A multi-agent spatiotemporal precise cooperative control method based on bias proportional guidance

Abstract

The present invention discloses a multi-agent spatiotemporal precise collaborative control method based on biased proportional guidance, which belongs to the field of multi-agent coordinated control and includes the following steps: S1, constructing a relative kinematic model and a multi-agent network communication structure; S2, generating a virtual guide point according to the multi-agent and target position information; S3, assigning each agent its own tracking virtual guide point; S4, designing an agent normal acceleration control with a terminal sight angle constraint, and using error feedback to design an agent tangential acceleration control; S5, updating the state information of each agent in the network communication structure in real time, and updating the normal acceleration and tangential acceleration of each agent, so that each agent reaches the corresponding virtual guide point at the same time. The present invention adopts the above-mentioned multi-agent spatiotemporal precise collaborative control method based on biased proportional guidance, which can autonomously coordinate the remaining time of each agent to finally reach consistency, thereby achieving terminal time consistency.

Description

Technical Field

[0001] The present invention relates to the field of multi-agent coordinated control, and in particular to a multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance. Background Art

[0002] In the modern and complex battlefield environment, the strategic position of intelligent agents that integrate control theory, communication technology, navigation technology, etc. is constantly improving. Therefore, it is of great significance to study the precise spatiotemporal collaborative control methods of multiple intelligent agents outside the radar detection range of enemy targets.

[0003] Multiple agents precisely coordinate their movement trajectories outside the enemy target radar detection range and simultaneously enter the enemy radar detection range. The heading angle of each agent when entering is equal to the line of sight angle, which can exert tremendous strategic pressure on the enemy target defense system, thereby greatly improving the destructive effectiveness of the multiple agents.

[0004] Spatiotemporal collaboration refers to the collaborative work between multiple agents, multiple systems or multiple tasks in the two dimensions of space and time. The existing methods for the spatiotemporal collaborative control of intelligent agents for targets are as follows:

[0005] Fahimi F. published a paper titled Sliding-mode formation control for underactuated surface vessels. He designed a formation controller for underactuated ships using the sliding mode control law. The controller design took into account the parameter uncertainty of the dynamic model and wave disturbances, and realized the arbitrary formation of ships around the target under wave disturbances.

[0006] Breivik M, Hovstein VE, Fossen TI disclosed in "Straight-line target tracking for unmanned surface vehicles" that for under-actuated intelligent agents, the parallel approach method commonly used in missile guidance methods is introduced into the design of intelligent agent linear tracking controller, so that the intelligent agent can track targets moving in a high-speed straight line.

[0007] Although the above research has achieved certain results, it is only applicable to the tracking or formation of a target by a single intelligent agent.

[0008] The existing research on multi-agent target collaborative control is as follows:

[0009] Glotzbach T, Schneider M, Otto P proposed a cooperative line of sight target tracking method in "Cooperative line of sight target tracking for heterogeneous unmanned marine vehicle teams: From theory topractice", and applied it to heterogeneous unmanned marine vehicle teams to achieve tracking of underwater moving targets.

[0010] It can be seen that the existing collaborative control methods for multi-agents have the following limitations:

[0011] 1. Most of them are aimed at realizing the collaborative tracking of relatively close targets, and there is little research on the collaborative control methods of multiple agents in complex combat scenarios against enemy targets;

[0012] 2. No terminal heading angle limit is set for each agent in the multi-agent system. Summary of the invention

[0013] To solve the above problems, the present invention provides a multi-agent spatiotemporal precise collaborative control method based on biased proportional guidance, which can autonomously coordinate their respective remaining times to eventually reach a consensus, thereby achieving terminal time consistency.

[0014] To achieve the above object, the present invention provides a multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance, comprising the following steps:

[0015] S1. Construct a relative kinematics model based on the relative motion relationship between the agent and the target, and build a multi-agent network communication structure;

[0016] S2. Obtain the position information of all agents in the multi-agent system through information interaction, and generate a virtual guidance point based on the multi-agent and target position information. The virtual guidance point is located at the boundary of the target perception range.

[0017] S3, assigning each intelligent agent a virtual tracking point according to the position information of each intelligent agent and the virtual guide point;

[0018] S4. Based on the relative kinematics model and multi-agent network communication structure constructed in S1, the biased proportional guidance method is used to design the normal acceleration control of the agent with the terminal sight angle constraint, and the error feedback is used to design the tangential acceleration control of the agent;

[0019] S5. Update the status information of each intelligent agent in the network communication structure in real time, and update the normal acceleration and tangential acceleration of each intelligent agent according to the real-time position relationship between each intelligent agent and the virtual guide point, so that each intelligent agent reaches the corresponding virtual guide point at the same time.

[0020] Preferably, step S1 specifically includes the following steps:

[0021] S11. Construct a relative kinematics model based on the relative motion relationship between the agent and the target:

[0022] S111. Assume that in the north-east coordinate system A multi-agent system consisting of agents is established. Kinematic model:

[0023] ;

[0024] In the formula, Representing an Agent The speed of Representing an Agent The heading angle; Representing an Agent The heading angular rate; Representing an Agent The rate of change of speed; , ) represents the agent The rate of change of position coordinates; and Respectively represent intelligent agents Normal acceleration and tangential acceleration;

[0025] S112. Assume that the position coordinates of the target in the north-east coordinate system are , a relative kinematics model is established based on the relative motion relationship between the agent representation and the target:

[0026] ;

[0027] In the formula, Representing an Agent The rate of change of relative distance to the target; Representing an Agent The relative distance to the target, and ; Agent The position coordinates of the center of mass; Representing an Agent Angular rate of sight to the target; Representing an Agent The line of sight angle to the target, and ; Representing an Agent The angle between the velocity direction and the line of sight, and ;

[0028] S12. Constructing a multi-agent network communication structure:

[0029] Constructing a multi-agent network topology ,in, Represents a vertex set, where each element represents an agent; represents a set of edges, where each edge represents the information interaction relationship between two agents;

[0030] If the network topology is an undirected topology, the condition that the network topology is a connected graph must be met. If the network topology is a directed topology, the condition that there is at least one directed spanning tree must be met.

[0031] Preferably, step S2 specifically includes the following steps:

[0032] S21, each agent performs the following information exchange with its neighbors in the network topology structure: sends its own location information to its neighbors, and receives location information sent by its neighbors. After n-1 information exchanges, each agent obtains the location information of all agents in the multi-agent structure;

[0033] S22, generating virtual group location points based on the location information of all agents in the multi-agent body ;

[0034] S23, generating a first virtual guide point according to the positional relationship between the virtual group position point and the target :

[0035] ;

[0036] In the formula, Indicates the target radar monitoring range; Indicates the sight angle between the virtual group position point and the target;

[0037] S24, continue to generate the first k Virtual guide points :

[0038] ;

[0039] In the formula, ; represents the angle between two adjacent agents and the target at the preset terminal moment, and , Indicates the distance between adjacent agents at the preset terminal moment.

[0040] Preferably, in step S22, the virtual group location point The expression is as follows:

[0041] ;

[0042] In step S23, the sight angle between the virtual group position point and the target .

[0043] Preferably, step S3 specifically includes the following steps:

[0044] S31. Arrange in descending order according to the initial sight angle of each agent and renumber them , so that the sight angles of each agent after renumbering satisfy ;

[0045] S32, according to the sight angle between each virtual guide point and the target Sort in descending order and renumber , so that the renumbered backsight angle satisfy ;

[0046] S33, assign the same number to the agent Assign virtual guide points , at this time the agent Assigned virtual guide points Virtual Guide Point .

[0047] Preferably, in step S32, the sight angle .

[0048] Preferably, step S4 specifically includes the following steps:

[0049] S41. Design of normal acceleration control of intelligent body with terminal sight angle constraint using bias proportional guidance method:

[0050] S411, according to the intelligent agent The current state of the virtual guidance point is used to estimate the remaining time to reach the corresponding virtual guidance point. :

[0051] ;

[0052] In the formula, Representing an Agent To the corresponding virtual guide point The relative distance of ; Representing an Agent With virtual guide point The lead angle is , and the remaining time represents the time interval between the current time and the estimated time of arrival at the virtual guidance point;

[0053] S412, calculate the agent according to the multi-agent network communication structure and the remaining time estimate of each agent Remaining time estimate Remaining time estimate with neighbors The co-consistency error between :

[0054] ;

[0055] In the formula, It is an intelligent agent Adjacency Matrix in Network Communication Topology The elements in Indicates all A collection of real matrices; if the agent Can receive intelligent The information sent is ,otherwise , and define ;

[0056] S413, according to the multi-agent network communication structure and the agent Remaining time collaborative consistency error, design agent The collaborative navigation ratio is:

[0057] ;

[0058] In the formula, is a constant greater than 0; is a constant greater than 0 and less than 1; The function definition is as follows:

[0059] ;

[0060] In the formula, express The independent variable in the function;

[0061] S414、Designing Intelligent Agents Normal acceleration with terminal sight angle constraint Control is:

[0062] ;

[0063] S42. Designing Intelligent Agents Using Error Feedback The tangential acceleration control:

[0064] ;

[0065] In the formula, is a constant greater than 0, is a constant greater than 1.

[0066] Preferably, in step S5, the agent state information includes position information and speed information of each agent.

[0067] Preferably, in step S5, each agent updates normal acceleration and shear velocity control according to its own real-time state information and neighbor's real-time state information;

[0068] In step S5, the network communication structure is updated according to the positional relationship between the agent and the virtual guide point. If there is an agent among the multiple agents that arrives at the virtual guide point, the information exchange between the agent that arrives at the virtual guide point and the remaining agents is stopped, and the network communication structure of the remaining multiple agents is updated; otherwise, the current network communication structure is maintained.

[0069] The present invention has the following beneficial effects:

[0070] 1. It can automatically generate virtual guidance points based on the location information of each agent and target, combined with the target radar detection range and the preset distance between adjacent agents, and assign the corresponding virtual guidance points to each agent;

[0071] 2. There is no need to specify the terminal time of each agent in advance. The terminal time is the time when each agent arrives at its own virtual guidance point. Multiple agents can autonomously coordinate their remaining time and finally reach a consensus, thus achieving terminal time consistency;

[0072] 3. The heading angle of each agent to the target at the terminal moment is equal to the sight angle, thereby increasing the attack capability of multiple agents;

[0073] In summary, the present invention enables multiple intelligent agents to autonomously coordinate and arrive at the target radar monitoring range at the same time within a fixed time in a combat scenario according to the initial conditions and target positions of each intelligent agent, thereby greatly improving the damage efficiency of multiple intelligent agents on enemy targets in a combat scenario.

[0074] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a multi-agent spatiotemporal precise collaborative control method based on bias proportional guidance of the present invention;

[0076] Figure 2 It is a schematic diagram of the angle relationship between the agent and the target in a multi-agent spatiotemporal precise cooperative control method based on bias proportional guidance of the present invention;

[0077] Figure 3 It is a schematic diagram of the multi-agent network communication structure of the simulation experiment of the present invention;

[0078] Figure 4 It is a schematic diagram of the change of motion trajectory of multiple intelligent agents in the simulation experiment of the present invention;

[0079] Figure 5 is a schematic diagram of the distance between each intelligent agent and the corresponding virtual guide point changing over time in the simulation experiment of the present invention;

[0080] Figure 6 It is a schematic diagram of the change of the lead angle between each intelligent agent and the corresponding virtual guide point over time in the simulation experiment of the present invention;

[0081] Figure 7 is a schematic diagram of the change of the sight angle between each intelligent agent and the corresponding virtual guide point over time in the simulation experiment of the present invention;

[0082] Figure 8 It is a schematic diagram of the change of the speed of each intelligent agent over time in the simulation experiment of the present invention. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical scheme and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention are further described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions.

[0084] It should be noted that the terms "include", "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or devices.

[0085] Like reference numerals and letters denote similar items in the following drawings, and thus, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0086] like Figure 1 and Figure 2 As shown, a multi-agent spatiotemporal precise collaborative control method based on bias proportional guidance includes the following steps:

[0087] S1. Construct a relative kinematics model based on the relative motion relationship between the agent and the target, and build a multi-agent network communication structure;

[0088] Step S1 specifically includes the following steps:

[0089] S11. Construct a relative kinematics model based on the relative motion relationship between the agent and the target:

[0090] S111. Assume that in the north-east coordinate system (the origin is fixed, the coordinate axes are fixed to point to the north and east, and it can be approximately regarded as an inertial coordinate system) A multi-agent system consisting of agents is established. Kinematic model:

[0091] ;

[0092] In the formula, Representing an Agent The speed of Representing an Agent The heading angle; Representing an Agent The heading angular rate; Representing an Agent The rate of change of speed; , ) represents the agent The rate of change of position coordinates; and Respectively represent intelligent agents Normal acceleration and tangential acceleration;

[0093] S112. Assume that the position coordinates of the target in the north-east coordinate system are , a relative kinematics model is established based on the relative motion relationship between the agent representation and the target:

[0094] ;

[0095] In the formula, Representing an Agent The rate of change of relative distance to the target; Representing an Agent The relative distance to the target, and ; Agent The position coordinates of the center of mass; Representing an Agent Angular rate of sight to the target; Representing an Agent The line of sight angle to the target, and ; Representing an Agent The angle between the velocity direction and the line of sight, and ;

[0096] S12. Constructing a multi-agent network communication structure:

[0097] Constructing a multi-agent network topology ,in, Represents a vertex set, where each element represents an agent; represents a set of edges, where each edge represents the information interaction relationship between two agents;

[0098] If the network topology is an undirected topology, the condition that the network topology is a connected graph must be met. If the network topology is a directed topology, the condition that there is at least one directed spanning tree must be met.

[0099] S2. Obtain the position information of all agents in the multi-agent system through information interaction, and generate a virtual guidance point based on the multi-agent and target position information. The virtual guidance point is located at the boundary of the target perception range.

[0100] Step S2 specifically includes the following steps:

[0101] S21, each agent performs the following information exchange with its neighbors in the network topology structure: sends its own location information to its neighbors, and receives location information sent by its neighbors. After n-1 information exchanges, each agent obtains the location information of all agents in the multi-agent structure;

[0102] S22, generating virtual group location points based on the location information of all agents in the multi-agent body ;

[0103] In step S22, the virtual group location point The expression is as follows:

[0104] ;

[0105] S23, generating a first virtual guide point according to the positional relationship between the virtual group position point and the target :

[0106] ;

[0107] In the formula, Indicates the target radar monitoring range; Indicates the sight angle between the virtual group position point and the target;

[0108] In step S23, the sight angle between the virtual group position point and the target .

[0109] S24, continue to generate the first kVirtual guide points :

[0110] ;

[0111] In the formula, ; represents the angle between two adjacent agents and the target at the preset terminal moment, and , Indicates the distance between adjacent agents at the preset terminal moment.

[0112] S3, assigning each intelligent agent a virtual tracking point according to the position information of each intelligent agent and the virtual guide point;

[0113] Step S3 specifically includes the following steps:

[0114] S31. Arrange in descending order according to the initial sight angle of each agent and renumber them , so that the sight angles of each agent after renumbering satisfy ;

[0115] S32, according to the sight angle between each virtual guide point and the target Sort in descending order and renumber , so that the renumbered backsight angle satisfy ;

[0116] S33, assign the same number to the agent Assign virtual guide points , at this time the agent Assigned virtual guide points Virtual Guide Point .

[0117] Preferably, in step S32, the sight angle .

[0118] S4. Based on the relative kinematics model and multi-agent network communication structure constructed in S1, the biased proportional guidance method is used to design the normal acceleration control of the agent with the terminal sight angle constraint, and the error feedback is used to design the tangential acceleration control of the agent;

[0119] Step S4 specifically includes the following steps:

[0120] S41. Design of normal acceleration control of intelligent body with terminal sight angle constraint using bias proportional guidance method:

[0121] S411, according to the intelligent agent The current state of the virtual guidance point is used to estimate the remaining time to reach the corresponding virtual guidance point. :

[0122] ;

[0123] In the formula, Representing an Agent To the corresponding virtual guide point The relative distance of ; Representing an Agent With virtual guide point The lead angle is , and the remaining time represents the time interval between the current time and the estimated time of arrival at the virtual guidance point;

[0124] S412, calculate the agent according to the multi-agent network communication structure and the remaining time estimate of each agent Remaining time estimate Remaining time estimate with neighbors The co-consistency error between :

[0125] ;

[0126] In the formula, It is an intelligent agent Adjacency Matrix in Network Communication Topology The elements in Indicates all A collection of real matrices; if the agent Can receive intelligent The information sent is ,otherwise , and define ;

[0127] S413, according to the multi-agent network communication structure and the agent Remaining time collaborative consistency error, design agent The collaborative navigation ratio is:

[0128] ;

[0129] In the formula, is a constant greater than 0; is a constant greater than 0 and less than 1; The function definition is as follows:

[0130] ;

[0131] In the formula, express The independent variable in the function;

[0132] S414、Designing Intelligent Agents Normal acceleration with terminal sight angle constraint Control is:

[0133] ;

[0134] S42, design tangential acceleration To adjust the agent The tangential acceleration of the agent with a large remaining time accelerates the agent with a small remaining time and slows down the agent, thereby speeding up the terminal time coordination speed. The error feedback is used to design the agent The tangential acceleration control:

[0135] ;

[0136] In the formula, is a constant greater than 0, is a constant greater than 1.

[0137] S5. Update the status information of each intelligent agent in the network communication structure in real time, and update the normal acceleration and tangential acceleration of each intelligent agent according to the real-time position relationship between each intelligent agent and the virtual guide point, so that each intelligent agent reaches the corresponding virtual guide point at the same time.

[0138] In step S5, the agent state information includes the position information and speed information of each agent.

[0139] Preferably, in step S5, each agent updates normal acceleration and shear velocity control according to its own real-time state information and neighbor's real-time state information;

[0140] In step S5, the network communication structure is updated according to the positional relationship between the agent and the virtual guide point. If there is an agent among the multiple agents that arrives at the virtual guide point, the information exchange between the agent that arrives at the virtual guide point and the remaining agents is stopped, and the network communication structure of the remaining multiple agents is updated; otherwise, the current network communication structure is maintained.

[0141] Simulation experiment

[0142] In this simulation experiment, MATLAB software is used to simulate the precise spatiotemporal coordination of four intelligent agents to arrive at their respective virtual guidance points at the same time.

[0143] Table 1 The initial state of each agent and the sequence number after renumbering

[0144] ;

[0145] Table 2 The initial values ​​of the virtual guidance point positions and the sight angles of each agent to the virtual guidance point

[0146] ;

[0147] Table 3 Statistics of enemy target position, radar monitoring range and preset adjacent agent distance

[0148] ;

[0149] Establish as Figure 3 The network communication structure shown is a directed graph and satisfies the condition of the existence of a directed spanning tree.

[0150] In the virtual software simulation experiment, the parameters in the normal acceleration and tangential acceleration control law are as follows: .

[0151] The simulation results of the precise spatiotemporal coordinated control of multiple intelligent agents are as follows: Figure 4-Figure 8 As shown, Figure 4 It means that each agent in the multi-agent system has successfully arrived at its corresponding virtual guidance point; Figure 5 Table 4 shows that each agent arrives at the corresponding virtual guidance point in a coordinated manner, and the error time does not exceed 0.05s; Figure 6 Indicates that the leading angle of each agent changes smoothly and converges to zero at the terminal moment, so that the heading angle at the terminal moment is equal to the line of sight angle; Figure 7 It means that the sight angle to the virtual guidance point will eventually be equal to the sight angle to the target by the agent under the control law; Figure 8 Indicates the change in the speed of each agent. The agent with a large remaining time accelerates, and the agent with a small remaining time decelerates.

[0152] Table 4 Statistics of the time it takes for each agent to reach the corresponding virtual guidance point in a coordinated manner

[0153] ;

[0154] According to the above virtual simulation experiment, the multi-agent spatiotemporal precise collaborative control method based on biased proportional guidance of the present invention can obtain the multi-agent and target position information through network communication, and can autonomously generate virtual guidance points and correspond to the agents respectively, without specifying the virtual guidance points, and the multi-agents can reach the virtual guidance points in a coordinated manner under the designed acceleration control law, thereby greatly improving the destructive effectiveness of multi-agent combat, and can be applied to scenarios such as unmanned boats fighting against the enemy at sea, unmanned vehicles fighting against the enemy on land, and unmanned aerial vehicles fighting against the enemy in the air, and has high research value and broad application potential.

[0155] Therefore, the present invention adopts the above-mentioned multi-agent spatiotemporal precise collaborative control method based on bias proportional guidance, which can autonomously coordinate their respective remaining times to finally reach a consensus, thereby achieving terminal time consistency.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance, characterized by: The following steps are involved: S1. Construct a relative kinematics model based on the relative motion relationship between the agent and the target, and build a multi-agent network communication structure; S2. Obtain the position information of all agents in the multi-agent system through information interaction, and generate a virtual guidance point based on the multi-agent and target position information. The virtual guidance point is located at the boundary of the target perception range. Step S2 specifically includes the following steps: S21, each agent performs the following information exchange with its neighbors in the network topology structure: sends its own location information to its neighbors, and receives location information sent by its neighbors. After n-1 information exchanges, each agent obtains the location information of all agents in the multi-agent structure; S22, generating virtual group location points based on the location information of all agents in the multi-agent body ; In step S22, it is assumed that there is A multi-agent system composed of agents, virtual group location point The expression is as follows: In the formula, Agent The position coordinates of the center of mass; In step S23, the sight angle between the virtual group position point and the target , where Indicates the position coordinates of the target in the north-east coordinate system; S23, generating a first virtual guide point according to the positional relationship between the virtual group position point and the target : In the formula, Indicates the target radar monitoring range; Indicates the sight angle between the virtual group position point and the target; S24, continue to generate the first k Virtual guide points : In the formula, ; represents the angle between two adjacent agents and the target at the preset terminal moment, and , Indicates the distance between adjacent agents at the preset terminal moment; S3, assigning each intelligent agent a virtual tracking point according to the position information of each intelligent agent and the virtual guide point; S4. Based on the relative kinematics model and multi-agent network communication structure constructed in S1, the biased proportional guidance method is used to design the normal acceleration control of the agent with the terminal sight angle constraint, and the error feedback is used to design the tangential acceleration control of the agent; S5, updating the state information of each agent in the network communication structure in real time, and updating the normal acceleration and tangential acceleration of each agent according to the real-time position relationship between each agent and the virtual guide point, so that each agent reaches the corresponding virtual guide point at the same time; Step S3 specifically includes the following steps: S31. Arrange in descending order according to the initial sight angle of each agent and renumber them , so that the sight angles of each agent after renumbering satisfy ; S32, according to the sight angle between each virtual guide point and the target Sort in descending order and renumber , so that the renumbered backsight angle satisfy ; S33, assign the same number to the agent Assign virtual guide points , at this time the agent Assigned virtual guide points Virtual Guide Point .

2. The multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance according to claim 1 is characterized by: Step S1 specifically includes the following steps: S11. Construct a relative kinematics model based on the relative motion relationship between the agent and the target: S111. Assume that in the north-east coordinate system A multi-agent system consisting of agents is established. Kinematic model: In the formula, Representing an Agent The speed of Representing an Agent The heading angle; Representing an Agent The heading angular rate; Representing an Agent The rate of change of velocity; , ) represents the agent The rate of change of position coordinates; and Respectively represent intelligent agents Normal acceleration and tangential acceleration; S112. Assume that the position coordinates of the target in the north-east coordinate system are , a relative kinematics model is established based on the relative motion relationship between the agent representation and the target: In the formula, Representing an Agent The rate of change of relative distance to the target; Representing an Agent The relative distance to the target, and ; Agent The position coordinates of the center of mass; Representing an Agent Angular rate of sight to the target; Representing an Agent The line of sight angle to the target, and ; Representing an Agent The angle between the velocity direction and the line of sight, and ; S12. Constructing a multi-agent network communication structure: Constructing a multi-agent network topology ,in, Represents a vertex set, where each element represents an agent; represents a set of edges, where each edge represents the information interaction relationship between two agents; If the network topology is an undirected topology, the condition that the network topology is a connected graph must be met. If the network topology is a directed topology, the condition that there is at least one directed spanning tree must be met.

3. The multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance according to claim 2 is characterized by: In step S32, the sight angle .

4. The multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance according to claim 3 is characterized by: Step S4 specifically includes the following steps: S41. Design of normal acceleration control of intelligent body with terminal sight angle constraint using bias proportional guidance method: S411, according to the intelligent agent The current state of the virtual guidance point is used to estimate the remaining time to reach the corresponding virtual guidance point. : In the formula, Representing an Agent To the corresponding virtual guide point The relative distance of ; Representing an Agent With virtual guide point The lead angle is , and the remaining time represents the time interval between the current time and the estimated time of arrival at the virtual guidance point; S412, calculate the agent according to the multi-agent network communication structure and the remaining time estimate of each agent Remaining time estimate Remaining time estimate with neighbors The co-consistency error between : In the formula, It is an intelligent agent Adjacency Matrix in Network Communication Topology The elements in Indicates all A collection of real matrices; if the agent Can receive intelligent The information sent is ,otherwise , and define ; S413, according to the multi-agent network communication structure and the agent Remaining time collaborative consistency error, design agent The collaborative navigation ratio is: In the formula, is a constant greater than 0; is a constant greater than 0 and less than 1; The function definition is as follows: In the formula, express The independent variable in the function; S414、Designing Intelligent Agents Normal acceleration with terminal sight angle constraint Control is: ； S42. Designing Intelligent Agents Using Error Feedback The tangential acceleration control: In the formula, is a constant greater than 0, is a constant greater than 1.

5. The multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance according to claim 4 is characterized by: In step S5, the agent state information includes the position information and speed information of each agent.

6. The multi-agent spatiotemporal precise coordinated control method based on bias proportional guidance according to claim 5 is characterized by: In step S5, each agent updates the normal acceleration and shear velocity control according to its own real-time state information and the real-time state information of its neighbors; In step S5, the network communication structure is updated according to the positional relationship between the agent and the virtual guide point. If there is an agent among the multiple agents that arrives at the virtual guide point, the information exchange between the agent that arrives at the virtual guide point and the remaining agents is stopped, and the network communication structure of the remaining multiple agents is updated; otherwise, the current network communication structure is maintained.

Images