Unmanned vehicle formation dynamic control method and system

Abstract

The invention discloses an unmanned vehicle formation dynamic control method and system. The method comprises the following steps: collecting perception information of a current environment through a sensor and communication equipment; analyzing and determining the formation with the highest efficiency under the current condition according to the collected sensing information; establishing a master-vehicle and slave-vehicle formation model, and enabling each unmanned vehicle to move according to a kinematical equation of the master-vehicle and slave-vehicle formation model; an error dynamic model is established, and an error control method is designed, so that the tracking error between the slave vehicle and the master vehicle is converged to a minimum neighborhood; when a task vehicle needs to drive away from the formation, the task vehicle predicts the tracks of other vehicles and performs obstacle avoidance driving based on a dynamic prediction obstacle avoidance control method; and performing formation switching according to the current environment conditions and task requirements. According to the invention, the formation generation and maintenance capability of the unmanned vehicle formation can be improved in a complex environment, the formation obstacle avoidance strategy is optimized, and the balance between the formation and an independent task is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned vehicle formation control, and in particular to an unmanned vehicle formation dynamic control method and system. Background Art

[0002] There is an increasing demand for unmanned vehicles to perform collaborative tasks in unstructured and complex terrain. However, when multiple unmanned vehicles perform tasks autonomously at the same time, they often face multiple challenges, such as mutual avoidance, unclear priorities, and confusion in driving routes. The use of platoon driving can not only effectively improve task efficiency and collaborative work capabilities, but also improve the ability of multiple vehicles to quickly arrive at the mission area. However, there are significant problems with the non-platoon multi-vehicle driving mode, including the lack of a collaborative mechanism and inefficient path planning.

[0003] In order to meet the needs of unmanned vehicle platooning, the control architecture has developed three main structures: centralized, distributed and hybrid. The centralized control structure is globally controlled by a central processing unit and is suitable for scenarios that require global planning, but there are bottlenecks in information interaction and environmental adaptability. The distributed control structure is decided by each unmanned vehicle, which has good environmental adaptability, but is insufficient in global planning. The hybrid control structure combines the advantages of centralized and distributed control and is suitable for multi-vehicle collaborative task scenarios in complex environments.

[0004] The realization of unmanned vehicle formation control mainly relies on formation algorithms, including virtual structure method, behavior-based method and leader-follower method. Although these algorithms have achieved multi-vehicle collaboration to a certain extent, there are still model limitations, such as insufficient formation maintenance ability and poor adaptability to dynamic environments. In addition, unmanned vehicle formations face complex obstacle environments during movement, so the design of formation obstacle avoidance strategies is crucial. Artificial potential field method, rotating potential field method and grid-based obstacle avoidance methods are widely used in traditional obstacle avoidance, but they are insufficient in maintaining formation integrity and dynamic adaptability.

[0005] Existing formation control algorithms find it difficult to take into account both formation stability and dynamic adaptability at the same time, especially in narrow terrain, where a single fixed formation often cannot meet traffic needs, reducing traffic efficiency; traditional obstacle avoidance algorithms (such as the artificial potential field method) are prone to local minima problems when faced with complex obstacle distributions, resulting in unreachable targets. In addition, autonomous obstacle avoidance methods are frequently used in formations. Although they ensure the safety of individual vehicles, they can easily destroy the integrity of the formation and affect the efficiency of the formation; when the mission vehicle needs to leave the formation to perform an independent mission, the existing methods find it difficult to achieve dynamic switching between the formation and independent tasks, which may lead to formation confusion. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for dynamic control of unmanned vehicle formations, which is used to solve at least one of the above technical problems, and can improve the formation generation and maintenance capabilities of unmanned vehicle formations in complex environments, optimize the formation obstacle avoidance strategy, and achieve a balance between formations and independent tasks. By dynamically sensing environmental information, the optimal formation is automatically determined, and efficient formation tracking is achieved based on the master vehicle and slave vehicle model and error control method, providing a dynamic control solution for task vehicle departure and formation switching.

[0007] The embodiment of the present invention is achieved as follows: A method for dynamic control of an unmanned vehicle formation comprises: collecting perception information of the current environment through sensors and communication equipment.

[0008] Based on the collected perception information, the most efficient formation in the current situation is analyzed and determined.

[0009] A master-car and slave-car formation model is established, so that each unmanned vehicle moves according to the kinematic equations of the master-car and slave-car formation model.

[0010] An error dynamic model is established and an error control method is designed to make the tracking error between the slave vehicle and the master vehicle converge to the minimum neighborhood.

[0011] When a mission vehicle needs to leave the formation, the mission vehicle predicts the trajectories of other vehicles and drives to the mission point based on the dynamic prediction obstacle avoidance control method to avoid obstacles.

[0012] Switch formations according to current environmental conditions and mission requirements.

[0013] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the collecting of perception information of the current environment through sensors and communication equipment includes: Each unmanned vehicle is equipped with several sensors, including at least one of a lidar, a camera, an inertial measurement unit and a GPS, to collect perception information of the current environment in real time.

[0014] The master vehicle and the slave vehicle share the sensing information through wireless communication.

[0015] The sensory information is integrated using a sensor data fusion algorithm.

[0016] Its technical effects are: through the combined use of multiple sensors, obstacles, road features and the positions of other vehicles can be identified more accurately in complex and dynamic environments; the master vehicle and the slave vehicle share perception information in real time through wireless communication to ensure that the entire formation has a consistent understanding of the current environment. Each vehicle not only relies on its own sensors, but can also use the perception information of other vehicles to avoid blind spots and information lags, and enhance dynamic adaptability in complex environments.

[0017] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the analyzing and determining the most efficient formation under the current situation based on the collected perception information includes: Conduct environmental constraint analysis and collect data on road width, obstacle information, terrain changes, and traffic flow.

[0018] Conduct task requirement analysis to determine the tasks to be performed, the time and location requirements of the tasks, the area size and shape of the tasks, and the collaboration requirements of the tasks.

[0019] A formation efficiency evaluation model is established, and the stability index, obstacle avoidance ability index and driving efficiency index used to evaluate the formation efficiency are set. The stability index is established by calculating the relative position error and speed difference between vehicles, the obstacle avoidance ability index is established by calculating the relative distance between vehicles and the reaction time of the obstacle avoidance algorithm, and the driving efficiency index is established by calculating the time to complete the task, the path length and the speed of the formation.

[0020] The quantities of the environmental constraint analysis and the task requirement analysis are input into the formation efficiency evaluation model, and the efficiency of different formations is evaluated to obtain the formation that best suits the current environment and task.

[0021] Its technical effects are: through environmental constraint analysis (including road width, obstacles, terrain and traffic flow) and mission requirement analysis (including mission type, time, location, area size and collaboration requirements), the unmanned vehicle formation can quickly adapt to different environmental conditions and mission scenarios; through stability indicators to examine the relative position error and speed difference between vehicles in the formation, maintain the formation structure in different environments and mission scenarios, enhance the stability and collaborative control of the formation, and improve the obstacle avoidance capability and safety of the formation.

[0022] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the establishment of a master vehicle and slave vehicle formation model so that each unmanned vehicle moves according to the kinematic equation of the master vehicle and slave vehicle formation model includes: Establish the kinematic equation of the main vehicle , in, is the position of the host vehicle, is the horizontal speed of the main vehicle, is the moving speed of the main vehicle in the vertical direction, is the heading angle of the main vehicle, is the linear speed of the main vehicle, is the angular velocity of the main vehicle.

[0023] Based on the position of the master vehicle, the kinematic equation of the slave vehicle is established. , in, is the desired position of the slave vehicle, is the expected relative distance between the slave vehicle and the master vehicle, is the expected relative angle of the slave vehicle relative to the master vehicle, is the heading angle of the slave vehicle.

[0024] Its technical effect is: setting the kinematic equation of the main vehicle as a reference, so that the slave vehicle follows the main vehicle to form a predetermined formation, ensuring that the slave vehicle can track the dynamic position and heading changes of the main vehicle in real time, realize precise formation control and maintain the formation structure, and the movement behavior of the slave vehicle is coordinated with the main vehicle in space, which is conducive to adapting to dynamic environmental changes, enhancing the safety of the formation and the ability to avoid collisions, and facilitating formation adjustment and flexibility.

[0025] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the establishment of the error dynamic model and the design of the error control method so that the tracking error between the slave vehicle and the master vehicle converges to the minimum neighborhood include: Calculate the actual position error of the slave vehicle , in, is the actual position of the slave vehicle.

[0026] Establishing the error dynamic model between the master vehicle and the slave vehicle , in, is the rate of change of the actual position error.

[0027] Design the sliding surface and control law so that the actual position error Converges to zero, making the slave vehicle follow the master vehicle.

[0028] The calculation formula of the sliding surface is: ,in, , , and is the weight coefficient used to determine the rate at which the error converges.

[0029] The calculation formula of the control law is: ,in, To control the gain, , is a symbolic function.

[0030] Its technical effect is: by establishing an error dynamic model, the actual position error of the slave vehicle is calculated in real time, and the speed and angular velocity of the slave vehicle are continuously adjusted under the action of the control law, so that the error converges to the minimum neighborhood, ensuring that the slave vehicle stably follows the main vehicle and accurately maintains the desired formation; through the guidance of the sliding mode control law, the movement adjustment of the slave vehicle no longer relies on simple proportional control, but is based on the design of the error dynamic model and the sliding surface, making the control adjustment smoother and more stable, preventing large fluctuations and drastic changes in the movement of the slave vehicle, improving the stability and accuracy of the formation control, and reducing unnecessary following interference to the main vehicle.

[0031] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, when a task vehicle needs to leave the formation, the task vehicle predicts the trajectory of other vehicles and performs obstacle avoidance driving based on the dynamic prediction obstacle avoidance control method, and heading to the task point includes: When a task vehicle needs to leave the formation, a trajectory prediction model for other vehicles is established ,in, is the position of the other vehicle in the two-dimensional space, are the x and y components of the velocity of the other vehicle, is the heading angle of the other vehicle.

[0032] Based on time step Update the state of the trajectory prediction model , To describe the state transfer matrix of vehicle dynamics, is the input matrix describing the effect of control input on the state, is the control input, , are the accelerations of the other vehicles in the x and y directions, is the steering angular velocity of the other vehicle.

[0033] Calculate the future time Trajectory prediction , is the position of the other vehicle at the current time t, , is the speed of the other vehicle, .

[0034] The current position of the vehicle for the mission , the target position of the mission vehicle and the predicted trajectories of the other vehicles , set the obstacle avoidance strategy to minimize the distance between the trajectory of the task vehicle and the target position , establish the constraints for the trajectory of the task vehicle to avoid collision with obstacles ,in, For a safe distance.

[0035] By using model predictive control, we can find the optimal control input , the acceleration and steering angular velocity of the task vehicle are obtained, where is the weight coefficient.

[0036] According to the obtained acceleration and steering angular velocity of the task vehicle, a path planning algorithm is executed to generate a driving trajectory of the task vehicle so that the task vehicle goes to the task point.

[0037] Its technical effects are: based on the time step state update and model predictive control (MPC) method, the task vehicle can quickly obtain the future trajectory, predict and dynamically adjust the control strategy. The task vehicle can flexibly and safely leave the formation and go to the mission point through trajectory prediction, obstacle avoidance planning and path optimization, which improves the real-time performance and flexibility of obstacle avoidance decisions, enhances the collaboration of the formation system, avoids interference with other vehicles, and realizes the organic combination of path planning and obstacle avoidance strategy.

[0038] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, when the task vehicle needs to leave the formation and head to the target position, the obstacle position information is updated in real time based on the perception information to correct the dynamic prediction obstacle avoidance control method.

[0039] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the switching of formations according to current environmental conditions and task requirements includes: Setting up the environmental conditions matrix ,in, Represents the state of the kth obstacle in the environment.

[0040] Setting up the task requirements matrix ,in, Represents the lth goal of the task.

[0041] Define the formation switching priority function ,in, is the environmental weight, is the task weight, Quantify the priority of environmental conditions, Quantify the value of the task condition priority.

[0042] Based on the value of the priority function, the current optimal formation is determined, wherein each formation is determined by the position relative to the main vehicle. The target offset , calculate the target position of each slave vehicle .

[0043] Calculate the target position error of the slave vehicle .

[0044] Design a proportional-differential control method to make the target position error converge to zero, from the vehicle line speed , from the vehicle angular velocity ,in, is the proportional gain, is the differential gain.

[0045] According to the obtained linear velocity and angular velocity of the slave vehicle, a path planning algorithm is executed to generate a driving trajectory of the slave vehicle and perform formation switching.

[0046] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, when the formation is switched, when the current environmental conditions and task requirements change, the priority function is recalculated to select a new optimal formation.

[0047] Each time the formation is switched, the target position and target position error of the slave vehicle are updated, and the movement of the slave vehicle is controlled according to the new formation.

[0048] Its technical effect is: by setting the environmental condition matrix and calculating the formation switching priority through the priority function, the formation can be automatically adjusted in a complex and uncertain environment, obstacles can be effectively avoided, and the environmental adaptability of the formation can be improved. The priority function combines the environmental weight and the task weight to automatically select the current optimal formation and realize the intelligent switching of the formation. The proportional-differential control method (PD control) is used to converge the target position error of the slave vehicle to zero, so that the slave vehicle can stably follow the main vehicle to switch formation, effectively reduce the target position error, and reduce the oscillation and instability caused by formation switching, optimize the motion control of the slave vehicle, and ensure the stability of the formation.

[0049] An unmanned vehicle formation dynamic control system, comprising: The perception information collection module is used to collect perception information of the current environment through sensors and communication devices.

[0050] The formation analysis module is used to analyze and determine the most efficient formation in the current situation based on the collected perception information.

[0051] The formation model establishment module is used to establish a master-car and slave-car formation model so that each unmanned vehicle moves according to the kinematic equations of the master-car and slave-car formation model.

[0052] The error control module is used to establish an error dynamic model and design an error control method so that the tracking error between the slave vehicle and the master vehicle converges to the minimum neighborhood.

[0053] The task execution obstacle avoidance control module is used to predict the trajectory of other vehicles when a task vehicle needs to leave the formation, and avoid obstacles based on the dynamic prediction obstacle avoidance control method to go to the task point.

[0054] The formation switching module is used to switch the formation according to the current environmental conditions and task requirements.

[0055] The beneficial effects of the embodiments of the present invention are: The present invention establishes a master-slave formation model and designs a control method based on an error dynamic model, thereby achieving the convergence of the tracking error between the slave and master vehicles, thereby ensuring the overall stability and coordination of the formation. During the formation movement, the error control module enables the slave vehicle to accurately follow the master vehicle, and achieves rapid convergence of the slave vehicle target position error through the proportional-differential control method, avoiding the formation instability problem caused by large tracking errors in traditional methods.

[0056] When the task vehicle needs to leave the formation and go to a specific task location, the present invention can predict the trajectory of other vehicles in real time and use the dynamic prediction obstacle avoidance control method for planning to avoid collisions between the task vehicle and other vehicles during independent action. At the same time, the task execution obstacle avoidance control module updates the obstacle position information in real time during the task vehicle's dynamic prediction obstacle avoidance, achieving efficient response to complex environments, greatly improving the autonomy and safety of the task vehicle, and is particularly suitable for dynamically changing scenes.

[0057] The present invention sets up an environmental condition matrix and a task requirement matrix, defines a formation switching priority function, and dynamically switches formations in combination with the current environment and task requirements, ensuring that the optimal formation mode is selected under different tasks and environments. This flexible formation switching mechanism enables the unmanned vehicle formation to efficiently adapt to environmental changes and meet task requirements, thereby improving the overall efficiency of the formation and the effectiveness of task completion.

[0058] The formation analysis module proposed in the present invention establishes a formation efficiency evaluation model by analyzing environmental constraints and mission requirements, combines multiple key indicators such as stability, obstacle avoidance and driving efficiency, and automatically evaluates and determines the formation that best suits the current environment. Compared with traditional fixed formations, the automated evaluation and decision-making mechanism of the present invention greatly improves the operational efficiency and environmental adaptability of the formation.

[0059] Environmental perception information is collected in real time through various sensors (such as lidar, camera, inertial measurement unit, GPS, etc.) of each unmanned vehicle, and information is shared through wireless communication. The sensor data fusion algorithm is combined for integration and processing to ensure the accuracy and real-time nature of environmental perception. This data fusion method significantly reduces the error and time lag of perception information, and improves the navigation and decision-making capabilities of unmanned vehicles in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 The figure is a flow chart of the unmanned vehicle formation dynamic control method of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0063] Please refer to Figure 1 The first embodiment of the present invention provides a method for dynamic control of an unmanned vehicle formation, which includes: collecting perception information of the current environment through sensors and communication equipment; analyzing and determining the most efficient formation under the current circumstances based on the collected perception information; establishing a master vehicle and slave vehicle formation model, so that each unmanned vehicle moves according to the kinematic equation of the master vehicle and slave vehicle formation model; establishing an error dynamic model and designing an error control method so that the tracking error between the slave vehicle and the master vehicle converges to a minimum neighborhood; when a task vehicle needs to leave the formation, the task vehicle predicts the trajectory of other vehicles, and performs obstacle avoidance driving based on a dynamic prediction obstacle avoidance control method to go to the task point; and switching the formation according to the current environmental conditions and task requirements.

[0064] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the collection of perception information of the current environment through sensors and communication equipment includes: each unmanned vehicle is equipped with a number of sensors, and the sensors include at least one of a lidar, a camera, an inertial measurement unit and a GPS, which collects perception information of the current environment in real time; the master vehicle and the slave vehicle share the perception information through wireless communication; and the sensor data fusion algorithm is used to integrate the perception information.

[0065] Its technical effects are: through the combined use of multiple sensors, obstacles, road features and the positions of other vehicles can be identified more accurately in complex and dynamic environments; the master vehicle and the slave vehicle share perception information in real time through wireless communication to ensure that the entire formation has a consistent understanding of the current environment. Each vehicle not only relies on its own sensors, but can also use the perception information of other vehicles to avoid blind spots and information lags, and enhance dynamic adaptability in complex environments.

[0066] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the analysis and determination of the most efficient formation under the current situation based on the collected perception information includes: conducting environmental constraint analysis, collecting road width, obstacle information, terrain changes and traffic flow data; conducting task requirement analysis, determining the tasks to be performed, the time and location requirements of the tasks, the area size and form of the tasks, and the collaboration requirements of the tasks; establishing a formation efficiency evaluation model, setting stability indicators, obstacle avoidance ability indicators and driving efficiency indicators for evaluating formation efficiency, the stability index is established by calculating the relative position error and speed difference between vehicles, the obstacle avoidance ability index is established by calculating the relative distance between vehicles and the reaction time of the obstacle avoidance algorithm, and the driving efficiency index is established by calculating the time to complete the task, the path length and speed of the formation; the quantities of the environmental constraint analysis and the task requirement analysis are input into the formation efficiency evaluation model, and the efficiency of different formations (such as regular triangles, inverted triangles, vertical formations, horizontal formations, etc.) is evaluated to obtain the formation that best suits the current environment and task.

[0067] Specifically, for environmental constraint analysis, calculate the minimum width of the formation , if the road width ≥Minimum width , then the formation is feasible, otherwise the formation needs to be adjusted, where For the i-th car, is the occupancy factor of the i-th vehicle in the convoy, which depends on the formation type and the spacing between vehicles; calculates the safe distance between the vehicle and the obstacle ,in, is the minimum safe distance of the vehicle, is the adjustment factor determined by obstacle type and environment, Obstacle density is used to measure the number of obstacles per unit distance; slope θ and path curvature κ are collected; traffic flow density is calculated ,in, is the traffic flow per unit length, is the length of the road section under consideration; for task requirement analysis, collect the maximum allowed time for the task , the target location of the task , width of the task area and shape and collaboration requirements; calculate relative position error ,in, and are the positions of the two cars i and j, is the Euclidean distance between the two vehicles; calculate the relative speed error ,in, and are the speeds of the two cars i and j, is the speed difference; calculate the minimum distance safety threshold between vehicles ; Calculate the threshold of obstacle avoidance reaction time ,in, is the maximum speed of the vehicle; calculate the task completion time ,in, is the total path length of the task, is the average speed; calculate the formation path length ; Establish a formation efficiency evaluation model .

[0068] The equilateral triangle formation is suitable for relatively open roads, and can better maintain the stability of the formation while facilitating task coordination; the inverted triangle formation is suitable for more compact task scenarios, and can improve the field of vision of the leading vehicle, allowing the rear vehicles to better follow the main vehicle; the vertical formation is more suitable for narrow roads and can effectively avoid obstacles ahead; the horizontal formation can increase the task coverage rate under certain specific task requirements (such as full coverage of a specific area).

[0069] Its technical effects are: through environmental constraint analysis (including road width, obstacles, terrain and traffic flow) and mission requirement analysis (including mission type, time, location, area size and collaboration requirements), the unmanned vehicle formation can quickly adapt to different environmental conditions and mission scenarios; through stability indicators to examine the relative position error and speed difference between vehicles in the formation, maintain the formation structure in different environments and mission scenarios, enhance the stability and collaborative control of the formation, and improve the obstacle avoidance capability and safety of the formation.

[0070] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the establishment of the master vehicle and slave vehicle formation model so that each unmanned vehicle moves according to the kinematic equation of the master vehicle and slave vehicle formation model includes: establishing the kinematic equation of the master vehicle ,in, is the position of the host vehicle, is the horizontal speed of the main vehicle, is the moving speed of the main vehicle in the vertical direction, is the heading angle of the main vehicle, is the linear speed of the main vehicle, is the angular velocity of the master vehicle; based on the position of the master vehicle, the kinematic equation of the slave vehicle is established ,in, is the desired position of the slave vehicle, is the expected relative distance between the slave vehicle and the master vehicle, is the expected relative angle of the slave vehicle relative to the master vehicle, is the heading angle of the slave vehicle.

[0071] Its technical effect is: setting the kinematic equation of the main vehicle as a reference, so that the slave vehicle follows the main vehicle to form a predetermined formation, ensuring that the slave vehicle can track the dynamic position and heading changes of the main vehicle in real time, realize precise formation control and maintain the formation structure, and the movement behavior of the slave vehicle is coordinated with the main vehicle in space, which is conducive to adapting to dynamic environmental changes, enhancing the safety of the formation and the ability to avoid collisions, and facilitating formation adjustment and flexibility.

[0072] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the establishment of the error dynamic model and the design of the error control method so that the tracking error between the slave vehicle and the master vehicle converges to the minimum neighborhood include: calculating the actual position error of the slave vehicle ,in, is the actual position of the slave vehicle; establishing a dynamic model of the error between the master vehicle and the slave vehicle ,in, is the rate of change of the actual position error; design the sliding surface and control law so that the actual position error Converges to zero, so that the slave vehicle follows the master vehicle; the calculation formula of the sliding surface is: ,in, , , and is the weight coefficient used to determine the rate of error convergence; the calculation formula of the control law is: ,in, To control the gain, , is a symbolic function.

[0073] The error dynamic model describes the law of error change. The sliding surface provides a target for guiding the error convergence for the control system, and the control law makes the error converge to zero by adjusting the speed and angular velocity of the slave vehicle, ensuring that the slave vehicle can always maintain the desired formation with the master vehicle.

[0074] Its technical effect is: by establishing an error dynamic model, the actual position error of the slave vehicle is calculated in real time, and the speed and angular velocity of the slave vehicle are continuously adjusted under the action of the control law, so that the error converges to the minimum neighborhood, ensuring that the slave vehicle stably follows the main vehicle and accurately maintains the desired formation; through the guidance of the sliding mode control law, the movement adjustment of the slave vehicle no longer relies on simple proportional control, but is based on the design of the error dynamic model and the sliding surface, making the control adjustment smoother and more stable, preventing large fluctuations and drastic changes in the movement of the slave vehicle, improving the stability and accuracy of the formation control, and reducing unnecessary following interference to the main vehicle.

[0075] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, when a task vehicle needs to leave the formation, the task vehicle predicts the trajectory of other vehicles and avoids obstacles based on the dynamic prediction obstacle avoidance control method to go to the task point, including: when a task vehicle needs to leave the formation, establish a trajectory prediction model for other vehicles ,in, is the position of the other vehicle in the two-dimensional space, are the x and y components of the velocity of the other vehicle, is the heading angle of the other vehicle; based on the time step Update the state of the trajectory prediction model , To describe the state transfer matrix of vehicle dynamics, is the input matrix describing the effect of control input on the state, is the control input, , are the accelerations of the other vehicles in the x and y directions, is the steering angular velocity of the other vehicle; calculated at the future time Trajectory prediction , is the position of the other vehicle at the current time t, , is the speed of the other vehicle, ; The current position of the mission vehicle , the target position of the mission vehicle and the predicted trajectories of the other vehicles , set the obstacle avoidance strategy to minimize the distance between the trajectory of the task vehicle and the target position , establish the constraints for the trajectory of the task vehicle to avoid collision with obstacles ,in, is a safe distance; the optimal control input is solved by the model predictive control method , the acceleration and steering angular velocity of the task vehicle are obtained, where is the weight coefficient; according to the obtained acceleration and steering angular velocity of the task vehicle, a path planning algorithm is executed to generate a driving trajectory of the task vehicle so that the task vehicle goes to the task point.

[0076] Its technical effects are: based on the time step state update and model predictive control (MPC) method, the task vehicle can quickly obtain the future trajectory, predict and dynamically adjust the control strategy. The task vehicle can flexibly and safely leave the formation and go to the mission point through trajectory prediction, obstacle avoidance planning and path optimization, which improves the real-time performance and flexibility of obstacle avoidance decisions, enhances the collaboration of the formation system, avoids interference with other vehicles, and realizes the organic combination of path planning and obstacle avoidance strategy.

[0077] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, when the task vehicle needs to leave the formation and head to the target position, the obstacle position information is updated in real time based on the perception information to correct the dynamic prediction obstacle avoidance control method.

[0078] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, the formation switching according to the current environmental conditions and task requirements includes: setting the environmental condition matrix ,in, Represents the state of the kth obstacle in the environment; sets the task requirement matrix ,in, Indicates the lth target of the task; defines the formation switching priority function ,in, is the environmental weight, is the task weight, Quantify the priority of environmental conditions, is a quantitative value of the task condition priority; based on the value of the priority function, the current optimal formation is determined, wherein each formation is determined by the position relative to the main vehicle The target offset , calculate the target position of each slave vehicle ; Calculate the target position error of the slave vehicle ; Design proportional-differential control method to make the target position error converge to zero, from the vehicle line speed , from the vehicle angular velocity ,in, is the proportional gain, is the differential gain; according to the obtained linear velocity and angular velocity of the slave vehicle, a path planning algorithm is executed to generate the driving trajectory of the slave vehicle and perform formation switching.

[0079] In a preferred embodiment of the present invention, in the above-mentioned unmanned vehicle formation dynamic control method, when the formation is switched, when the current environmental conditions and task requirements change, the priority function is recalculated and a new optimal formation is selected; each time the formation is switched, the target position and target position error of the slave vehicle are updated, and the movement of the slave vehicle is controlled according to the new formation.

[0080] Its technical effect is: by setting the environmental condition matrix and calculating the formation switching priority through the priority function, the formation can be automatically adjusted in a complex and uncertain environment, obstacles can be effectively avoided, and the environmental adaptability of the formation can be improved. The priority function combines the environmental weight and the task weight to automatically select the current optimal formation and realize the intelligent switching of the formation. The proportional-differential control method (PD control) is used to converge the target position error of the slave vehicle to zero, so that the slave vehicle can stably follow the main vehicle to switch formation, effectively reduce the target position error, and reduce the oscillation and instability caused by formation switching, optimize the motion control of the slave vehicle, and ensure the stability of the formation.

[0081] The second embodiment of the present invention provides a dynamic control system for an unmanned vehicle formation, which includes: a perception information acquisition module, which is used to collect perception information of the current environment through sensors and communication equipment; a formation analysis module, which is used to analyze and determine the most efficient formation under the current circumstances based on the collected perception information; a formation model establishment module, which is used to establish a master vehicle and slave vehicle formation model, so that each unmanned vehicle moves according to the kinematic equation of the master vehicle and slave vehicle formation model; an error control module, which is used to establish an error dynamic model and design an error control method so that the tracking error between the slave vehicle and the master vehicle converges to a minimum neighborhood; a task execution obstacle avoidance control module, which is used when a task vehicle needs to leave the formation, the task vehicle predicts the trajectory of other vehicles, and performs obstacle avoidance driving based on a dynamic prediction obstacle avoidance control method to go to the task point; a formation switching module, which is used to switch the formation according to the current environmental conditions and task requirements.

[0082] The computer program product of the unmanned vehicle formation dynamic control method and device provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the previous method embodiments. The specific implementation can be found in the method embodiments, which will not be repeated here.

[0083] Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, it can execute the above-mentioned unmanned vehicle formation dynamic control method, thereby improving the formation generation and maintenance capabilities of the unmanned vehicle formation in a complex environment, optimizing the formation obstacle avoidance strategy, and achieving a balance between the formation and independent tasks.

[0084] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that can be executed by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0085] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for dynamic control of an unmanned vehicle formation, characterized in that: include: Collecting sensory information about the current environment through sensors and communication devices; Analyze and determine the most efficient formation under the current circumstances based on the collected perception information; Establishing a master-car-slave-car formation model, so that each unmanned vehicle moves according to the kinematic equation of the master-car-slave-car formation model; Establish error dynamic model and design error control method to make the tracking error between the slave vehicle and the master vehicle converge to the minimum neighborhood; When a task vehicle needs to leave the formation, the task vehicle predicts the trajectory of other vehicles and drives to the task point based on the dynamic prediction obstacle avoidance control method to avoid obstacles; Switch formations according to current environmental conditions and mission requirements.

2. The unmanned vehicle formation dynamic control method according to claim 1, characterized in that: The sensing information of the current environment collected by the sensor and the communication device includes: Each unmanned vehicle is equipped with a number of sensors, including at least one of a laser radar, a camera, an inertial measurement unit, and a GPS, to collect perception information of the current environment in real time; The master vehicle and the slave vehicle share the sensing information via wireless communication; The sensory information is integrated using a sensor data fusion algorithm.

3. The unmanned vehicle formation dynamic control method according to claim 1, characterized in that: The analyzing and determining the most efficient formation in the current situation based on the collected perception information includes: Conduct environmental constraint analysis to collect road width, obstacle information, terrain changes, and traffic flow data; Conduct task requirement analysis to determine the tasks to be performed, the time and location requirements of the tasks, the size and shape of the task area, and the collaborative requirements of the tasks; Establish a formation efficiency evaluation model, set stability index, obstacle avoidance index and driving efficiency index for evaluating formation efficiency, the stability index is established by calculating the relative position error and speed difference between vehicles, the obstacle avoidance index is established by calculating the relative distance between vehicles and the reaction time of the obstacle avoidance algorithm, and the driving efficiency index is established by calculating the time to complete the task, the path length and speed of the formation; The quantities of the environmental constraint analysis and the task requirement analysis are input into the formation efficiency evaluation model, and the efficiency of different formations is evaluated to obtain the formation that best suits the current environment and task.

4. The unmanned vehicle formation dynamic control method according to claim 1, characterized in that: The step of establishing a master-car-slave-car formation model so that each unmanned vehicle moves according to the kinematic equation of the master-car-slave-car formation model includes: Establish the kinematic equation of the main vehicle , in, is the position of the host vehicle, is the horizontal speed of the main vehicle, is the moving speed of the main vehicle in the vertical direction, is the heading angle of the main vehicle, is the linear speed of the main vehicle, is the angular velocity of the main vehicle; Based on the position of the master vehicle, the kinematic equation of the slave vehicle is established. , in, is the desired position of the slave vehicle, is the expected relative distance between the slave vehicle and the master vehicle, is the expected relative angle of the slave vehicle relative to the master vehicle, is the heading angle of the slave vehicle.

5. The unmanned vehicle formation dynamic control method according to claim 4, characterized in that: The establishment of the error dynamic model and the design of the error control method so that the tracking error between the slave vehicle and the master vehicle converges to a minimum neighborhood includes: Calculate the actual position error of the slave vehicle , in, is the actual position of the slave vehicle; Establishing the error dynamic model between the master vehicle and the slave vehicle , in, is the rate of change of the actual position error; Design the sliding surface and control law so that the actual position error Converge to zero, so that the slave vehicle follows the master vehicle; The calculation formula of the sliding surface is: ,in, , , and is the weight coefficient used to determine the rate of error convergence; The calculation formula of the control law is: ,in, To control the gain, , is a symbolic function.

6. The unmanned vehicle formation dynamic control method according to claim 1, characterized in that: When a task vehicle needs to leave the formation, the task vehicle predicts the trajectory of other vehicles and performs obstacle avoidance driving based on the dynamic prediction obstacle avoidance control method to go to the task point, including: When a task vehicle needs to leave the formation, a trajectory prediction model for other vehicles is established ,in, is the position of the other vehicle in the two-dimensional space, are the x and y components of the velocity of the other vehicle, is the heading angle of the other vehicle; Based on time step Update the state of the trajectory prediction model , To describe the state transfer matrix of vehicle dynamics, is the input matrix describing the effect of control input on the state, is the control input, , are the accelerations of the other vehicles in the x and y directions, is the steering angular velocity of the other vehicle; Calculate the future time Trajectory prediction , is the position of the other vehicle at the current time t, , is the speed of the other vehicle, ; The current position of the vehicle for the mission , the target position of the mission vehicle and the predicted trajectories of the other vehicles , set the obstacle avoidance strategy to minimize the distance between the trajectory of the task vehicle and the target position , establish the constraints for the trajectory of the task vehicle to avoid collision with obstacles ,in, For safe distance; By using model predictive control, we can find the optimal control input , the acceleration and steering angular velocity of the task vehicle are obtained, where is the weight coefficient; According to the obtained acceleration and steering angular velocity of the task vehicle, a path planning algorithm is executed to generate a driving trajectory of the task vehicle so that the task vehicle goes to the task point.

7. The unmanned vehicle formation dynamic control method according to claim 6, characterized in that: When the task vehicle needs to leave the formation and head to the target position, the obstacle position information is updated in real time based on the perception information, and the dynamic prediction obstacle avoidance control method is corrected.

8. The unmanned vehicle formation dynamic control method according to claim 1, characterized in that: The switching of formations according to current environmental conditions and task requirements includes: Setting up the environmental conditions matrix ,in, Represents the state of the kth obstacle in the environment; Setting up the task requirements matrix ,in, represents the lth goal of the task; Define the formation switching priority function ,in, is the environmental weight, is the task weight, Quantify the priority of environmental conditions, Quantify the value of the task condition priority; Based on the value of the priority function, the current optimal formation is determined, wherein each formation is determined by the position relative to the main vehicle. The target offset , calculate the target position of each slave vehicle ; Calculate the target position error of the slave vehicle ; Design a proportional-differential control method to make the target position error converge to zero, from the vehicle line speed , from the vehicle angular velocity ,in, is the proportional gain, is the differential gain; According to the obtained linear velocity and angular velocity of the slave vehicle, a path planning algorithm is executed to generate a driving trajectory of the slave vehicle and perform formation switching.

9. The unmanned vehicle formation dynamic control method according to claim 8, characterized in that: When switching formations, if the current environmental conditions and task requirements change, the priority function is recalculated to select a new optimal formation; Each time the formation is switched, the target position and target position error of the slave vehicle are updated, and the movement of the slave vehicle is controlled according to the new formation.

10. A dynamic control system for an unmanned vehicle formation, characterized in that: include: A perception information collection module is used to collect perception information of the current environment through sensors and communication devices; A formation analysis module, used to analyze and determine the most efficient formation in the current situation based on the collected perception information; A formation model building module is used to build a master-slave formation model so that each unmanned vehicle moves according to the kinematic equation of the master-slave formation model; The error control module is used to establish the error dynamic model and design the error control method so that the tracking error between the slave vehicle and the master vehicle converges to the minimum neighborhood; The task execution obstacle avoidance control module is used for predicting the trajectory of other vehicles when a task vehicle needs to leave the formation, and performing obstacle avoidance driving based on the dynamic prediction obstacle avoidance control method to reach the task point; The formation switching module is used to switch the formation according to the current environmental conditions and task requirements.

Images