An inclusive control method for a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuous maintenance

Abstract

The present invention belongs to the field of intelligent unmanned cluster system control, and specifically relates to an inclusive control method for a multi-leader intelligent unmanned cluster system with obstacle avoidance and continuity maintenance, comprising: constructing an intelligent unmanned cluster system; taking each individual in the system as an intelligent agent, dividing the intelligent agents into leaders and followers, and establishing dynamic equations for the leaders and followers; setting a collision avoidance inclusive control protocol, and using a time sedimentation function to conditionally restrict the collision avoidance inclusive control protocol; the followers use the conditionally restricted collision avoidance inclusive control protocol to obtain status information of the leader; the followers perform collision avoidance and connectivity maintenance based on the obtained status information; the follower control protocol in the system of the present invention is designed based on estimated information obtained through an observer, and through this protocol, the followers can complete inclusive control.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent unmanned cluster system control, and in particular relates to an inclusive control method for a multi-leader intelligent unmanned cluster system with obstacle avoidance and continuous maintenance. Background Art

[0002] In recent years, the consistency problem of intelligent unmanned swarm systems has attracted widespread attention in various research fields. Consistency is a key and fundamental issue in the distributed cooperative control of swarm systems. System consistency refers to the interaction of agents under effective control algorithms to achieve the same state value. Furthermore, with the rapid development of various disciplines, the consistency problem has been widely applied to fields such as unmanned systems, intelligent traffic control, and smart grids.

[0003] When solving distributed consensus tracking problems, the states and outputs of network subsystems are typically synchronized with a reference trajectory, and the desired reference trajectory is typically set by the leader's state trajectory. Once consistency is achieved in a leader-follower system, only a few followers can obtain the leader's state information. Therefore, an observer is required to obtain the leader's real-time state information so that all followers can track the leader's information and better achieve the control objective. Considering the implementation of inclusive control in practical engineering applications, the safety of individuals in unmanned swarm systems and the maintenance of connections must be addressed. Ensuring individual safety and avoiding collisions with obstacles within individuals and their environment are fundamental requirements. Convergence rate is crucial for evaluating the performance of a control system. Traditional consensus only requires system convergence, with no time limit imposed, which is inconsistent with practical engineering applications. Therefore, researchers have begun studying the problem of achieving control objectives within a limited timeframe. Based on the above analysis, it is essential to implement inclusive control for unmanned swarm systems in complex environments within a specified timeframe. Summary of the Invention

[0004] In order to solve the problems existing in the above-mentioned prior art, the present invention proposes an inclusive control method for a multi-leader intelligent unmanned cluster system with obstacle avoidance and continuity maintenance, the method comprising: constructing an intelligent unmanned cluster system; taking each individual in the system as an intelligent agent, dividing the intelligent agents into leaders and followers, and establishing dynamic equations of leaders and followers; setting a collision avoidance inclusive control protocol, and using a time sedimentation function to conditionally restrict the collision avoidance inclusive control protocol; the followers use the conditionally restricted collision avoidance inclusive control protocol to obtain the status information of the leader; and the followers perform collision avoidance and connectivity maintenance based on the obtained status information.

[0005] Preferably, the intelligent unmanned swarm system is composed of a second-order dynamic equation, including n followers and m leaders; the dynamic equation of the leader includes:

[0006] x i =v i

[0007] v i =g i

[0008] i={n+1,n+2,...,n+m}

[0009] Among them, x i ∈R n ,v i ∈R n is the status information of the i-th leader, g i is the control input status.

[0010] Preferably, the collision avoidance containment control protocol includes: an effective exponential potential field function based on estimated distance constraints and a distributed control protocol.

[0011] Furthermore, the expression of the effective exponential potential field function based on the estimated distance constraint is:

[0012]

[0013]

[0014]

[0015] Among them, R col is the minimum distance for collision between agents, R con is the minimum distance to maintain connectivity, x ij The distance between two individual homes.

[0016] Preferably, the formula for obtaining the leader's status information includes:

[0017]

[0018]

[0019] in, and are the position and speed estimates of the i-th agent within the convex hull formed by the leader, and γ≥d1 are parameters chosen by the user.

[0020] Beneficial effects of the present invention

[0021] 1. This invention uses a distance-based potential field function. It considers scenarios involving both internal and external collision avoidance and internal connectivity maintenance. This is closer to practical application than most existing results. The proposed potential field function not only maintains safe contact with neighboring agents but also allows agents to avoid obstacles in the environment.

[0022] 2. The multi-leader-follower system studied in this invention is different from the leaderless or single leader-follower system. In this system, not all followers can know the status information of the leader. Instead, they obtain the real-time information of the leader through the designed observer.

[0023] 3. The follower control protocol in the system of the present invention is designed based on the estimated information obtained by the observer. Through this protocol, the follower can achieve inclusive control.

[0024] 4. The present invention achieves control objectives within a specified time. Unlike finite and fixed time, convergence within a specified time is not affected by the system's initial state or the upper limit of the settling time. In other words, the system can achieve the corresponding control objectives within a specified time value arbitrarily assigned by the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a system control flow chart of an embodiment of the present invention;

[0026] Figure 2 A communication topology diagram of an individual unmanned cluster system according to an embodiment of the present invention;

[0027] Figure 3 A diagram showing the evolution of the position state of the convex hull of the leader estimated by the observer within a specified time period according to an embodiment of the present invention;

[0028] Figure 4 A velocity state evolution diagram of the leader convex hull estimated by the observer within a specified time period according to an embodiment of the present invention;

[0029] Figure 5 A state evolution diagram of a follower in an environment with obstacles within a specified time period according to an embodiment of the present invention;

[0030] Figure 6 A diagram showing the evolution of a follower's position state in an environment with obstacles within a specified time period according to an embodiment of the present invention;

[0031] Figure 7 A diagram showing the evolution of the speed of a follower in an environment with obstacles within a specified time period according to an embodiment of the present invention;

[0032] Figure 8 This is a state evolution diagram of the position tracking error of a follower in an environment with obstacles according to an embodiment of the present invention;

[0033] Figure 9 FIG. 1 is a state evolution diagram of the speed tracking error of a follower in an environment with obstacles according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0035] A method for inclusive control of a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuity maintenance includes: constructing an intelligent unmanned swarm system; treating each individual in the system as an intelligent agent, dividing the intelligent agents into leaders and followers, and establishing dynamic equations for the leaders and followers; setting a collision avoidance inclusive control protocol, and using a time sedimentation function to conditionally restrict the collision avoidance inclusive control protocol; the followers use the conditionally restricted collision avoidance inclusive control protocol to obtain the leader's status information; and the followers perform collision avoidance and connectivity maintenance based on the obtained status information.

[0036] The topological structure in the intelligent unmanned cluster system referred to in the present invention is set with two constraints. One is that the directed topological graph is assumed to be a connected graph, and the other is that there is a spanning tree with the leader as the root in the communication topology of the intelligent agent, that is, there is at least one path from the root node in the topological graph to any node; the intelligent agent referred to is each individual in the unmanned cluster system.

[0037] A specific implementation method of an inclusive control method for a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuous maintenance, such as Figure 1 As shown, the method includes but is not limited to the following steps:

[0038] S1. Construct a system model and consider an unmanned swarm system consisting of n followers and m leaders, which is composed of second-order dynamic equations. The follower dynamic equation is:

[0039] x i =v i ,v i =u i , i={1,2,...,n}

[0040] Among them, x i ∈R n ,v i ∈R n and u i ∈R n are the position state, velocity state and control input state of the i-th agent respectively.

[0041] The leader's dynamic equation is as follows:

[0042] x i =v i , vi =g i , i={n+1,n+2,...,n+m}

[0043] Among them, x i ∈R n ,v i ∈R n is the status information of the i-th leader. g i is the control input state and is continuous and bounded within time t.

[0044] And set up a communication topology that meets some specific conditions, such as Figure 1 As shown, these conditions are: first, it is assumed that the topology graph is a connected graph; second, there is a spanning tree with the leader as the root in the communication topology of the intelligent agent, that is, there is at least one path from the root node to any node in the topology graph.

[0045] S2. Considering the actual size of the intelligent agents in the system and the limited communication range, as well as the scenarios where there are obstacles outside the system, an effective exponential potential field function based on the estimated distance constraint is designed to achieve the collision resistance and connectivity maintenance of the intelligent agents in the unmanned swarm system within a specified time.

[0046]

[0047]

[0048]

[0049] Among them, R col is the minimum distance for collision between agents, R con is the minimum distance to maintain connectivity. x ij is the distance between the two agents.

[0050] The following formula can be obtained by derivatizing the designed potential field function:

[0051]

[0052] Among them, col (x ij ) represents the collision avoidance function between two individuals, x i represents the position of the i-th agent, x j represents the position of the jth agent, o con (x ij ) represents the function of maintaining connectivity between two individuals, R col is the minimum distance for collision between agents, R con is the minimum distance to maintain connectivity, x ij is the distance between two individuals.

[0053] The repulsive force is designed to act on the intelligent agent along the negative gradient direction of the potential field function, so that it stays away from the collision avoidance area and thus safely completes the control goal.

[0054] S3. In a multi-leader-follower system, not all followers can obtain the leader's information. Therefore, followers need to use observers to track the leader's status information in real time.

[0055]

[0056]

[0057] For followers, and is the position and speed estimate of the convex hull formed by the i-th agent on the leader. When the i-th agent is the leader itself, and holds true, k represents a parameter greater than zero, represents the first-order derivative of the specified time function, η represents the exponential time function, a ij Represents the adjacency matrix of the topological graph, l represents the number of leaders, n represents the number of followers, and sign represents the sign function. and γ≥d1 are parameters chosen by the user;

[0058] Specify the time sedimentation function as:

[0059]

[0060] Wherein, h represents a constant greater than zero, T represents the time specified by the user, t0 represents the initial time, t represents the current time, and t1 = t0 + T.

[0061] The state estimates of position and velocity obtained by the estimator act on the following control protocol, and an estimator-based inclusive control protocol is proposed.

[0062] S4. The inclusion control protocol is combined with the potential field function to obtain a new collision avoidance inclusion control protocol, and the specified time settlement function is used to constrain the protocol so that all followers can complete collision avoidance and connectivity maintenance within any pre-allocated time and enter the convex hull formed by the leader.

[0063]

[0064]

[0065]

[0066] Where k1>0, is the information state of the leader observed by the observer; L1 is the Laplace matrix of the topology graph of the unmanned cluster system; ξ is the specified time function, is the tracking error, It is the repulsive force exerted on the intelligent agents in the unmanned swarm system.

[0067] The expression for specifying the time function is:

[0068]

[0069] The tracking error is:

[0070]

[0071]

[0072] This example considers an intelligent unmanned swarm system consisting of n agents and three leaders. For the study of unmanned swarm systems, we first abstract the system into a network system consisting of multiple nodes. The individuals in the system are considered nodes, and the complex relationships between nodes are considered as the communication between agents. The essence of a network system is represented by a graph. Given a graph G r =<v,ε>, where v represents the vertex set of the graph, usually expressed as v={1,2,...,n,n+1,...,n+m}, the first n nodes are considered followers in the system, and the last m nodes are leaders; ε={(v i ,v j )|v i ,v j ∈v} is the set of edges connecting vertices. If (v i ,v j )∈v and (v j ,v i )∈v, then we call the graph G r is an undirected graph, where the order group (v i ,v j ) represents the information flow from individual j to individual i, that is, the information exchange between individuals i and j. The neighbor set N of individual i i It consists of individuals other than individual i who have information exchanges with it, that is, N i ={v j |(v j ,v i )∈v}. Vertex v i The degree of deg(v i ) is N i The cardinality of .

[0073] For unmanned cluster systems with dynamically changing network topology, proximity graph is a useful tool to describe communication relationships. Its adjacency matrix A consists of elements a ij When j∈N i , a ij > 0, otherwise a ij =0, and a ii = 0, where i, j∈{1,2,...,n,n+1,...,n+m}. The degree matrix D is defined as D=diag(D1,D2,...,D n ,D n+1 ,...,D n+m ),in, is the in-degree of node i. In a leader-follower system, given a new graph Among them, v f ={v1,...,v n}and This graph is used to describe the information exchange between all followers, which is graph G r Similarly, the graph For an undirected graph when a ij =a ji ,j≠i,i,j∈{1,2,...,n}. The Laplace matrix of the system Where L1∈R n×n , L2∈R n×m .

[0074] In order to verify the effectiveness of the proposed intelligent unmanned swarm system inclusive control for achieving obstacle avoidance and connectivity maintenance within a specified time, Matlab was used for simulation verification. Figure 2 The experimental topology is an unmanned cluster system consisting of 9 nodes. Nodes L1 to L3 represent leaders, and nodes F1 to F6 represent followers. Regarding the parameter values in the system, the parameters in the observer are: h = 3, k = 3, c = 2, γ = 7, and T = 0.2s. The nonlinear function of the leader is: x = sin(t), and the initial value is: x L (t0) = [0] T ,ν L =[1] T The initial values of the follower's position and velocity are: F (t0) = [4.5, 2.5, 0.5, -1, -3, -3.5] and ν F =[3.5,1.5,0,2,4,2.5]. Through the set observer, the follower can observe the convex hull information of the leader within the specified time. The simulation results are as follows Figure 3 and Figure 4 shown.

[0075] Secondly, consider that in real applications, there are more or less obstacles in the environment. In this experiment, we set three static obstacles and set the agent's neighbors as dynamic obstacles. There are three leaders and six followers in the system. The nonlinear functions of the three leaders are: x1 = sin(t), x2 = sin(t)-1, x3 = sin(t+1). The initial position and velocity of the leaders are: x L (t0)=[0 0,-1 -1,1 2] T ,ν L =[1 1,1 1,1 1] T ; The initial values of the follower's position and velocity are: x F (t0)=[4.5 2.5,2.5 3.5,0.5 1.5,-1 1,-3 -2.5,-3.5 -1.5], ν F (t0)=[3 3,1.5 1.5,0 0,-2 -2,-3 -3,-2.5 -2.5]. The parameters in the controller are: k1=1, R col =0.5, R con =2, T=0.8s, it can be concluded from the simulation results that Figure 5 The figure shows the 3D state motion trajectories of all agents in an environment with obstacles. The blue ones are three static obstacles. It can be seen that the agents perform avoidance actions at the locations with obstacles. After this action, the agents are able to enter the convex hull of the leader again. This demonstrates the effectiveness of our proposed potential field function and control protocol. Figure 6 and Figure 7 ,From the 2-D perspective, it can be observed that the follower ,completed the tracking within the specified time, and achieved ,position and velocity tracking within 0.8s. Figure 8 and Figure 9 The corresponding position and velocity tracking errors are shown in the figure. As can be seen from the figure, the tracking error converges to zero and stabilizes within 0.8 seconds. This demonstrates that the followers in the unmanned swarm system achieved collision avoidance, connectivity maintenance, and consistent control objectives within the specified time.

[0076] Simulation experiments have verified that the designed control protocol, which includes both obstacle avoidance and inclusive control, enables multiple following agents in a system with random initial states to quickly avoid collisions, maintain connectivity, and achieve inclusive control within a specified timeframe. Compared to convergence methods with asymptotic, finite, or fixed time, this method converges faster and more stably, making it more practical for unmanned swarm systems.

[0077] It should be noted that those skilled in the art will appreciate that all or part of the processes in the above method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0078] The above embodiments further illustrate the purpose, technical solutions and advantages of the present invention in detail. It should be understood that the above embodiments are only preferred implementation plans of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made to the present invention within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An inclusive control method for a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuous maintenance, characterized in that: include: Build an intelligent unmanned swarm system; treat each individual in the system as an intelligent agent, divide the intelligent agents into leaders and followers, and establish dynamic equations for leaders and followers; Set up a collision avoidance inclusive control protocol and use a time sedimentation function to conditionally restrict the collision avoidance inclusive control protocol; the follower uses the conditionally restricted collision avoidance inclusive control protocol to obtain the leader's status information; Followers perform collision avoidance and connectivity maintenance based on the acquired status information; The collision avoidance control protocol includes: an effective exponential potential function based on estimated distance constraints and a distributed control protocol; The expression of the effective exponential potential field function based on the estimated distance constraint is: in, R col is the minimum distance for collision between agents, R con is the minimum distance to maintain connectivity, x ij is the distance between two individual homes; Distributed control protocols include: Where k1 is a constant greater than 0, is the information state of the leader observed by the observer; ξ is the specified time function; θ is the tracking error; is the repulsive force on the intelligent agent in the unmanned swarm system, representing the repulsive force; It is an intermediate variable composed of tracking error, Laplace matrix and exponential function. L1 represents the Laplace matrix of the topology graph of the unmanned cluster system. represents the position tracking error, represents the derivative of a specified time function, represents the velocity tracking error, λ1 represents the first eigenvalue of the Laplace matrix, represents an intermediate variable consisting of a specified time function and tracking error, Represents the second-order derivative function of the specified time function; The time sedimentation function is: Where h is a constant greater than zero, T is the time specified by the user, t0 is the initial time, and t is the current time.

2. The inclusive control method of a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuous maintenance according to claim 1 is characterized in that: The intelligent unmanned swarm system consists of a second-order dynamic equation consisting of n followers and m leaders. The leader's dynamic equation includes: x i =v i v i =g i i={n+1,n+2,...,n+m} The follower dynamic equations include x i =v i v i =u i i={1,2,...,n} Among them, x i and v i is the status information of the i-th leader, g i is the control input status.

3. The inclusive control method of a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuous maintenance according to claim 1, characterized in that: Intermediate parameters The expression is: Among them, col (x ij ) represents the collision avoidance function between two individuals, x i represents the position of the i-th agent, x j represents the position of the jth agent, o con (x ij ) represents the function of maintaining connectivity between two individuals, R col is the minimum distance for collision between agents, R con is the minimum distance to maintain connectivity, x ij is the distance between two individuals.

4. The inclusive control method of a multi-leader intelligent unmanned swarm system with obstacle avoidance and continuous maintenance according to claim 1, characterized in that: The formula for obtaining the leader's status information includes: in, and are the position and speed estimates of the i-th agent within the convex hull formed by the leader, and γ≥d1 are parameters selected by the user, k represents a parameter greater than zero, represents the first-order derivative function of the specified time function, η represents the specified time function, a ij Represents the adjacency matrix of the topological graph, l represents the number of leaders, n represents the number of followers, and sign represents the sign function.

Images