A Cooperative Control Method for Unmanned Vehicle Fleets Based on Regional Visual Relative Positioning

Abstract

The present invention discloses a method for cooperative motion control of an unmanned vehicle fleet based on regional vision relative positioning, belonging to the field of intelligent control technology, including: collecting visual information of the surrounding environment corresponding to any unmanned vehicle in real time, performing regional vision relative positioning, and obtaining the relative position and motion state of any unmanned vehicle in the fleet; transmitting the relative positions and motion states corresponding to all unmanned vehicles in the fleet to the control module based on a communication sharing mode; designing and generating a cooperative control strategy based on the relative positions, adjusting the cooperative control strategy based on the motion states, and generating the overall motion path of the fleet; monitoring the motion state of any unmanned vehicle, and generating a motion instruction for any unmanned vehicle in real time based on the overall motion path and visual information; any unmanned vehicle receives the motion instruction based on the communication sharing mode and adjusts its motion state to complete the cooperative motion control of the fleet. The accuracy of the cooperative motion control of the unmanned vehicle fleet can be improved through the present invention.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent control, and particularly relates to a method for motion cooperative control of an unmanned vehicle fleet based on regional vision relative positioning. Background Art

[0002] In recent years, with the development of computer science and artificial intelligence, unmanned vehicles have been studied more and more widely. An unmanned vehicle fleet composed of multiple unmanned vehicles based on unmanned vehicles has also been widely studied. Cooperative control is a problem that unmanned vehicles must solve, and the unmanned vehicle fleet also faces great challenges in cooperative control, mainly including the following points: in terms of information cognition, the incompleteness and time-variability of vehicle states; in terms of control, control constraints brought by complex environments and actuator constraints.

[0003] As the core problem that the unmanned vehicle fleet needs to solve, cooperative control is mainly divided into distributed and centralized. In the distributed multi-vehicle cooperative algorithm, vehicles transmit and exchange information through V2V (Vehicle to Vehicle) communication to complete driving decisions. However, V2V communication often has communication delays and information loss, so the stability and robustness of the distributed algorithm need to be improved. The centralized multi-vehicle cooperative algorithm collects the information of all vehicles to the central processor for centralized processing, and then conveys the information to each vehicle. Its stability and robustness are much greater than those of the distributed; at the same time, in the centralized cooperative algorithm, multiple vehicles can cooperate to complete perception, decision-making and planning functions, improving the adaptability of the vehicle to the environment.

[0004] Therefore, in actual situations, it is necessary to improve the accuracy of the self-positioning of unmanned vehicles and accurately complete the motion cooperative control of the unmanned vehicle fleet. Summary of the Invention

[0005] To solve the above problems, the present invention provides a method for motion cooperative control of an unmanned vehicle fleet based on regional vision relative positioning to solve the problems in the prior art.

[0006] To achieve the above invention purpose, the present invention proposes a method for motion cooperative control of an unmanned vehicle fleet based on regional vision relative positioning, including:

[0007] S1: Using a camera device to collect visual information of the surrounding environment corresponding to any unmanned vehicle in real time, and combining vehicle-mounted sensor data for regional vision relative positioning to obtain the relative position and motion state of any said unmanned vehicle in the fleet;

[0008] S2: Setting a communication sharing mode, and transmitting the relative position and the motion state corresponding to all the unmanned vehicles in the fleet to the control module based on the communication sharing mode;

[0009] S3: The control module designs and generates a cooperative control strategy based on the relative positions, adjusts the cooperative control strategy based on the motion states, and generates the overall motion path of the vehicle fleet.

[0010] S4: Monitor the motion state of any one of the driverless vehicles, and generate a motion instruction for any one of the driverless vehicles in real time based on the overall motion path and the visual information.

[0011] S5: Any one of the driverless vehicles receives the motion instruction based on the communication sharing mode, and adjusts the motion state based on the motion instruction to complete the motion cooperative control of the vehicle fleet.

[0012] Further, obtaining the relative position and motion state of any one of the driverless vehicles in the vehicle fleet includes the following steps:

[0013] Extract regional features based on the visual information, match them with the regional map to obtain the target features of the preceding vehicle, perform regional visual relative positioning based on the target features, and generate the relative position based on the regional map and the regional visual relative positioning.

[0014] Obtain the driving speed and driving angle based on the in-vehicle sensor data, and set the driving speed and the driving angle as the motion state.

[0015] Further, generating the overall motion path of the vehicle fleet includes the following steps:

[0016] Update the target features of the preceding vehicle, update the regional features that match the regional map, mark the updated target features and regional features based on the relative positions, generate the arrangement order of all the driverless vehicles in the vehicle fleet, and combine the motion states corresponding to all the driverless vehicles based on the arrangement order to generate the overall motion path.

[0017] Further, dynamically adjusting the motion instruction of any one of the driverless vehicles based on the overall motion path and the visual information includes the following steps:

[0018] Collect the relative positions and the motion states at preset time intervals, respectively generate a position change sequence and a motion change sequence, construct a prediction model based on a neural network, input both the position change sequence and the motion change sequence into the prediction model, and the prediction model outputs the predicted position and predicted speed corresponding to any one of the driverless vehicles within the prediction time based on the change amount corresponding to the preset time interval.

[0019] Generate an estimated trajectory based on the predicted position and the predicted speed. The control module obtains the estimated trajectories corresponding to all the unmanned vehicles, obtains the overlap degree between the estimated trajectory and the cooperative control strategy based on the arrangement order, adjusts the estimated trajectory of any one of the unmanned vehicles based on the overlap degree and the overall movement path to generate a standard trajectory, generates an adjustment instruction based on the standard trajectory, and sets the adjustment instruction as the adjusted movement instruction.

[0020] Further, the updating of the target features of the leading vehicle includes the following steps:

[0021] Obtain the basic information of the leading vehicle based on the target features, and determine the relative distance, relative angle, and relative speed between the leading vehicle and the unmanned vehicle;

[0022] Use the LK optical flow method to detect target feature points in the target features, use the FAST algorithm to extract features from the regional map and then match the target feature points, and use the successfully matched map feature points to calculate the position of the unmanned vehicle;

[0023] Construct a database, update the database using the map feature points, and send the updated database to the control module.

[0024] Further, the LK optical flow method matches the regional map corresponding to the unmanned vehicle and the leading vehicle to generate a matching map. If the matching map exists, use the RANSAC algorithm to output the matching effect of the matching map. If the matching effect is greater than the first threshold, it is determined as a map matching feature point; otherwise, it is determined as a non-matching feature point. If the matching map does not exist, use the LK optical flow method to extract feature points from the regional map of the unmanned vehicle and determine it as the map matching feature point; set the map matching feature point as the map feature point.

[0025] Further, the completion of the motion cooperative control of the vehicle fleet includes:

[0026] Real-time calculate the actual position of the vehicle, the position of the leading vehicle, and the driving length of the vehicle. Subtract the driving length of the vehicle from the position of the leading vehicle to obtain the real-time vehicle speed of the vehicle. Estimate the vehicle speed and the heading angle of the unmanned vehicle based on the real-time vehicle speed of the vehicle to generate new feature points;

[0027] Send the new feature points to the control module, and repeat this step until the vehicle fleet reaches the destination.

[0028] Further, the setting of the communication sharing mode includes:

[0029] Set the communication frequency and information tags, design the communication technology based on the number of the vehicle fleets, establish a communication topology map based on the communication frequency and the communication technology, and transmit the data information corresponding to the information tags based on the communication topology map;

[0030] Obtain the transmission performance of the data information, optimize and adjust the communication topology map based on the transmission performance, and set the adjusted communication topology map as the communication sharing mode.

[0031] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0032] The present invention first obtains the relative positions and motion states of the unmanned vehicles, can quickly generate the overall motion path of the vehicle fleets, then monitors the motion states of the unmanned vehicles through a prediction model, can accurately and real-time generate motion instructions, and finally adjusts the motion states of the corresponding unmanned vehicles through the motion instructions, and can intelligently complete the motion cooperative control of the vehicle fleets.

[0033] The present invention also calculates the position changes between the feature points through regional vision relative positioning, and can accurately calculate the relative positions of the unmanned vehicles. Description of the Drawings

[0034] Figure 1 is a step flow chart of a method for motion cooperative control of an unmanned vehicle fleet based on regional vision relative positioning according to the present invention;

[0035] Figure 2 is a schematic diagram of the communication sharing mode in the present invention. Detailed Embodiments

[0036] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0037] It can be understood that the terms "first", "second", etc. used in the present application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish the first element from another element. For example, without departing from the scope of the present application, the first xx script may be called the second xx script, and similarly, the second xx script may be called the first xx script.

[0038] As Figure 1 shown, a method for motion cooperative control of an unmanned vehicle fleet based on regional vision relative positioning includes:

[0039] S1: Use a camera device to collect visual information of the surrounding environment corresponding to any unmanned vehicle in real time, and combine it with in-vehicle sensor data for regional visual relative positioning to obtain the relative position and motion state of any unmanned vehicle in the vehicle fleet.

[0040] Specifically, when the unmanned vehicle performs cooperative motion in the vehicle fleet, it is necessary to analyze the images in the front, rear, left, and right directions of the unmanned vehicle, that is, the surrounding environment. In this embodiment, the visual information includes but is not limited to video sets or image sets captured by the camera device, and the in-vehicle sensor data includes but is not limited to data collected and generated by speed sensors, direction angle sensors, and positioning sensors, etc. Regional visual relative positioning is a technical means that uses visual information to determine the relative position of an object in space, mainly relying on cameras and other visual sensors to capture image data, extract feature information from it, and then calculate the relative position and pose between objects. The relative position refers to the current spatial position of the unmanned vehicle in the vehicle fleet, and the motion state includes but is not limited to state data such as driving speed and steering wheel angle.

[0041] S2: Set up a communication sharing mode, and transmit the relative positions and motion states corresponding to all unmanned vehicles in the vehicle fleet to the control module based on the communication sharing mode.

[0042] Specifically, in this embodiment, the control module refers to a remote control terminal, and the communication sharing mode refers to the way of data communication transmission. As Figure 2 shown, there is a communication connection between any two unmanned vehicles, and all communication connection lines form the communication sharing mode H1.

[0043] S3: The control module designs and generates a cooperative control strategy based on the relative position, adjusts the cooperative control strategy based on the motion state, and generates the overall motion path of the vehicle fleet.

[0044] Specifically, in this embodiment, the cooperative control strategy refers to the planned way of the vehicle fleet for motion coordination. For example, motion formation and motion following, etc. Adjusting the cooperative control strategy is to intelligently modify the cooperative control strategy according to the motion states of all unmanned vehicles to avoid a single cooperative control strategy affecting the cooperative motion of the vehicle fleet. The overall motion path refers to the route mode generated after the path planning of all unmanned vehicles by the cooperative control strategy.

[0045] S4: Monitor the motion state of any unmanned vehicle, and generate a motion instruction for any unmanned vehicle in real time based on the overall motion path and visual information.

[0046] Specifically, in this embodiment, the motion instruction refers to the instruction information used to control the motion mode of the unmanned vehicle. The visual information collected in real time is changing. Therefore, the position of the unmanned vehicle can be located according to the visual information, and by comparing with the overall motion path, a motion instruction can be generated in a timely manner.

[0047] S5: Any unmanned vehicle receives a motion instruction based on the communication sharing mode, and adjusts its motion state based on the motion instruction to complete the motion cooperative control of the vehicle fleet.

[0048] Specifically, in this embodiment, the motion instruction can be received quickly and accurately by the communication sharing mode, avoiding communication interference and ensuring that the motion instruction is accurately transmitted to the corresponding unmanned vehicle. Each unmanned vehicle controls the data information of the motion state according to the motion instruction, so that all unmanned vehicles in the vehicle fleet complete the operation of motion cooperative control.

[0049] As a preferred technical solution of the present invention, obtaining the relative position and motion state of any unmanned vehicle in the vehicle fleet includes the following steps:

[0050] Extract the regional features based on the visual information, match them with the regional map to obtain the target features of the vehicle in front, perform regional visual relative positioning based on the target features, and generate the relative position based on the regional map and the regional visual relative positioning.

[0051] Obtain the driving speed and driving angle based on the in-vehicle sensor data, and set the driving speed and driving angle as the motion state.

[0052] Specifically, in the present invention, the regional feature refers to the image data intercepted at the current moment in the visual information, the regional map refers to the spatial image data corresponding to the position of the unmanned vehicle at the current moment, the vehicle in front refers to the unmanned vehicle in front of the current unmanned vehicle, and there may be no target features of the vehicle in front. The target feature refers to the image data belonging to the unmanned vehicle.

[0053] The driving angle refers to the wheel angle controlled by the steering wheel, and the data corresponding to the driving speed and driving angle are set as the data corresponding to the motion state.

[0054] As a preferred technical solution of the present invention, generating the overall motion path of the vehicle fleet includes the following steps:

[0055] Update the target features of the vehicle in front, update the regional features matching the regional map, arrange the updated target features and regional features based on the regional map to generate the arrangement order of all unmanned vehicles in the vehicle fleet, and combine the motion states corresponding to all unmanned vehicles based on the arrangement order to generate the overall motion path.

[0056] Specifically, in this embodiment, since all the driverless vehicles in the fleet are in real-time motion, it is necessary to update the target features and regional features. The relative positions of the regional features and target features corresponding in the regional map can be marked using the technical method corresponding to regional vision relative positioning. Subsequently, the marked regional features and target features can be sorted through feature point matching, and thus the arrangement order of all the driverless vehicles can be obtained, where the arrangement order refers to the spatial position distribution of each driverless vehicle in the fleet. The overall motion path of the fleet can be obtained in real time from the arrangement order and the motion state.

[0057] As a preferred technical solution of the present invention, dynamically adjusting the motion instruction of any driverless vehicle based on the overall motion path and visual information includes the following steps:

[0058] Collect the relative position and motion state based on a preset time interval, respectively generate a position change sequence and a motion change sequence, construct a prediction model based on a neural network, input both the position change sequence and the motion change sequence into the prediction model, and the prediction model outputs the predicted position and predicted speed corresponding to any driverless vehicle within the prediction time based on the change amount corresponding to the preset time interval;

[0059] Generate an estimated trajectory based on the predicted position and predicted speed, the control module obtains the estimated trajectories corresponding to all the driverless vehicles, obtains the coincidence degree between the estimated trajectory and the collaborative control strategy based on the arrangement order, adjusts the estimated trajectory of any driverless vehicle based on the coincidence degree and the overall motion path to generate a standard trajectory, generates an adjustment instruction based on the standard trajectory, and sets the adjustment instruction as the adjusted motion instruction.

[0060] Specifically, in this embodiment, in order to achieve the intelligence and timeliness of the collaborative control of the fleet motion, a prediction model can be used to perform predictive analysis on the historical motion data of the driverless vehicles. Among them, both the position change sequence and the motion change sequence are historical motion data, and an LSTM (Long Short-Term Memory) neural network can be used to construct the prediction model.

[0061] The deviation degree between the estimated trajectory of the driverless vehicle and the path mode set in the collaborative control strategy can be compared through the coincidence degree. Among them, the coincidence degree can be calculated by the Euclidean distance.

[0062] As a preferred technical solution of the present invention, updating the target features of the leading vehicle includes the following steps:

[0063] Obtain the basic information of the leading vehicle based on the target features, and determine the relative distance, relative angle, and relative speed between the leading vehicle and the driverless vehicle.

[0064] Detect target feature points in the target features using the LK optical flow method, extract features from the regional map using the FAST algorithm, and then match the target feature points. Use the successfully matched map feature points to calculate the position of the unmanned vehicle.

[0065] Build a database, update the database using the map feature points, and send the updated database to the control module.

[0066] Specifically, in this embodiment, the basic information refers to the information composed of the speed change, position change, etc. of the target features of the vehicle in front. The relative distance, relative angle, and relative speed all refer to the correlation between the current unmanned vehicle and the vehicle in front. They can be calculated and generated through the target features and time interval using the technical methods corresponding to regional vision relative positioning.

[0067] The LK optical flow method (Lucas-Kanade) is a classic computer vision technology that can be used to estimate the motion information of pixel points in an image sequence. The target feature points refer to the feature points corresponding to the motion information detected and generated among all the feature points included in the target features. FAST (Features from Accelerated Segment Test) is a corner detection algorithm that can be used to extract feature points from an image. The map feature points refer to the feature points in the regional map that match the target feature points.

[0068] The database refers to the database composed of all map feature points. Since the vehicle fleet is moving, the map feature points will update the database in real time.

[0069] As a preferred technical solution of the present invention, the LK optical flow method matches the regional map corresponding to the unmanned vehicle and the vehicle in front to generate a matching map. If the matching map exists, use the RANSAC algorithm to output the matching effect of the matching map. If the matching effect is greater than the first threshold, determine it as a map matching feature point; otherwise, determine it as a non-matching feature point. If the matching map does not exist, use the LK optical flow method to extract feature points from the regional map of the unmanned vehicle and determine it as a map matching feature point. Set the map matching feature points as map feature points.

[0070] Specifically, in this embodiment, the RANSAC (RANdom Sample Consensus) algorithm is an iterative method used to estimate the parameters of a mathematical model from a set of data containing outliers. It can effectively remove outliers from the matching point pairs and estimate a reliable geometric transformation model, which is very important for the positioning and map matching of unmanned vehicles. The matching effect can be set according to the matching accuracy. If the matching effect is greater than the first threshold, set the feature points on the matching map as map matching feature points.

[0071] As a preferred technical solution of the present invention, the implementation of the motion cooperative control of the vehicle fleet includes:

[0072] Calculate the actual position of the vehicle, the position of the vehicle in front, and the driving length of the vehicle in real time. Subtract the driving length of the vehicle from the position of the vehicle in front to obtain the real-time vehicle speed of the vehicle. Estimate the vehicle speed and the heading angle of the driverless vehicle based on the real-time vehicle speed, and generate new feature points.

[0073] Send the new feature points to the control module, and repeat this step until the vehicle fleet reaches the destination.

[0074] Specifically, in this embodiment, after receiving the motion instruction, the driverless vehicle still monitors the actual position of the vehicle itself and the position of the vehicle in front. The driving length of the vehicle refers to the vehicle model length. The position of the vehicle in front changes in real time. Based on the difference between the real-time change value of the position of the vehicle in front and the driving length of the vehicle, as well as the time interval, the real-time vehicle speed of the vehicle can be calculated. Then, based on the change direction of the feature points, the vehicle speed estimation and the heading angle estimation of the driverless vehicle can be estimated. Set the feature points after the real-time change as the new feature points and transmit them to the database of the control module until the motion of the vehicle fleet ends.

[0075] As a preferred technical solution of the present invention, setting the communication sharing mode includes:

[0076] Set the communication frequency and information tags, design the communication technology based on the number of vehicle fleets, establish a communication topology diagram based on the communication frequency and communication technology, and transmit the data information corresponding to the information tags based on the communication topology diagram;

[0077] Obtain the transmission performance of the data information, optimize and adjust the communication topology diagram based on the transmission performance, and set the adjusted communication topology diagram as the communication sharing mode.

[0078] Specifically, the communication frequency refers to the transmission frequency, the information tag refers to the tag information corresponding to the data types to be transmitted, and the communication technology is designed as V2X (Vehicle-to-Everything) communication technology, which can directly exchange status information and control information. The communication topology diagram is a graphical representation describing the communication connection method between driverless vehicles, including but not limited to star topology, mesh topology, etc.

[0079] If the transmission performance is less than the preset value, modify the communication topology diagram to generate a new communication sharing mode.

[0080] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the indications of the arrows, these steps are not necessarily executed in the sequence indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0081] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The above program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0082] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0083] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention.

[0084] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for collaborative motion control of an unmanned vehicle fleet based on regional visual relative positioning, characterized in that, Including: S1: Use a camera device to collect visual information of the surrounding environment corresponding to any unmanned vehicle in real time, and combine vehicle-mounted sensor data for regional visual relative positioning to obtain the relative position and motion state of any unmanned vehicle in the vehicle fleet; S2: Set a communication sharing mode, and transmit the relative positions and motion states corresponding to all unmanned vehicles in the vehicle fleet to the control module based on the communication sharing mode; S3: The control module designs and generates a cooperative control strategy based on the relative position, adjusts the cooperative control strategy based on the motion state, and generates the overall motion path of the vehicle fleet, including updating the target features of the vehicle in front, obtaining the basic information of the vehicle in front based on the target features, determining the relative distance, relative angle, and relative speed between the vehicle in front and the unmanned vehicle; using the LK optical flow method to detect target feature points in the target features, using the FAST algorithm to extract features from the regional map and then match the target feature points, and using the successfully matched map feature points to calculate the relative position of the unmanned vehicle; construct a database, update the database using the map feature points, and send the updated database to the control module; update the regional features of the matching regional map, mark the updated target features and regional features based on the relative position, generate the arrangement order of all unmanned vehicles in the vehicle fleet, combine the motion states corresponding to all unmanned vehicles based on the arrangement order, and generate the overall motion path; the LK optical flow method matches the regional map of the unmanned vehicle and the vehicle in front to generate a matching map. If the matching map exists, use the RANSAC algorithm to output the matching effect of the matching map. If the matching effect is greater than the first threshold, it is determined as a map matching feature point, otherwise, it is determined as a non-matching feature point; if the matching map does not exist, use the LK optical flow method to extract feature points from the regional map of the unmanned vehicle and determine it as a map matching feature point; Set the map matching feature point as the map feature point; S4: Monitor the motion state of any unmanned vehicle, and dynamically generate a motion instruction for any unmanned vehicle based on the overall motion path and visual information; S5: Any unmanned vehicle receives the motion instruction based on the communication sharing mode, and adjusts the motion state based on the motion instruction to complete the motion cooperative control of the vehicle fleet.

2. The method according to claim 1, wherein Obtaining the relative position and motion state of any unmanned vehicle in the vehicle fleet includes the following steps: Extract regional features based on visual information, match them with the regional map, obtain the target features of the vehicle in front, perform regional visual relative positioning based on the target features, and generate the relative position based on the regional map and regional visual relative positioning; Obtain the driving speed and driving angle based on the vehicle-mounted sensor data, and set the driving speed and driving angle as the motion state.

3. The method according to claim 2, wherein Dynamically adjusting the motion instruction of any unmanned vehicle based on the overall motion path and visual information includes the following steps: Collect the relative position and motion state based on a preset time interval, respectively generate a position change sequence and a motion change sequence, construct a prediction model based on a neural network, input both the position change sequence and the motion change sequence into the prediction model, and the prediction model outputs the predicted position and predicted speed of any unmanned vehicle within the prediction time based on the change amount corresponding to the preset time interval; Generate an estimated trajectory based on the predicted position and predicted speed. The control module obtains the estimated trajectories corresponding to all unmanned vehicles, obtains the overlap degree between the estimated trajectory and the cooperative control strategy based on the arrangement order, adjusts the estimated trajectory of any unmanned vehicle based on the overlap degree and the overall motion path, generates a standard trajectory, generates an adjustment instruction based on the standard trajectory, and sets the adjustment instruction as the adjusted motion instruction.

4. The method according to claim 3, wherein Completing the motion cooperative control of the vehicle fleet includes: Real-time calculate the actual position of the vehicle, the position of the vehicle in front, and the driving length of the vehicle. Subtract the driving length of the vehicle from the position of the vehicle in front to obtain the real-time vehicle speed of the vehicle. Estimate the vehicle speed and heading angle of the unmanned vehicle according to the real-time vehicle speed of the vehicle, and generate a new feature point; Send the new feature point to the control module, and repeat this step until the vehicle fleet reaches the destination.

5. The method according to claim 4, wherein Setting the communication sharing mode includes: Set the communication frequency and information tags, design the communication technology based on the number of the vehicle fleet, establish a communication topology map based on the communication frequency and the communication technology, and transmit the data information corresponding to the information tags based on the communication topology map; Obtain the transmission performance of the data information, optimize and adjust the communication topology map based on the transmission performance, and set the adjusted communication topology map as the communication sharing mode.

Images