Unmanned cluster target search control method and device based on local information, and medium

Abstract

The application provides a local information-based unmanned cluster target search control method and device, and a medium, and belongs to the technical field of group robot control. The method comprises the following steps: when a group robot explores a target, a tracking cluster is formed by allocating a set tracking number of intelligent agents to track the target, so that multiple targets can be tracked at the same time; a local coordinate system is established through the tracking cluster, local information is obtained through the local coordinate system, a comprehensive concentration field about obstacles and targets is generated through a GRN upper-layer network model, the combined velocity of the potential field force of each intelligent agent in the tracking cluster and the gradient information of each point in the comprehensive concentration field are calculated, so as to determine the running speed of each intelligent agent in the tracking cluster, so that the intelligent agent can realize rapid tracking of the target while avoiding collision with the obstacles, so as to complete group control. Through calculation based on local information, the calculation amount is greatly reduced, and the tracking efficiency and tracking accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of swarm robot control, in particular to a target search control method and device based on local information, and a medium. BACKGROUND

[0002] Swarm robot system tracking technology involves multiple frontier fields, including artificial intelligence, machine learning, deep learning, computer vision, and sensor technology. Through these technologies, intelligent agents can achieve efficient target tracking in applications such as autonomous driving, drone tracking, and robot navigation.

[0003] In the prior art, complex tracking tasks are converted into a partially observable Markov decision process (POMDP) problem, and an online solver is used to handle uncertainty and decision-making in dynamic environments, thereby enabling intelligent agents to effectively track moving targets. However, the intelligent agent of this method mainly relies on cameras for visual odometry and landmark detection, which is easily disturbed by environmental factors. In real-time applications, the algorithm needs to handle a large number of possible states and observations, resulting in high computational complexity, and the response of the intelligent agent may be delayed. SUMMARY

[0004] The main purpose of the embodiments of the present application is to provide a target search control method and device based on local information for unmanned clusters, which can greatly reduce the amount of calculation and improve tracking efficiency and accuracy.

[0005] To achieve the above purpose, one aspect of the embodiments of the present application provides a target search control method based on local information for unmanned clusters, which comprises:

[0006] When the swarm robot explores the target, a set number of intelligent agents are allocated from the swarm robot to form a tracking cluster to track the target;

[0007] Through the tracking cluster, a local coordinate system is established, and according to the local coordinate system, the first position information of the target, the second position information of the explored obstacles, and the third position information of each intelligent agent in the tracking cluster are determined;

[0008] According to the first position information and the second position information, a comprehensive concentration field of the obstacles and the target is generated through a GRN upper layer network model;

[0009] According to the third position information, a first set of intelligent agents having an attractive relationship with the current intelligent agent and a second set of intelligent agents having a repulsive relationship with the current intelligent agent are determined, and the combined velocity of the potential field force of each intelligent agent in the tracking cluster is calculated according to the first set, the second set, and the third position information.

[0010] The gradient information of each point in the comprehensive concentration field is determined through differential calculation, the running speed of each agent in the tracking cluster is determined according to the gradient information and the resultant velocity, and the agents in the tracking cluster are controlled to track the target according to the running speed.

[0011] Further, the local coordinate system is established by the tracking cluster, specifically including:

[0012] In the tracking cluster, three agents not on the same straight line are selected, and the three agents are respectively taken as a first coordinate agent, a second coordinate agent and a third coordinate agent.

[0013] The first coordinate agent is taken as a coordinate center o, a straight line pointing from the first coordinate agent to the second coordinate agent is taken as an x-axis, and a straight line perpendicular to the x-axis and pointing to one side of the third coordinate agent is taken as a y-axis, so as to establish the local coordinate system xoy.

[0014] Further, the third position information of each agent in the tracking cluster is determined according to the local coordinate system, specifically including:

[0015] The first distance from the first coordinate agent to the second coordinate agent, the second distance from the remaining agents to the first coordinate agent, and the third distance from the remaining agents to the second coordinate agent are obtained.

[0016] Based on the local coordinate system, the third position information is determined according to the first distance, the second distance and the third distance.

[0017] Further, for the current agent R i , the resultant velocity of the potential field force of each agent in the tracking cluster is calculated according to the first set, the second set and the third position information, specifically including:

[0018] According to the first set G and the third position information, the attractive force U att (d ik ) suffered by the current agent R i is calculated through an attractive field calculation formula, wherein the first set G is d ik is the distance between the current agent R i and G k , k = 1, 2, …, h, and h is the number of agents in the first set G;

[0019] According to the second set F and the third position information, the repulsive force U i suffered by the current agent R i is calculated through a repulsive field calculation formula.req (d ij ), where the second set F is d ij For the current intelligent agent R i With F j The distance, j = 1, 2, ..., l, where l is the number of agents in the second set F;

[0020] According to the gravity U att (d ik ) and the repulsive force U req (d ij ), calculate the current agent R i The resultant velocity V of the potential field force ip (t), where i = 1, 2, ..., m, m is the number of agents in the tracking cluster, to calculate the combined velocity of each agent in the tracking cluster.

[0021] Furthermore, the determination of gradient information at each point within the integrated concentration field through differential calculation specifically includes:

[0022] Central difference is performed on the non-boundary points within the comprehensive concentration field to determine the gradient information of the non-boundary points;

[0023] Forward and backward differences are performed on the boundary points within the comprehensive concentration field to determine the gradient information of the boundary points.

[0024] Furthermore, regarding the current intelligent agent R... i The step of determining the running speed of each agent in the tracking cluster based on the gradient information and the combined velocity specifically includes:

[0025] Based on the current intelligent agent R i The third location information is used to determine the gradient information of the corresponding point within the comprehensive concentration field;

[0026] Based on the gradient information at the corresponding point and the current agent R i The running speed of the current agent is determined by the corresponding combined speed, where i = 1, 2, ..., m, and m is the number of agents in the tracking cluster, so as to calculate the running speed of each agent in the tracking cluster.

[0027] Furthermore, the unmanned swarm target search control method based on local information also includes:

[0028] The swarm of robots is divided into several search groups, each of which includes several intelligent agents, and each search group starts from a corresponding initial position.

[0029] The intelligent agents in each search group perform decentralized exploration in the unknown environment according to a Gaussian random walk strategy to explore the target.

[0030] Further, when the swarm robots explore the target, assigning the set tracking number of intelligent agents from the swarm robots to form a tracking cluster specifically comprises:

[0031] When the target is explored, for the current search group, the current search group communicates with the search group that explores the target to determine the tracking number of intelligent agents that currently track the target;

[0032] It is determined whether the tracking number of intelligent agents is less than the set tracking number.

[0033] If yes, according to the tracking number of intelligent agents and the set tracking number, a distribution number is determined, and the distribution number of intelligent agents is separated from the current search group to track the target to form the tracking cluster, and the remaining intelligent agents in the current search group perform decentralized exploration according to the Gaussian random walk strategy.

[0034] To achieve the above object, another aspect of the embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned target search control method for a swarm of unmanned agents based on local information when executing the computer program.

[0035] To achieve the above object, still another aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the above-mentioned target search control method for a swarm of unmanned agents based on local information when executed by a processor.

[0036] The embodiment of the present application at least has the following beneficial effects: the present application provides a target search control method and device for a swarm of unmanned agents based on local information, and a medium, which forms a tracking cluster by assigning a set tracking number of intelligent agents to track the target, so that multiple targets can be tracked at the same time; a local coordinate system is established, local information is obtained through the local coordinate system, a GRN upper network model is introduced to construct a comprehensive concentration field, the gradient information of each point in the comprehensive concentration field and the resultant velocity of the potential field force of each intelligent agent are calculated, so as to determine the running speed of each intelligent agent in the tracking cluster, so that the intelligent agents realize rapid tracking of the target while avoiding collision with obstacles, to complete swarm control, and through calculation based on local information, the calculation amount is greatly reduced, and the tracking efficiency and tracking accuracy are improved. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1is a flow diagram of a local information-based unmanned cluster target search control method according to an embodiment of the present application;

[0038] Figure 2 is a comprehensive concentration field diagram of a local information-based unmanned cluster target search control method according to an embodiment of the present application;

[0039] Figure 3 is a diagram of a tracking cluster tracking multiple target exploration phases of a local information-based unmanned cluster target search control method according to an embodiment of the present application;

[0040] Figure 4 is a diagram of a tracking cluster tracking multiple target tracking phases of a local information-based unmanned cluster target search control method according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application. When the following description refers to the accompanying drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with embodiments of the present application, but are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0042] It can be understood that the terms "first", "second", and the like used herein can be used to describe various concepts in this document, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "when" or "when" or "in response to determining".

[0043] The terms "at least one", "multiple", "each", "any", and the like used herein include one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by a person skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0045] In order to facilitate the understanding of the inventive concept of the present application, before the detailed description of the embodiments of the present application, the English abbreviations (terms) related to the embodiments of the present application are first described, and the English abbreviations (terms) related to the embodiments of the present application are applicable to the following explanations.

[0046] Gene Regulatory Network (GRN) is a network describing the interaction relationship between genes and their regulatory factors in a biological body. The gene expression mechanism in biological morphogenesis can be applied to group robot control.

[0047] As described in the background, in the prior art, the agent mainly relies on the camera for visual odometry and landmark detection, which is easily disturbed by environmental factors, and in real-time applications, because the algorithm needs to process a large number of possible states and observations, the computational complexity is high, and the response of the agent may be delayed.

[0048] Therefore, the embodiments of the present application provide a target search control method and equipment based on local information for unmanned clusters, and a medium, which greatly reduces the amount of calculation and improves the tracking efficiency and tracking accuracy.

[0049] Figure 1 is an optional flowchart of the target search control method based on local information for unmanned clusters provided by the embodiments of the present application, Figure 1 The method in the above embodiment can include but is not limited to S100 to S500.

[0050] S100, when the group robot explores the target, a set number of agents are allocated from the group robot to form a tracking cluster to track the target;

[0051] S200, a local coordinate system is established through the tracking cluster, and the first position information of the target, the second position information of the explored obstacle and the third position information of each agent in the tracking cluster are determined according to the local coordinate system;

[0052] S300, according to the first position information and the second position information, a comprehensive concentration field of the obstacle and the target is generated through a GRN upper network model;

[0053] S400, according to the third position information, a first set of agents having an attractive relationship with the current agent and a second set of agents having a repulsive relationship with the current agent are determined, and the combined velocity of the potential field force of each agent in the tracking cluster is calculated according to the first set, the second set and the third position information;

[0054] S500, gradient information of each point in the integrated concentration field is determined through differential calculation, and running speed of each agent in the tracking swarm is determined according to the gradient information and the resultant velocity, so as to control the agents in the tracking swarm to track the target according to the running speed.

[0055] The S100 to S500 shown in the embodiments of the present application can track multiple targets at the same time by allocating the set number of agents to form a tracking swarm to track the target, establish a local coordinate system, obtain local information through the local coordinate system, introduce a GRN upper network model to construct an integrated concentration field, and calculate the resultant velocity of the potential field force of each agent and the gradient information of each point in the integrated concentration field, so as to determine the running speed of each agent in the tracking swarm, so that the agents can quickly track the target while avoiding collision with obstacles, thereby completing group control.

[0056] In S100, when the target is explored, the set number of agents is allocated from the swarm robots, and the set number of agents forms a tracking swarm to track the target.

[0057] In an embodiment, the swarm robots are equally divided into several search groups, and each search group starts from a set initial position and disperses in different directions to explore. During the search process, when a search group explores a target, the search group that explores the target communicates with the agents near the target to determine the number M of agents currently tracking the target. It is judged whether the number M of agents meets the set number N of tracking agents. When the number M of agents is less than the set number N of tracking agents, the number (N-M) of agents is allocated from the swarm robots to the search group that explores the target to form a tracking swarm, so that the number of agents in the tracking swarm meets the set number of tracking agents, so as to ensure that the target is fully tracked.

[0058] By fixing the number of agents tracking each target, a tracking swarm is formed, so that when the agent resources are sufficient, multiple targets can be tracked at the same time, and each target can be fully monitored.

[0059] In S200, when the tracking swarm is tracking, a local coordinate system covering the target and all obstacles needs to be established to determine the coordinates of the target and the obstacles as inputs of the gene regulatory network.

[0060] In the tracking swarm, a local coordinate system is established to determine the coordinates of the target, the coordinates of the explored obstacles and the coordinates of each agent in the tracking swarm, i.e., to determine the first position information, the second position information and the third position information, so as to obtain local information and reduce dependence on global information.

[0061] In S300, for the tracking cluster, the first position information of the target and the second position information of the obstacle are taken as inputs of the GRN upper-layer network model through the local coordinate system established in S200, a comprehensive concentration field based on the position information of the target and the position information of the obstacle is generated, so as to realize obstacle avoidance and tracking of the target in the tracking phase of the tracking cluster.

[0062] In S400, when the agent is tracking the target, the information of the obstacle has been integrated in the comprehensive concentration field, so that the obstacle avoidance in the tracking process can be solved through the collision avoidance of the agent.

[0063] Through the third position information of each agent in the tracking cluster, for the current agent in the tracking cluster, all agents having an attractive relationship and a repulsive relationship with the current agent are counted, so as to determine a first set and a second set. The first set is the agent having an attractive relationship with the current agent, and the second set is the agent having a repulsive relationship with the current agent.

[0064] Through the first set, the second set and the third position information, the resultant velocity of the potential field force of each agent in the tracking cluster is determined, so as to introduce the artificial potential field method on the basis of the comprehensive concentration field to realize obstacle avoidance, and further improve the efficiency of tracking the target.

[0065] In S500, since the comprehensive concentration field is continuous, but is discrete when stored on the terminal, the gradient information of each point in the comprehensive concentration field is calculated by using difference instead of gradient.

[0066] Through the resultant velocity of the potential field force superposition and the gradient information of each point obtained in S400, the running speed of each agent in the tracking cluster is determined, and the agents in the tracking cluster are controlled to track the target according to the running speed, so that based on the gradient method of the environment, the agents can move to the place with high concentration gradient, that is, move to the first position information of the target, so as to realize tracking of the target and obstacle avoidance at the same time.

[0067] In some embodiments of S200 of the present application, through the tracking cluster, the local coordinate system is established, and the third position information of each agent in the tracking cluster is determined according to the local coordinate system, which includes the following steps:

[0068] S210, in the tracking cluster, three agents not on the same straight line are selected, and the three agents are respectively taken as a first coordinate agent, a second coordinate agent and a third coordinate agent.

[0069] S220, taking the first coordinate agent as the coordinate center o, a straight line pointing from the first coordinate agent to the second coordinate agent is taken as the x-axis, and a straight line perpendicular to the x-axis and pointing to one side of the third coordinate agent is taken as the y-axis, so as to establish a local coordinate system xoy.

[0070] S230, obtain a first distance from the first coordinate agent to the second coordinate agent, a second distance from the rest of the agents to the first coordinate agent, and a third distance from the rest of the agents to the second coordinate agent.

[0071] S240, determine the third position information of the rest of the agents according to the first distance, the second distance, and the third distance.

[0072] In S210, three agents not on the same straight line are selected from the tracking swarm, which are respectively taken as the first coordinate agent R1, the second coordinate agent R2, and the third coordinate agent R3.

[0073] In S220, the first coordinate agent R1 is taken as the coordinate center o, and the straight line connecting the first coordinate agent R1 and the second coordinate agent R2 is taken as the x-axis. A straight line perpendicular to the x-axis and pointing to one side of the third coordinate agent R3 is taken as the y-axis to establish a local coordinate system xoy.

[0074] In S230, a first distance d 12 between the first coordinate agent R1 and the second coordinate agent R2 is obtained. 1i A second distance d 2i from the rest of the agents to the first coordinate agent is obtained. 12 1i 2i

[0075] In S240, the third position information of the agents is determined according to the first distance d

[0076]

[0077] wherein i = 1, 2, …, m, m is the number of agents in the tracking swarm, and θ is the angle between the agent and the x-axis. Through S220, the third position information of the first coordinate agent R1 (0, 0) and the third position information of the second coordinate agent R2 (d 12 , 0) can also be quickly determined.

[0078] Through S210 to S240, a local coordinate system xoy is established among the agents in the tracking swarm, which covers the target and all obstacles, so as to obtain local information and realize swarm control based on local information.

[0079] ​​​In some embodiments of the S300 of the present application, the generating, according to the first position information and the second position information, a comprehensive concentration field about the obstacles and the target through the GRN upper-layer network model comprises the following steps:

[0080] S310, generating, according to the first position information p1 of the target and the second position information p2 of the explored obstacles, n a first concentration field g1 about the first position information p1 and the second position information p2 and a second concentration field g2 about the second position information p2.

[0081] S320, generating a comprehensive concentration field through the first concentration field g1 and the second concentration field g2.

[0082] In the above embodiments, the tracking cluster determines the first position information p1 of the target and the second position information p2 of the explored obstacles through the local coordinate system xoy, and the expression is as follows:

[0083]

[0084] wherein p is a total matrix containing the first position information p1 and the second position information p2; (T x ,T y ) represents the first position information p1 of the target assigned, (O s,x ,O s,y )(s = 1, 2, …, z) represents the second position information p2 of the explored obstacles, and z is the number of explored obstacles.

[0085] Through the GRN upper-layer concentration field calculation formula:

[0086]

[0087]

[0088] wherein g1 is a concentration field formed by combining the target position information and the obstacle position information, i.e. the first concentration field about the first position information p1 and the second position information p2; g2 is a concentration field formed by the obstacle position, i.e. the second concentration field about the second position information p2; g3 represents a comprehensive concentration field fused with g1 and g2, corresponding to the generation of the tracking mode. θ1, θ2 are threshold values of the S-type function. k is a concentration field influence factor.

[0089] Through S310 and S320, the tracking cluster tracking the target takes the position information of the explored obstacles and targets in the local coordinate system xoy as the input of the GRN upper network model, and generates a comprehensive concentration field based on the target and obstacle position information. The gene regulation network model is introduced as the selection basis of the agent tracking, and the comprehensive concentration field generated by the detected target information and obstacle information is taken as the reference of the agent tracking target, so that flexible collision avoidance of the obstacles and rapid tracking of the target are realized.

[0090] In some embodiments of S400 of the present application, the calculation process of the resultant velocity of the potential field force specifically comprises:

[0091] S410, according to the third position information, the agent existing the attractive force field with the current agent is counted to form a first set, and the agent existing the repulsive force field with the current agent is counted to form a second set.

[0092] In this embodiment, when tracking the target, since the information of the obstacles has been integrated in the comprehensive concentration field, only the collision avoidance of the agents in the tracking cluster and the agents inside and outside the cluster needs to be concerned in the tracking cluster. According to the tracking cluster, for the current agent R i , all the agents existing the attractive force field and the repulsive force field with it are counted, and the expression is as follows:

[0093]

[0094] Wherein, G is the agent set existing the attractive force field with the current agent R i , that is, the first set, and F is the agent set existing the repulsive force field with the current agent R i , that is, the second set.

[0095] S420, for the current agent R i , according to the first set G and the third position information, the attractive force U att (d ik ) of the current agent R i is calculated through the attractive force field calculation formula.

[0096] In this embodiment, for the current agent R i , the distance d ik between the current agent R i and G k is determined through the third position information, d ik is a vector, and the direction is that the agent G k in the first set G points to the current agent R i , G kFor the elements in the first set G, that is, the agents in the first set G, k = 1, 2, …, h, h is the number of agents in the first set G, the gravitational field calculation formula is:

[0097]

[0098] The current agent R i receives the gravitational force U att (d ik ) where d att is the maximum safe distance, and the distance d i between the current agent R ik and other agents is greater than d att , then the gravitational force appears, and η is the proportional gain coefficient.

[0099] S430, for the current agent R i , according to the second set F and the third position information, the repulsive force field calculation formula is used to calculate the repulsive force U req (d ij ) received by the current agent R i .

[0100] In this embodiment, for the current agent R i , the distance d ij between the current agent R i and F j is determined by the third position information, d ij is a vector, and the direction is that the agent F j in the second set F points to the current agent R i , F j is an element in the second set F, that is, an agent in the second set F, j = 1, 2, …, l, l is the number of agents in the first set F, and the repulsive force field calculation formula is:

[0101]

[0102] The repulsive force U req (d ij ) received by the current agent R i is calculated, where d req is the minimum safe distance, and the distance d ij between the current agent R i and other agents is less than d att , then the repulsive force appears, and λ is the proportional gain coefficient.

[0103] S440, for the current agent R i , according to the gravitational force U att (d ik ) and the repulsive force dUreq (d ij ), the resultant velocity V i of the potential field force of the current agent R ip (t) is calculated.

[0104] In this embodiment, for the current agent R i , after the calculation through S420 and S430, the attractive force U att (d ik ) and the repulsive force U req (d jj ) are obtained, the resultant force field can be calculated as:

[0105]

[0106] Wherein, i = 1, 2, …, m, m is the number of agents in the tracking cluster, G k is an element in the first set G, that is, an agent in the first set G, F j is an element in the second set F, that is, an agent in the second set F.

[0107] Through the resultant force field, the potential field force and the velocity are solved, and the formula is as follows:

[0108]

[0109] V ip (t) = E(i)

[0110] Wherein, F(i) is the resultant force of the resultant force field, V ip (t) is the resultant velocity of the current agent R i superimposed by the potential field force, and so on, so as to determine the resultant velocity of the potential field force of each agent.

[0111] Through S410 to S440, the gene regulation network model is introduced to realize the tracking of the target, and the artificial potential field method is introduced for collision avoidance between agents, further improving the efficiency of exploration and tracking. On the basis of the comprehensive concentration field, the artificial potential field method is used for obstacle avoidance, reducing the amount of obstacle avoidance calculation.

[0112] Referring to Figure 2 , in some embodiments of S500 of the present application, the running speed determination process of each agent in the tracking cluster specifically includes:

[0113] S510, the central difference is performed on the non-boundary point in the comprehensive concentration field, and the gradient information of the non-boundary point is determined.

[0114] In this embodiment, referring to Figure 2, the target is in the positive gradient of the concentration field, and the obstacle is in the negative gradient of the concentration field, so the environment-based gradient method can not only track the target, but also avoid obstacles.

[0115] Since the integrated concentration field is continuous, but the integrated concentration field is discrete when stored on the terminal, the difference is used instead of the gradient, and the central difference method is used for non-boundary points in the integrated concentration field. For a certain point (x p ,y q ) in the integrated concentration field, the concentration gradient formula of the point is as follows:

[0116]

[0117] Where g(x,y) is the concentration field generated by the upper layer, is the concentration gradient at point (x p ,y q ), that is, the gradient information of point (x p ,y q ); g(x p+1 ,y q ), g(x p-1 ,y q ) are the next point and the previous point of g(x p ,y q ) along the positive direction of the X axis of the integrated concentration field.

[0118] g(x p ,y q+1 ), g(x p ,y q-1 ) are the next point and the previous point of g(x p ,y qw ) along the positive direction of the Y axis of the integrated concentration field.

[0119] S520, forward difference and backward difference are performed on the boundary points in the integrated concentration field to determine the gradient information of the boundary points.

[0120] In this embodiment, when (x p ,y q ) is located on the boundary of the integrated concentration field, forward difference and backward difference are needed. For the X axis of the integrated concentration field, the gradient formula when on the left and right boundaries is as follows:

[0121]

[0122] Where, when x p is located on the left boundary of the positive direction of the X axis of the integrated concentration field, the forward difference method is used, and when x pWhen the right boundary of the comprehensive concentration field in the positive direction of the X axis is located, the backward difference method is used, and the y of the boundary point on the left boundary of the X axis of the comprehensive concentration field is greater than the y of the boundary point on the right boundary of the X axis of the comprehensive concentration field q . q .

[0123] For the Y axis of the comprehensive concentration field, the gradient formula when the left and right boundaries are as follows:

[0124]

[0125] wherein, when y q When the left boundary of the comprehensive concentration field in the positive direction of the Y axis is located, the forward difference method is used, and when y q When the right boundary of the comprehensive concentration field in the positive direction of the Y axis is located, the backward difference method is used, and the x of the boundary point on the left boundary of the Y axis of the comprehensive concentration field is less than the x of the boundary point on the right boundary of the Y axis of the comprehensive concentration field p . p .

[0126] Through S510 and S520, the gradient information of the entire comprehensive concentration field can be obtained after the boundary points and non-boundary points of the comprehensive concentration field are differentiated.

[0127] S530, for the current agent, according to the third position information of the current agent R i , determine the gradient information of the corresponding point in the comprehensive concentration field.

[0128] In this embodiment, for the current agent R i , according to the third position information of the current agent R i , determine the gradient of the point corresponding to the current agent R i in the coordinate system xoy in the local coordinate system That is, the gradient information of the corresponding point

[0129] S540, according to the gradient information of the corresponding point and the combined velocity corresponding to the current agent, determine the running speed of the current agent.

[0130] In this embodiment, for the current agent R i , according to the gradient information of the corresponding point determined by S540 and the combined velocity V i (t) of the current agent R ip obtained by S400, the running speed of the current agent is determined, and the speed formula is as follows:

[0131]

[0132] Wherein, a is the speed weight of the agent tracking the target, and so on, so as to determine the running speed of each agent in the tracking cluster, so that each agent can move to the place where the concentration gradient is high, that is, to the target position, so as to realize tracking the target and avoiding obstacles at the same time.

[0133] Through S510 to S540, the gene regulation network is introduced to build the concentration field, and the running speed of each agent is determined through the gradient method of the environment, so as to avoid collision with obstacles while quickly approaching the target.

[0134] Referring to Figure 3 and Figure 4 In some embodiments of the present application, the unmanned cluster target search control method further comprises:

[0135] S101, the group robots are divided into several search groups, each search group includes several agents, and each search group starts from the corresponding set initial position.

[0136] In this embodiment, in the initial stage, the group robots are equally divided into several search groups, and each search group starts from the corresponding set initial position and disperses in different directions to search for the target in the unknown environment, and there are several agents in a single search group.

[0137] S102, the agents in each search group perform dispersed exploration in the unknown environment according to the Gaussian random walk strategy to explore the target.

[0138] In this embodiment, each search group adopts the Gaussian random walk strategy to search when exploring in the unknown environment.

[0139] In order to maintain the formation, the step length extracted by the agents in the same group follows the Gaussian distribution and is consistent. For the current agent R i The random walk speed at time t is:

[0140] V ig (t)=(∈ 1,t ,∈ 2,t )

[0141] Wherein, V ig (t) represents the speed vector of the current agent R i in the group, and ∈ 1,t and ∈ 2,t are Gaussian distribution N(mu, sigma 2, μ is the mean of each move, determines the distance from the initial position after multiple moves, and σ is the variance of each move, determines the deviation of this move from the last move. The larger σ is, the greater the search breadth is, and vice versa. By adjusting these two parameters, the speed and coverage of the agent searching for the target in the unknown environment can be effectively improved, and the adaptability to diversified terrain can be enhanced.

[0142] Exemplarily, with reference to Figure 3 and Figure 4 , the unmanned aerial vehicle cluster is taken as the group robot, the unmanned aerial vehicles start from the set initial positions, disperse in different directions to search, search for the target in the unknown environment according to the Gaussian random walk strategy, and track two targets.

[0143] Through S101 and S102, the Gaussian random walk is introduced, the search ability of the agent in the unknown environment is enhanced, the search efficiency for the target is improved, and the agent can quickly explore the target in the complex and changeable unknown environment.

[0144] S103, distance information between the current agent and other agents is obtained, a third set of agents having an attractive relationship with the current agent and a fourth set of agents and obstacles having a repulsive relationship with the current agent are determined according to the distance information, and the combined velocity of the agents in a plurality of search groups is calculated according to the third set, the fourth set and the distance information.

[0145] In this embodiment, when the agent performs the exploration task, the repulsive field is used to realize the collision avoidance between different groups, between agents in the same group and between the agent and obstacles. In order to simplify the calculation process, all obstacles are equivalent to agents of other groups, so that only the interaction between agents of the same group and agents of different groups needs to be considered. In order to maintain the formation, the agents in the same group need an additional attractive field to achieve this purpose. For the current agent R i , all individuals having an attractive field and a repulsive field with the current agent are counted, and the expression is as follows:

[0146]

[0147] , G' is the set of agents having an attractive field with the current agent R i , that is, the third set, t is the number of agents in the third set G', F' is the set of agents and obstacles having a repulsive field with the current agent R i , that is, the fourth set, n is the number of agents in the fourth set F', and e is the number of obstacles in the fourth set F'.

[0148] Specifically, the current agent R iThe distance to other agents or obstacles is less than the minimum safety distance d req repulsion will occur; the current agent R i The distance to other agents in the group is greater than the maximum holding distance d att attraction will occur.

[0149] For the current agent R i , the attraction and repulsion are calculated by the attraction field calculation formula and the repulsion field calculation formula according to the third set, the fourth set and the distance information, so as to calculate the combined velocity of the potential field force of the agents in the search groups in the exploration stage.

[0150] In this embodiment, the calculation process of the combined velocity of the potential field force of the agents in the exploration stage is similar to that of the agents in the tracking stage S420 to S440, and therefore will not be described in detail in this embodiment. The second set in the tracking stage can be updated by using the fourth set in the exploration stage to reduce the calculation amount.

[0151] S104, according to the combined velocity of the potential field force of the agents in the search groups in the exploration stage, and combining the Gaussian random walk strategy, the total velocity of the agents in the search groups when exploring the unknown environment is obtained.

[0152] In this embodiment, for the current agent R i , the total velocity formula is:

[0153] V i (t)=V ip (t)+V ig (t).

[0154] Through S101 to S104, when exploring the environment, the group robots are grouped, the Gaussian random walk strategy is introduced for walking, and the artificial potential field method is combined for obstacle avoidance, so as to maintain the formation while avoiding obstacles, so as to improve the coverage and randomness of the search, and improve the search efficiency.

[0155] In some embodiments of the present application S100, when the group robots explore the target, the set tracking number of agents is allocated from the group robots to form a tracking cluster, specifically comprising:

[0156] S110, when the target is explored, for the current search group, the current search group communicates with the search group that explores the target to determine the tracking number of agents that track the current target.

[0157] In this embodiment, in the search process, after the search group explores the target, the search group that explores the target communicates with the current search group to determine the tracking number M of the agents that participate in tracking the target.

[0158] That is, when the target is explored, the agent of the search group that explores the target communicates with the agents near the target to determine the number M of agents currently participating in tracking the target.

[0159] Wherein, the current search group can be a group that is separated from the search group that explores the target by a set distance, that is, the agents near the target can be agents that are separated from the agents in the search group that explores the target by a set distance.

[0160] S120, determining whether the number of agents tracking is less than the set number of tracking.

[0161] In this embodiment, it is determined whether the number M of agents tracking is less than the set number N of tracking.

[0162] S130, if yes, determining the number of allocation according to the number of agents tracking and the set number of tracking, and allocating the number of agents in the current search group to track the target to form a tracking cluster, and the remaining agents in the current search group disperse to explore according to the Gaussian random walk strategy.

[0163] In this embodiment, when the number M of agents tracking is less than the set number N of tracking, the number of allocation (N-M) is determined, (N-M) number of agents in the current search group are allocated to the search group that explores the target to form a tracking cluster, so that the agents in the tracking cluster satisfy the set number of tracking, to ensure that the target is fully tracked.

[0164] The remaining agents in the current search group will disperse to explore according to the Gaussian random walk strategy, that is, the agents that are not allocated to track the target will continue to disperse to explore.

[0165] If the number of agents in the current search group does not satisfy the number of allocation (N-M), all available agents in the current search group are allocated to track the target, and the current search group continues to communicate with other search groups within the set communication range to supplement the agents, until the number of agents participating in tracking the target reaches the set number N of tracking.

[0166] Through S110 and S130, before tracking, the number of agents tracking each target is fixed to form a tracking cluster, so that when the agent resources are sufficient, multiple targets can be tracked at the same time, and each target is ensured to be fully monitored.

[0167] The embodiment of the application also provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned target search control method of the unmanned cluster based on local information when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0168] It can be understood that the contents in the above method embodiments are all applicable to the present device embodiments, the present device embodiments specifically implement the functions same as those of the above method embodiments, and achieve the same beneficial effects as those of the above method embodiments.

[0169] The electronic device comprises:

[0170] The processor can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, and is configured to execute related programs to implement the technical solutions provided by the embodiments of the present application.

[0171] The memory can be implemented in the form of a ROM (Read Only Memory), a static storage device, a dynamic storage device, or a RAM (Random Access Memory). The memory can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are stored in the memory and are called and executed by the processor to implement the local information-based unmanned cluster target search control method of the embodiments of the present application.

[0172] The input / output interface is configured to realize information input and output.

[0173] The communication interface is configured to realize communication interaction between the device and other devices. The communication can be realized in a wired manner (for example, a USB, a network cable, etc.) or in a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.).

[0174] The bus is configured to transmit information between various components (for example, the processor, the memory, the input / output interface, and the communication interface) of the device.

[0175] The processor, the memory, the input / output interface, and the communication interface are connected to each other through the bus to realize communication connection within the device.

[0176] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The computer program is executed by the processor to implement the above local information-based unmanned cluster target search control method.

[0177] It can be understood that the contents in the above method embodiments are all applicable to the present storage medium embodiments, the present storage medium embodiments specifically implement the functions same as those of the above method embodiments, and achieve the beneficial effects same as those of the above method embodiments.

[0178] The memory, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory disposed remotely relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0179] It can be understood by those skilled in the art that all or some steps in the above disclosed method, functions of modules / units in the system and the device can be implemented as software, firmware, hardware and appropriate combinations thereof.

[0180] The terms "first", "second", "third", "fourth" and the like in the description of the application and in the claims of the foregoing drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so clad can interchange, where appropriate, so that the embodiments of the application described herein can be carried out in sequences other than those illustrated or otherwise described herein. Moreover, the terms "comprise" and "have", and any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or apparatus that comprises a list of steps or units not necessarily limited to those clearly identified as such, but can include other not clearly recited or inherent to such processes, methods, products or apparatus.

[0181] The preferred embodiments of the application are specifically described above, but the application is not limited to the above-described embodiments, and those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the present application.

Claims

1. A local information-based unmanned cluster target search control method, characterized by, The method comprises: When the group of robots explores the target, a set of tracking agents is formed by allocating a set number of agents from the group of robots to track the target; Through the tracking set, a local coordinate system is established, and first position information of the target, second position information of an obstacle explored, and third position information of each agent in the tracking set are determined according to the local coordinate system; According to the first position information and the second position information, a comprehensive concentration field about the obstacle and the target is generated through a GRN upper network model; According to the third position information, a first set of agents having an attractive relationship with the current agent and a second set of agents having a repulsive relationship with the current agent are determined, and a combined velocity of the potential field force of each agent in the tracking set is calculated according to the first set, the second set and the third position information; Through difference calculation, gradient information of each point in the comprehensive concentration field is determined, and running speed of each agent in the tracking set is determined according to the gradient information and the combined velocity, and each agent in the tracking set is controlled to track the target according to the running speed.

2. The local information-based control method of a swarm of unmanned targets for search according to claim 1, characterized in that, The local coordinate system is established through the tracking set, which specifically comprises: In the tracking set, three agents not on the same straight line are selected, and the three agents are respectively taken as a first coordinate agent, a second coordinate agent and a third coordinate agent; Taking the first coordinate agent as the coordinate center o, a straight line pointing from the first coordinate agent to the second coordinate agent is taken as the x-axis, and a straight line perpendicular to the x-axis and pointing to one side of the third coordinate agent is taken as the y-axis, so as to establish the local coordinate system xoy.

3. The local information-based control method of a swarm of unmanned targets for search according to claim 2, characterized in that, The third position information of each agent in the tracking set is determined according to the local coordinate system, which specifically comprises: The first distance from the first coordinate agent to the second coordinate agent, the second distance from the remaining agents to the first coordinate agent, and the third distance from the remaining agents to the second coordinate agent are obtained; Based on the local coordinate system, the third position information is determined according to the first distance, the second distance and the third distance.

4. The local information-based control method of a swarm of unmanned targets for search according to claim 1, characterized in that, For the current agent R i , the calculating the resultant velocity of the potential force of each agent in the tracking cluster according to the first set, the second set and the third position information specifically comprises: According to the first set G and the third position information, the current agent R is calculated by a gravitational field calculation formula i The gravity U received att (d ik ), wherein the first set G is d ik The distance between the current agent R i and G k , k = 1, 2, …, h, h is the number of agents in the first set G; According to the second set F and the third position information, the current agent R is calculated by a repulsive force field calculation formula i The repulsive force U received by the agent R req (d ij , wherein the second set F is d ij The distance between the current agent R i and F j , j = 1, 2, …, l, l is the number of agents in the second set F; According to the attractive force U att (d ik ) and the repulsive force U req (d ij ), the resultant velocity V i (t) of the potential field force of the current agent R ip (i) is calculated, wherein i=1, 2, …, m, m is the number of agents in the tracking cluster, so as to calculate the resultant velocity of each agent in the tracking cluster.

5. The local information-based control method of a swarm of unmanned targets for search according to claim 1, characterized in that, The gradient information of each point in the comprehensive concentration field is determined through difference calculation, which specifically comprises: The gradient information of the non-boundary point is determined by central difference of the non-boundary point in the comprehensive concentration field; The gradient information of the boundary point is determined by forward difference and backward difference of the boundary point in the comprehensive concentration field.

6. The local information-based control method of a swarm of unmanned targets for search according to claim 1, characterized in that, For the current agent R i , the determining the running speed of each agent in the tracking cluster according to the gradient information and the combined speed specifically comprises: According to the third position information of the current agent R i , gradient information of a corresponding point in the comprehensive concentration field is determined. According to the gradient information of the corresponding point and the current agent R i According to the corresponding combined velocity, the running speed of the current agent is determined, wherein i=1, 2, …, m, m is the number of agents in the tracking cluster, so as to calculate the running speed of each agent in the tracking cluster.

7. The local information-based control method of a swarm of unmanned targets for search according to claim 1, characterized in that, Further comprising: The group of robots is divided into a plurality of search groups, each search group comprising a plurality of agents, and each search group starts from a corresponding initial position; The agents in each search group perform dispersed exploration in an unknown environment according to a Gaussian random walk strategy to explore the target.

8. The local information-based control method of a swarm of unmanned targets for search according to claim 7, characterized in that, When the group of robots explores the target, a set of tracking agents is formed by allocating a set number of agents from the group of robots to track the target, which specifically comprises: When the target is explored, for the current search group, the current search group communicates with the search group that explores the target, determines the number of agents currently tracking the target; It is judged whether the number of agents is less than the set tracking number; If yes, according to the number of agents and the set tracking number, the number of agents is determined, and the number of agents is separated from the current search group to track the target to form the tracking cluster, and the remaining agents in the current search group disperse according to the Gaussian random walk strategy.

9. An electronic device, comprising: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the local information-based unmanned cluster target search control method of any one of claims 1-8 when executing the computer program.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the local information-based unmanned cluster target search control method of any one of claims 1-8.

Images