Multi-agent cooperative control method, device and equipment and storage medium

Abstract

The application relates to the technical field of multi-agent cooperative coverage control, and discloses a multi-agent cooperative control method, device and equipment and a storage medium, the method comprising the following steps: constructing a first / second type supervision area according to initial / primary position coordinates of the first / second type agent; constructing a common supervision area according to the first type supervision area and the second type supervision area, and determining optimal / advantage target positions of the first / second type agent according to all area points in the common supervision area; updating the first / second type supervision area when the first / second type agent moves towards the optimal / advantage target position; returning to the step of constructing the common supervision area according to the first type supervision area and the second type supervision area according to the updated first / second type supervision area until a coverage loss value corresponding to the common supervision area meets a preset optimal coverage loss value. The application aims to improve the coverage capability of heterogeneous agent cooperative control.

Description

Technical Field

[0001] This application relates to the field of multi-agent cooperative control technology, and in particular to a multi-agent cooperative control method, apparatus, device, and storage medium. Background Technology

[0002] With the rapid development of IoT and AI technologies, multi-agent collaborative coverage control technology has shown significant application value in fields such as environmental monitoring and disaster relief.

[0003] Current multi-agent cooperative coverage control technologies are mainly geared towards homogeneous agent systems, which are insufficient to meet the needs of multi-type agent cooperative response in real-world scenarios. For example, in complex scenarios such as forest fires, it is necessary to deploy heterogeneous agents such as fire fighting and air purification simultaneously. Existing methods lack optimized design for cooperative coverage of heterogeneous agents, resulting in insufficient coverage and low response efficiency.

[0004] Therefore, improving the coverage capability of collaborative control of heterogeneous intelligent agents is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] The main objective of this application is to provide a multi-agent cooperative control method, apparatus, device, and storage medium, which aims to improve the coverage capability of heterogeneous agent cooperative control.

[0006] To achieve the above objectives, this application provides a multi-agent cooperative control method, which includes:

[0007] A first-type surveillance region is constructed based on the initial position coordinates of the first-type intelligent agent, and a second-type surveillance region is constructed based on the original position coordinates of the second-type intelligent agent;

[0008] A common regulatory region is constructed based on the first type of regulatory region and the second type of regulatory region, and the optimal target position of the first type of intelligent agent and the optimal target position of the second type of intelligent agent are determined based on all the regional points in the common regulatory region.

[0009] The first type of surveillance region is updated when the first type of intelligent agent moves toward the optimal target position, and the second type of surveillance region is updated when the second type of intelligent agent moves toward the optimal target position.

[0010] Based on the updated first type of regulatory area and the updated second type of regulatory area, return to the step of constructing a common regulatory area based on the first type of regulatory area and the second type of regulatory area, until the coverage loss value corresponding to the common regulatory area meets the preset optimal coverage loss value.

[0011] In one embodiment, the step of determining the optimal target location of the first type of intelligent agent and the optimal target location of the second type of intelligent agent based on all area points in the common monitoring area includes:

[0012] Determine a first distance from a regional target point in the common regulatory area to the first type of intelligent agent, and a second distance from the regional target point to the second type of intelligent agent, wherein the regional target point is any one of all regional points;

[0013] Based on the comparison results of the first distance and the second distance, the first regulatory sub-region of the first type of intelligent agent in the common regulatory region and the second regulatory sub-region of the second type of intelligent agent in the common regulatory region are determined;

[0014] Based on all second-type intelligent agents adjacent to the first-type intelligent agent, all first-type intelligent agents adjacent to the second-type intelligent agent, the first regulatory sub-region, and the second regulatory sub-region, the optimal target position of the first-type intelligent agent and the optimal target position of the second-type intelligent agent are determined.

[0015] In one embodiment, the step of determining the optimal target position of the first type of intelligent agent and the optimal target position of the second type of intelligent agent based on all second type intelligent agents adjacent to the first type of intelligent agent, all first type intelligent agents adjacent to the second type of intelligent agent, the first monitoring sub-region, and the second monitoring sub-region includes:

[0016] Based on the union operation of all second-type intelligent agents adjacent to the first-type intelligent agent with the first regulatory sub-region and the second regulatory sub-region respectively, the maximum fusion sub-region and the minimum fusion sub-region of the first-type intelligent agent are obtained.

[0017] By performing a union operation on all first-type intelligent agents adjacent to the second-type intelligent agent with the second regulatory sub-region and the first regulatory sub-region respectively, the maximum and minimum cooperative sub-regions of the second-type intelligent agent are obtained.

[0018] The optimal target position of the first type of agent is determined based on the maximum fusion sub-region and the minimum fusion sub-region, and the optimal target position of the second type of agent is determined based on the maximum cooperative sub-region and the minimum cooperative sub-region.

[0019] In one embodiment, the step of determining the optimal target location of the first type of agent based on the largest fusion sub-region and the smallest fusion sub-region includes:

[0020] Integrating the preset environmental field function over the maximum fusion sub-region yields the maximum environmental field integral mass of the first type of agent, and the maximum centroid position corresponding to the maximum environmental field integral mass is determined.

[0021] Integrating the environmental field function over the minimum fusion sub-region yields the minimum environmental field integral mass of the first type of agent, and the minimum centroid position corresponding to the minimum environmental field integral mass is determined.

[0022] The optimal target position of the first type of intelligent agent is obtained by performing a weighted average calculation based on the preset weight priority, the maximum environmental field integral mass, the maximum centroid position, the minimum environmental field integral mass, and the minimum centroid position.

[0023] In one embodiment, the step of determining the optimal target position of the second type of agent based on the maximum cooperative sub-region and the minimum cooperative sub-region includes:

[0024] Integrating the environmental field function over the largest cooperative sub-region yields the highest environmental field integral quality of the second type of agent, and the highest centroid position corresponding to the highest environmental field integral quality is determined.

[0025] Integrating the environmental field function over the minimum cooperative sub-region yields the lowest environmental field integral quality of the second type of agent, and the lowest centroid position corresponding to the lowest environmental field integral quality is determined.

[0026] The optimal target position of the second type of agent is obtained by performing a weighted average calculation based on the preset weight priority, the highest environmental field integral quality, the highest centroid position, the lowest environmental field integral quality, and the lowest centroid position.

[0027] In one embodiment, the step of determining a first regulatory sub-region of the first type of intelligent agent in the common regulatory region and a second regulatory sub-region of the second type of intelligent agent in the common regulatory region based on the comparison result of the first distance and the second distance includes:

[0028] The first distance and the second distance are compared to obtain the comparison result;

[0029] If the comparison result is that the first distance is less than or equal to the second distance, then the area in the common regulatory area where the first distance is less than or equal to the second distance is taken as the first regulatory sub-area;

[0030] If the comparison result is that the second distance is less than or equal to the first distance, then the area in the common regulatory area where the second distance is less than or equal to the first distance is designated as the second regulatory sub-area.

[0031] In one embodiment, the multi-agent cooperative control method includes:

[0032] Determine the first dynamic motion model of the first type of intelligent agent and the second dynamic motion model of the second type of intelligent agent;

[0033] The first type of intelligent agent is driven to move to the optimal target position according to the first dynamic motion model, and the second type of intelligent agent is driven to move to the optimal target position according to the second dynamic motion model.

[0034] Furthermore, to achieve the above objectives, this application also provides a multi-agent cooperative control device, which includes:

[0035] The construction module is used to construct a first type of regulatory region based on the initial position coordinates of the first type of intelligent agent, and to construct a second type of regulatory region based on the original position coordinates of the second type of intelligent agent;

[0036] A common regulatory region is constructed based on the first type of regulatory region and the second type of regulatory region, and the optimal target position of the first type of intelligent agent and the optimal target position of the second type of intelligent agent are determined based on all the regional points in the common regulatory region.

[0037] The first type of surveillance region is updated when the first type of intelligent agent moves toward the optimal target position, and the second type of surveillance region is updated when the second type of intelligent agent moves toward the optimal target position.

[0038] Based on the updated first type of regulatory area and the updated second type of regulatory area, return to the step of constructing a common regulatory area based on the first type of regulatory area and the second type of regulatory area, until the coverage loss value corresponding to the common regulatory area meets the preset optimal coverage loss value.

[0039] Each functional module of the multi-agent cooperative control device of this application implements the steps of the multi-agent cooperative control method of this application as described above during operation.

[0040] In addition, to achieve the above objectives, this application also provides a multi-agent cooperative control device, which includes a memory, a processor, and a multi-agent cooperative control program stored in the memory and executable on the processor. When the multi-agent cooperative control program is executed by the processor, it implements the steps of the above-described multi-agent cooperative control method.

[0041] In addition, to achieve the above objectives, this application also provides a storage medium, which is a computer-readable storage medium, on which a multi-agent cooperative control program is stored, and when the multi-agent cooperative control program is executed by a processor, it implements the steps of the multi-agent cooperative control method described above.

[0042] This application provides a multi-agent cooperative control method that achieves adaptive optimization of coverage effects for heterogeneous agents through dynamic cooperation and closed-loop feedback mechanisms. Specifically, this application constructs a first-type monitoring region based on the initial position coordinates of a first-type agent and a second-type monitoring region based on the original position coordinates of a second-type agent, thereby effectively constructing independent monitoring regions for each type of agent. Next, a common monitoring region is defined by the intersection of the first-type and second-type monitoring regions, allowing the functional requirements of different types of agents to be coupled and analyzed within the common monitoring region. Furthermore, by calculating the coverage status of all points within the common monitoring region in real time, the moving target positions of the two types of agents (i.e., the optimal target position of the first-type agent and the optimal target position of the second-type agent) are simultaneously optimized, ensuring that the two types of agents maintain their functional coverage characteristics during cooperative movement. The coverage capability is enhanced through dynamic position adjustment. When the first and second types of agents move to the optimal target positions, the independent monitoring areas corresponding to the different types of agents are updated simultaneously. The common monitoring area is then reconstructed iteratively to continuously optimize the target position decisions of the two types of agents. This allows each type of agent to reassess its coverage status after each move. By dynamically adjusting the boundaries of the monitoring area and the target position, the coverage loss value of the common monitoring area is gradually reduced until the preset optimal coverage loss value is met. This enables different types of agents (i.e., heterogeneous agents) to autonomously weigh the core responsibility area and collaborative coverage needs, achieving efficient coverage and optimized resource allocation in environmentally sensitive areas, thereby significantly improving the coverage capability of heterogeneous agent collaborative control. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating the first embodiment of the multi-agent cooperative control method of this application;

[0044] Figure 2 This is a schematic diagram of the motion trajectory of heterogeneous multi-agent in a two-dimensional environment in an embodiment of this application;

[0045] Figure 3 This is a schematic diagram showing the final positions of heterogeneous multi-agent systems in an embodiment of the present invention;

[0046] Figure 4 This is the coverage loss function involved in the embodiments of this application. A schematic diagram of the curve showing the change over time;

[0047] Figure 5 This is a schematic diagram of the structure of the multi-agent cooperative control device involved in the embodiments of this application;

[0048] Figure 6 This is a schematic diagram of the structure of the multi-agent collaborative control device involved in the embodiments of this application.

[0049] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0050] This application provides a multi-agent cooperative control method, referring to... Figure 1 As shown, Figure 1 This is a flowchart illustrating the first embodiment of the multi-agent cooperative control method of this application.

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0052] Environmental monitoring plays an irreplaceable and vital role in modern social development. As a fundamental component of the environmental protection system, environmental monitoring not only provides crucial data support for scientific research but also offers decision-making basis in key areas such as ecosystem protection, public health maintenance, disaster early warning, and sustainable development. With the deep application of advanced technologies such as the Internet of Things and artificial intelligence, the strategic value of intelligent environmental monitoring systems in environmental governance and sustainable development is becoming increasingly prominent. In recent years, multi-agent collaborative coverage control technology has emerged as a significant breakthrough in the field of environmental monitoring. Through the precise coverage of key areas by intelligent agent clusters, it has achieved comprehensive and real-time environmental monitoring. This technology has been successfully applied to several important scenarios, including forest fire early warning and pollutant diffusion tracking. In these applications, monitoring high-risk areas such as those with abnormal temperatures and elevated pollutant concentrations is particularly important. Existing technologies mainly divide the monitoring area by intelligent agents, acquire local environmental field data, and achieve global coverage through collaborative movement, thereby ensuring the accuracy and timeliness of monitoring data. In this model, emergencies within the area are typically responded to by a single responsible intelligent agent.

[0053] However, in real-world monitoring scenarios, handling certain emergencies requires the coordinated response of multiple agents. For example, in forest fires, it may be necessary to deploy multiple agents simultaneously, such as those for firefighting and rescue, and air purification. These scenarios present new challenges to coverage control technology: how to optimize the coordinated coverage of multiple agents to achieve a more efficient joint response capability after coverage. While existing research has proposed schemes for multi-agent coordinated coverage, these mainly target homogeneous agent systems. When faced with complex scenarios requiring the coordination of multiple functional agents, such as fire scenes requiring heterogeneous agents for firefighting and rescue, and air purification, the applicability of existing methods is clearly insufficient.

[0054] Therefore, in order to overcome the technical defects of the above-mentioned traditional methods, this application provides a multi-agent cooperative control method, apparatus, device and storage medium.

[0055] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, mobile phone, database system, etc., or a device capable of performing the above functions, such as a multi-agent collaborative control device. The following description uses a multi-agent collaborative control device as an example to illustrate this embodiment and the subsequent embodiments.

[0056] The multi-agent cooperative control method of this application includes the following implementation steps S10 to S40.

[0057] Step S10: Construct a first type of regulatory region based on the initial position coordinates of the first type of intelligent agent, and construct a second type of regulatory region based on the original position coordinates of the second type of intelligent agent.

[0058] In this embodiment, a set of position coordinates for all first-type agents is obtained, which includes the position coordinates of each first-type agent at time [time value missing]. The initial position coordinates are determined; next, the preset coverage area of ​​the intelligent agent is determined. And detect the coverage area of ​​the intelligent agent. any region point within To the first type of intelligent agent exist Initial position coordinates at time distance Whether it does not exceed the area point To other first-type intelligent agents exist Distance from the initial position coordinates at time 1 Subsequently, all those meeting the distance requirement... Less than or equal to area points The set of points constitutes the first type of regulatory area. This enables the effective construction of independent regulatory regions for each type of first-class intelligent agent.

[0059] Similarly, obtain the agent position set of all second-type agents, which includes the position of each second-type agent at time [time value missing]. The original location coordinates are determined; next, the preset coverage area of ​​the intelligent agent is determined. And detect the coverage area of ​​the intelligent agent. any region point within To the second type of intelligent agent exist Original position coordinates at time distance Whether it does not exceed the area point To other second-type intelligent agents exist Distance from the original position coordinates at time [time] Subsequently, all those meeting the distance requirement... Less than or equal to area points The set of points constitutes the second type of regulatory area. This enables the effective construction of independent regulatory regions for each second-type intelligent agent.

[0060] It should be noted that the expression for the first type of regulatory area is:

[0061]

[0062] in, Represents the first type of intelligent agent The first type of regulatory area, also known as the first A first-type intelligent agent (i.e., a first-type intelligent agent) ) Always within the coverage area of ​​the intelligent agent The radiation area under the responsibility of the superior; Indicates the coverage area of ​​the intelligent agent Any point within the region; Represents the first type of intelligent agent exist The initial position coordinates at time 1, also known as the first-type agent. exist The position at that moment; Represents other first-type intelligent agents exist The initial position coordinates at time t, also known as the coordinates of other first-type intelligent agents. exist The position at that moment; This can be understood as the number of first-type intelligent agents.

[0063] The expression for the second type of regulatory area is:

[0064]

[0065] in, Represents the second type of intelligent agent The second type of regulatory area, also known as the third A second type of intelligent agent (i.e., a second type of intelligent agent) ) Always within the coverage area of ​​the intelligent agent The radiation area under the responsibility of the superior; Indicates the coverage area of ​​the intelligent agent Any point within the region; Represents the second type of intelligent agent exist The original position coordinates at time t, also known as the second-type agent. exist The position at that moment; Represents other second-type intelligent agents exist The original position coordinates at time t, also known as the original position coordinates of other first-type intelligent agents. exist The position at that moment; This can be understood as the number of second-type intelligent agents.

[0066] Other Type I intelligent agents Represents the first type of intelligent agent Neighboring agents of the same type, i.e., all other first-type agents The set of neighbors of the same type constitutes ;in, Represents the first type of intelligent agent The neighbor set of the same type consists of all other first-type agents. composition; Represents other first-type intelligent agents The first type of regulatory area; Indicates the empty set; These other first-type intelligent agents represent Type 1 regulatory area With the first type of intelligent agent Type 1 regulatory area There is an intersection.

[0067] Other Type II intelligent agents Represents the second type of intelligent agent Neighboring agents of the same type, i.e., all other agents of the second type. The set of neighbors of the same type constitutes ;in, Represents the second type of intelligent agent The neighbor set of the same type consists of all other second-type agents. composition; Represents other second-type intelligent agents The second type of regulatory area; Indicates the empty set; These other second-type intelligent agents Second type of regulatory area With the second type of intelligent agent Second type of regulatory area There is an intersection.

[0068] The first type of intelligent agent and the second type of intelligent agent can be understood as robots with different functional requirements. For example, if the application scenario is a forest fire early warning scenario, the first type of intelligent agent can be a fire fighting and rescue robot, and the second type of intelligent agent can be understood as an air purification robot.

[0069] Step S20: Construct a common regulatory region based on the first type of regulatory region and the second type of regulatory region, and determine the optimal target position of the first type of intelligent agent and the optimal target position of the second type of intelligent agent based on all the regional points in the common regulatory region.

[0070] In this embodiment, the first A first-type intelligent agent (i.e., a first-type intelligent agent) ) and the A first-type intelligent agent (i.e., a second-type intelligent agent) Taking the first type of intelligent agent as an example... Type 1 regulatory area and second type of intelligent agent Second type of regulatory area The intersection between them serves as the basis for two types of intelligent agents (i.e., Type I intelligent agents). and second type of intelligent agent The shared regulatory area; next, for each point in this shared regulatory area. Calculate its value to the first type of intelligent agent At Distance from the location at any given time and to the second type of intelligent agent At Distance from the location at any given time Subsequently, based on distance With distance The comparison results between them can accurately determine the first type of intelligent agent. The first regulatory sub-region dominated within the shared regulatory area, and the second type of intelligent agent. The second regulatory sub-region dominated within the shared regulatory area; finally, based on the first regulatory sub-region and the first type of intelligent agent. The optimal target location for the first type of intelligent agent is determined by the adjacent heterogeneous neighbor set, and based on the second supervisory sub-region and the second type of intelligent agent... The set of adjacent heterogeneous neighbors determines the optimal target location for the second type of agent.

[0071] In other words, this application uses dynamic computation to calculate two types of heterogeneous intelligent agents (i.e., the first type of intelligent agent). and second type of intelligent agent In each area within the shared regulatory region The distance to the current positions of both parties is calculated, and the regulatory sub-regions dominated by each party are accurately divided based on the distance comparison results. Then, combined with the collaborative relationship of adjacent heterogeneous neighbors, the optimal target positions of the first type of intelligent agent and the second type of intelligent agent are optimized and determined respectively. This effectively improves the collaboration efficiency and spatial allocation rationality of multiple intelligent agents in overlapping regulatory areas, avoids regulatory conflicts and resource waste, and enhances the adaptability of multiple intelligent agents to complex environments through dynamic position optimization. Ultimately, it maximizes the regulatory coverage and improves the overall performance of task execution.

[0072] It should be noted that the expression for a shared regulatory area is: ;in, Represents the first type of intelligent agent and second type of intelligent agent A shared regulatory area; Represents the first type of intelligent agent The first type of regulatory area; Represents the second type of intelligent agent The second type of regulatory area.

[0073] Type I intelligent agent The set of adjacent heterogeneous neighbors is ;in, Represents the first type of intelligent agent The heterogeneous neighbor set is composed of all the first-type intelligent agents. Adjacent second-type agents composition; Represents all agents of the first type. Adjacent second-type agents exist The second type of regulatory area at any given time; Indicates the empty set; These represent the first type of intelligent agent. Adjacent second-type agents exist Second type of regulatory area at any time With the first type of intelligent agent exist The first type of regulatory area at any time There is an intersection.

[0074] Type I intelligent agent The set of adjacent heterogeneous neighbors is ;in, Represents the second type of intelligent agent The heterogeneous neighbor set is composed of all the second-type intelligent agents. Adjacent first-type intelligent agents composition; Represents all agents of the second type. Adjacent first-type intelligent agents exist The first type of regulatory area at any given time; Indicates the empty set; These represent the second type of intelligent agent. Adjacent first-type intelligent agents exist The first type of regulatory area at any time With the second type of intelligent agent exist The first type of regulatory area at any time There is an intersection.

[0075] Step S30: Update the first type of monitoring region when the first type of intelligent agent moves toward the optimal target position, and update the second type of monitoring region when the second type of intelligent agent moves toward the optimal target position.

[0076] In this embodiment, by continuously optimizing and adjusting the overlapping regulatory space during the dynamic movement of two types of heterogeneous intelligent agents, their respective independent regulatory areas are updated in real time based on their actual position coordinates at the current moment. Specifically, when the first type of intelligent agent moves towards the optimal target position, its actual position coordinates also change. Updating its first-type regulatory area based on its current position coordinates reflects its current coverage capability in real time. Similarly, when the second type of intelligent agent moves, its second-type regulatory area is dynamically reconstructed. This real-time area update mechanism, which accompanies position changes, ensures that the division of the common regulatory area always maintains strong consistency with the actual spatial distribution of the two types of heterogeneous intelligent agents. This effectively solves the regulatory lag problem caused by traditional static partitioning, significantly improves the real-time performance, flexibility, and resource utilization efficiency of multi-agent collaborative regulation in complex dynamic environments, and provides an instant spatial awareness basis for scenarios such as obstacle avoidance and emergency task response.

[0077] Step S40: Based on the updated first type of regulatory area and the updated second type of regulatory area, return to the step of constructing a common regulatory area based on the first type of regulatory area and the second type of regulatory area, until the coverage loss value corresponding to the common regulatory area meets the preset optimal coverage loss value.

[0078] In this embodiment, the step of constructing a common regulatory region based on the updated first-type and second-type regulatory regions is returned. That is, through a cyclical iterative mechanism, the independent and common regulatory regions of the two heterogeneous agents are continuously updated and reconstructed, achieving self-optimization of collaborative regulation in a dynamic environment. Specifically, each time the common regulatory region is reconstructed based on the latest first-type and second-type regulatory regions, the coverage loss value of the common regulatory region is evaluated in real time. If the coverage loss value does not reach the preset optimal coverage loss value, a new round of optimization of the common regulatory region is triggered until the coverage loss value corresponding to the new common regulatory region converges to the preset optimal coverage loss value. This significantly improves the accuracy and robustness of multi-agent collaborative regulation in complex scenarios, effectively solves the adaptability defects of traditional static strategies, and achieves optimal allocation of regulatory resources by minimizing coverage loss.

[0079] It should be noted that the coverage loss function model for the shared regulatory area is as follows:

[0080]

[0081] in, This indicates the coverage loss value of the jointly regulated area. This indicates the preset coverage area of ​​the intelligent agent. This represents each area point within the shared regulatory area. To the first type of intelligent agent At The distance to the current location at any given time (i.e., the first distance); This represents each area point within the shared regulatory area. To the first type of intelligent agent At The distance to the location at that moment (i.e., the second distance); This represents a weighting function (i.e., an environmental field function) used to assign weights to different regions of points. Different levels of importance.

[0082] also, The fusion function represents the maximum and minimum response distances, and its specific form is as follows:

[0083]

[0084] in, This indicates the preset weight priority, and .

[0085] In summary, this application provides a multi-agent cooperative control method that achieves adaptive optimization of coverage effects for heterogeneous agents through dynamic cooperation and closed-loop feedback mechanisms. Specifically, this application constructs a first-type monitoring region based on the initial position coordinates of a first-type agent and a second-type monitoring region based on the original position coordinates of a second-type agent, thereby effectively constructing independent monitoring regions for each type of agent. Next, a common monitoring region is defined by the intersection of the first-type and second-type monitoring regions, allowing the functional requirements of different types of agents to be coupled and analyzed within the common monitoring region. Furthermore, by calculating the coverage status of all points within the common monitoring region in real time, the moving target positions of the two types of agents (i.e., the optimal target position of the first-type agent and the optimal target position of the second-type agent) are simultaneously optimized, ensuring that the two types of agents maintain their functional coverage characteristics while simultaneously optimizing their coverage during cooperative movement. The coverage capability is enhanced through dynamic position adjustment. When the first and second types of agents move to the optimal target positions, the independent monitoring areas corresponding to the different types of agents are updated simultaneously. The common monitoring area is then reconstructed iteratively to continuously optimize the target position decisions of the two types of agents. This allows each type of agent to reassess its coverage status after each move. By dynamically adjusting the boundaries of the monitoring area and the target position, the coverage loss value of the common monitoring area is gradually reduced until the preset optimal coverage loss value is met. This enables different types of agents (i.e., heterogeneous agents) to autonomously weigh the core responsibility area and collaborative coverage needs, achieving efficient coverage and optimized resource allocation in environmentally sensitive areas, thereby significantly improving the coverage capability of heterogeneous agent collaborative control.

[0086] Furthermore, based on the first embodiment of the multi-agent cooperative control method of this application, a second embodiment of the multi-agent cooperative control method of this application is proposed. In some feasible embodiments, the above step S20: constructing a common regulatory region based on the first type of regulatory region and the second type of regulatory region, and determining the optimal target position of the first type of intelligent agent and the optimal target position of the second type of intelligent agent based on all area points in the common regulatory region, further includes the following implementation steps S201~S203.

[0087] Step S201: Determine the first distance from the regional target point in the common monitoring area to the first type of intelligent agent, and the second distance from the regional target point to the second type of intelligent agent, wherein the regional target point is any one of the regional points.

[0088] In this embodiment, regional target points belonging to the common regulatory area are calculated. First distance to the first type of intelligent agent And calculate regional target points belonging to the common regulatory area. The second distance to the second type of intelligent agent It can accurately quantify the spatial relationship between the first / second type of intelligent agent and any point in the common regulatory area, providing accurate data support for collaborative coverage control among multiple intelligent agents.

[0089] It should be noted that the target point in a region is any one of the region points.

[0090] Step S202: Based on the comparison result of the first distance and the second distance, determine the first regulatory sub-region of the first type of intelligent agent in the common regulatory region, and the second regulatory sub-region of the second type of intelligent agent in the common regulatory region.

[0091] In this embodiment, by dividing the first regulatory sub-region dominated by the first type of intelligent agent in the common regulatory region and the second regulatory sub-region dominated by the second type of intelligent agent in the common regulatory region based on the comparison results of the first distance and the second distance, the common regulatory region can be scientifically and efficiently segmented.

[0092] In a specific embodiment, the first A first-type intelligent agent (first-type intelligent agent) ) and the A first-type intelligent agent (a second-type intelligent agent) For example, the first type of intelligent agent In a shared regulatory area The first sub-region under its jurisdiction is The specific expression for the first regulatory sub-region is: Second type of intelligent agent The second regulatory sub-region, dominated by the joint regulatory area, is The specific expression for this second regulatory sub-region is: .

[0093] Step S203: Based on all second-type intelligent agents adjacent to the first-type intelligent agent, all first-type intelligent agents adjacent to the second-type intelligent agent, the first regulatory sub-region, and the second regulatory sub-region, determine the optimal target position of the first-type intelligent agent and the optimal target position of the second-type intelligent agent.

[0094] In this embodiment, based on the first regulatory sub-region and the first type of intelligent agent The optimal target location for the first type of intelligent agent is determined by the adjacent heterogeneous neighbor set, and based on the second supervisory sub-region and the second type of intelligent agent... The set of adjacent heterogeneous neighbors determines the optimal target location for the second type of intelligent agent, thereby fully considering the agent's own regulatory responsibilities and the collaborative needs of surrounding heterogeneous intelligent agents, and achieving precise optimization of the agent's deployment location.

[0095] Furthermore, in some feasible embodiments, the above step S203: determining the optimal target position of the first type of intelligent agent and the optimal target position of the second type of intelligent agent based on all second type intelligent agents adjacent to the first type of intelligent agent, all first type intelligent agents adjacent to the second type of intelligent agent, the first regulatory sub-region and the second regulatory sub-region, further includes the following implementation steps S2031~S2033.

[0096] Step S2031: Perform a union operation on all adjacent second-type intelligent agents of the first type with the first regulatory sub-region and the second regulatory sub-region respectively to obtain the maximum fusion sub-region and the minimum fusion sub-region of the first type of intelligent agent.

[0097] In this embodiment, all agents of the first type will be involved. Adjacent second-type agents heterogeneous neighbor set With the first regulatory sub-region Perform a union operation to obtain the smallest fused sub-region of the first type of agent; and then combine this heterogeneous neighbor set. With the second regulatory sub-region Perform a union operation to obtain the largest fused sub-region of the first type of intelligent agent.

[0098] It should be noted that the smallest fusion sub-region is The largest fused sub-region is ,in, .

[0099] Step S2032: Based on the union operation of all first-type intelligent agents adjacent to the second-type intelligent agent with the second regulatory sub-region and the first regulatory sub-region respectively, the maximum cooperative sub-region and the minimum cooperative sub-region of the second-type intelligent agent are obtained.

[0100] In this embodiment, it will be composed of all agents of the second type. Adjacent first-type intelligent agents heterogeneous neighbor set With the second regulatory sub-region Perform a union operation to obtain the minimum cooperative sub-region of the second type of agent; and then combine this heterogeneous neighbor set. With the first regulatory sub-region Perform a union operation to obtain the largest cooperative sub-region of the second type of intelligent agent.

[0101] It should be noted that the smallest cooperative sub-region is The largest collaborative sub-region is ,in, .

[0102] Step S2033: Determine the optimal target position of the first type of agent based on the maximum fusion sub-region and the minimum fusion sub-region, and determine the optimal target position of the second type of agent based on the maximum cooperative sub-region and the minimum cooperative sub-region.

[0103] In this embodiment, the optimal target position of the first type of intelligent agent is determined by the maximum and minimum fusion sub-regions, and the optimal target position of the second type of intelligent agent is determined by the maximum and minimum cooperative sub-regions, thus realizing the optimal calculation of the deployment position of the first / second type of intelligent agents.

[0104] Furthermore, in some feasible embodiments, the above step S2033: determining the optimal target position of the first type of agent based on the largest fusion sub-region and the smallest fusion sub-region, further includes the following implementation steps A10~A30.

[0105] Step A10: Integrate the preset environmental field function over the maximum fusion sub-region to obtain the maximum environmental field integral mass of the first type of agent, and determine the position of the maximum centroid corresponding to the maximum environmental field integral mass.

[0106] In this embodiment, the preset environmental field function is integrated over the largest fusion sub-region to obtain the maximum environmental field integral mass of the first type of intelligent agent, and the position of the largest centroid corresponding to the maximum environmental field integral mass is determined, so that the high-value target (such as the core area of ​​a fire when the first type of intelligent agent is a fire fighting and rescue robot) can be focused on the area with the strongest coverage capability of the first type of intelligent agent.

[0107] It should be noted that the maximum environmental field integral mass of the first type of intelligent agent is Maximum environmental field integral mass Corresponding maximum centroid position Its specific expression is as follows:

[0108]

[0109] Step A20: Integrate the environmental field function over the minimum fusion sub-region to obtain the minimum environmental field integral mass of the first type of agent, and determine the minimum centroid position corresponding to the minimum environmental field integral mass.

[0110] In this embodiment, the environmental field function is integrated over the smallest fusion sub-region to obtain the minimum environmental field integral mass of the first type of intelligent agent, and the minimum centroid position corresponding to the minimum environmental field integral mass is determined, thereby effectively capturing key risk points in the weak coverage area (such as the location of trapped personnel when the first type of intelligent agent is a fire-fighting and rescue robot).

[0111] It should be noted that the minimum environmental field integral mass of the first type of intelligent agent is Minimum environmental field integral mass Corresponding minimum centroid position Its specific expression is as follows:

[0112]

[0113] Step A30: Calculate the optimal target position of the first type of agent by performing a weighted average calculation based on the preset weight priority, the maximum environmental field integral mass, the maximum centroid position, the minimum environmental field integral mass, and the minimum centroid position.

[0114] In this embodiment, the maximum environmental field integral quality and the maximum centroid position, as well as the minimum environmental field integral quality and the minimum centroid position, are calculated by weighted average calculation using preset weight priorities. This ensures that the optimal target position of the first type of intelligent agent not only prioritizes the key area with the maximum environmental field integral quality to release the strongest coverage performance of the first type of intelligent agent, but also takes into account the risk coverage requirements of weak areas by adjusting the weights according to task requirements (such as fire extinguishing priority or rescue priority).

[0115] It should be noted that the algorithm expression for calculating the weighted average of the first type of intelligent agents is as follows:

[0116]

[0117] in, This indicates the preset weight priority. Represents the first type of intelligent agent The optimal target location.

[0118] Furthermore, in some other feasible embodiments, the above step S2033: determining the optimal target position of the second type of agent based on the maximum cooperative sub-region and the minimum cooperative sub-region may also include the following implementation steps B10~B30.

[0119] Step B10: Integrate the environmental field function over the largest cooperative sub-region to obtain the highest environmental field integral quality of the second type of agent, and determine the highest centroid position corresponding to the highest environmental field integral quality.

[0120] In this embodiment, the preset environmental field function is integrated over the largest cooperative sub-region to obtain the highest environmental field integral quality of the second type of intelligent agent, and the highest centroid position corresponding to the highest environmental field integral quality is determined. This allows the focus to be placed on high-value targets in the area where the second type of intelligent agent has the strongest coverage capability (such as the peak area of ​​toxic gas concentration when the second type of intelligent agent is an air purification robot).

[0121] It should be noted that the highest environmental field integral quality of the second type of intelligent agent is Highest environmental field integral mass Corresponding highest centroid position Its specific expression is as follows:

[0122]

[0123] Step B20: Integrate the environmental field function over the minimum cooperative sub-region to obtain the minimum environmental field integral quality of the second type of agent, and determine the minimum centroid position corresponding to the minimum environmental field integral quality.

[0124] In this embodiment, the preset environmental field function is integrated over the smallest cooperative sub-region to obtain the lowest environmental field integral quality of the second type of intelligent agent, and the lowest centroid position corresponding to the lowest environmental field integral quality is determined, thereby effectively capturing key risk points in the weak coverage area (such as the location of a hidden trapped person when the second type of intelligent agent is an air purification robot).

[0125] It should be noted that the minimum environmental field integral mass of the second type of intelligent agent is Minimum environmental field integral mass Corresponding lowest centroid position Its specific expression is as follows:

[0126]

[0127] Step B30: Calculate the optimal target position of the second type of agent by performing a weighted average calculation based on the preset weight priority, the highest environmental field integral quality, the highest centroid position, the lowest environmental field integral quality, and the lowest centroid position.

[0128] In this embodiment, the highest environmental field integral quality and the highest centroid position, as well as the lowest environmental field integral quality and the lowest centroid position, are calculated by weighted average calculation through preset weight priorities. This ensures that the optimal target position of the second type of intelligent agent is both prioritized to point to the key area with the highest environmental field integral quality to maximize the effectiveness of the second intelligent agent, and also takes into account the risk coverage requirements of weak areas through weight adjustment according to task requirements (such as purification priority or rescue priority).

[0129] It should be noted that the algorithm expression for calculating the weighted average of the second type of intelligent agents is as follows:

[0130]

[0131] in, This indicates the preset weight priority. Represents the second type of intelligent agent The optimal target position.

[0132] Furthermore, in some feasible embodiments, the above step S202: determining the first regulatory sub-region of the first type of intelligent agent in the common regulatory region and the second regulatory sub-region of the second type of intelligent agent in the common regulatory region based on the comparison result of the first distance and the second distance may also include the following implementation steps S2021~S2023.

[0133] Step S2021: Compare the first distance and the second distance to obtain a comparison result.

[0134] In this embodiment, by comparing in real time the distances (i.e., the first distance and the second distance) from each point within the shared monitoring area to the two types of heterogeneous intelligent agents, a dynamic decision-making basis is provided for the spatial responsibility allocation of the two types of heterogeneous intelligent agents. In other words, by comparing the first distance and the second distance, a spatial proximity criterion is constructed, transforming the abstract cooperative coverage problem into a quantifiable geometric relationship analysis, which significantly improves the system response efficiency.

[0135] Step S2022: If the comparison result is that the first distance is less than or equal to the second distance, then the area in the common supervision area where the first distance is less than or equal to the second distance is taken as the first supervision sub-area.

[0136] In this embodiment, when the comparison result is that the first distance is less than or equal to the second distance, the area containing all the points that satisfy the condition that the first distance is less than or equal to the second distance is divided into the first regulatory sub-region, thereby realizing the granting of spatial dominance to the first type of intelligent agent.

[0137] Step S2023: If the comparison result is that the second distance is less than or equal to the first distance, then the area in the common supervision area where the second distance is less than or equal to the first distance is taken as the second supervision sub-area.

[0138] In this embodiment, when the comparison result is that the second distance is less than or equal to the first distance, the region containing all the points that satisfy the condition that the second distance is less than or equal to the first distance is divided into the second regulatory sub-region, thereby realizing the granting of spatial dominance to the second type of intelligent agent.

[0139] Furthermore, in some feasible embodiments, the multi-agent cooperative control method may also include the following implementation steps C10~C20.

[0140] Step C10: Determine the first dynamic motion model of the first type of intelligent agent and the second dynamic motion model of the second type of intelligent agent.

[0141] In this embodiment, a first dynamic motion model for a first type of intelligent agent and a second dynamic motion model for a second type of intelligent agent are determined.

[0142] It should be noted that the expression for the first dynamic motion model of the first type of intelligent agent is:

[0143]

[0144] in, Indicates the first indivual Type of intelligent agent Speed ​​of motion at any given moment; Indicates the first indivual Type of intelligent agent The input speed value at any given time.

[0145] The expression for the second dynamic motion model of the second type of intelligent agent is:

[0146]

[0147] in, Indicates the first indivual Type of intelligent agent Speed ​​of motion at any given moment; Indicates the first indivual Type of intelligent agent The input speed value at any given time.

[0148] Step C20: Drive the first type of intelligent agent to move to the optimal target position according to the first dynamic motion model, and drive the second type of intelligent agent to move to the optimal target position according to the second dynamic motion model.

[0149] In this embodiment, the first type of intelligent agent outputs the input velocity value through the first dynamic motion model. Drive to the current optimal target position, while enabling the second type of intelligent agent to output the input velocity value through the second dynamic motion model. Drive to the current optimal target position.

[0150] It should be noted that the input speed value Changes with optimal target position and Type I intelligent agents exist Initial position coordinates at time The input speed value changes accordingly. The transformation expression is as follows:

[0151]

[0152] in, Indicates the first indivual Type of intelligent agent The input speed value at any given time; Indicates the first indivual Type of intelligent agent The optimal target position at any given time; Represents the first type of intelligent agent exist The initial position coordinates at that moment; express Control gain coefficient of the type of intelligent agent.

[0153] Input speed value The change with the optimal target position and second type of intelligent agent exist Original position coordinates at time The input speed value changes accordingly. The transformation expression is as follows:

[0154]

[0155] in, Indicates the first indivual Type of intelligent agent The input speed value at any given time; Indicates the first A second-type intelligent agent in The optimal target position at any given time; Represents the second type of intelligent agent exist The original position coordinates at that moment; express Control gain coefficient of the type of intelligent agent.

[0156] In a specific embodiment, this application utilizes real-time position error (i.e., optimal target position). with initial position coordinates The difference, and the optimal target position Compared with the original position coordinates The difference) is used as the control input (i.e., the input speed value). And input speed value The direct basis for generation, combined with independently adjustable control gain coefficients. as well as It can drive two types of heterogeneous agents to reach their respective updated "current best / optimal target positions" synchronously with optimal dynamic response (such as exponential convergence) in parallel and reliably, which significantly improves the task execution efficiency, target tracking accuracy and real-time performance of heterogeneous agents in dynamic environments.

[0157] Furthermore, in another embodiment, consider a The two-dimensional environment (i.e., the pre-defined coverage area of ​​the intelligent agent) can be modeled as follows: .

[0158] Specifically, this embodiment uses two first-type intelligent agents and four second-type intelligent agents to achieve collaborative coverage of the entire region, and sets the parameters as follows: , Next, the coverage loss function established using the above parameters is used to measure the quality of coverage. A higher value of this function indicates worse coverage, while a lower value indicates better coverage.

[0159]

[0160] The initial positions of the two types of heterogeneous agents are randomly distributed, and the result of cooperative coverage is as follows: Figure 2 and Figure 3 As shown. Figure 2 This is a schematic diagram of the motion trajectories of heterogeneous multi-agent systems in a two-dimensional environment, as described in the embodiments of this application. Figure 2 Different sector-shaped arc contour lines represent the distribution of the environmental field. Figure 2 The label 10 in the text represents a first-type intelligent agent (i.e., first-type intelligent agent 10), that is to say Figure 2The image shows the motion trajectories of two Type I intelligent agents 10, in which... Figure 2 The cross sign at the beginning of agent 10 of type I indicates its initial position. Figure 2 The dotted end of the first type of intelligent agent 10 indicates its final position. Figure 2 The dashed trajectory of Type I intelligent agent 10 and the historical trajectory of Type I intelligent agent; Figure 2 The label 20 in the text represents a second-type agent (i.e., second-type agent 20), that is to say Figure 2 The image shows the motion trajectories of four Type II agents 20, among which... Figure 2 The cross sign at the end of the second type of intelligent agent 20 indicates its initial position. Figure 2 The dotted end of the second type of intelligent agent 20 indicates its final position. Figure 2 The dashed trajectory of the second type of intelligent agent 20 represents the historical trajectory of the second type of intelligent agent. Figure 3 This is a schematic diagram showing the final positions of heterogeneous multi-agent systems in an embodiment of the present invention. Figure 3 Different sector-shaped arc contour lines represent the distribution of the environmental field. Each hollow circle represents the final position of a Type 1 intelligent agent, and each solid circle represents the final position of a Type 2 intelligent agent. Bold diagonal lines represent the monitoring area of ​​a Type 1 intelligent agent, and intersecting lines represent the monitoring area of ​​a Type 2 intelligent agent. Combined with... Figure 2 and Figure 3 It can be seen that the two types of heterogeneous intelligent agents worked together to cover the entire area. Figure 4 This is the coverage loss function involved in the embodiments of this application. A schematic diagram showing the change over time. Under the distributed cooperative coverage control method (i.e., multi-agent cooperative control method) proposed in this application, the coverage loss function... It decreases monotonically over time, eventually reaching its minimum value, at which point the coverage effect is optimal.

[0161] In addition, this application also provides a multi-agent cooperative control device, please refer to... Figure 5 , Figure 5 This is a schematic diagram of the multi-agent cooperative control device involved in the embodiments of this application. The multi-agent cooperative control device provided in this application includes:

[0162] The construction module H01 is used to construct a first type of regulatory region based on the initial position coordinates of the first type of intelligent agent, and to construct a second type of regulatory region based on the original position coordinates of the second type of intelligent agent;

[0163] The location determination module H02 is used to construct a common regulatory area based on the first type of regulatory area and the second type of regulatory area, and to determine the optimal target location of the first type of intelligent agent and the optimal target location of the second type of intelligent agent based on all area points in the common regulatory area.

[0164] Update module H03 is used to update the first type of monitoring area when the first type of intelligent agent moves toward the optimal target position, and to update the second type of monitoring area when the second type of intelligent agent moves toward the optimal target position;

[0165] The loop module H04 is used to return to the step of constructing a common regulatory area based on the first type of regulatory area and the second type of regulatory area based on the updated first type of regulatory area and the updated second type of regulatory area, until the coverage loss value corresponding to the common regulatory area meets the preset optimal coverage loss value.

[0166] In addition, this application also provides a multi-agent cooperative control device. Please refer to... Figure 6 , Figure 6 This is a schematic diagram of the structure of the multi-agent cooperative control device involved in the embodiments of this application. Specifically, the device in the embodiments of this application can be a device that runs the multi-agent cooperative control method locally.

[0167] This application provides a multi-agent cooperative control device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the multi-agent cooperative control device method in the first embodiment described above.

[0168] The following is for reference. Figure 6 The diagram illustrates a structural schematic suitable for implementing the multi-agent cooperative control device of the embodiments of this application. The multi-agent cooperative control device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The multi-agent collaborative control device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0169] like Figure 6 As shown, the multi-agent cooperative control device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the multi-agent cooperative control device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output interface (i.e., an I / O interface) 1006 is also connected to the bus. Typically, the following devices can be connected to the input / output interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the multi-agent collaborative control device to communicate wirelessly or wiredly with other devices to exchange data. Although a multi-agent collaborative control device with various devices is shown in the figure, it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0170] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0171] The multi-agent cooperative control device provided in this application, employing the multi-agent cooperative control device method described in the above embodiments, can solve the technical problem of low reliability in multi-agent cooperative control devices. Compared with the prior art, the beneficial effects of the multi-agent cooperative control device provided in this application are the same as those of the multi-agent cooperative control device method described in the above embodiments, and other technical features in this multi-agent cooperative control device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0172] Furthermore, this application provides a computer-readable storage medium. This computer-readable storage medium stores a multi-agent cooperative control program, which, when executed by a processor, implements the steps of the aforementioned multi-agent cooperative control method.

[0173] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0174] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0175] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0176] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A multi-agent cooperative control method, characterized in that, The multi-agent cooperative control method comprises: constructing a first type of supervision area according to initial position coordinates of the first type of agent, and constructing a second type of supervision area according to original position coordinates of the second type of agent; constructing a common supervision area according to the first type of supervision area and the second type of supervision area, and determining a first distance from a region target point in the common supervision area to the first type of agent and a second distance from the region target point to the second type of agent according to all region points in the common supervision area, the region target point being any one of all region points; determining a first supervision sub-area of the first type of agent in the common supervision area and a second supervision sub-area of the second type of agent in the common supervision area according to a comparison result of the first distance and the second distance; performing a set operation on the first supervision sub-area and the second supervision sub-area respectively by all second type of agents adjacent to the first type of agent to obtain a maximum fusion sub-area and a minimum fusion sub-area of the first type of agent; performing a set operation on the second supervision sub-area and the first supervision sub-area respectively by all first type of agents adjacent to the second type of agent to obtain a maximum cooperative sub-area and a minimum cooperative sub-area of the second type of agent; determining a best target position of the first type of agent according to the maximum fusion sub-area and the minimum fusion sub-area, and determining an optimal target position of the second type of agent according to the maximum cooperative sub-area and the minimum cooperative sub-area; updating the first type of supervision area when the first type of agent moves towards the best target position, and updating the second type of supervision area when the second type of agent moves towards the optimal target position; returning to the step of constructing a common supervision area according to the first type of supervision area and the second type of supervision area according to the updated first type of supervision area and the updated second type of supervision area until a coverage loss value corresponding to the common supervision area meets a preset optimal coverage loss value.

2. The multi-agent cooperative control method of claim 1, wherein, The step of determining a best target position of the first type of agent according to the maximum fusion sub-area and the minimum fusion sub-area comprises: integrating a preset environmental field function on the maximum fusion sub-area to obtain a maximum environmental field integral quality of the first type of agent, and determining a maximum centroid position corresponding to the maximum environmental field integral quality; integrating the environmental field function on the minimum fusion sub-area to obtain a minimum environmental field integral quality of the first type of agent, and determining a minimum centroid position corresponding to the minimum environmental field integral quality; performing weighted average calculation according to a preset weight priority, the maximum environmental field integral quality, the maximum centroid position, the minimum environmental field integral quality, and the minimum centroid position to obtain the best target position of the first type of agent.

3. The multi-agent cooperative control method of claim 2, wherein, The step of determining the optimal target position of the second type agent according to the maximum synergy sub-region and the minimum synergy sub-region comprises: integrating the environmental field function on the maximum synergy sub-region to obtain the highest environmental field integral quality of the second type agent, and determining the highest centroid position corresponding to the highest environmental field integral quality; integrating the environmental field function on the minimum synergy sub-region to obtain the lowest environmental field integral quality of the second type agent, and determining the lowest centroid position corresponding to the lowest environmental field integral quality; performing weighted average calculation according to the preset weight priority, the highest environmental field integral quality, the highest centroid position, the lowest environmental field integral quality and the lowest centroid position to obtain the optimal target position of the second type agent.

4. The multi-agent cooperative control method of claim 1, wherein, The step of determining the first supervision sub-region of the first type agent in the common supervision region and the second supervision sub-region of the second type agent in the common supervision region according to the comparison result of the first distance and the second distance comprises: comparing the first distance and the second distance to obtain a comparison result; if the comparison result is that the first distance is less than or equal to the second distance, then the region in the common supervision region where the first distance is less than or equal to the second distance is taken as the first supervision sub-region; if the comparison result is that the second distance is less than or equal to the first distance, then the region in the common supervision region where the second distance is less than or equal to the first distance is taken as the second supervision sub-region.

5. The multi-agent cooperative control method of claim 1, wherein, The multi-agent cooperative control method comprises: determining a first dynamic motion model of the first type agent and a second dynamic motion model of the second type agent; driving the first type agent to move to the optimal target position according to the first dynamic motion model, and driving the second type agent to move to the optimal target position according to the second dynamic motion model.

6. A multi-agent cooperative control device, characterized by comprising: The multi-agent cooperative control device comprises: a construction module configured to construct a first type supervision region according to the initial position coordinates of the first type agent, and construct a second type supervision region according to the original position coordinates of the second type agent; The position determining module is configured to: construct a common supervision area according to the first type of supervision area and the second type of supervision area, and determine a first distance from a region target point in the common supervision area to the first type of agent and a second distance from the region target point to the second type of agent according to all region points in the common supervision area, the region target point being any one of the region points; determine a first supervision sub-area of the first type of agent in the common supervision area and a second supervision sub-area of the second type of agent in the common supervision area according to a comparison result of the first distance and the second distance; perform a set operation on the first supervision sub-area and the second supervision sub-area respectively by all second type of agents adjacent to the first type of agent, to obtain a maximum fusion sub-area and a minimum fusion sub-area of the first type of agent; perform a set operation on the second supervision sub-area and the first supervision sub-area respectively by all first type of agents adjacent to the second type of agent, to obtain a maximum coordination sub-area and a minimum coordination sub-area of the second type of agent; determine an optimal target position of the first type of agent according to the maximum fusion sub-area and the minimum fusion sub-area, and determine an optimal target position of the second type of agent according to the maximum coordination sub-area and the minimum coordination sub-area; The updating module is configured to update the first type of supervision area when the first type of agent moves towards the optimal target position, and update the second type of supervision area when the second type of agent moves towards the optimal target position. The loop module is configured to return to the step of constructing the common supervision area according to the first type of supervision area and the second type of supervision area according to the updated first type of supervision area and the updated second type of supervision area, until a coverage loss value corresponding to the common supervision area meets a preset optimal coverage loss value.

7. A multi-agent cooperative control device, characterized by comprising: The multi-agent cooperative control device includes a memory, a processor, and a multi-agent cooperative control program stored on the memory and executable on the processor, and the processor implements the steps of the multi-agent cooperative control method according to any one of claims 1 to 5 when executing the multi-agent cooperative control program.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a multi-agent cooperative control program, and the multi-agent cooperative control program implements the steps of the multi-agent cooperative control method according to any one of claims 1 to 5 when executed by the processor.

Images