vehicle and the predicted moving trajectory of the first vehicle to a grid space to obtain trajectory densities corresponding to each grid in the grid space; the grid space comprises a plurality of grids obtained by gridding the space-time in which the first vehicle is located, each grid corresponding to a time point and a space region; the processing unit, for calculating the second probability of the first vehicle colliding with the second vehicle according to the trajectory densities corresponding to each grid in the grid space; the processing unit, for calculating the risk distribution of each region in the environment in which the first vehicle is located by using a deep potential network according to the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle; and the processing unit, for adjusting the second probability according to the risk distribution of each region in the environment in which the first vehicle is located to obtain the first probability of the first vehicle colliding with the second vehicle.

[0020] In some embodiments, the processing unit, for controlling the first vehicle to travel according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, comprises: the processing unit, for determining a planned travel trajectory of the first vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle; the processing unit, for controlling the first vehicle to travel according to the planned travel trajectory; and / or the processing unit, for controlling the first vehicle to perform a risk-avoiding operation when the first perception information of any of the other vehicles includes a collision signal predicting that any of the other vehicles will collide with the first vehicle soon.

[0021] In some embodiments, the first vehicle comprises: a tractor and a trailer; the acquisition unit is further configured to acquire travel state information of the first vehicle and position information of the trailer of the first vehicle; the processing unit is further configured to, if the first vehicle is in a stationary state, send third perception information outward at a preset signal transmission power, the third perception information comprising the travel state information of the first vehicle and the position information of the trailer of the first vehicle; and the processing unit is further configured to, if the first vehicle is in a moving state, determine a signal transmission power according to a moving speed of the first vehicle, and send the third perception information outward according to the signal transmission power, the third perception information comprising the travel state information of the first vehicle and the position information of the trailer of the first vehicle.

[0022] In some implementations, if a first probability of a collision between the first vehicle and the second vehicle is greater than a probability threshold according to a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle, the third perception information carries a collision signal predicting that the first vehicle and the second vehicle are about to collide and an identifier of the second vehicle.

[0023] Compared with the prior art, the technical solutions provided by the embodiments of the present disclosure have the following advantages:

[0024] In the method provided by the present disclosure, on the one hand, during the driving of the vehicle, the state of the sending vehicle can be sent to the nearby receiving vehicle (i.e., the first vehicle) through the sending of the perception information. Specifically, when the sending vehicle includes a vehicle (i.e., the second vehicle) with a trailer, the perception information sent by the second vehicle can include the driving state information (for example, the driving state information can include positioning, speed, acceleration, etc.) of the second vehicle and the position information of the trailer of the second vehicle. In this way, the first vehicle can determine in a timely manner that the second vehicle is a vehicle with a trailer according to the received perception information from the second vehicle, and predict the driving trajectory of the second vehicle according to the driving state information carried in the perception information and the position information of the trailer of the second vehicle, and then control the first vehicle to drive. On the other hand, in the process of predicting the driving trajectory of the second vehicle according to the received perception information, the first vehicle can determine the predicted driving trajectory of the tractor of the second vehicle and the predicted driving trajectory of the trailer of the second vehicle as two independent individuals, so as to determine a more accurate driving trajectory of the second vehicle, so as to control the first vehicle to drive according to the predicted driving trajectory of the tractor of the second vehicle and the predicted driving trajectory of the trailer of the second vehicle, and then achieve the purpose of safely controlling the first vehicle to drive. BRIEF DESCRIPTION OF DRAWINGS

[0025] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure.

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the accompanying drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0027] Figure 1 One of the structural schematic diagrams of a vehicle provided by the embodiments of the present disclosure;

[0028] Figure 2One of the schematic diagrams of a vehicle driving state provided by an embodiment of the present disclosure;

[0029] Figure 3 One of the flow schematic diagrams of a vehicle control method provided by an embodiment of the present disclosure;

[0030] Figure 4 One of the structural schematic diagrams of a vehicle provided by an embodiment of the present disclosure;

[0031] Figure 5 One of the flow schematic diagrams of a vehicle control method provided by an embodiment of the present disclosure;

[0032] Figure 6 One of the schematic diagrams of communication between vehicles provided by an embodiment of the present disclosure;

[0033] Figure 7 One of the flow schematic diagrams of a vehicle control method provided by an embodiment of the present disclosure;

[0034] Figure 8 One of the flow schematic diagrams of a vehicle control method provided by an embodiment of the present disclosure;

[0035] Figure 9 One of the flow schematic diagrams of a vehicle control method provided by an embodiment of the present disclosure;

[0036] Figure 10 One of the flow schematic diagrams of a vehicle control method provided by an embodiment of the present disclosure;

[0037] Figure 11 One of the structural schematic diagrams of a vehicle controller provided by an embodiment of the present disclosure;

[0038] Figure 12 One of the structural schematic diagrams of a vehicle controller provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0039] In order to enable a person skilled in the art to more clearly understand the above-mentioned purposes, features and advantages of the present disclosure, the schemes of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0040] In the following description, many specific details are set forth in order to fully understand the present disclosure, but the present disclosure can also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some of the embodiments of the present disclosure, not all the embodiments.

[0041] The related technologies involved in the embodiments of the present disclosure will be introduced below in combination with examples:

[0042] Currently, the vehicle with a trailer is widely used in various logistics transportation scenarios as a common transportation tool. Figure 1 As shown in the figure, the vehicle with a trailer includes a powered tractor and a trailer providing a load space, and the tractor and the trailer can be connected through a towing connection device. Due to the special vehicle structure, the motion trajectory of the vehicle with a trailer is more complex when driving. For example, in the scenario of vehicle A overtaking vehicle B as shown in the figure, at this time, although the tractor of vehicle A has safely overtaken the position of vehicle B, the trailer of vehicle A may collide with vehicle B. Figure 2

[0043] Therefore, when a vehicle (hereinafter referred to as a first vehicle) is driving, if there is a vehicle with a trailer (hereinafter referred to as a second vehicle) in the driving environment, how to safely control the first vehicle to drive and avoid traffic accidents is a problem to be solved at present.

[0044] In view of the above problems, in the embodiments of the present disclosure, on the one hand, during the driving of the vehicle, the vehicles can transmit the state of the sending vehicle to the nearby receiving vehicle (hereinafter referred to as the first vehicle) by sending perception information. Specifically, when the sending vehicle includes a vehicle with a trailer (hereinafter referred to as a second vehicle), the perception information sent by the second vehicle can include the driving state information (for example, the driving state information can include positioning, speed, acceleration, etc.) of the second vehicle and the position information of the trailer of the second vehicle. In this way, the first vehicle can determine in time that the second vehicle is a vehicle with a trailer according to the received perception information from the second vehicle, and predict the driving trajectory of the second vehicle according to the driving state information carried in the perception information and the position information of the trailer of the second vehicle, and then control the first vehicle to drive.

[0045] On the other hand, in the process of predicting the driving trajectory of the second vehicle according to the received perception information, the first vehicle can determine the predicted driving trajectory of the tractor of the second vehicle and the predicted driving trajectory of the trailer of the second vehicle as two independent individuals, so as to determine a more accurate driving trajectory of the second vehicle, so as to control the first vehicle to drive according to the predicted driving trajectory of the tractor of the second vehicle and the predicted driving trajectory of the trailer of the second vehicle, and then achieve the purpose of safely controlling the first vehicle to drive.

[0046] Based on the above considerations, the embodiments of the present disclosure provide a vehicle control method, as shown in the figure. Figure 3 ​As shown, in the method, on one hand, the first vehicle can receive perception information shared by other vehicles (hereinafter referred to as first perception information). The other vehicles include the second vehicle, and the second vehicle includes a tractor and a trailer. The first perception information of the second vehicle includes driving state information of the second vehicle and position information of the trailer of the second vehicle (i.e., S101). On the other hand, the first vehicle can determine a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle according to the first perception information of the other vehicles (i.e., S102). Further, the first vehicle can control the first vehicle to drive according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle (i.e., S103).

[0047] The specific implementation process of the vehicle control method provided by the embodiments of the present disclosure will be introduced below in combination with examples.

[0048] Firstly, the application scenario of the vehicle control method provided by the embodiments of the present disclosure will be introduced. The vehicle control method provided by the embodiments of the present disclosure can be applied in various vehicles. For example, Figure 4 As shown in the structural schematic diagram of a vehicle provided by the present disclosure. The vehicle 20 can include an intelligent driving system 21 for realizing automatic driving and intelligent driving, a steering system 22 for controlling vehicle steering, and a power braking system 23 for controlling vehicle power output and brake braking.

[0049] Among them, the intelligent driving system 21 can include a vehicle controller 211, a vehicle-mounted sensor 212, and a communication module 213, etc.

[0050] The vehicle-mounted sensor 212 is used to detect the driving environment around the vehicle. In actual application, the vehicle-mounted sensor 212 can include a camera, an ultrasonic radar, a millimeter wave radar, and a laser radar, etc.

[0051] The communication module 213 is used to make the vehicle communicate with other vehicles according to the method provided by the embodiments of the present disclosure.

[0052] The vehicle controller 211 is used to control the vehicle to drive according to the method provided by the embodiments of the present disclosure.

[0053] The communication process between vehicles in the method provided by the embodiments of the present disclosure will be introduced below taking the process of the vehicle 20 sending perception information to other vehicles as an example. Specifically, in the case that the vehicle 20 is a vehicle with a trailer (i.e., the vehicle 20 includes a tractor and a trailer), as shown in Figure 5 As shown, the method can include:

[0054] S301, the vehicle 20 acquires driving state information of the vehicle 20 and position information of a trailer of the vehicle 20.

[0055] For example, the vehicle 20 can acquire the driving state information of the vehicle 20 and the position information of the trailer of the vehicle 20 by collecting data of the vehicle 20 and the surrounding environment in real time through the on-board sensor 212.

[0056] For example, the driving state information of the vehicle 20 can include positioning information, trajectory information, speed vector information, acceleration information of the vehicle, and obstacle information (including coordinates, volume, and object type of the obstacle) perceived by the vehicle 20 through the on-board sensor.

[0057] S302, the vehicle 20 sends perception information (hereinafter referred to as perception information x) to a vehicle other than the vehicle 20.

[0058] For example, the perception information x can include the driving state information of the vehicle 20 and the position information of the trailer of the vehicle 20.

[0059] Specifically, the perception information x can include a plurality of fields. Each field is used to carry part of the driving state information of the vehicle 20 and the position information of the trailer of the vehicle 20.

[0060] For example, the perception information x can include a "location" field for carrying the positioning information of the vehicle 20, a "path" field for carrying the trajectory information of the vehicle 20, a "velocity" field for carrying the speed vector information of the vehicle 20, an "acceleration" field for carrying the acceleration information of the vehicle 20, and an "obstacle location" field for carrying the obstacle information detected by the vehicle 20. In addition, the perception information x can also include a "trailer location" field for carrying the position information of the trailer of the vehicle 20.

[0061] In some implementations, the perception information x can also include collision information predicting that the vehicle 20 will soon collide with another vehicle (for example, vehicle B) and an identification of the vehicle B.

[0062] For example, in one aspect, the vehicle 20 can receive the perception information from other vehicles according to the method provided by the embodiments of the present disclosure. In another aspect, the vehicle 20 can determine the predicted driving trajectory of the vehicle B (specifically, when the vehicle B is a vehicle with a trailer, the predicted driving trajectories of the towing vehicle and the trailer of the vehicle B can be determined respectively) according to the received perception information and the perception information detected by the on-board sensors of the vehicle 20, and then calculate the probability of collision between the vehicle B and the vehicle according to the predicted driving trajectory of the vehicle B, and carry the collision information and the identifier of the vehicle B in the perception information x when the probability is greater than the probability threshold.

[0063] For example, the "crash" field for carrying the collision information and the identifier of the vehicle B can also be included in the perception information x.

[0064] Through the above implementation manner, when the vehicle 20 predicts that the vehicle will collide with the vehicle B, the perception information carrying the collision signal and the identifier of the vehicle B can be sent to other vehicles, so that when other vehicles receive the perception information, corresponding measures can be taken in time to avoid the occurrence of the collision accident or reduce the loss of the collision accident. For example, when the vehicle B receives the perception information, the vehicle B can perform risk avoidance operations (such as emergency braking, emergency steering, etc.) in time to avoid collision with the vehicle 20. For another example, when other vehicles except the vehicle B receive the perception information, the other vehicles can change the driving trajectory and the like to keep a safe distance from the vehicle 20 and the vehicle B, thereby ensuring the safety of the vehicle.

[0065] In one implementation manner, the vehicle 20 can send the perception information x to vehicles other than the vehicle 20 through vehicle-to-everything (V2X).

[0066] For example, as shown in Figure 6 , in one aspect, the vehicle 20 can send the perception information x to the outside through the communication module 213 (such as a telematics box (T-box)) of the vehicle 20, so as to send the perception information x to the cloud server through the mobile communication network. In another aspect, the cloud server can send the perception information x to vehicles near the vehicle 20 through the mobile communication network.

[0067] In some implementation manners, S302 can specifically include:

[0068] S3021, if the vehicle 20 is in a stationary state, the vehicle 20 sends the perception information x to the outside according to the preset transmission power.

[0069] S3022, if the vehicle 20 is in a moving state, the vehicle 20 determines the signal transmission power according to the moving speed of the vehicle 20, and transmits the perception information x according to the signal transmission power.

[0070] For example, in one aspect, when it is determined that the vehicle 20 is in a stationary state (for example, it is determined that the vehicle 20 is in a stationary state when the moving speed of the vehicle 20 within 5 seconds is less than 0.5 m / s), the vehicle-mounted terminal of the vehicle 20 is automatically converted into a temporary Road Side Unit (RSU), and a TD-LTE time slot allocation algorithm is used to optimize the channel resource utilization. The perception information x is processed by a multi-source fusion process through an Extended Kalman Filter (EKF) algorithm to generate a Basic Safety Message (BSM) message package conforming to the SAE J2735 standard. The data package is broadcast at a frequency of 10 Hz and according to a preset transmission power through a MIMO 4x4 antenna array after Turbo encoding, and the effective coverage radius reaches 300 m.

[0071] On the other hand, when it is determined that the vehicle 20 is in a moving state, the vehicle 20 can use an Adaptive Power Control (APC) algorithm to dynamically adjust the transmission power according to the moving speed (Δv). The signal transmission power satisfies the following formula (1):

[0072] P_tx=P_base+k·log(1+|Δv| / v_ref) Formula (1)

[0073] Wherein, P_base=23dBm, k=5.2, v_ref=15m / s. At the same time, LDPC channel coding and QPSK modulation technology are applied to ensure that the Bit Error Rate (BER) is less than 10 -6 .

[0074] In the above implementation, the perception information can be transmitted according to the signal transmission power corresponding to the state of the vehicle. Specifically, in the case where the vehicle is in a moving state, when the moving speed of the vehicle is greater, the signal transmission power of the vehicle for transmitting the perception information to the outside can be increased, so that the perception information of the vehicle is transmitted to other vehicles in a larger range nearby; on the contrary, when the moving speed of the vehicle is smaller, the signal transmission power of the vehicle for transmitting the perception information to the outside can be reduced to reduce the electromagnetic interference of the environment. In the case where the vehicle is in a stationary state, the vehicle can transmit the perception information to the outside according to a preset signal transmission power to simplify the process of transmitting the perception information.

[0075] The following describes a vehicle control method provided by the embodiments of the present disclosure, taking the process of a first vehicle after receiving perception information from other vehicles as an example. As shown in FIG. 3, the method can include the following steps. Figure 7

[0076] S401, the first vehicle receives perception information (hereinafter referred to as first perception information) shared by other vehicles.

[0077] Taking the first vehicle as an example, the first vehicle can receive the first perception information shared by other vehicles through the communication module 213. Figure 4 Taking the vehicle 20 in FIG. 1 as an example, the first vehicle can receive the first perception information shared by other vehicles through the communication module 213.

[0078] The process in which the other vehicles send the first perception information to the first vehicle can refer to the content of S301-S302 above.

[0079] Specifically, the other vehicles can include a second vehicle, and the second vehicle includes a tractor and a trailer. The first perception information of the second vehicle includes the driving state information of the second vehicle and the position information of the trailer of the second vehicle.

[0080] In addition, the other vehicles can also include a vehicle other than the second vehicle (referred to as a third vehicle). When the third vehicle is a vehicle with a trailer, the first perception information of the third vehicle can include the driving state information of the third vehicle and the position information of the trailer of the third vehicle. When the third vehicle is a conventional vehicle without a trailer, the first perception information of the third vehicle can only include the driving state information of the third vehicle.

[0081] In some implementations, when any vehicle predicts that the vehicle will soon collide with the first vehicle, the vehicle can carry a collision signal predicting that any other vehicle will soon collide with the first vehicle in the first perception information sent by the vehicle. In this way, the first vehicle can control the first vehicle to perform a risk-avoiding operation (such as emergency braking, emergency steering, etc.) after detecting the collision signal in the first perception information sent by the vehicle.

[0082] Therefore, the method can further include: when the first perception information of any other vehicle includes a collision signal predicting that any other vehicle will soon collide with the first vehicle, the first vehicle controls the first vehicle to perform a risk-avoiding operation.

[0083] S402, the first vehicle determines the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle according to the first perception information of the other vehicles.

[0084] In the embodiments of the present disclosure, the predicted moving track of a vehicle can be understood as the possible moving track of the vehicle in a future time period (such as the next 5 seconds or the next 10 seconds, etc.).​

[0085] For example, after obtaining the first perception information of the other vehicle, the first vehicle can fuse the first perception information of the other vehicle and the perception information (hereinafter referred to as second perception information) detected by the vehicle-mounted sensor of the first vehicle, and predict the moving track of the tractor and the trailer of the second vehicle, and further determine the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle.

[0086] For example, after obtaining the first perception information of the other vehicle, the first vehicle can first verify the data integrity of the first perception information by using a corresponding verification algorithm (such as a Cyclic Redundancy Check 32 (CRC32) algorithm or the like).

[0087] Then, the first vehicle converts the coordinate system of various information in the first perception information into the Universal Transverse Mercator (UTM) coordinate corresponding to the first vehicle by coordinate system normalization. For example, the first vehicle can convert the various information in the first perception information of the second vehicle into the UTM coordinate corresponding to the first vehicle by using a quaternion attitude solution based on the tractor-trailer articulation point. For another example, the obstacle information carried in the first perception information is usually data in polar coordinates (with the other vehicle as the pole), so the first vehicle can first convert the obstacle information carried in the first perception information from polar coordinate data to Cartesian coordinate, and then convert it into the UTM coordinate corresponding to the first vehicle.

[0088] After converting the coordinates of various information in the first perception information into the UTM coordinate corresponding to the first vehicle, the first perception information and the second perception information detected by the vehicle-mounted sensor of the first vehicle can be fused by using a preset fusion strategy. For example, the preset fusion strategy can include an exception handling mechanism. Specifically, the exception handling mechanism can include a multi-level confidence check. The multi-level confidence check includes a sensor level check, that is, to eliminate data exceeding the physical limit value (such as speed > 200 km / h); a vehicle level, that is, to compare the first perception information and the second perception information from multiple vehicles, and to adopt the data when the data of multiple vehicles are consistent; and a system level, that is, to verify the accuracy of the data by historical trajectory prediction. In addition, the exception handling mechanism can also include establishing a data traceability log.

[0089] In some implementations, S402 can specifically include the following S4021-S4022:

[0090] S4021, the first vehicle extracts feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle from the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle.

[0091] In some designs, as shown in Figure 8 S4021 can specifically include the following S40211-S40212:

[0092] S40211, the first vehicle constructs a spatio-temporal relationship dynamic graph of the first vehicle and the other vehicles according to the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle.

[0093] In the spatio-temporal relationship dynamic graph, nodes corresponding to the tractor of the second vehicle, the trailer of the second vehicle, and the first vehicle are included. In addition, nodes corresponding to the other vehicles except the second vehicle and nodes corresponding to other objects (such as pedestrians, fixed obstacles, etc.) except vehicles can also be included in the spatio-temporal relationship dynamic graph.

[0094] For example, the first vehicle can first fuse the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle according to the above description. Then, the spatio-temporal relationship dynamic graph of the first vehicle and the other vehicles is constructed according to the fused information.

[0095] In actual implementation, the edge weight between two nodes in the spatio-temporal relationship dynamic graph can be determined by the distance between the nodes corresponding vehicles, the communication delay, and the sensor confidence.

[0096] S40212, the first vehicle extracts feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle in the spatio-temporal relationship dynamic graph.

[0097] Specifically, the first vehicle can use a dynamic graph convolutional network (DGCN) to extract feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle in the spatio-temporal relationship dynamic graph.

[0098] In the above design, by regarding the tractor of the second vehicle and the trailer of the second vehicle as a node in the spatio-temporal relationship dynamic graph respectively when constructing the spatio-temporal relationship dynamic graph, the feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle can be more accurately extracted.

[0099] S4022, the first vehicle inputs the feature information into a trajectory prediction model to obtain a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle output by the trajectory prediction model.

[0100] The trajectory prediction model can be a Social-Wasserstein Graph Double-Attention Network (Social-WaGDAT) model.

[0101] In actual application, the first vehicle can input the feature information obtained through S4021 and feature information obtained through other manners (for example, feature information of road structure extracted by a Deep Residual Network-18 (ResNet-18), and feature information reflecting the historical moving trajectory of the first vehicle or other vehicles, etc.) as the input of the trajectory prediction model, so as to obtain the predicted moving trajectory of the tow vehicle of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle output by the trajectory prediction model.

[0102] In addition, the result output by the trajectory prediction model can also include the predicted moving trajectory of other vehicles in addition to the predicted moving trajectory of the tow vehicle of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle.

[0103] For example, in the process of trajectory prediction, one or more associated targets corresponding to the first vehicle (wherein the associated targets can include the tow vehicle of the second vehicle, the trailer of the second vehicle, and other vehicles except the second vehicle) can be determined according to the feature information obtained through S4021 and the feature information obtained through other manners by using the Hungarian algorithm, etc. Then, after inputting the feature information obtained through S4021 and the feature information obtained through other manners into the trajectory prediction model, the predicted moving trajectory corresponding to each associated target output by the trajectory prediction model can be obtained. In addition, the predicted moving trajectory of the first vehicle can also be included in the output result of the trajectory prediction model.

[0104] In addition, in actual application, the trajectory prediction model can output multiple candidate trajectories for the first vehicle and one or more associated targets of the first vehicle, respectively. For example, the trajectory prediction model can output 20 candidate trajectories for the first vehicle, and the trajectory prediction model also outputs 20 candidate trajectories for each associated target of the first vehicle, and so on.

[0105] S403, the first vehicle controls the first vehicle to travel according to the predicted moving trajectory of the tow vehicle of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle.

[0106] For example, after obtaining the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle, on one hand, the first vehicle can determine a moving track of itself according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle, and then control the first vehicle to travel according to the moving track; on the other hand, the first vehicle can also determine whether the second vehicle is likely to collide with the first vehicle according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle, and if a collision is likely to occur, control the first vehicle to perform a risk-avoiding operation such as emergency braking, emergency steering, etc.

[0107] In an implementation manner, S403 can specifically include steps S4031-S4032.

[0108] S4031, the first vehicle calculates a first probability of a collision between the first vehicle and the second vehicle according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle.

[0109] S4032, the first vehicle controls the first vehicle to perform a risk-avoiding operation in a case where the first probability is greater than a probability threshold.

[0110] The risk-avoiding operation can specifically include emergency braking, emergency steering, etc.

[0111] In addition, in some designs, the risk-avoiding operation can also include carrying a collision signal indicating that the first vehicle and the second vehicle are about to collide and a representation of the second vehicle in the perception information when the first vehicle sends the perception information to other vehicles. The process in which the first vehicle sends the perception information can refer to the processes of S301-S302.

[0112] The implementation process of calculating the first probability (i.e., S4031) will be introduced in two designs as follows:

[0113] In the first design, as shown in FIG. 4, S4031 can include S40311-S40312. Figure 9

[0114] S40311, the first vehicle maps the predicted moving track of the tractor of the second vehicle, the predicted moving track of the trailer of the second vehicle, and the predicted moving track of the first vehicle to a grid space to obtain a trajectory density corresponding to each grid in the grid space.

[0115] ​The grid space includes a plurality of grids obtained by gridding the space-time in which the first vehicle is located, and each grid corresponds to a time point and a space region.

[0116] It can be understood that when the trajectory prediction model can obtain a plurality of candidate trajectories corresponding to the tractor of the second vehicle, the trailer of the second vehicle, and the first vehicle respectively (for example, the trajectory prediction model can output 20 candidate trajectories for the first vehicle, and the trajectory prediction model also outputs 20 candidate trajectories for the tractor of the second vehicle and the trailer of the second vehicle respectively), the plurality of candidate trajectories corresponding to the tractor of the second vehicle, the trailer of the second vehicle, and the first vehicle can be respectively mapped to the grid space to obtain the trajectory density corresponding to each grid in the grid space.

[0117] S40312, the first vehicle calculates a first probability of collision between the first vehicle and the second vehicle according to the trajectory density corresponding to each grid in the grid space.

[0118] For example, the U-Net architecture in the paper: Convolutional Networks for Biomedical Image Segmentation can be used to calculate the first probability of collision between the first vehicle and the second vehicle according to the trajectory density corresponding to each grid in the grid space.

[0119] In the second design, as shown in Figure 10 S4031 can include S40313-S40316:

[0120] S40313, the first vehicle maps the predicted moving trajectory of the tractor of the second vehicle, the predicted moving trajectory of the trailer of the second vehicle, and the predicted moving trajectory of the first vehicle to the grid space to obtain the trajectory density corresponding to each grid in the grid space.

[0121] The grid space includes a plurality of grids obtained by gridding the space-time in which the first vehicle is located, and each grid corresponds to a time point and a space region.

[0122] The specific implementation process of S40313 can refer to the corresponding description of S40311 described above.

[0123] S40314, the first vehicle calculates a second probability of collision between the first vehicle and the second vehicle according to the trajectory density corresponding to each grid in the grid space.

[0124] For example, the U-Net architecture in the paper: Convolutional Networks for Biomedical Image Segmentation can be used to calculate the second probability of the first vehicle colliding with the second vehicle according to the trajectory density corresponding to each grid in the grid space.

[0125] S40315, the first vehicle calculates the risk distribution of each region in the environment where the first vehicle is located by using a deep potential network according to the first perception information of the other vehicles and the second perception information detected by the on-board sensor of the first vehicle.

[0126] It can be understood that in actual application, the content of S40315 can be executed at other time points before S40316 is executed. For example, S40315 can be executed before S40313 and S40314, or S40315 can be executed synchronously during the execution of S40313 and S40314.

[0127] S40316, the first vehicle adjusts the second probability according to the risk distribution of each region in the environment where the first vehicle is located, to obtain the first probability of the first vehicle colliding with the second vehicle.

[0128] In some possible designs, the first vehicle can use eXtreme Gradient Boosting (XGBoost) to adjust the second probability according to the risk distribution of each region in the environment where the first vehicle is located and other information (such as local traffic regulation information), to obtain the first probability of the first vehicle colliding with the second vehicle.

[0129] In the above design, it is considered that the road includes regions with relatively high risks such as intersections and ramp entrances, and also includes other regions with relatively low risks. Therefore, the deep potential network can be used to calculate the risk distribution of each region in the environment where the first vehicle is located. Then, the second probability is adjusted according to the risk distribution of each region in the environment where the first vehicle is located, so that the obtained first probability can more accurately reflect the probability of the first vehicle colliding with the second vehicle.

[0130] In some other implementations, S403 can specifically include the following steps S4033-S4034:

[0131] S4033, the first vehicle determines the planned driving trajectory of the first vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle.

[0132] For example, the first vehicle can determine a driving trajectory that does not conflict with the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, and take the driving trajectory as the planned driving trajectory of the first vehicle.

[0133] S4034, the first vehicle controls the first vehicle to drive according to the planned driving trajectory.

[0134] For example, the first vehicle can control the steering system and the power brake system of the first vehicle so that the first vehicle drives according to the planned driving trajectory.

[0135] It can be understood that the above mainly introduces the execution process of the first vehicle after the first vehicle receives the first perception information shared by other vehicles through the implementation process of S401-S403. In actual application, on the one hand, the first vehicle can receive the first perception information shared by other vehicles and control the first vehicle to drive through S401-S403; on the other hand, the first vehicle can also send the perception information to the outside according to the process of S301-S302. The process of the first vehicle sending the perception information to the outside can refer to the content of S301-S302, and the repeated content will not be described here.

[0136] Based on the above method embodiment, the device provided by the embodiment of the present disclosure is described below. As shown in Figure 11 The structure of the vehicle controller provided by the embodiment of the present disclosure is shown in FIG. 5. Specifically, the vehicle controller 50 can be a chip or a system on chip, for example, the vehicle controller 50 can be a vehicle controller 211 in Figure 4 The vehicle controller 50 can be used to realize the functions of the first vehicle or the vehicle 20 in the method provided by the embodiment of the present disclosure. Specifically, the vehicle controller 50 can include:

[0137] The acquisition unit 501 is configured to receive the first perception information shared by other vehicles, wherein the other vehicles include the second vehicle, the second vehicle includes the tractor and the trailer, and the first perception information of the second vehicle includes the driving state information of the second vehicle and the position information of the trailer.

[0138] The processing unit 502 is configured to determine the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle according to the first perception information of the other vehicles.

[0139] The processing unit 502 is further configured to control the first vehicle to drive according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle.

[0140] In some embodiments, the processing unit 502 is configured to determine, according to the first perception information of the other vehicle, a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle, including:

[0141] The processing unit 502 is configured to extract, from the first perception information of the other vehicle and the second perception information detected by the on-board sensor of the first vehicle, feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle.

[0142] The processing unit 502 is configured to input the feature information into the trajectory prediction model to obtain the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle output by the trajectory prediction model.

[0143] In some embodiments, the processing unit 502 is configured to extract, from the first perception information of the other vehicle and the second perception information detected by the on-board sensor of the first vehicle, feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle, including:

[0144] The processing unit 502 is configured to construct a spatio-temporal relationship dynamic graph of the first vehicle and the other vehicle according to the first perception information of the other vehicle and the second perception information detected by the on-board sensor of the first vehicle; the spatio-temporal relationship dynamic graph includes nodes corresponding to the tractor of the second vehicle, the trailer of the second vehicle and the first vehicle one by one.

[0145] The processing unit 502 is configured to extract feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle in the spatio-temporal relationship dynamic graph.

[0146] In some embodiments, the processing unit 502 is configured to control the first vehicle to travel according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, including:

[0147] The processing unit 502 is configured to calculate a first probability of collision between the first vehicle and the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle.

[0148] The processing unit 502 is configured to control the first vehicle to perform a risk-avoiding operation in a case where the first probability is greater than a probability threshold.

[0149] In some embodiments, the processing unit 502 is configured to calculate a first probability of collision between the first vehicle and the second vehicle according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, including:

[0150] The processing unit 502 is configured to map the predicted moving track of the tractor of the second vehicle, the predicted moving track of the trailer of the second vehicle, and the predicted moving track of the first vehicle to a grid space to obtain a track density corresponding to each grid in the grid space; the grid space includes a plurality of grids obtained by gridding a space-time in which the first vehicle is located, and each grid corresponds to a time point and a space region.

[0151] The processing unit 502 is configured to calculate a first probability of collision between the first vehicle and the second vehicle according to the track density corresponding to each grid in the grid space.

[0152] In some implementations, the processing unit 502 is configured to calculate the first probability of collision between the first vehicle and the second vehicle according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle, including:

[0153] The processing unit 502 is configured to map the predicted moving track of the tractor of the second vehicle, the predicted moving track of the trailer of the second vehicle, and the predicted moving track of the first vehicle to a grid space to obtain a track density corresponding to each grid in the grid space; the grid space includes a plurality of grids obtained by gridding a space-time in which the first vehicle is located, and each grid corresponds to a time point and a space region.

[0154] The processing unit 502 is configured to calculate a second probability of collision between the first vehicle and the second vehicle according to the track density corresponding to each grid in the grid space.

[0155] The processing unit 502 is configured to calculate a risk distribution of each region in an environment in which the first vehicle is located by using a deep potential network according to the first perception information of the other vehicle and second perception information detected by the on-board sensor of the first vehicle.

[0156] The processing unit 502 is configured to adjust the second probability according to the risk distribution of each region in the environment in which the first vehicle is located to obtain the first probability of collision between the first vehicle and the second vehicle.

[0157] In some implementations, the processing unit 502 is configured to control the first vehicle to travel according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle, including:

[0158] The processing unit 502 is configured to determine a planned travel track of the first vehicle according to the predicted moving track of the tractor of the second vehicle and the predicted moving track of the trailer of the second vehicle.

[0159] The processing unit 502 is configured to control the first vehicle to travel according to the planned travel track.

[0160] And / or,

[0161] The processing unit 502 is further configured to control the first vehicle to perform a safety operation when the first perception information of any other vehicle includes a collision signal predicting that a collision between the first vehicle and any other vehicle will occur soon.

[0162] In some implementations, the first vehicle includes a tractor and a trailer.

[0163] The acquisition unit 502 is further configured to acquire driving state information of the first vehicle and position information of the trailer of the first vehicle.

[0164] The processing unit 502 is further configured to transmit the third perception information according to a preset signal transmission power if the first vehicle is in a static state, and the third perception information includes the driving state information of the first vehicle and the position information of the trailer of the first vehicle.

[0165] The processing unit 502 is further configured to determine the signal transmission power according to the moving speed of the first vehicle if the first vehicle is in a moving state, and transmit the third perception information according to the signal transmission power, and the third perception information includes the driving state information of the first vehicle and the position information of the trailer of the first vehicle.

[0166] In some implementations, if the first probability of a collision between the first vehicle and the second vehicle calculated according to the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle is greater than the probability threshold, the third perception information carries a collision signal predicting that a collision between the first vehicle and the second vehicle will occur soon and an identifier of the second vehicle.

[0167] The vehicle controller 50 provided by the embodiments of the present disclosure can perform part or all of the steps of the above method, and the implementation principle and technical effects are similar, which will not be described here.

[0168] Based on the same inventive concept, the embodiments of the present disclosure further provide another vehicle controller. Figure 12 As shown in the structural schematic diagram of the vehicle controller provided by the embodiments of the present disclosure, Figure 12 The vehicle controller provided by the embodiments of the present disclosure includes a memory 601 and a processor 602, the memory 601 is used to store a computer program, and the processor 602 is used to execute any method provided by the above embodiments when executing the computer program.

[0169] Based on the same inventive concept, the embodiments of the present disclosure further provide a computer readable storage medium, and the computer readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program causes a computing device to implement the method provided by the above embodiments.

[0170] Based on the same inventive concept, the embodiments of the present disclosure further provide a computer program product, which, when running on a computer, causes a computing device to implement the method provided by the above-mentioned embodiments.

[0171] Those skilled in the art will appreciate that embodiments of the present disclosure can be provided as methods, systems, or computer program products. Accordingly, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure can take the form of a computer program product on one or more computer-usable storage media (including volatile and non-volatile computer-readable media) having computer-usable program code embodied in the medium.

[0172] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can be any conventional processor.

[0173] The memory can include non-persistent memory in computer-readable media, random access memory (RAM), and / or non-volatile memory, etc., such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0174] The computer-readable media include non-transitory and transitory, removable and non-removable storage media. The storage media can be implemented by any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile discs (DVDs) or other optical storage, magnetic cassettes, magnetic disks storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to computing devices. According to the definition herein, computer-readable media do not include transitory media, such as modulated data signals and carrier waves.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, and are not intended to limit the present disclosure. Although the present disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions described in the foregoing embodiments, or make equivalent replacements to some or all of the technical features, without departing from the scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A vehicle control method, characterized in that: The method is applied to a first vehicle and includes: Receiving first perception information shared from other vehicles, the other vehicles including a second vehicle, the second vehicle including a tractor and a trailer, the first perception information of the second vehicle including: driving state information of the second vehicle and position information of the trailer; determining a predicted moving trajectory of a tractor of the second vehicle and a predicted moving trajectory of a trailer of the second vehicle based on the first perception information of the other vehicle; The first vehicle is controlled to travel according to the predicted movement trajectory of the tractor of the second vehicle and the predicted movement trajectory of the trailer of the second vehicle.

2. The method according to claim 1, characterized in that The determining, based on the first perception information of the other vehicle, a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle includes: extracting feature information corresponding to a tractor of the second vehicle and a trailer of the second vehicle from the first perception information of the other vehicle and second perception information detected by a vehicle-mounted sensor of the first vehicle; The feature information is input into a trajectory prediction model to obtain a predicted moving trajectory of the tractor of the second vehicle and a predicted moving trajectory of the trailer of the second vehicle output by the trajectory prediction model.

3. The method according to claim 2, characterized in that Extracting feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle from the first perception information of the other vehicle and the second perception information detected by the onboard sensor of the first vehicle includes: constructing a spatiotemporal relationship dynamic graph between the first vehicle and the other vehicles based on the first perception information of the other vehicle and the second perception information detected by the onboard sensor of the first vehicle; the spatiotemporal relationship dynamic graph including nodes corresponding one-to-one to the tractor of the second vehicle, the trailer of the second vehicle, and the first vehicle; Feature information corresponding to the tractor of the second vehicle and the trailer of the second vehicle in the spatiotemporal relationship dynamic graph is extracted.

4. The method according to claim 1, wherein The controlling the first vehicle to travel according to the predicted movement trajectory of the tractor of the second vehicle and the predicted movement trajectory of the trailer of the second vehicle includes: calculating a first probability of a collision between the first vehicle and the second vehicle based on a predicted movement trajectory of the tractor of the second vehicle and a predicted movement trajectory of the trailer of the second vehicle; When the first probability is greater than a probability threshold, the first vehicle is controlled to perform a risk avoidance operation.

5. The method according to claim 4, characterized in that The calculating, based on the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, a first probability of collision between the first vehicle and the second vehicle includes: Mapping the predicted movement trajectory of the tractor of the second vehicle, the predicted movement trajectory of the trailer of the second vehicle, and the predicted movement trajectory of the first vehicle to a grid space to obtain a trajectory density corresponding to each grid in the grid space; the grid space includes: a plurality of grids obtained by rasterizing the space-time of the first vehicle, each grid corresponding to a time point and a spatial area; A first probability of collision between the first vehicle and the second vehicle is calculated according to trajectory densities corresponding to each grid in the grid space.

6. The method according to claim 4, characterized in that The calculating, based on the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle, a first probability of collision between the first vehicle and the second vehicle includes: Mapping the predicted movement trajectory of the tractor of the second vehicle, the predicted movement trajectory of the trailer of the second vehicle, and the predicted movement trajectory of the first vehicle to a grid space to obtain a trajectory density corresponding to each grid in the grid space; the grid space includes: a plurality of grids obtained by rasterizing the space-time of the first vehicle, each grid corresponding to a time point and a spatial area; calculating a second probability of collision between the first vehicle and the second vehicle based on trajectory densities corresponding to each grid in the grid space; Calculating, using a deep potential network, a risk distribution of each area in an environment where the first vehicle is located based on the first perception information of the other vehicle and the second perception information detected by an onboard sensor of the first vehicle; The second probability is adjusted according to the risk distribution of each area in the environment where the first vehicle is located to obtain a first probability of collision between the first vehicle and the second vehicle.

7. The method according to any one of claims 1 to 6, characterized in that The controlling the first vehicle to travel according to the predicted movement trajectory of the tractor of the second vehicle and the predicted movement trajectory of the trailer of the second vehicle includes: determining a planned driving trajectory of the first vehicle based on the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle; controlling the first vehicle to travel according to the planned travel trajectory; and / or, The method further comprises: When the first perception information of any other vehicle includes a collision signal predicting that any other vehicle is about to collide with the first vehicle, the first vehicle is controlled to perform a risk avoidance operation.

8. The method according to any one of claims 1 to 6, characterized in that The first vehicle includes: a tractor and a trailer, and the method further includes: Acquiring driving status information of the first vehicle and location information of the trailer of the first vehicle; If the first vehicle is stationary, third sensing information is sent outward according to a preset transmission power; the third sensing information includes the driving state information of the first vehicle and the position information of the trailer of the first vehicle; If the first vehicle is in a moving state, the signal transmission power is determined according to the moving speed of the first vehicle, and third perception information is sent out according to the signal transmission power; the third perception information includes the driving status information of the first vehicle and the position information of the trailer of the first vehicle.

9. The method according to claim 8, characterized in that If the first probability of a collision between the first vehicle and the second vehicle calculated based on the predicted moving trajectory of the tractor of the second vehicle and the predicted moving trajectory of the trailer of the second vehicle is greater than a probability threshold, then the third perception information carries: a collision signal predicting that the first vehicle and the second vehicle are about to collide and an identification of the second vehicle.

10. A vehicle controller, characterized in that: include: a memory and a processor, wherein the memory is used to store a computer program; The processor is configured to, when executing the computer program, enable the vehicle control to implement the vehicle control method provided by any one of claims 1 to 9.

Images