Path planning and control method for UAV swarm based on tunnel preset performance
An adaptive control law constructed using a tunnel preset performance strategy and a distributed learning strategy solves the problems of overshoot and uncertainty in UAV swarm path planning, achieving higher-precision path tracking control.
Patent Information
- Application Number
- CN202410343637.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-25
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-03-25
AI Technical Summary
Existing path planning control methods suffer from neglecting performance constraints in practical applications, leading to significant uncertainties and overshoot, which are particularly difficult to address effectively in multi-UAV collaborative control.
A tunnel-preset performance strategy is adopted to transform the path error, construct a kinematic control law, and combine neural networks and distributed learning strategies to build an adaptive control law. A distributed path controller is designed to realize the path tracking control of the UAV swarm.
It effectively reduces overshoot, lowers uncertainty, and improves the path tracking accuracy of UAV swarms in both transient and steady-state processes.
Smart Images

Figure CN118331297B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned aerial vehicle cluster control, and in particular to a path planning and control method for an unmanned aerial vehicle cluster based on preset tunnel performance. Background Art
[0002] In the current field of robotic automatic control technology, with the continuous development of UAV and other robotic technologies, the demand for collaborative control is growing. Nowadays, multi-UAV collaborative control has become one of the research hotspots. Given its wide application in various engineering fields, it mainly studies target tracking, trajectory tracking and path planning. In particular, path-based collaborative control aims to coordinate the motion of all UAVs along several parameterized paths.
[0003] Many researchers have proposed various algorithms for path planning control methods, such as path maneuver control methods designed for path updating to reduce control effort; path guidance consensus maneuver control methods based on leader-follower schemes for SISO nonlinear systems; and modular distributed maneuver controllers based on neural network predictors for nonlinear multi-agent systems. However, the above path maneuver control methods are based on ideal system performance, but in practical applications, performance constraints cannot be easily ignored, which may include overshoot and tracking error limits. In recent years, preset performance control (PPC) has provided an effective method to enhance stability and performance constraints by combining error transformation with preset performance, but for distributed path maneuver control, these methods may face loose performance constraints due to the existence of distributed funnel boundaries and symmetric global performance functions.
[0004] Therefore, the traditional path planning control method has the problem of uncertainty because the performance constraints are ignored. Summary of the invention
[0005] Based on this, in order to solve the above technical problems, a UAV cluster path planning and control method based on tunnel preset performance is provided, which can effectively reduce overshoot and reduce uncertainty.
[0006] A method for controlling a cluster of unmanned aerial vehicles (UAVs) based on preset tunnel performance, wherein the cluster of UAVs includes a virtual leader and a follower, and the method includes:
[0007] For the follower, a path error is defined, and the path error is transformed by using a tunnel preset performance strategy to construct a kinematic control law based on the tunnel preset performance;
[0008] Determine the path parameter variables of the virtual leader, define the path variable dynamics and parameter errors, introduce feedback laws for each of the path parameter variables, and construct a path update law;
[0009] For the follower, a velocity tracking error at a dynamic level is defined, a neural network and a state estimator are introduced, and a distributed learning strategy is used to construct a DNP-based adaptive control law;
[0010] The kinematic control law, the path update law, and the DNP-based adaptive control law are used as control signals, and a distributed path controller is designed based on the control signals;
[0011] According to the distributed path controller, the UAV cluster path tracking control is performed.
[0012] In one of the embodiments, for the follower, a path error is defined, including:
[0013] Define the distributed path error of the i-th follower as Right now
[0014] Derivative of the path error yields
[0015] Where M represents the number of followers, NM represents the number of virtual leaders, and p r,j (θ j ) represents the parameterized path of the jth virtual leader, and p r,j (θ j ) is bounded and satisfies Conditions; a i,j (p i -p j ) represents the position difference between the ith follower and other followers, a i,j (p i -p r,j (θ j )) represents the position difference between the i-th follower and the virtual leader; a i,j represents the communication connection weight, d i represents the sum of the communication connection weights between followers, that is,
[0016] In one embodiment, the path error is transformed using a preset tunnel performance strategy, including:
[0017] The distributed path error designed to meet the preset tunnel performance strategy constraints is: in, Ei,k and is a time-varying function, in, Performance functions
[0018] Substitute Among them, G i,k (δ i,k ): (-∞, ∞)→(-1, 1) represents an increasing transformation function, and δi,k corresponds to a function equivalent to E i,k The unconstrained transformation error of
[0019] Transformation function G i,k (δ i,k ) is represented by G i,k (δ i,k )=(exp(δ i,k )-exp(-δ i,k )) / (exp(δ i,k )+exp(-δ i,k ));
[0020] By G i,k (δ i,k ) to obtain the transformed error
[0021] In one embodiment, a kinematic control law based on preset tunnel performance is constructed, including:
[0022] Construct a kinematic control law based on the preset performance of the tunnel i for: in, is the positive definite control gain matrix;
[0023] Substituting into δ i The dynamics of
[0024] In one embodiment, the path parameter variables of the virtual leader are determined, the path variable dynamics and parameter errors are defined, and feedback rules are introduced for each of the path parameter variables to construct a path update law, including:
[0025] Define the dynamics of the path parameter variables of the jth virtual leader as
[0026] Based on the leader-follower structure, the design parameter error is:
[0027] Introducing the feedback law, the path update law is constructed as:
[0028] in, is a gain, and are two tuning parameters.
[0029] In one embodiment, for the follower, the velocity tracking error at the dynamic level is defined as: E i,2 =v i -α i ; Calculate E i,2 The dynamics of is:
[0030] In one embodiment, the method further comprises:
[0031] The ideal dynamic control law is determined as: in, is a positive definite control gain matrix.
[0032] In one embodiment, a neural network and a state estimator are introduced, including:
[0033] Using Neural Networks in, represents an optional weight vector, p is the number of neurons, is a known smooth activation function, is the input vector, The time-varying approximation error satisfies
[0034] Introducing the neural network, the actual dynamic control law is determined as
[0035] In one embodiment, a distributed learning strategy is used to construct a DNP-based adaptive control law: in, is the adaptive gain, is an additional positive definite tuning parameter matrix used to shape the transient learning behavior of the neural network, σ i ∈(0, 1) is a correction factor.
[0036] The above-mentioned UAV cluster path planning control method based on tunnel preset performance constructs a kinematic control law by introducing a tunnel preset performance strategy, which can limit the overshoot of the distributed path error in transient and steady-state processes; a distributed learning strategy is used to construct an adaptive control law based on DNP, so that the generalization ability of the distributed path controller is enhanced under the distributed learning strategy, and the uncertainty in the robot dynamics is effectively estimated, which can effectively reduce the overshoot and reduce the uncertainty. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 This is an application environment diagram of a UAV cluster path planning and control method based on tunnel preset performance in one embodiment;
[0038] Figure 2 It is a flowchart of a method for controlling a cluster path of unmanned aerial vehicles based on preset tunnel performance in one embodiment;
[0039] Figure 3 Schematic diagram of the path update law for three virtual leaders;
[0040] Figure 4 Schematic diagram of the comparison between DNP and neural predictor;
[0041] Figure 5 Schematic diagram of follower control input based on tunnel preset performance policy TPP;
[0042] Figure 6 The schematic diagram of the curve trajectory of the virtual leader and six followers based on the tunnel preset performance strategy TPP and DNP control;
[0043] Figure 7 Schematic diagram of the circular trajectory of the virtual leader and six followers based on the tunnel preset performance strategy TPP and DNP control;
[0044] Figure 8 Schematic diagram of the comparison between TPP-based UAV tracking error analysis and traditional preset performance control (PPC) error analysis. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0046] The UAV cluster path planning and control method based on tunnel preset performance provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Figure 1 As shown, the application environment includes a virtual leader 110 and a plurality of followers 120. In this embodiment, the followers 120 may include 6, namely UAV1-6. The UAV cluster path planning control method based on the preset performance of the tunnel can be implemented by simulation software, wherein the simulation software can be MATLAB2022, and the operating environment is Windows.
[0047] In one embodiment, Figure 2 As shown, a UAV cluster path planning control method based on tunnel preset performance is provided, comprising the following steps:
[0048] Step 202 , for the follower, a path error is defined, and the path error is transformed using a tunnel preset performance strategy to construct a kinematic control law based on the tunnel preset performance.
[0049] Before constructing the kinematic control law, a controller design may be performed first, and the controller may control the motion of the virtual leader and the followers. In this embodiment, the drone cluster includes at least one virtual leader, and the virtual leader has a path leading to each follower.
[0050] In order to achieve the distributed path error E of UAV cluster under multi-parameter path guidance i,1 =[E i,x , E i,y , E i,z ], the distributed coordination under the condition of , the following distributed path error of the ith follower can be defined as Right now Derivative of the path error yields Where M represents the number of followers, NM represents the number of virtual leaders, and p r,j (θ j ) represents the parameterized path of the jth virtual leader, and p r,j (θ j ) is bounded and satisfies Conditions; a i,j (p i -p j ) represents the position difference between the ith follower and other followers, a i,j (p i -p r,j (θ j )) represents the position difference between the i-th follower and the virtual leader; a i,j represents the communication connection weight, and di represents the sum of the communication connection weights between followers, that is,
[0051] In one embodiment, in order to improve the transient and steady-state performance of the UAV, a distributed path error that satisfies the tunnel preset performance policy TPP constraint can be designed. Specifically, the designed distributed path error that satisfies the tunnel preset performance policy constraint is: Among them, E i,k and is a time-varying function, in, Performance functions Substitute Among them, G i,k (δi,k ): (-∞, ∞)→(-1, 1) represents an increasing transformation function, and δi,k corresponds to a function equivalent to E i,k The unconstrained transformation error of i,k (δ i,k ) is represented by G i,k (δ i,k )=(exp(δ i,k )-exp(-δ i,k )) / (exp(δ i,k )+exp(-δ i,k )); by G i,k (δ i,k ) to obtain the transformed error
[0052] In one embodiment, in order to stabilize the error dynamics, a kinematic control law α based on the preset performance of the tunnel is constructed. i for: in, is a positive definite control gain matrix; Substituting into δ i The dynamics of
[0053] Step 204, determine the path parameter variables of the virtual leader, define the path variable dynamics and parameter errors, and introduce feedback laws for each path parameter variable to construct a path update law.
[0054] Different from trajectory tracking, path control has an additional control degree of freedom. Therefore, it is necessary to introduce feedback rules for each path variable to achieve update and synchronization between various virtual leaders.
[0055] In one embodiment, the dynamics of the path parameter variables of the jth virtual leader can be defined as Based on the leader-follower structure, the design parameter error is: Introducing the feedback law, the path update law is constructed as: in, is a gain, and are two tuning parameters. In one embodiment, the path update rules of the three virtual leaders are as follows: Figure 3 As shown, represents the path update law of the first virtual leader, represents the path update law of the second virtual leader, Represents the path update law of the third virtual leader.
[0056] Step 206, for the follower, define the speed tracking error at the dynamic level, introduce a neural network and a state estimator, and use a distributed learning strategy to construct a DNP-based adaptive control law.
[0057] Specifically, for tracking UAVs, we can first define the velocity tracking error at the dynamic level as: E i,2 =v i -α i ; Calculate E i,2 The dynamics of is: Then, the ideal dynamic control law can be proposed as: in, is a positive definite control gain matrix.
[0058] In this embodiment, since the ideal motion control law contains uncertain nonlinear terms f, u i cannot be calculated directly, so some adaptive estimators can be introduced to identify f. Specifically, the neural network that can be used is in represents an optional weight vector, where p is the number of neurons, is a known smooth activation function, is the input vector, The time-varying approximation error satisfies
[0059] By introducing neural networks, a practical dynamic control law is proposed:
[0060] In existing path control methods, neural predictors are usually used to adjust the weights of neural networks. These neural predictors rely on local learning strategies. To further improve the prediction ability of neural predictors, in one embodiment, a distributed learning strategy can be used to construct a DNP-based adaptive control law:
[0061]
[0062] in is the adaptive gain, is an additional positive definite tuning parameter matrix used to shape the transient learning behavior of the neural network, σ i ∈(0, 1) is a correction factor.
[0063] The comparison between the traditional method of controlling drones using neural predictors and the drone cluster path planning control method DNP based on tunnel preset performance in this embodiment is as follows: Figure 4As shown, taking one of the follower drones as an example, it can be seen that the drone cluster path planning control method DNP based on tunnel preset performance in this embodiment improves the prediction ability of the neural predictor.
[0064] Step 208 , using the kinematic control law, the path update law, and the DNP-based adaptive control law as control signals, and designing a distributed path controller based on the control signals.
[0065] By selecting a distributed path controller, the kinematic control law, path update law, and DNP-based adaptive control law can all be used as control signals to achieve eventually uniformly bounded (UUB) in the full closed-loop system, and actual multi-UAV cluster distributed path planning control can be realized. In this embodiment, the error dynamics of the distributed path closed-loop system is obtained as:
[0066]
[0067] Step 210, performing path tracking control of the drone cluster according to the distributed path controller.
[0068] When performing path tracking control of a UAV cluster, the control inputs of the six followers are as follows: Figure 5 In one embodiment, the provided UAV cluster path planning control method based on tunnel preset performance is experimentally verified. The trajectories of the virtual leader and six followers based on the tunnel preset performance strategies TPP and DNP are shown in FIG. Figure 6 , Figure 7 As shown, the six followers follow the virtual leader to perform curved or circular motion.
[0069] When the virtual leader and followers move according to the path planning control of the UAV cluster based on the preset performance of the tunnel, the tracking error analysis of each UAV is compared with the error analysis of the traditional preset performance control (PPC), such as Figure 8 As shown, taking one of the follower drones as an example, it can be seen that the drone cluster path planning control error based on the preset tunnel performance is significantly smaller.
[0070] It should be understood that, although the various steps in the above-mentioned flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above-mentioned flow chart may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.
[0071] In one embodiment, a computer device is provided, comprising a processor, a memory, a network interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for controlling the path planning of a drone cluster based on preset tunnel performance is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
[0072] Those skilled in the art will understand that the structure of the computer device is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0073] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of a method for controlling a path planning of a drone cluster based on preset tunnel performance are implemented.
[0074] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of a method for controlling a path planning of a drone cluster based on preset tunnel performance are implemented.
[0075] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an 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 (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0076] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this specification.
[0077] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A method for controlling the path planning of a drone cluster based on tunnel preset performance, wherein the drone cluster includes a virtual leader and a follower, characterized in that: The method comprises: For the follower, a path error is defined, including: defining the following distributed path error of the i-th follower as Right now Derivative of the path error yields Where M represents the number of followers, NM represents the number of virtual leaders, and p r,j (θ j ) represents the parameterized path of the jth virtual leader, and p r,j (θ j ) is bounded and satisfies Conditions; a i,j (p i -p j ) represents the position difference between the ith follower and other followers, a i,j (p i -p r,j (θ j )) represents the position difference between the i-th follower and the virtual leader; a i,j represents the communication connection weight, d i represents the sum of the communication connection weights between followers, that is, The path error is transformed by using a preset tunnel performance strategy, including: designing a distributed path error that satisfies the preset tunnel performance strategy constraint as: Among them, E ˉi,k and is a time-varying function, in, Performance functions Substitute Among them, G i,k (δ i,k ):(-∞,∞)→(-1,1) represents an increasing transformation function, and δi,k corresponds to a function equivalent to E i,k The unconstrained transformation error of i,k (δ i,k Represented as G i,k (δ i,k )=(exp(δ i,k )-exp(-δ i,k )) / (exp(δ i,k )+exp(-δ i,k )); by G i,k (δ i,k ) to obtain the transformed error Constructing a kinematic control law based on the preset performance of the tunnel, including: constructing a kinematic control law α based on the preset performance of the tunnel i for: in, is a positive definite control gain matrix; Substituting into δ i The dynamics of Determine the path parameter variables of the virtual leader, define the path variable dynamics and parameter errors, introduce feedback rules for each of the path parameter variables, and construct a path update law, including: define the dynamics of the path parameter variables of the jth virtual leader as Based on the leader-follower structure, the design parameter error is: Introducing the feedback law, the path update law is constructed as: in, is a gain, and are two tuning parameters; For the follower, a velocity tracking error at a dynamic level is defined, a neural network and a state estimator are introduced, and a distributed learning strategy is used to construct a DNP-based adaptive control law; The kinematic control law, the path update law, and the DNP-based adaptive control law are used as control signals, and a distributed path controller is designed based on the control signals; According to the distributed path controller, the UAV cluster path tracking control is performed.
2. The method for controlling the path planning of a drone cluster based on preset tunnel performance according to claim 1 is characterized in that: For the follower, the velocity tracking error at the dynamic level is defined as: E i,2 =v i -α i ; Calculate E i,2 The dynamics of is:
3. The method for controlling the path planning of a drone cluster based on preset tunnel performance according to claim 2 is characterized in that: The method further comprises: The ideal dynamic control law is determined as: in, is a positive definite control gain matrix.
4. The method for controlling the path planning of a drone cluster based on preset tunnel performance according to claim 3 is characterized in that: Introducing neural networks and state estimators, including: Using Neural Networks in, represents an optional weight vector, p is the number of neurons, is a known smooth activation function, is the input vector, The time-varying approximation error satisfies Introducing the neural network, the actual dynamic control law is determined as 5. The method for controlling the path planning of a drone cluster based on preset tunnel performance according to claim 4 is characterized in that: The distributed learning strategy is used to construct the adaptive control law based on DNP: in, is the adaptive gain, is an additional positive definite tuning parameter matrix used to shape the transient learning behavior of the neural network, σ i ∈(0, 1) is a correction factor.
