A tracking control method for multi-agent systems based on finite time
By designing a limited time observer and distributed adaptive sliding mode control method in a multi-agent system, the consistency tracking problem under external interference and unknown state information is solved, and the rapid and bounded convergence effect is achieved, and the system is improved robustness and stability.
Patent Information
- Application Number
- CN202411208819.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-30
AI Technical Summary
The prior art has failed to effectively solve the problem of finite time consistency tracking in multiagent systems in case of external interference and unknown agent state information, especially in terms of convergence speed and robustness.
By constructing the network topology diagram and dynamic model of multi-agent system, a finite time observer and a distributed adaptive sliding mode control method are designed, and a consistent bounded system is used to verify that the system is realized in a limited time.
Under the situation of external interference and unknown state information, the limited time consistency tracking of multi-agent systems is realized, which improves the robustness and stability of the system and ensures convergence speed and efficiency.
Smart Images

Figure CN118707859B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of multi-agent systems, and in particular to a multi-agent system tracking control method based on finite time. Background Art
[0002] In recent years, the cooperative control of multi-agent systems has been widely used in robot formation control, unmanned aerial vehicle combat control, etc., and has gradually become a hot topic in the fields of control theory, statistical physics, applied mathematics, computer science, etc. The consistency problem is the primary problem to be solved in the realization of cooperative control of multi-agent systems. Its essence is to use local information to design distributed control protocols to ultimately make the states of all agents consistent. However, in practical applications, especially in some systems with high control accuracy, there are requirements for convergence speed. Compared with asymptotic consistency, finite-time consistency has more obvious advantages. It can not only ensure that the system obtains faster convergence speed, but also show better robustness in the face of external interference. Therefore, it is of great significance to study the finite-time tracking problem of multi-agent systems.
[0003] In actual industrial systems, external environmental disturbances and system modeling deviations affect the stability of the system. Therefore, it is particularly important to quickly and effectively achieve finite-time bounded tracking of multi-agent systems and design more efficient controllers to improve the robustness of the system. It is worth noting that in the prior art, fully distributed controllers are usually used to achieve consistency tracking of multi-agents when the relative states are known, but the consistency tracking problem when the relative states are unknown is not considered. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and to provide a multi-agent system tracking control method based on finite time, comprising the following steps:
[0005] S1. Construct the network topology of the multi-agent system and the dynamic model of the leader and follower agents;
[0006] S2. According to the dynamic model and network topology, determine the observation error of the multi-agent system and design a finite time observer, as well as the parameter matrix information when the agent system meets the preset assumptions;
[0007] S3, calculate the convergence time of the multi-agent system observation to be consistent in finite time;
[0008] S4. Design a distributed adaptive sliding mode control method. According to Lyapunov stability theory, verify that the multi-agent system can achieve finite-time uniform boundedness under the designed distributed adaptive sliding mode control method.
[0009] Preferably, the network topology diagram constructed in step S1 includes: constructing an undirected graph in represents the node set of the undirected graph Π, N is the total number of nodes; is the edge set of the undirected graph Π, is the adjacency matrix of the undirected graph Π, represents the edge weight from communication node i to communication node j. If there is no connecting edge between communication node i and communication node j, then =0; is the representation of the undirected graph Π and the The agent has a set of neighboring agents with which to interact; according to the undirected graph Π, the degree matrix and the Laplace matrix L, where , , ,matrix is positive definite if and only if the graph With the root node Has a directed spanning tree.
[0010] Preferably, the physical implementation process of the dynamic model of the leading intelligent agent and the following intelligent agent constructed according to the network topology diagram in step S1 is:
[0011] Determine a dynamic model of a leader agent in a multi-agent system, wherein the dynamic model of the leader agent is expressed as:
[0012]
[0013] in and Represent the state information and measurement output of the leader agent respectively. represents the control input of the known leader agent; where n, p and m are spatial dimensions and t is a time variable;
[0014] Determine the dynamics model of the following agent, and the dynamics model of the following agent is expressed as:
[0015]
[0016] in It is The state information of the following agent, It is A control input for the following agent, It is The measured output of the following agent, is a known external disturbance.
[0017] Preferably, the assumptions include:
[0018] The information exchange topology formed by the follower agent and the leader agent is undirected and connected; It is controllable. It is considerable;
[0019] Each follower agent cannot obtain its own state information, but can obtain its own output information;
[0020] The control input of the leader agent is known, and the neighbor agents that have direct information interaction with the leader agent can obtain the state information of the leader agent, but the follower agents that have no direct information interaction with the leader agent cannot obtain the state information of the leader agent.
[0021] Preferably, the observation error is expressed as: ,
[0022] in It is The estimated value of each agent's own state, coupling gain The following conditions are met:
[0023]
[0024] Among them, when following the agent and the following agent When there is information interaction, then, , otherwise ;
[0025]
[0026] Among them, when following the agent When there is information interaction with the leader agent, , otherwise .
[0027] Preferably, the finite time observer is expressed as:
[0028]
[0029] in It is The estimated value of each agent's own state, is the gain matrix to be designed, and is the coupling coefficient, sign() is the sign function; is a positive definite symmetric matrix that satisfies the following inequalities, as shown below:
[0030]
[0031] When the gain matrix satisfy When Estimated state With leadership status In a finite time, the parameter matrix information includes a positive definite symmetric matrix and the gain matrix .
[0032] Preferably, let , , then the error system is:
[0033]
[0034] The convergence time of the computational observation finite time consistency is expressed as:
[0035]
[0036] in Represent the maximum and minimum characteristic roots of H and P respectively,
[0037] ,when .
[0038] Preferably, the implementation process of step S4 is:
[0039] S41, when When The state error with the leader agent is:
[0040]
[0041] S42. The ideal sliding surface is proposed, and the following integral surface is introduced, which has the following form:
[0042]
[0043] in is a real matrix designed to ensure is a non-singular positive definite symmetric matrix ,in , , construct a sliding mode controller that can make the closed-loop system enter the ideal sliding mode surface (such as (c is a constant));
[0044] S43. Design of distributed tracking control protocol is expressed as:
[0045]
[0046]
[0047] when hour, have to ,in is a time-varying function that satisfies ,in is a very small positive number;
[0048] S44. Define tracking error, expressed as:
[0049]
[0050] S45, calculating and tracking the convergence time of the finite time bound, the convergence time is expressed as:
[0051] .
[0052] S46. Verify that the multi-agent system tracking is bounded in finite time:
[0053] Construct the following Lyapunov equation , and we can get
[0054]
[0055]
[0056] because ,so ,but ;in , that is, existence time , ,when hour .
[0057] Beneficial effects of the present invention: In the case of external interference and unknown agent state information, the present invention constructs a finite time observer to observe the state of each follower agent, and realizes that the observed state of the agent system is consistent with the leader state after the convergence time. The observation error is defined according to the network topology diagram between multiple agents, and the integral sliding surface is designed based on the observation error. A controller is constructed for each agent, and the system is linearly transformed. Compensation control is added to the sliding control, and finally tracking of finite time boundedness is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying creative labor, other drawings obtained based on these drawings still belong to the scope of the present invention.
[0059] Figure 1 is a flow chart of the method of the present invention;
[0060] Figure 2 A network topology diagram of the multi-agent system of the present invention;
[0061] Figure 3 It is a trajectory diagram of the observation state and the leadership state of the follower of the present invention;
[0062] Figure 4 It is the movement trajectory diagram of the follower and the leader of the present invention. DETAILED DESCRIPTION
[0063] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.
[0064] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two non-identical entities with the same name or non-identical parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be understood as limitations on the embodiments of the present invention. The subsequent embodiments will not explain this one by one.
[0065] The directions and positions mentioned in the present invention, such as "upper", "lower", "front", "back", "left", "right", "inside", "outside", "top", "bottom", "side", etc., are only for reference to the directions or positions of the drawings. Therefore, the directions and positions used are for explaining and understanding the present invention, but not for limiting the scope of protection of the present invention.
[0066] like Figures 1 to 4 As shown in the embodiment of the present invention, the implementation process of the present invention includes the following steps:
[0067] S1. Constructing a network topology diagram of a multi-agent system and a dynamic model of a leader agent and a follower agent; the multi-agent system includes 1 leader and N followers;
[0068] S2. According to the dynamic model and network topology, determine the observation error of the multi-agent system and design a finite time observer, as well as the parameter matrix information when the agent system meets the preset assumptions;
[0069] S3, calculate the convergence time of the multi-agent system observation to be consistent in finite time;
[0070] S4. Design a distributed adaptive sliding mode control method. According to Lyapunov stability theory, verify that the multi-agent system tracking can achieve finite-time uniformly bounded tracking under the action of the designed distributed adaptive sliding mode control method.
[0071] In this embodiment, the multi-agent system can be understood as a network system including multiple agents with certain perception, communication, calculation and execution capabilities connected through communication and other means. The network topology diagram clearly presents the connection between the agents in the multi-agent system. The leader agent can be understood as the target agent. The dynamic model can be understood as a way to characterize the multi-agent system as a dynamic model.
[0072] The network topology constructed in step S1 includes: constructing an undirected graph in represents the node set of the undirected graph Π, N is the total number of nodes; is the edge set of the undirected graph Π, is the adjacency matrix of the undirected graph Π, represents the edge weight from communication node i to communication node j. If there is no connecting edge between communication node i and communication node j, then =0; is the representation of the undirected graph Π and the The agent has a set of neighboring agents with which to interact; the degree matrix is determined according to the undirected graph Π and the Laplace matrix L, where , , ,matrix is positive definite if and only if the graph With the root node Has a directed spanning tree.
[0073] Step S1 implements the dynamic model of the leader agent and the follower agent based on the network topology diagram as follows:
[0074] Determine a dynamic model of a leader agent in a multi-agent system, wherein the dynamic model of the leader agent is expressed as:
[0075]
[0076] in and Represent the state information and measurement output of the leader agent respectively. represents the control input of the known leader agent; where n, p and m are spatial dimensions and t is a time variable;
[0077] Determine the dynamic model of the following agent, which is expressed as:
[0078]
[0079] in It is The state information of the following agent, It is A control input for the following agent, It is The measured output of the following agent, is a known external disturbance;
[0080] The topological graph of information exchange formed by follower agents and leader agents is undirected and connected; It is controllable. It is considerable;
[0081] Each follower agent cannot obtain its own state information, but can obtain its own output information;
[0082] The control input of the leader agent is known, and the neighbor agents that have direct information interaction with the leader agent can obtain the state information of the leader agent, but the follower agents that have no direct information interaction with the leader agent cannot obtain the state information of the leader agent.
[0083] The observation error is expressed as: ,
[0084] in It is The estimated value of each agent's own state, coupling gain The following conditions are met:
[0085]
[0086] Among them, when following the agent and the following agent When there is information interaction, , otherwise ;
[0087]
[0088] Among them, when following the agent When there is information interaction with the leader agent, , otherwise .
[0089] The finite-time observer is expressed as:
[0090]
[0091] in It is The estimated value of each agent's own state, is the gain matrix to be designed, and is the coupling coefficient, sign() is the sign function; is a positive definite symmetric matrix that satisfies the following inequalities, as shown below:
[0092]
[0093] in, It is controllable. is considerable, when the gain matrix satisfy When Estimated state With leadership status In a finite time, the parameter matrix information includes a positive definite symmetric matrix and the gain matrix .
[0094] make , , then the error system is:
[0095]
[0096] The convergence time of the computational observation finite time consistency is expressed as:
[0097]
[0098] in Represent the maximum and minimum characteristic roots of H and P respectively.
[0099] ,when .
[0100] The implementation process of step S4 is:
[0101] S41, when When The state error with the leader agent is:
[0102]
[0103]
[0104] S42. Introduce the following integral surface and design the ideal sliding surface, which is in the following form:
[0105]
[0106] in is a real matrix designed to ensure is a non-singular positive definite symmetric matrix ,in , ; Construct a sliding mode controller that can make the closed-loop system enter the ideal sliding mode surface (such as (c is a constant));
[0107] S43, design a distributed tracking control protocol represented as;
[0108]
[0109]
[0110] when hour, have to ,in is a time-varying function that satisfies ,in is a very small positive number;
[0111] S44, defining a tracking error, wherein the tracking error is expressed as:
[0112]
[0113] S45, calculating and tracking the convergence time of the finite time bound, the convergence time is expressed as:
[0114]
[0115] S46. Verify that the multi-agent system is bounded in finite time tracking: According to Lyapunov stability theory, it can be verified that the multi-agent system composed of follower agents and leader agents can achieve finite time consistent boundedness under the action of distributed adaptive sliding mode control method. The proof process is as follows:
[0116] Proof: Order The following error system is obtained
[0117]
[0118] because
[0119]
[0120] make have to
[0121]
[0122] assumed in is a positive definite matrix, so Non-singular.
[0123] Construct the following Lyapunov equation The derivative is
[0124] , further we can get
[0125]
[0126] because ,so ,but ;in , that is, existence time , ,when hour .
[0127] Based on the above analysis, the state of the multi-agent system is bounded in finite time tracking.
[0128] The effectiveness of the implementation scheme is illustrated by a simulation case.
[0129] Multi-agent system. It consists of 1 leader and 4 followers. Its network topology is given by Figure 2 Given. According to the communication topology, the Laplace matrix L and degree matrix can be calculated The specific expression is:
[0130] ,
[0131] For the model of the multi-agent system, the dynamic model of the leader agent is described as follows:
[0132]
[0133] The dynamic model of the following agent is:
[0134]
[0135] The coefficient matrices are as follows:
[0136] , , ,
[0137] The interference signal is ;
[0138] Satisfy the inequality The positive definite symmetric matrix The values are as follows
[0139]
[0140] At the initial moment, the control input of the leader agent and the state quantity of the follower agent of the system are:
[0141] .
[0142] The initial disturbance is .
[0143] After 0.35 seconds, the absolute value of the error between the observed state of the agent and the state of the leader agent remains within 0.10, and the trajectory diagram is as follows: Figure 3 shown.
[0144] Satisfy the inequality (in ) is a positive definite symmetric matrix The values are as follows:
[0145]
[0146] Controller:
[0147]
[0148] Initial Value:
[0149]
[0150] At the same time from Figure 4 The trajectory diagram is available, and the finite time tracking can be achieved after 5 seconds.
[0151] The finite-time multi-agent system tracking control method proposed in the present invention has a fast response speed and can solve the problem of external disturbance in the tracking control of the multi-agent system. The method can ensure that the system completes convergence within a bounded time, improves the efficiency of the equipment to a certain extent, reduces the cost, and enhances the robustness and stability of the system, and realizes the preservation of connectivity; it can be widely used in the consistency realization problem of a class of multi-agent systems with unknown interference and state information.
[0152] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as ROM / RAM, disk, CD-ROM, etc.
[0153] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.
[0154] It should be noted that the embodiments of the present invention can be implemented by hardware, software or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in processor control codes, such as such codes provided on a programmable memory or a data carrier such as an optical or electronic signal carrier.
[0155] In addition, although the operation of the method of the present invention is described in a particular order in the accompanying drawings, this does not require or imply that these operations must be performed in this particular order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the steps described in the flow chart can change the order of execution. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step for execution, and / or one step can be decomposed into multiple steps for execution. It should also be noted that the features and functions of two or more devices according to the present invention can be embodied in one device. On the contrary, the features and functions of a device described above can be further divided into being embodied by multiple devices.
[0156] Although the invention has been described with reference to several specific embodiments, it should be understood that the invention is not limited to the specific embodiments disclosed. The invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims
1. A finite time based multi-agent system tracking control method, characterized in that: The method comprises the following steps: S1. Construct the network topology of the multi-agent system and the dynamic model of the leader and follower agents; The kinetic model implementation process is: Determine a leader agent dynamics model in a multi-agent system, wherein the leader agent dynamics model is expressed as: in and Represent the state information and measurement output of the leader agent, respectively. represents the control input of the known leader agent; where n, p and m are spatial dimensions and t is a time variable; Determine the dynamics model of the following agent, which is expressed as: in It is The state information of the following agent, It is A control input for the following agent, It is The measured output of the following agent, is a known external disturbance; S2. According to the dynamic model and network topology, determine the observation error of the multi-agent system and design a finite time observer, as well as the parameter matrix information when the multi-agent system meets the preset assumptions; The observation error is expressed as: in It is The estimated value of each agent's own state, coupling gain The following conditions are met: Among them, when following the agent and the following agent When there is information interaction, , otherwise ; Among them, when following the agent When there is information interaction with the leader agent, , otherwise ; The finite-time observer is expressed as: in It is The estimated value of each agent's own state, is the gain matrix to be designed, and is the coupling coefficient, sign() is the sign function; including is a positive definite symmetric matrix satisfying the following inequalities: When the gain matrix satisfy When Estimated state With leadership status In a finite time, the parameter matrix information includes a positive definite symmetric matrix and the gain matrix ; The assumptions include: the information exchange topology graph formed by the follower agent and the leader agent is undirectedly connected; It is controllable. It is considerable; Each follower agent cannot obtain its own state information, but can obtain its own output information; The control input of the leader agent is known, and the neighbor agents that have direct information interaction with the leader agent can obtain the state information of the leader agent, but the follower agents that do not have direct information interaction with the leader agent cannot obtain the state information of the leader agent; S3, calculate the convergence time of the multi-agent system observation to be consistent in finite time; S4. Design a distributed adaptive sliding mode control method. According to Lyapunov stability theory, verify that the multi-agent system can achieve finite-time bounded tracking under the action of the distributed adaptive sliding mode control method.
2. The finite time based multi-agent system tracking control method according to claim 1 is characterized in that: The network topology constructed in step S1 includes: constructing an undirected graph ,in represents the node set of the undirected graph Π, N is the total number of nodes; is the edge set of the undirected graph Π, is the adjacency matrix of the undirected graph Π, represents the edge weight from communication node i to communication node j. If there is no connecting edge between communication node i and communication node j, then =0; is the representation of the undirected graph Π and the The agent has a set of neighboring agents with which to interact; according to the undirected graph Π, the degree matrix and the Laplace matrix L, where , , ,matrix is positive definite if and only if the graph With the root node Has a directed spanning tree.
3. The finite time based multi-agent system tracking control method according to claim 1 is characterized in that: make , , then the error system is: The convergence time of the computational multi-agent system observation finite time consistency is expressed as: in Represent the maximum and minimum characteristic roots of H and P respectively, ,when .
4. The finite time based multi-agent system tracking control method according to claim 3 is characterized in that: The implementation process of step S4 is: S41, when When The state error with the leader agent is: S42. Introduce the following integral surface and design the ideal sliding surface, which is in the following form: in is a real matrix designed to ensure is a non-singular positive definite symmetric matrix ,in , ; Construct a sliding mode controller that can make the closed-loop system enter the ideal sliding mode surface (such as ,in is a constant); S43. Design of distributed tracking control protocol is expressed as: when hour, have to ,in is a time-varying function that satisfies ,in is a very small positive number; S44, defining a tracking error, wherein the tracking error is expressed as: S45, calculating and tracking the convergence time of the finite time bound, the convergence time is expressed as: S46. Verify that the multi-agent system tracking is bounded in finite time: Construct the following Lyapunov equation , and we can get because ,so ,but ;in , that is, existence time , ,when hour .
