Intelligent agent tracking control system and method under switching symbol diagram

By constructing the leadership-follower dynamics model and topology model of switching symbolograms in a multi-agent system, the tracking control protocol is designed, and the problem that fixed topology cannot cope with communication link attacks or noise is achieved. The efficient multi-agent system is achieved Tracking control.

CN119045329APending Publication Date: 2024-11-29JIANGNAN UNIV

Patent Information

Application Number
CN202411167370.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

When existing multi-agent systems face communication link attacks or noise, the accuracy and real-timeness of tracking control cannot be guaranteed, and the fixed topology cannot effectively deal with this situation.

Method used

A agent tracking control system under the switching symbol diagram was designed. By constructing the leadership-follower dynamics model and switching topology model of the second-order multi-agent system, the leader follow-up tracking control protocol was designed, and the Laplace matrix was used. and Liyapunov stability, realize tracking control under the switching symbol map.

Benefits of technology

This method reduces the connectivity requirements of the communication topology, can effectively deal with communication link attacks or noise, ensures the accuracy and real-timeness of multi-agent tracking control, and has a simple control method, only the control gain needs to meet relevant conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119045329A_ABST
    Figure CN119045329A_ABST
Patent Text Reader

Abstract

The invention discloses an agent tracking control system and method under a switching symbol diagram, and belongs to the technical field of multi-agent system tracking control. The method comprises the following steps: constructing a leader-follower dynamic model of a second-order multi-agent system; constructing a switching topology model by taking each agent as a communication node and taking a communication link of the adjacent agent as an edge; designing a leader following tracking control protocol; constructing a tracking error system between the leader and the follower; constructing an equivalent system by using matrix transformation and Kronecker product properties; and a tracking control result under the switching symbol diagram is given by using Lyapunov stability. By designing the control method which is simple and novel, the problem of binary tracking control of the second-order multi-agent system under the conditions that a communication link can be changed and constant connection does not need to be guaranteed is solved, the requirement for reducing the connectivity of the communication link is met, the control method is simple, and only the gain needs to be controlled to meet related conditions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an intelligent agent tracking control system and method under a switching symbol diagram, belonging to the technical field of multi-agent system tracking control. Background Art

[0002] A multi-agent system is a computing system composed of multiple agents interacting in the same environment. In recent years, multi-agent systems have been widely used in drones, sensor networks, and other fields. In graph theory, the communication links between agents and adjacent agents are regarded as nodes and edges respectively. In order to characterize adjacent agents with cooperative relationships, researchers proposed the concept of unsigned graphs and defined the weights of communication links as positive. However, in real systems, agents cooperate with each other, but also compete with each other due to contradictions and differences. Therefore, some scholars have established the theory of symbolic graphs and applied it to related problems, and the role of cooperation-competition interaction has also been explained accordingly. The weights of edges considered in symbolic graphs can be either positive or negative.

[0003] As the basis of cooperative control of multi-agent systems, the tracking problem has become one of the hot issues that many scholars have paid attention to. Under the control protocol, multiple agent systems can track a pre-given target. So far, some scholars have conducted relevant research on the tracking problem. If a symbolic graph is used to characterize a multi-agent system that includes both friendly cooperation and confrontational competition, then under the control protocol, the state of the agent can eventually tend to two parts with the same modulus value but opposite signs. Researchers have summarized it as a binary tracking control problem. At present, many existing works focus on the binary tracking control problem of multi-agent systems connected under fixed symbolic graphs.

[0004] In actual systems, agents are usually represented as second-order dynamic systems with information representing position and velocity. CN118068696A invention discloses a binary tracking control method for a second-order nonlinear multi-agent system, the method comprising: for a second-order nonlinear multi-agent system including followers and leaders, an adaptive control protocol is designed according to the dynamic model of the follower and the dynamic model of the leader, and when the undirected connected graph corresponding to the second-order nonlinear multi-agent system is a structural equilibrium graph, the binary tracking control of the leader-follower is realized. In order to realize the binary tracking control of the multi-agent system, the scheme usually assumes that the communication topology graph is always connected or has a spanning tree (the spanning tree refers to the existence of a root node in the communication topology graph composed of all agents, and this root node agent can reach any agent, that is, the communication information can be transmitted to any agent). In addition, due to inevitable attacks and noise, the communication link for information transmission is easy to change. If a fixed topology is used, it is impossible to effectively deal with the impact of communication link attacks or noise, resulting in the inability to guarantee the accuracy and real-time performance of the tracking control.

[0005] Therefore, it is more realistic and challenging to consider the reduced communication topology requirements, that is, the multi-agent control system and method that does not require constant connectivity or a spanning tree and whose communication topology can be switched at the same time. Summary of the invention

[0006] In order to solve the problem that the current fixed topology cannot effectively deal with communication link attacks or noise, resulting in the inability to ensure the accuracy and real-time performance of multi-agent tracking, the present invention provides an agent tracking control system and method under a switching symbolic graph, and the technical solution is as follows:

[0007] A tracking control method of an intelligent agent system under a switching symbol graph of the present invention comprises:

[0008] Step 1: Construct a second-order multi-agent system leader-follower dynamics model;

[0009] Step 2: Take each agent system as a communication node and the adjacent agent communication links as edges to build a switching topology model;

[0010] Step 3: Design a leader-follower tracking control protocol based on the dynamic model and the switching topology model;

[0011] The leader-follower tracking control protocol is expressed as:

[0012]

[0013] Where u0(t)∈R represents the control protocol of the virtual leader, k>0 represents the gain parameter, v0(t∈R represents the speed of the virtual leader, and u i (t)∈R represents the control protocol of the ith agent in the follower system model, i=1,2,…,N, N represents the number of agents in the follower system model; v i (t) represents the speed of the i-th agent; N i (t) represents the neighbor set of the i-th agent; a ij (t) represents the weight of the edge between the i-th agent and the j-th neighbor node; x i (t)∈R represents the position of the ith agent, a io (t) represents the weight of the edge between the i-th agent and the virtual leader;

[0014] Step 4: Construct a tracking error system between leaders and followers;

[0015] Step 41: Order Then the multi-agent system under the control protocol is transformed into:

[0016]

[0017] Pick The above system is transformed into:

[0018]

[0019] in, l ij (t) is the element of the Laplacian matrix;

[0020] Step 42: Define the tracking error between the leader and the follower as e i (t) = η i (t)-σ i η0(t)(i∈w) and e(t)=(e1(t),e2(t),…,e N (t)) T , then the tracking error system is:

[0021]

[0022] Among them, H δ(t) =L δ(t) +D δ(t) , L δ(t) The communication topology symbol diagram of the multi-agent system The Laplace matrix, D δ(t) =diag{|a 10 (t)|,|a 20 (t)|,…,|a N0 (t)|}, contains the absolute weights of all edges from the virtual leader; I N represents the identity matrix;

[0023] Step 5: According to the tracking error system, an equivalent system is constructed using matrix transformation and Kronecker product properties;

[0024] The equivalent system is:

[0025]

[0026] in:

[0027]

[0028] Step 6: Based on the equivalent system, using Lyapunov stability, the tracking control result under the switching sign diagram is given;

[0029] When the parameter k in the control protocol in step 3 satisfies the following formula, the second-order multi-agent system in the switching symbol diagram realizes tracking control:

[0030]

[0031] Among them, l ii (t) is the element in the i-th row and i-th column of the Laplace matrix.

[0032] Optionally, in the second-order multi-agent system leader-follower dynamics model constructed in step 1, the follower system model is:

[0033]

[0034] in, Represents the position state x of the i-th agent i The derivative of (t), represents the speed v of the i-th agent i The derivative of (t);

[0035] The virtual leadership model is:

[0036]

[0037] in, represents the derivative of the virtual leader position state x0(t), Represents the derivative of the virtual leader's velocity v0(t).

[0038] Optionally, the process of constructing the switching topology model in step 2 includes:

[0039] Consider an infinite time series with bounded, non-overlapping and adjacent time intervals. Let the infinite time series be:

[0040] [T0,T1), [T1,T2), …, [T m ,T m+1 ), …

[0041] Where T0=0,0<T m+1 -T m <T, T>0;

[0042] At each time interval [T m ,T m+1 ), there exists a finite time series with non-overlapping subintervals, expressed as:

[0043]

[0044] in, and τ is the minimum residence time;

[0045] when When considering the topological structure of the multi-agent system Switching is performed when δ(t):[0,+∞)→{1,...,h}(h>1) is used as the switching signal, so the symbolic diagram of the multi-agent system is represented as follows: and Indicates time [T m ,T m+1 )The joint topology of multi-agent systems within.

[0046] Optionally, in step 3, the virtual leader only performs traction control on some followers, and there is an edge between the virtual leader and the i-th agent, then a i0 (t)≠0, otherwise a i0 (t)=0.

[0047] A tracking control system of an intelligent agent system under a switching symbol diagram of the present invention comprises:

[0048] A multi-agent system model building module, configured to build a second-order multi-agent system leader-follower dynamics model;

[0049] A switching topology model building module is configured to use each agent system as a communication node and adjacent agent communication links as edges to build a switching topology model;

[0050] A control protocol design module is configured to design a leader-follower tracking control protocol according to the dynamic model and the switching topology model;

[0051] The leader-follower tracking control protocol is expressed as:

[0052]

[0053] Where u0(t)∈R represents the control protocol of the virtual leader, k>0 represents the gain parameter, v0(t∈R represents the speed of the virtual leader, and u i (t)∈R represents the control protocol of the ith agent in the follower system model, i=1,2,…,N, N represents the number of agents in the follower system model; v i (t) represents the speed of the i-th agent; N i (t) represents the neighbor set of the i-th agent; a ij (t) represents the weight of the edge between the i-th agent and the j-th neighbor node; x i (t)∈R represents the position of the ith agent, a io (t) represents the weight of the edge between the i-th agent and the virtual leader;

[0054] The error system building module is configured to build a tracking error system between leaders and followers. The building process includes:

[0055] make Then the multi-agent system under the control protocol is transformed into:

[0056]

[0057] Pick The above system is transformed into:

[0058]

[0059] in, l ij (t) is the element of the Laplacian matrix;

[0060] Define the tracking error between the leader and the follower as e i (t) = η i (t)-σ i η0(t)(i∈w) and e(t)=(e1(t),e2(t),…,e N (t)) T , then the tracking error system is:

[0061]

[0062] Among them, H δ(t) =L δ(t) +D δ(t) , L δ(t) The communication topology symbol diagram of the multi-agent system The Laplace matrix, D δ(t) =diag{|a 10 (t)|,|a 20 (t)|,…,|a N0 (t)|}, contains the absolute weights of all edges from the virtual leader; I N represents the identity matrix;

[0063] An equivalent system construction module is configured to construct an equivalent system according to the tracking error system by using matrix transformation and Kronecker product property;

[0064] The equivalent system is:

[0065]

[0066] in:

[0067]

[0068] A tracking control module is configured to provide a tracking control result under a switching sign diagram based on the equivalent system and using Lyapunov stability;

[0069] When the parameter k in the control protocol satisfies the following formula, the second-order multi-agent system in the switching symbol diagram realizes tracking control:

[0070]

[0071] Among them, l ii (t) is the element in the i-th row and i-th column of the Laplace matrix.

[0072] Optionally, in the multi-agent system leader-follower dynamics model, the follower system model is:

[0073]

[0074] in, Represents the position state x of the i-th agent i The derivative of (t), represents the speed v of the i-th agent i The derivative of (t);

[0075] The virtual leadership model is:

[0076]

[0077] in, represents the derivative of the virtual leader position state x0(t), Represents the derivative of the virtual leader's velocity v0(t).

[0078] Optionally, the process of constructing the switching topology model includes:

[0079] Consider an infinite time series with bounded, non-overlapping and adjacent time intervals. Let the infinite time series be:

[0080] [T0,T1), [T1,T2), …, [T m ,T m+1 ), …

[0081] Where T0=0,0<T m+1 -T m <T, T>0;

[0082] At each time interval [T m ,T m+1 ), there exists a finite time series with non-overlapping subintervals, expressed as:

[0083]

[0084] in, and τ is the minimum residence time;

[0085] when When considering the topological structure of the multi-agent system Switching is performed when δ(t):[0,+∞)→{1,...,h}(h>1) is used as the switching signal, so the symbolic diagram of the multi-agent system is represented as follows: and Indicates time [T m ,T m+1 )The joint topology of multi-agent systems within.

[0086] Optionally, if there is a time t when the virtual leader has an edge to the i-th agent, then a i0 (t)≠0, otherwise a i0 (t)=0.

[0087] The present invention provides an electronic device, comprising a memory and a processor;

[0088] The memory is used to store computer programs;

[0089] The processor is used to implement the tracking control method as described in any one of the above items when executing the computer program.

[0090] The present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the tracking control method as described in any one of the above items is implemented.

[0091] The beneficial effects of the present invention are:

[0092] With regard to the tracking control problem, the communication topology diagram of the second-order multi-agent system in the present invention has a switching structure. Adjacent agents do not need to be connected all the time. As long as there is a spanning tree within a fixed time interval, the connectivity requirements are greatly reduced. The present invention solves the binary tracking control problem of the second-order multi-agent system in which the communication link can change and does not need to be guaranteed to be connected all the time by designing a simple and novel control method, thereby achieving the requirement of reducing the connectivity of the communication link. Compared with the existing solutions using a fixed topology, the present invention can effectively deal with the impact of communication link attacks or noise, thereby ensuring the accuracy and real-time performance of the multi-agent tracking control. Moreover, the control method of the present invention is simple, and it is only necessary to control the gain to meet the relevant conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0094] Figure 1 It is a flow chart of the tracking control method of the intelligent body system under the switching symbol diagram of the present invention.

[0095] Figure 2 It is a communication topology requirement diagram under time interval in the present invention.

[0096] Figure 3 This is a switching topology diagram of the multi-agent system in the second embodiment of the present invention.

[0097] Figure 4 This is a switching topology signal diagram of the multi-agent system in the second embodiment of the present invention.

[0098] Figure 5 It is a position and velocity diagram of the multi-agent system in the second embodiment of the present invention.

[0099] Figure 6 This is a tracking error diagram of the multi-agent system in the second embodiment of the present invention. DETAILED DESCRIPTION

[0100] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0101] Embodiment 1:

[0102] This embodiment aims at the tracking control problem of the second-order multi-agent system under the switching symbol graph and proposes an agent tracking control method under the switching symbol graph. Figure 1 , the method comprising:

[0103] Step 1: Construct a second-order multi-agent system leader-follower dynamics model;

[0104] Consider a second-order multi-agent system with a virtual leader. The follower system model is as follows:

[0105]

[0106] Among them, x i (t)∈R and v i (t)∈R represents the position and velocity of the ith agent, u i (t)∈R represents the control protocol of the ith agent, and i=1,2,…,N, and Respectively represent the position state x of the i-th agent i (t) and speed v i The derivative of (t).

[0107] The virtual leadership model is as follows:

[0108]

[0109] Similar to the follower, assume that x0(t)∈R represents the position of the virtual leader, v0(t)∈R represents the speed of the virtual leader, and u0(t)∈R represents the control protocol of the virtual leader. represents the derivative of the virtual leader position state x0(t), Represents the derivative of the virtual leader's velocity v0(t).

[0110] Step 2: Take each agent system as a communication node and the adjacent agent communication links as edges to construct a switching topology model under loose connectivity requirements.

[0111] use Describe the communication topology symbolic graph of multiple agents, where w = (ω1, ω2, …, ω N ) represents a set of points, represents the edge set, A=(a ij ) N×N Representation diagram The adjacency matrix of i ,ω j )∈ε ij ≠0, and a ii =0(i=1,2,…,N); In addition, if the weight is positive, then a ij >0, then ω i and ω j is cooperative, otherwise, ω i and ω j is competitive; node ω i The neighbor set is described as N i ={ω j ∈w:(ω j ,ω i )∈ε}.

[0112] If the adjacency matrix A is symmetric, then the sign graph is an undirected graph, otherwise the graph is a directed graph; for the point sequence ω1,ω2,…,ω N , if (ω i ,ω j )∈ε, then ω i and ω j There is a path between these two nodes. An undirected graph is said to be connected if there is a path between any two nodes. On the other hand, if a directed graph has a root node that can reach any other node through a path, then the graph has a spanning tree. It is worth noting that the Laplacian matrix of the symbolic graph θ is defined as L = (l ij )N×N ,in And l ij =-a ij (i≠j).

[0113] Definition 1: Symbolic graph is structurally balanced, that is, if the symbol graph The nodes can be divided into two non-overlapping groups w1 and w2, that is, w1∪w2=w and When i ,ω j ∈w s (s∈{1,2}), there is a ij ≥0; when ω i ∈w s ,ω j ∈w t ,s≠t,(s,t∈{1,2}),a ij ≤0.

[0114] Lemma 1: If the symbolic graph is structurally balanced, then there exists a canonical transformation matrix T∈{diag{σ1,σ2,…,σ N}|σ i ∈{1,-1}}, so that all elements in TAT are non-negative, and when ω i ∈w1, there is σ i =1; when ω i ∈w2, there is σ i =-1.

[0115] Consider an infinite time series with bounded, non-overlapping, and adjacent time intervals. Let the infinite time series be [T0,T1), [T1,T2), …, [T m ,T m+1 ), ..., satisfying T0 = 0 and 0 < T m+1 -T m <T(T>0), and at each time interval [T m ,T m+1 ), there exists a finite time series with non-overlapping subintervals satisfy and (τ is the minimum residence time), such as Figure 2 As shown. When considering the topological structure of the multi-agent system Switching is performed when δ(t):[0,+∞)→{1,...,h}(h>1) is used as the switching signal, so the symbolic graph of the multi-agent system (1)-(2) can be expressed as and Indicates that the time [Tm ,T m+1 ) is the joint topology of the multi-agent system (1)-(2) within a network.

[0116] The present invention proposes basic assumptions and definitions for this second-order multi-agent system with a virtual leader.

[0117] Assumption 1: The communication topology graph formed by the multi-agent system (1)-(2) is structurally balanced, that is, the agents can be divided into two fixed groups w1 and w2, which satisfy the requirement that agents in the same group are cooperative, while agents in different groups are competitive.

[0118] Assumption 2: At each time interval [T m ,T m+1 ),m=0,1,2,…, symbol diagram Have a spanning tree whose virtual leader is the root node.

[0119] Definition 2: The multi-agent system implements leader-follower binary tracking control, that is, for all i∈w1, and holds, and for all i∈w2, and Established.

[0120] Note 1: Due to the influence of network attacks, noise and other factors, the communication topology of the network is prone to time-varying. Therefore, it is assumed here that the multi-agent system has a switching joint connection topology.

[0121] Step 3: Design the leader-follower tracking control protocol based on the constructed system model and topology model;

[0122] Consider the multi-agent leader-follower bipartite tracking control under the switching symbol graph. The multi-agent system consists of a follower system (1) and a virtual leader (2). The virtual leader is an isolated agent that can pass information to some followers. The control protocol of leader-follower is designed as follows:

[0123]

[0124] Where i∈w={1,2,…,N}, k>0 represents the gain parameter, N i (t) and a ij (t) is the same as N described in step 2 i and a ij Similarly, this indicates that the multi-agent system has a switching communication topology. In addition, for the i-th agent, the control protocol (3) depends on its own speed information and the position information of the neighboring agents. If there is an edge from the leader to the follower at time t, then a i0(t)≠0, otherwise a i0 (t) = 0. In addition, if the structural equilibrium condition is satisfied, that is, if i∈w1, there is a i0 (t)≥0, and if i∈w2, there is a i0 (t)≤0.

[0125] Lemma 2: Consider non-negative f(t) is uniformly continuous, μ is a positive constant; if but

[0126] Lemma 3: Consider a graph containing N nodes is undirected, structurally balanced, and the graph A new node is included in addition to the N nodes, labeled 0. If the graph If we have a spanning tree with node 0 as the root node, then the matrix H = L + D is positive definite. Where L is the graph The Laplace matrix of 10 |,|a 20 |,…,|a N0 |}, a i0 Represents the weight from node 0 to the i-th node.

[0127] Step 4: Construct a tracking error system between leaders and followers;

[0128] The main goal is to achieve leader-follower bipartite tracking control of the multi-agent system (1)-(2) under the switching sign graph through the control protocol (3). In order to facilitate the definition of the error system, first let Then the multi-agent system (1)-(2) under the control protocol (3) is transformed into:

[0129]

[0130] Pick Then (4) is transformed into:

[0131]

[0132] Where i = 1, 2, ..., N, l ij (t) is the element of the Laplacian matrix.

[0133] On the other hand, based on Assumption 1, the symbol graph If the structural equilibrium condition is satisfied, then Lemma 1 exists a transformation matrix T, and T = diag{σ1,σ2,…,σ N}, so that when ω i ∈w1, there is σ i =1; when ω i ∈w2, there is σi =-1.

[0134] The following definition implements the leader-follower binary tracking control error as e i (t) = η i (t)-σ i η0(t)(i∈w), and e(t)=(e1(t),e2(t),…,e N (t)) T , the error system is:

[0135]

[0136] Among them, H δ(t) =L δ(t) +D δ(t) , L δ(t) for The Laplace matrix, D δ(t) =diag{|a 10 (t)|,|a 20 (t)|,,|a N0 (t)|}, which contains the absolute weights of all edges from the leader.

[0137] Note 2: The multi-agent system realizes the leader-follower binary tracking control, which can be transformed into the corresponding error system Right now This means and It holds true, that is, Definition 2 can realize the leader-follower bipartite tracking control of the multi-agent system (1)-(2).

[0138] Step 5: Based on the constructed tracking error system, use matrix transformation and Kronecker product properties to construct a simple equivalent system;

[0139] As mentioned above, if The leader-follower binary tracking control can be realized. The present invention performs stability analysis on the error system (6), and makes full use of the properties of matrix transformation and Kronecker product to provide sufficient conditions for realizing the leader-follower binary tracking control. First, an orthogonal matrix is ​​given: satisfy Therefore, the definition Then from the error system (6), we have

[0140]

[0141] in

[0142] Step 6: Based on the constructed equivalent system, using Lyapunov stability, the tracking control results under the switching sign diagram are given.

[0143] Considering that assumptions 1 and 2 hold, for the multi-agent system (1)-(2) with switching communication topology, under the control protocol (3), if the parameter k in (3) satisfies Then the leader-follower binary tracking control can be realized.

[0144] When t∈[T m ,T m+1 ),m=0,1,2,..., select Lyapunov function as:

[0145]

[0146] Then along the trajectory of system (7), the derivative of V(t) is:

[0147]

[0148] in,

[0149] also, in

[0150] Therefore, when k satisfies When Established, thus is positive semidefinite, so we have:

[0151]

[0152] Because H δ(t) is semi-positive definite, then In addition, combined and V(t)≥0, we can get V(t) has an upper bound, then according to the monotonic bounded theorem, exists, so according to the Cauchy convergence criterion, for any ε>0, there exists a number M ε > 0, for all m > M ε , there is |V(T m+1 )-V(T m )|<ε, which means Combination Then there is That is In addition, because Available in is a constant matrix.

[0153] therefore, At r=0,1,…,l m -1, which means Then we can get:

[0154]

[0155] According to Assumption 2 and Lemma 3, the matrix is positive definite. In addition, since V(t) is bounded, ξ(t) is bounded. At the same time, based on system (7), ξ(t) is bounded. Therefore, is uniformly continuous, and Lemma 2 shows that Right now So we can get Thus there is and It holds true. From Definition 2, we know that the bipartite tracking control of the leader-follower system (1)-(2) is realized.

[0156] Note 3: Conditions l ii (t) represents the degree of the ith agent at time t. If the weight of any edge, including the edge from the virtual leader, is +1 or -1, that is, |a ij (t)|=1, then ensure Established.

[0157] Embodiment 2:

[0158] This embodiment considers a second-order multi-agent system consisting of 4 followers and 1 virtual leader as follows:

[0159]

[0160] Where i=1,2,3,4, k=2, the agents are divided into w1={1,2}, w2={3,4}, and according to Assumption 1, the two groups satisfy that the agents in the same group are cooperative, while the agents in different groups are competitive.

[0161] Consider the multi-agent system topology diagram as Figure 3 As shown, at any time interval [T m ,T m+1 ), six switching topologies are given, namely The solid line and dashed line indicate that the weight of an edge is +1 or -1 respectively.

[0162] from Figure 3 As can be seen from the symbol diagram and There exists a spanning tree with the virtual leader as the root node, which is consistent with the proposed assumption 2. Combined with Note 3, we know that k = 2 guarantees is established, thereby satisfying the control conditions required in the present invention.

[0163] Figure 4The switching topology signal δ(t) of the multi-agent system (12) is given, which means that the topology switching order of the multi-agent system (12) is

[0164] From Definition 2, if and If it is established, the leader-follower binary tracking control can be realized. Figure 5 Given the position x of the multi-agent system i (t) and speed vi(t)(i=0,1,2,3,4), it can be seen from the figure that x i (t) in w1 will converge to x0(t), x i (t) will converge to -x0(t) in w2, and Figure 6 Given the multi-agent error system e ij (t)(i=1,2,3,4;j=1,2), where e i1 (t) = x i (t)-σ i x0(t) and It can be seen from the figure thereby and Based on this, the leader-follower binary tracking control of the multi-agent system (12) is realized.

[0165] Embodiment three:

[0166] This embodiment provides an intelligent agent tracking control system under a switching symbol diagram, including:

[0167] A multi-agent system model building module, configured to build a second-order multi-agent system leader-follower dynamics model;

[0168] A switching topology model building module is configured to use each agent system as a communication node and adjacent agent communication links as edges to build a switching topology model;

[0169] A control protocol design module is configured to design a leader-follower tracking control protocol according to the dynamic model and the switching topology model;

[0170] The leader-follower tracking control protocol is expressed as:

[0171]

[0172] Where u0(t)∈R represents the control protocol of the virtual leader, k>0 represents the gain parameter, v0(t∈R represents the speed of the virtual leader, and u i(t)∈R represents the control protocol of the ith agent in the follower system model, i=1,2,…,N, N represents the number of agents in the follower system model; v i (t) represents the speed of the i-th agent; N i (t) represents the neighbor set of the i-th agent; a ij (t) represents the weight of the edge between the i-th agent and the j-th neighbor node; x i (t)∈R represents the position of the ith agent, a io (t) represents the weight of the edge between the i-th agent and the virtual leader;

[0173] The error system building module is configured to build a tracking error system between leaders and followers. The building process includes:

[0174] make Then the multi-agent system under the control protocol is transformed into:

[0175]

[0176] Pick The above system is transformed into:

[0177]

[0178] in, l ij (t) is the element of the Laplacian matrix;

[0179] Define the tracking error between the leader and the follower as e i (t) = η i (t)-σ i η0(t)(i∈w) and e(t)=(e1(t),e2(t),…,e N (t)) T , then the tracking error system is:

[0180]

[0181] Among them, H δ(t) =L δ(t) +D δ(t) , L δ(t) The communication topology symbol diagram of the multi-agent system The Laplace matrix, D δ(t) =diag{|a 10 (t)|,|a 20 (t)|,…,|a N0 (t)|}, contains the absolute weights of all edges from the virtual leader; I Nrepresents the identity matrix;

[0182] An equivalent system construction module is configured to construct an equivalent system according to the tracking error system by using matrix transformation and Kronecker product property;

[0183] The equivalent system is:

[0184]

[0185] in:

[0186]

[0187] A tracking control module is configured to provide a tracking control result under a switching sign diagram based on the equivalent system and using Lyapunov stability;

[0188] When the parameter k in the control protocol satisfies the following formula, the second-order multi-agent system in the switching symbol diagram realizes tracking control:

[0189]

[0190] Among them, l ii (t) is the element in the i-th row and i-th column of the Laplace matrix.

[0191] Some steps in the embodiments of the present invention may be implemented using software, and the corresponding software program may be stored in a readable storage medium, such as a CD or a hard disk.

[0192] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An agent tracking control method under a switching sign diagram, characterized in that: The method comprises: Step 1: Construct a second-order multi-agent system leader-follower dynamics model; Step 2: Take each agent system as a communication node and the adjacent agent communication links as edges to build a switching topology model; Step 3: Design a leader-follower tracking control protocol based on the dynamic model and the switching topology model; The leader-follower tracking control protocol is expressed as: Where u0(t)∈R represents the control protocol of the virtual leader, k>0 represents the gain parameter, v0(t∈R represents the speed of the virtual leader, and u i (t)∈R represents the control protocol of the ith agent in the follower system model, i=1,2,…,N, N represents the number of agents in the follower system model; v i (t) represents the speed of the i-th agent; N i (t) represents the neighbor set of the i-th agent; a ij (t) represents the weight of the edge between the i-th agent and the j-th neighbor node; x i (t)∈R represents the position of the ith agent, a io (t) represents the weight of the edge between the i-th agent and the virtual leader; Step 4: Construct a tracking error system between leaders and followers; Step 41: Order Then the multi-agent system under the control protocol is transformed into: Pick The above system is transformed into: in, l ij (t) is the element of the Laplacian matrix; Step 42: Define the tracking error between the leader and the follower as e i (t) = η i (t)-σ i η0(t)(i∈w) and e(t)=(e1(t),e2(t),…,e N (t)) T , then the tracking error system is: Among them, H δ(t) =L δ(t) +D δ(t) , L δ(t) The communication topology symbol diagram of the multi-agent system The Laplace matrix, Dδ (t) =diag{|a 10 (t)|,|a 20 (t)|,…,|a N0 (t)}, contains the absolute weights of all edges from the virtual leader; I N represents the identity matrix; Step 5: According to the tracking error system, an equivalent system is constructed using matrix transformation and Kronecker product properties; The equivalent system is: in: Step 6: Based on the equivalent system, using Lyapunov stability, the tracking control result under the switching sign diagram is given; When the parameter k in the control protocol in step 3 satisfies the following formula, the second-order multi-agent system in the switching symbol diagram realizes tracking control: Among them, l ii (t) is the element in the i-th row and i-th column of the Laplace matrix.

2. The tracking control method according to claim 1, characterized in that: In the second-order multi-agent system leader-follower dynamics model constructed in step 1, the follower system model is: in, Represents the position state x of the i-th agent i The derivative of (t), represents the speed v of the i-th agent i The derivative of (t); The virtual leadership model is: in, represents the derivative of the virtual leader position state x0(t), Represents the derivative of the virtual leader's velocity v0(t).

3. The tracking control method according to claim 1, characterized in that: The process of constructing the switching topology model in step 2 includes: Consider an infinite time series with bounded, non-overlapping and adjacent time intervals. Let the infinite time series be: [T0,T1),[T1,T2),…,[T m ,T m+1 ),… Where T0=0,0<T m+1 -T m <T, T>0; At each time interval [T m ,T m+1 ), there exists a finite time series with non-overlapping subintervals, expressed as: in, and τ is the minimum residence time; when When considering the topological structure of the multi-agent system Switching is performed when δ(t):[0,+∞)→{1,...,h}(h>1) is used as the switching signal, so the symbolic diagram of the multi-agent system is represented as follows: and Indicates time [T m ,T m+1 )The joint topology of multi-agent systems within.

4. The tracking control method according to claim 3, characterized in that: In step 3, the virtual leader only performs traction control on some followers. If there is an edge between the virtual leader and the i-th agent, then a i0 (t)≠0, otherwise a i0 (t)=0.

5. An intelligent agent tracking control system under a switching symbol diagram, characterized in that: The system comprises: A multi-agent system model building module, configured to build a second-order multi-agent system leader-follower dynamics model; A switching topology model building module is configured to use each agent system as a communication node and adjacent agent communication links as edges to build a switching topology model; A control protocol design module is configured to design a leader-follower tracking control protocol according to the dynamic model and the switching topology model; The leader-follower tracking control protocol is expressed as: Where u0(t)∈R represents the control protocol of the virtual leader, k>0 represents the gain parameter, v0(t∈R represents the speed of the virtual leader, and u i (t)∈R represents the control protocol of the ith agent in the follower system model, i=1,2,…,N, N represents the number of agents in the follower system model; v i (t) represents the speed of the i-th agent; N i (t) represents the neighbor set of the i-th agent; a ij (t) represents the weight of the edge between the i-th agent and the j-th neighbor node; x i (t)∈R represents the position of the ith agent, a io (t) represents the weight of the edge between the i-th agent and the virtual leader; The error system building module is configured to build a tracking error system between leaders and followers. The building process includes: make Then the multi-agent system under the control protocol is transformed into: Pick The above system is transformed into: in, l ij (t) is the element of the Laplacian matrix; Define the tracking error between the leader and the follower as e i (t) = η i (t)-σ i η0(t)(i∈w) and e(t)=(e1(t),e2(t),…,e N (t)) T , then the tracking error system is: Among them, H δ(t) =L δ(t) +D δ(t) , L δ(t) The communication topology symbol diagram of the multi-agent system The Laplace matrix, Dδ (t) =diag{|a 10 (t)|,|a 20 (t)|,…,|a N0 (t)}, contains the absolute weights of all edges from the virtual leader; I N represents the identity matrix; An equivalent system construction module is configured to construct an equivalent system according to the tracking error system by using matrix transformation and Kronecker product property; The equivalent system is: in: A tracking control module is configured to provide a tracking control result under a switching sign diagram based on the equivalent system and using Lyapunov stability; When the parameter k in the control protocol satisfies the following formula, the second-order multi-agent system in the switching symbol diagram realizes tracking control: Among them, l ii (t) is the element in the i-th row and i-th column of the Laplace matrix.

6. The tracking control system according to claim 5, characterized in that: In the multi-agent system leader-follower dynamics model, the follower system model is: in, Represents the position state x of the i-th agent i The derivative of (t), represents the speed v of the i-th agent i The derivative of (t); The virtual leadership model is: in, represents the derivative of the virtual leader position state x0(t), Represents the derivative of the virtual leader's velocity v0(t).

7. The tracking control system according to claim 5, characterized in that: The construction process of the switching topology model includes: Consider an infinite time series with bounded, non-overlapping and adjacent time intervals. Let the infinite time series be: [T0,T1),[T1,T2),…,[T m ,T m+1 ),… Where T0=0,0<T m+1 -T m <T, T>0; At each time interval [T m ,T m+1 ), there exists a finite time series with non-overlapping subintervals, expressed as: in, and τ is the minimum residence time; when When considering the topological structure of the multi-agent system Switching is performed when δ(t):[0,+∞)→{1,...,h}(h>1) is used as the switching signal, so the symbolic diagram of the multi-agent system is represented as follows: and Indicates time [T m ,T m+1 )The joint topology of multi-agent systems within.

8. The tracking control system according to claim 5, characterized in that: The virtual leader only performs traction control on some followers. If there is an edge between the virtual leader and the i-th agent, then a i0 (t)≠0, otherwise a i0 (t)=0.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the tracking control method according to any one of claims 1 to 4 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the tracking control method according to any one of claims 1 to 4 is implemented.

Citation Information

Patent Citations

  • Bipartite consistency control method for second-order nonlinear multi-agent system

    CN118068696A

Cited By

  • Finite time consistency control method of leader-follower multi-agent system

    CN120972584A