Collaborative inspection method and device based on heterogeneous multi-machine combination, terminal equipment and storage medium

Abstract

The invention discloses a collaborative inspection method and device based on heterogeneous multi-machine combination, terminal equipment and a storage medium, and the method comprises the steps: constructing an undirected connected graph of an inspection region according to a collection point and a supply point, constructing a plurality of ant colonies according to the number of robots, initial positions and initial resources, and carrying out the construction of a plurality of ant colonies according to the undirected connected graph; and performing a plurality of population exploration operations on the ant colonies to enable each ant colony to generate a plurality of path sets, determining a target path set from the plurality of path sets, and finally determining the inspection task of each inspection robot according to the target path set. Therefore, the problem of endurance of the robot is solved by constructing the undirected connected graph together with the supply points and the collection points, and the calculation efficiency of the algorithm is improved by constructing a plurality of ant colonies to perform population exploration operation together in a concurrent mode, so that the problem that when large-scale inspection is performed in current electric power intelligent inspection, the algorithm is difficult to carry out is solved. And the defects of insufficient computing power and insufficient endurance of the robot are overcome.

Description

Technical Field

[0001] The present invention relates to the field of intelligent inspection technology, and in particular to a collaborative inspection method, device, terminal equipment, and storage medium based on heterogeneous multi-machine combination. Background Art

[0002] In recent years, with the acceleration of power system infrastructure construction, the scale of power systems has continued to increase, which has brought many problems and challenges to power inspection. Power inspection aims to ensure the safe and stable operation of the power system by regularly inspecting, testing and evaluating power equipment (such as substations, transmission lines, distribution lines, etc.).

[0003] There are basically two types of scenarios for intelligent power inspection. The first scenario: When obstacles are known, ground robots conduct traversal inspections or fixed-point inspections of the areas to be inspected for power equipment, such as substation inspection scenarios. The second scenario: Use large drones or helicopters to inspect transmission lines or towers. When inspecting in this scenario, it is necessary to plan one or more optimized paths in advance for the inspection sequence of multiple tower targets, and solve the problem of multi-drone and multi-target inspections under the premise of obstacle avoidance. In the path planning of power drones, traditional search algorithms are more used for obstacle avoidance path planning in substations. Intelligent optimization algorithms are widely used in power inspection path planning.

[0004] However, traditional search algorithms have always had shortcomings such as high computational complexity, slow computing speed, and weak processing capabilities for high-dimensional space. Intelligent optimization algorithms may be a suitable choice for dealing with these problems, but as the inspection coverage increases, the computational complexity of intelligent optimization algorithms also increases exponentially. Furthermore, robots cannot operate for long periods of time due to energy consumption issues, so the robot's power problem also makes it difficult for robots to conduct large-scale inspections. Therefore, current power intelligent inspections do not have large-scale application capabilities. Summary of the invention

[0005] The embodiments of the present invention provide a collaborative inspection method, apparatus, terminal device, and storage medium based on heterogeneous multi-machine cooperation. The problem of robot endurance is solved by constructing an undirected connected graph together with supply points and collection points. Secondly, a concurrent method is adopted to improve the computational efficiency of the algorithm by constructing several ant colonies to perform population exploration operations together, thereby solving the defects of insufficient computing power and insufficient robot endurance in current smart power inspection during large-scale inspections.

[0006] An embodiment of the present invention provides a collaborative inspection method based on heterogeneous multi-machine combination, including:

[0007] Obtain several collection points to be inspected in the inspection area, several supply points for supplying the inspection robot, the initial positions, initial resources, and number of robots of the inspection robots;

[0008] According to the collection points and the replenishment points, a connection path between each collection point and each replenishment point is determined, and according to the connection path, an undirected connected graph of the inspection area is constructed; wherein the collection points and the replenishment points are nodes of the undirected connected graph;

[0009] Constructing a plurality of ant colonies, and initializing a plurality of male ants in the ant colonies according to the number of the robots, the initial positions, and the initial resources; wherein each of the ant colonies includes a plurality of male ants equal to the number of the robots, and in each of the ant colonies, the male ants correspond one to one with the inspection robots;

[0010] Adopting an ant colony algorithm, according to the undirected connected graph, the initial positions of the male ants in the male ants and the initial resources, the population exploration operation is repeatedly performed on the male ants, until a target path set is determined from the path sets generated by the male ants in the population exploration operation; wherein the path set is composed of the valid paths that the male ants in the male ants in the ant colony pass through after traversing all the collection points in the undirected connected graph, and the valid path is the path that each male ant in the male ants moves from one node to another node; the target path set is the path set with the shortest completion time and the highest task balance rate among the path sets;

[0011] The inspection task of the inspection robot corresponding to each male ant is determined according to the effective path explored by each male ant in the target path set.

[0012] Furthermore, the population exploration operation includes:

[0013] Obtaining a global optimal solution, a global path pheromone, and a global expected reward table; wherein the global optimal solution is initially empty, the global path pheromone is used to record the pheromone on each valid path in the path set generated by each ant colony, and is initially a preset constant value; the global expected reward table is used to record the reward value of each male ant in each ant colony exploring the corresponding valid path, and is initially empty;

[0014] Inputting the undirected connectivity graph, the global path pheromone, and the global expected reward table into an ant colony that has not performed a node exploration operation, so that the ant colony, according to the undirected connectivity graph, the global path pheromone, the global expected reward table, the initial position of the male ants, and the initial resources, repeatedly performs the node exploration operation until the ant colony has explored all the collection points and the male ants are all at the supply point;

[0015] When it is confirmed that a target ant colony has explored all the collection points and the male ants are all at the supply point, a path set of the target ant colony is obtained;

[0016] Calculate a first performance indicator of the path set, and compare the performance indicator with a second performance indicator of the global optimal solution. When it is determined that the first performance indicator of the path set is greater than the second performance indicator of the current global optimal solution, update the global optimal solution to the path set; wherein the first performance indicator and the second performance indicator are composed of completion time and task balance rate;

[0017] According to the valid paths in the path set, updating the global path pheromone and the global expected reward table;

[0018] Determining whether the total number of times the ant colonies perform population exploration operations is not less than a preset number threshold;

[0019] If yes, re-perform the population exploration operation on the target ant colony according to the current global optimal solution, the current global path pheromone, and the current global expected reward table;

[0020] If not, the global optimal solution is used as the target path set.

[0021] Furthermore, the node exploration operation includes:

[0022] Obtain the current global path pheromone, the current global expected reward table, the current taboo table, the current position of each male ant, and the current resources; wherein, initially, the current position of each male ant and the current resources are the initial position and initial resources corresponding to each male ant, and the taboo table is used to record the collection points that the ant colony has explored, and is empty initially;

[0023] Determining a target male ant for exploring an effective path according to the current position and current resources of each male ant;

[0024] Determine a target node to be explored according to the target male ant, the global path pheromone, and the global expected reward table; wherein the target node is a collection point or a supply point;

[0025] When the target node is determined to be a collection point, the remaining resources of the target male ant after reaching the target node from the current position are calculated, and the target supply point closest to the target node is determined;

[0026] After determining that the target node is a supply point, or determining that the target male ant can reach the target node from the current position and can reach the target supply point by relying on the remaining resources, generating a valid path according to the current position and the target node, taking the target node as the new current position of the target male ant, taking the remaining resources as the new current resources of the target male ant, and adding the target node to the taboo table;

[0027] Determine whether all collection points are recorded in the current taboo table, and whether the male ants are at the supply points;

[0028] If so, a path set is generated based on several valid paths;

[0029] If not, the next round of node exploration operation is performed.

[0030] Furthermore, after calculating the remaining resources of the target male ant after reaching the target node from the current position and determining the target supply point closest to the target node, it also includes:

[0031] After determining that the target male ant reaches the target node from the current position, when it is impossible to reach the target supply point by relying on the remaining resources, the global path pheromone and the global expected reward table are updated according to the current valid paths, the taboo table is cleared, and the node exploration operation is re-executed.

[0032] Further, the updating of the global path pheromone and the global expected reward table includes:

[0033] Obtaining the path pheromone of each valid path in the global path pheromone;

[0034] Traversing the male ants, and calculating the pheromones left by each male ant on each valid path when the male ant passes through the valid path;

[0035] When the traversal is completed, the sum of the pheromones on each of the valid paths is calculated;

[0036] According to the sum of the pheromones of each valid path and the corresponding path pheromone, updating the path pheromone of each valid path in the global path pheromone;

[0037] The effective path is input into a preset reward prediction model so that the reward prediction model outputs the expected reward value of each male ant exploring the corresponding effective path, and updates the expected reward table according to the expected reward value; wherein the reward prediction model is a model constructed using a Q learning algorithm.

[0038] Further, determining a target node to be explored according to the target male ant, the global path pheromone, and the global expected reward table includes:

[0039] Take several collection points that are not recorded in the current taboo table, as well as idle supply points, as initial candidate nodes;

[0040] According to the current resources of the target male ant, determining several initial candidate nodes that the target male ant can reach as candidate nodes;

[0041] Calculate the probability of the target male ant colliding with other male ants when it arrives at each candidate node from the current position;

[0042] Randomly generate a probability value, and when the probability value is less than a preset probability, randomly determine a target node from the candidate nodes by using a roulette wheel method;

[0043] When the probability value is not less than a preset probability, a target node is determined from the candidate nodes according to the conflict probability, the global path pheromone, and the global expected reward table.

[0044] Further, determining a target node from the candidate nodes according to the conflict probability, the global path pheromone, and the global expected reward table includes:

[0045] Determine the working time for the male ant to reach each of the candidate nodes from the current position and perform the inspection task according to the current position of the male ant and the candidate nodes;

[0046] Calculating the heuristic information of each candidate node according to the working time and the conflict probability;

[0047] The transition probability of each candidate node is determined according to the following formula:

[0048]

[0049] in, is the transition probability of candidate node j, τ ij (t) is the path pheromone between the current position of the male ant and the candidate node j, η ij is the heuristic information of candidate node j, ζ ij is the expected reward of the male ant from the current position to the candidate node j, α is the preset pheromone heuristic factor, β is the preset expectation heuristic factor, γ is the preset reward heuristic factor, is the set of candidate nodes.

[0050] Another embodiment of the present invention provides a collaborative inspection device based on heterogeneous multi-machine combination, including:

[0051] The data acquisition module is used to obtain a number of collection points to be inspected in the inspection area, a number of supply points for supplying the inspection robot, the initial positions of the inspection robots, the initial resources, and the number of robots;

[0052] A connectivity graph construction module, used to determine connectivity paths between the collection points and the replenishment points according to the collection points and the replenishment points, and to construct an undirected connectivity graph of the inspection area according to the connectivity paths; wherein the collection points and the replenishment points are nodes of the undirected connectivity graph;

[0053] An ant colony construction module, used to construct a plurality of ant colonies, and initialize a plurality of male ants in the ant colonies according to the number of the robots, the initial positions, and the initial resources; wherein each of the ant colonies contains a plurality of male ants equal to the number of the robots, and in each of the ant colonies, the male ants correspond one to one with the inspection robots;

[0054] A population exploration module is used to use an ant colony algorithm to repeatedly perform population exploration operations on a number of ant colonies according to the undirected connected graph, initial positions of a number of male ants in the number of ant colonies, and initial resources, until a target path set is determined from a number of path sets generated by the ant colonies in a number of population exploration operations; wherein the path set is composed of a number of valid paths that a number of male ants in an ant colony traverse through all collection points in the undirected connected graph, and the valid path is a path that each male ant in the ant colony moves from one node to another node; the target path set is a path set with the shortest completion time and the highest task balancing rate among a number of path sets;

[0055] The task allocation module is used to determine the inspection task of the inspection robot corresponding to each male ant according to the effective path explored by each male ant in the target path set.

[0056] Another embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, a collaborative inspection method based on heterogeneous multi-machine collaboration as described in any one of the embodiments is implemented.

[0057] Another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute a collaborative inspection method based on heterogeneous multi-machine collaboration as described in any of the above embodiments.

[0058] The following beneficial effects are achieved by implementing the present invention:

[0059] The present invention discloses a collaborative inspection method, device, terminal equipment, and storage medium based on heterogeneous multi-machine combination. The method constructs an undirected connected graph of the inspection area according to the collection points and the supply points, constructs several ant colonies according to the number of robots, the initial position, and the initial resources, and performs several population exploration operations on the ant colonies according to the undirected connected graph, so that each ant colony generates several path sets, and confirms a target path set from the several path sets, and finally determines the inspection task of the inspection robot corresponding to each male ant according to the effective path explored by each male ant in the target path set. Therefore, the present invention constructs an undirected connected graph with the supply points together, so that the robot's endurance problem can be considered conditionally while solving the inspection task allocation problem later. Secondly, a concurrent method is adopted to improve the calculation efficiency of the algorithm by constructing several ant colonies to perform population exploration operations together, solving the problems of insufficient computing power and robot endurance when the current power intelligent inspection is conducting large-scale inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flowchart of a collaborative inspection method based on heterogeneous multi-machine combination provided by an embodiment of the present invention.

[0061] Figure 2 It is a structural schematic diagram of a collaborative inspection device based on heterogeneous multi-machine combination provided by an embodiment of the present invention.

[0062] Figure 3 It is a schematic diagram of the structure of an octree voxel map provided by an embodiment of the present invention.

[0063] Figure 4 This is a schematic diagram of inspection task allocation provided by an embodiment of the present invention.

[0064] Figure 5 It is a flowchart of a population exploration operation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0067] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0068] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0069] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0070] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0071] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0072] See also Figure 1 , is a flow chart of a collaborative inspection method based on heterogeneous multi-machine combination provided by an embodiment of the present invention, including:

[0073] S1. Obtain several collection points to be inspected in the inspection area, several supply points for supplying the inspection robot, the initial positions, initial resources, and number of robots of the inspection robots to be inspected;

[0074] In a preferred embodiment of the present invention, taking the large-scale joint inspection of transmission, transformation and distribution as an example, all the locations that need to be inspected and photographed are used as robot collection points. The collection point information includes but is not limited to: the ID of the collection point, the longitude and latitude and altitude coordinates of the collection point, and the shooting angle requirements required by the collection point (including pitch angle, roll angle and yaw angle); the supply point is a robot airport or charging pile and other equipment that can automatically charge the robot. The supply point defines its longitude and latitude coordinates and a list of robots that are allowed to charge; at the same time, all available robot IDs, robot models, power levels, etc. are obtained. Secondly, the list of collection points where the robot can perform inspections and the list of supply points that can be entered are clarified.

[0075] S2. Determine, based on the collection points and the replenishment points, the connection paths between the collection points and the replenishment points, and construct an undirected connected graph of the inspection area based on the connection paths; wherein the collection points and the replenishment points are nodes of the undirected connected graph;

[0076] In a preferred embodiment of the present invention, trajectory information between points is drawn based on the coordinate information of the collection points and the supply points. The trajectory information includes the trajectory length, speed limit, etc. Specifically, an octree voxel map is first constructed based on the inspection area. Especially in irregular terrains such as mountains, lakes, etc., the octree voxel map can achieve efficient map data processing. Multi-resolution representation and efficient storage are achieved, and persistent storage and fast loading are supported. In order to solve the storage problem of large-area maps, a hierarchical distributed storage mechanism is selected. The octree is defined in layers according to different resolutions and distributedly stored on different servers. The octree voxel map is as follows: Figure 3 As shown in the figure, a large cube is continuously divided into eight even blocks until it becomes the smallest block. The entire large block can be regarded as the root node, and the smallest block can be regarded as the "leaf node". Therefore, in the octree, when going up one layer from the next layer of nodes, the volume of the map can be expanded eight times. When all the child nodes of a block are occupied or unoccupied, there is no need to expand this node.

[0077] According to the distribution of collection points and supply points, a connected graph connecting all collection points is planned. It can be drawn manually or automatically generated by the algorithm. A search-based path planning algorithm such as A* and fast random tree can be used to generate a trajectory between two points. Among them, an undirected connected graph is drawn according to the collection points, supply points, and trajectory lines, in which all collection points are Supply point It means that the node V of the graph is V 1 ∪V 2 , the trajectory line is represented by E = {e1, e2, ...} to represent the edge of the graph. This undirected graph can be represented as<V,E> .

[0078] S3, constructing a plurality of ant colonies, and initializing a plurality of male ants in the ant colonies according to the number of the robots, the initial positions, and the initial resources; wherein each of the ant colonies includes a plurality of male ants equal to the number of the robots, and in each of the ant colonies, the male ants correspond one to one with the inspection robots;

[0079] In a preferred embodiment of the present invention, N populations are defined, each of which has a queen ant that manages M ants. The M ants correspond to M dispatchable robots, and the initial positions of the M ants correspond to the location nodes of the robots. In order to solve large-scale computing problems, the ant colonies are distributed and deployed to various algorithm servers, and several algorithm servers share a central data server. The central data server is used to store global path pheromones, global expected reward tables, and global optimal solutions. Population exploration is distributed and deployed to algorithm servers, and population exploration can be efficiently executed in parallel.

[0080] S4, using an ant colony algorithm, according to the undirected connected graph, the initial positions of the male ants in the male ants and the initial resources, repeatedly performing population exploration operations on the male ants, until a target path set is determined from the path sets generated by the male ants in the male ants in the ant colony in the population exploration operations; wherein the path set is composed of the valid paths that the male ants in the male ants in the ant colony pass through after traversing all the collection points in the undirected connected graph, and the valid path is the path that each male ant in the male ants moves from one node to another node; the target path set is the path set with the shortest completion time and the highest task balance rate among the path sets;

[0081] In a preferred embodiment of the present invention, according to Figure 5 , the population exploration of ant colonies can be simply understood as collecting points V 1 Requires all to be traversed. The robot is running out of power and needs to go to the supply point V that supports the ability in time. 2 Charge in time.

[0082] The problem can be further described as:

[0083] 1) All collection points are required to be visited at least once.

[0084] 2) The nodes in the supply point can be visited not at all or multiple times.

[0085] 3) The robot has capability constraints and can only perform some collection point tasks.

[0086] 4) A supply point can only supply one robot at a time, and due to capacity constraints, only some robots can be supplied. Assume that the current layout of supply points is sufficient to meet the task execution of all collection points.

[0087] 5) The robot needs to recharge at the recharge point and complete all the collection point tasks by leapfrogging. The robot's power consumption is related to the moving distance and the number of collection point operations. For example, 1 grid of power can move 10 kilometers or complete 5 collection tasks.

[0088] 6) By default, robots start from a supply point and return to the supply point at the end.

[0089] Preferably, the population exploration operation includes:

[0090] S41, obtaining a global optimal solution, a global path pheromone, and a global expected reward table; wherein the global optimal solution is initially empty, the global path pheromone is used to record the pheromone on each valid path in the path set generated by each ant colony, and is initially a preset constant value; the global expected reward table is used to record the reward value of each male ant in each ant colony exploring the corresponding valid path, and is initially empty; the taboo table is used to record the collection points that the ant colony has explored, and is initially empty;

[0091] In a preferred embodiment of the present invention, at the beginning, the global path pheromone and the global expected reward table are initialized. Each queen ant is assigned M male ants, and the M male ants are placed at the initial node position of the robot, and the M male ants are initialized according to the current remaining power of the robot. Each queen ant initializes a taboo table, which records the collection points that the queen ant has commanded the male ants to visit, reflecting the memory of the ants, so that the ants will not take the same path repeatedly. Initialize the global path pheromone and fill it with a relatively large value. Initialize the global expected reward table and set it to 0. Initialize an energy consumption matrix table, which records the minimum energy consumption required for the robot to go from collection point i to collection point j. The energy consumption is calculated as follows:

[0092] E(i,j)=e1(L(i,j)+e1min{L(j,l)|l∈V 2})+e2T(j);

[0093] Where L(i,j) represents the time required to move from collection point i to collection point j, min{L(j,l)|l∈V 2} represents the minimum moving time required for the robot to move from the collection point j to the nearest supply station l, e1 is the energy consumption required for the robot to move per unit time; T(j) is the operation time of the robot at the collection point j, and e2 is the energy consumption required for the robot to operate per unit time. If the energy consumption e1 and e2 calculated for each robot are different, initialize the energy consumption matrix table {E1(i,j),E2(i,j),……,E M (i,j)}.

[0094] S42, inputting the undirected connectivity graph, the global path pheromone, and the global expected reward table into an ant colony that has not performed a node exploration operation, so that the ant colony, according to the undirected connectivity graph, the global path pheromone, the global expected reward table, the initial position of the male ants, and the initial resources, repeatedly performs the node exploration operation until the ant colony has explored all the collection points and the male ants are all at the supply point;

[0095] In a preferred embodiment of the present invention, according to the deployment mode of the algorithm server, the current global path information, the global expected reward table, etc. are pushed to the algorithm server, and the node exploration is performed in parallel.

[0096] Preferably, the node exploration operation includes:

[0097] S421, obtaining the current global path pheromone, the current global expected reward table, the current taboo table, the current position of each male ant, and the current resources; wherein, initially, the current position of each male ant and the current resources are the initial position and initial resources corresponding to each male ant, and the taboo table is used to record the collection points that the ant colony has explored, and is empty initially;

[0098] S422, determining a target male ant for exploring an effective path according to the current position and current resources of each male ant;

[0099] In a preferred embodiment of the present invention, the moving ants are selected according to the Max-Min rule. First, the taboo table is used to identify whether all the collection points have been explored. If there are still unexplored collection points, the male ant with the smallest end time of the currently allocated exploration node is selected for allocation according to the Max-Min rule. If all the collection points have been explored, the male ant that is not currently docked at the supply point and has the smallest remaining energy consumption is selected for allocation according to the Max-Min rule.

[0100] S423, determining a target node to be explored according to the target male ant, the global path pheromone, and the global expected reward table; wherein the target node is a collection point or a supply point;

[0101] Preferably, determining a target node to be explored according to the target male ant, the global path pheromone, and the global expected reward table includes:

[0102] S4231, taking several collection points not recorded in the current taboo table and idle supply points as initial candidate nodes;

[0103] S4232, determining, according to the current resources of the target male ant, several initial candidate nodes that the target male ant can reach as candidate nodes;

[0104] S4233, calculating the probability of the target male ant colliding with other male ants when it arrives at each candidate node from the current position;

[0105] S4234, randomly generating a probability value, and when the probability value is less than a preset probability, randomly determining a target node from the candidate nodes by using a roulette wheel method;

[0106] S4235. When the probability value is not less than a preset probability, determine a target node from the candidate nodes according to the conflict probability, the global path pheromone, and the global expected reward table.

[0107] Preferably, determining a target node from the candidate nodes according to the conflict probability, the global path pheromone, and the global expected reward table includes:

[0108] S42351, determining the working time for the male ant to reach each of the candidate nodes from the current position and perform the inspection task according to the current position of the male ant and the candidate nodes;

[0109] S42352, calculating the heuristic information of each candidate node according to the working time and the conflict probability;

[0110] S42353. Determine the transition probability of each candidate node according to the following formula:

[0111]

[0112] in, is the transition probability of candidate node j, τ ij (t) is the path pheromone between the current position of the male ant and the candidate node j, η ij is the heuristic information of candidate node j, ζ ij is the expected reward of the male ant from the current position to the candidate node j, α is the preset pheromone heuristic factor, β is the preset expectation heuristic factor, γ is the preset reward heuristic factor, is the set of candidate nodes.

[0113] In a preferred embodiment of the present invention, first, the next exploration point that the male ant can choose is identified. The exploration point can only be a collection point that is not in the taboo table, and the remaining power of the robot corresponding to the male ant is sufficient to meet the minimum energy consumption requirement in the energy consumption matrix table.

[0114] Then, the probability c of the exploration point that the male ant can choose to conflict with the exploration path of other male ants is calculated. ij (t).

[0115] Finally, the optimal next exploration point is selected from the optional exploration points. There are two specific selection methods. The p value (p∈(0,1)) is a probability value preset in advance.

[0116] The ant has a 1-p probability of randomly selecting an exploration node that satisfies the endurance constraint in a roulette wheel manner;

[0117] The ant has a probability of p to select the next exploration point based on the global path pheromone and the global expected reward table. At time t, the queen ant sends the male ant k at node i to explore the next node j based on the path pheromone τ ij (t), inspiration information η ij and expected return ζ ij The transition probability is as follows:

[0118]

[0119] in, is the set of nodes that the kth node of the hth male ant is allowed to explore; α is the pheromone heuristic factor; β is the expectation heuristic factor; γ is the reward heuristic factor; d ij It is the time required for the operation from node i to target node j, including the moving time and the inspection operation time (i.e., the collection time).

[0120] S424, when the target node is determined to be a collection point, calculating the remaining resources of the target male ant after reaching the target node from the current position, and determining the target supply point closest to the target node;

[0121] Preferably, after calculating the remaining resources of the target male ant after reaching the target node from the current position and determining the target supply point closest to the target node, the method further includes:

[0122] S4241. After determining that the target male ant has reached the target node from the current position, but cannot reach the target supply point by relying on the remaining resources, update the global path pheromone and the global expected reward table based on the current valid paths, clear the taboo table, and re-execute the node exploration operation.

[0123] S425, after determining that the target node is a supply point, or determining that the target male ant can reach the target supply point by relying on the remaining resources after reaching the target node from the current position, generating a valid path according to the current position and the target node, taking the target node as the new current position of the target male ant, taking the remaining resources as the new current resources of the target male ant, and adding the target node to the taboo table;

[0124] S426, determining whether all the collection points are recorded in the current taboo table, and whether the male ants are all at the supply point;

[0125] S427, if yes, generate a path set based on a number of valid paths;

[0126] S428: If not, proceed to the next round of node exploration operation.

[0127] In a preferred embodiment of the present invention, if the selected target node is not sufficient to reach the next supply point, the current queen ant's overall exploration fails and the exploration stops; otherwise, the exploration continues and the taboo table is modified to record the collection points that have been explored: if the current exploration node belongs to a collection point, the collection point is added to the taboo table; if all collection points have been traversed and the last node of all male ants is a supply point, the exploration ends; otherwise, the node exploration operation continues.

[0128] It should be noted that during confirmation, it is necessary to identify whether there are conflicts among the trajectories of all male ants under the jurisdiction of the current queen ant. If there are conflicts, the trajectories are adjusted based on the game strategy to resolve the conflicts.

[0129] S43, when it is confirmed that a target ant colony has explored all the collection points and the male ants are all at the supply point, obtaining a path set of the target ant colony;

[0130] S44, calculating a first performance indicator of the path set, and comparing the performance indicator with a second performance indicator of the global optimal solution, and when it is determined that the first performance indicator of the path set is greater than the second performance indicator of the current global optimal solution, updating the global optimal solution to the path set; wherein the first performance indicator and the second performance indicator are composed of completion time and task balance rate;

[0131] In a preferred embodiment of the present invention, it is determined whether the effective path obtained by the current queen ant commanding the male ants to explore is better than the global optimal solution, that is, it is determined whether the path set is better than the global optimal solution. If so, the global optimal solution is updated, and the effective paths obtained by the queen ant exploration are compared, and the one with the best indicators is selected and recorded, including path information and evaluation indicators. Using the minimum and maximum completion time and task balance rate as indicators, it can be defined as:

[0132] minz = w1f1 + w2f2;

[0133] Among them, z represents the final evaluation index, w1 and w2 are weights, f1 represents the normalized minimum and maximum completion time, and f2 represents the normalized task balance rate.

[0134] S45, updating the global path pheromone and the global expected reward table according to the valid paths in the path set;

[0135] Preferably, the updating of the global path pheromone and the global expected reward table includes:

[0136] S451, obtaining the path pheromone of each valid path in the global path pheromone;

[0137] S452, traversing the male ants, and calculating the pheromones left by each male ant on each valid path when the male ant passes through the valid path;

[0138] S453, when the traversal is completed, calculating the sum of the pheromones on each of the valid paths;

[0139] S454, updating the path pheromone of each valid path in the global path pheromone according to the sum of the pheromones of each valid path and the corresponding path pheromone;

[0140] S455. Input the effective path into a preset reward prediction model so that the reward prediction model outputs the expected reward value of each male ant exploring the corresponding effective path, and updates the expected reward table according to the expected reward value; wherein the reward prediction model is a model constructed using a Q learning algorithm.

[0141] In a preferred embodiment of the present invention, the path pheromone τ ij (t) is updated according to the following formula:

[0142] τ ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t);

[0143]

[0144] Where 1-ρ represents the volatile pheromone. When the queen ant finds an effective path, all the paths that her male ants pass through will leave pheromones.

[0145] Expected Returnζ hijIt indicates that the queen ant issues an instruction to let the h-th male ant get the expected reward from going from node i to node j. The Q value Q(h,i,j) in Q-learning is used for calculation here.

[0146] Q-Learning is a reinforcement learning algorithm. It is a model-free prediction algorithm used to solve the Markov decision process (MDP) problem. The core idea of ​​the Q-Learning algorithm is to evaluate the expected utility of taking an action in a given state by learning an action value function (Q function). The Q-Learning algorithm can be used to train the decision-making ability of the queen ant.

[0147] The Q table is defined as follows:

[0148]

[0149]

[0150] The definition of state: the male ant currently commanded by the queen ant and the node currently explored are taken as the state. The choice of action: refers to the node explored by the male ant. The update of the Q value of the hth male ant from the current exploration node i to the next exploration node j is:

[0151] Q(h,i,j)=Q(h,i,j)+λ(r(h,i,j)+μ·max j′ Q(h,j,j ′ )-Q(h,i,j));

[0152] Among them, λ is the learning rate, μ is the discount rate, i is the last exploration position of the hth male ant, j is the new exploration node currently selected by the hth male ant, and j ′ is the next new exploration point that the hth male ant can choose after exploring j. r(h,i,j) is the immediate reward, max j′ Q(h,j,j ′ ) is the maximum cumulative reward value that the h-th male ant can obtain after exploring j.

[0153]

[0154] Among them, r1>r2>0>r3. If r3 is a negative number, it means that the current path is invalid and a certain reward needs to be deducted. The r1 value is the largest, which means the reward is the largest.

[0155] S46, determining whether the total number of times that the ant colonies perform population exploration operations is not less than a preset number threshold;

[0156] In a preferred embodiment of the present invention, when the total number of population exploration operations has exceeded the maximum number of iterations (number threshold), or the optimal solution has remained unchanged for a certain period of time, the population exploration can be exited, the results can be output and the algorithm can be terminated; otherwise, the population exploration can continue.

[0157] S47, if yes, re-perform the population exploration operation on the target ant colony according to the current global optimal solution, the current global path pheromone, and the current global expected reward table;

[0158] S48: If not, taking the global optimal solution as the target path set.

[0159] S5. Determine the inspection task of the inspection robot corresponding to each male ant according to the effective path explored by each male ant in the target path set.

[0160] In a preferred embodiment of the present invention, Figure 4 The figure shows a schematic diagram of the task allocation of heterogeneous multi-machine joint inspection. The robot can be a drone, a wheeled robot, a quadruped robot, or a biped robot.

[0161] Robot 1 starts from supply point 1, follows the ABC sequence, returns to supply point 1 for charging, then continues to collect DE, and finally returns to supply point 1 for charging.

[0162] Robot 2 starts from supply point 2, collects photos in the order of FG, recharges at supply point 2, continues to collect HJ, and finally docks at supply point 3.

[0163] Robot 3 starts from supply point 3, collects photos in the order of JK, and then goes to supply point 4 for charging.

[0164] Robot 4 starts from supply point 4, visits LM to collect photos, returns to supply point 4, then continues to collect OP and finally stops at supply point 2.

[0165] This embodiment provides a collaborative inspection method based on heterogeneous multi-machine combination, by constructing an undirected connected graph of the inspection area according to the collection points and the supply points, constructing several ant colonies according to the number of robots, the initial position, and the initial resources, and performing several population exploration operations on the ant colonies according to the undirected connected graph, so that each ant colony generates several path sets, and confirms a target path set from the several path sets, and finally determines the inspection task of the inspection robot corresponding to each male ant according to the effective path explored by each male ant in the target path set. Therefore, the present invention constructs an undirected connected graph with the supply points and the collection points, so that the robot's endurance problem can be considered conditionally while solving the problem of inspection task allocation in the future. Secondly, a concurrent method is adopted to improve the calculation efficiency of the algorithm by constructing several ant colonies to perform population exploration operations together, solving the problems of insufficient computing power and robot endurance when the current power intelligent inspection is conducting large-scale inspections.

[0166] See also Figure 2 , is a schematic diagram of a structure of a collaborative inspection device based on heterogeneous multi-machine combination provided by an embodiment of the present invention, including:

[0167] The data acquisition module is used to obtain a number of collection points to be inspected in the inspection area, a number of supply points for supplying the inspection robot, the initial positions of the inspection robots, the initial resources, and the number of robots;

[0168] A connectivity graph construction module, used to determine connectivity paths between the collection points and the replenishment points according to the collection points and the replenishment points, and to construct an undirected connectivity graph of the inspection area according to the connectivity paths; wherein the collection points and the replenishment points are nodes of the undirected connectivity graph;

[0169] An ant colony construction module, used to construct a plurality of ant colonies, and initialize a plurality of male ants in the ant colonies according to the number of the robots, the initial positions, and the initial resources; wherein each of the ant colonies contains a plurality of male ants equal to the number of the robots, and in each of the ant colonies, the male ants correspond one to one with the inspection robots;

[0170] A population exploration module is used to use an ant colony algorithm to repeatedly perform population exploration operations on a number of ant colonies according to the undirected connected graph, initial positions of a number of male ants in the number of ant colonies, and initial resources, until a target path set is determined from a number of path sets generated by the ant colonies in a number of population exploration operations; wherein the path set is composed of a number of valid paths that a number of male ants in an ant colony traverse through all collection points in the undirected connected graph, and the valid path is a path that each male ant in the ant colony moves from one node to another node; the target path set is a path set with the shortest completion time and the highest task balancing rate among a number of path sets;

[0171] The task allocation module is used to determine the inspection task of the inspection robot corresponding to each male ant according to the effective path explored by each male ant in the target path set.

[0172] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0173] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0174] Another preferred embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements a collaborative inspection method based on heterogeneous multi-machine collaboration as described in any one of the above embodiments.

[0175] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0176] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0177] The memory can be used to store the computer program, and the processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0178] Another preferred embodiment of the present invention provides a storage medium, the storage medium is a computer-readable storage medium, the computer program is stored in the computer-readable storage medium, and the computer program, when executed by the processor, can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0179] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A collaborative inspection method based on heterogeneous multi-machine combination, characterized in that: include: Obtain several collection points to be inspected in the inspection area, several supply points for supplying the inspection robot, the initial positions, initial resources, and number of robots of the inspection robots; According to the collection points and the replenishment points, a connection path between each collection point and each replenishment point is determined, and according to the connection path, an undirected connected graph of the inspection area is constructed; wherein the collection points and the replenishment points are nodes of the undirected connected graph; Constructing a plurality of ant colonies, and initializing a plurality of male ants in the ant colonies according to the number of the robots, the initial positions, and the initial resources; wherein each of the ant colonies includes a plurality of male ants equal to the number of the robots, and in each of the ant colonies, the male ants correspond one to one with the inspection robots; Adopting an ant colony algorithm, according to the undirected connected graph, the initial positions of the male ants in the male ants and the initial resources, the population exploration operation is repeatedly performed on the male ants, until a target path set is determined from the path sets generated by the male ants in the population exploration operation; wherein the path set is composed of the valid paths that the male ants in the male ants in the ant colony pass through after traversing all the collection points in the undirected connected graph, and the valid path is the path that each male ant in the male ants moves from one node to another node; the target path set is the path set with the shortest completion time and the highest task balance rate among the path sets; The inspection task of the inspection robot corresponding to each male ant is determined according to the effective path explored by each male ant in the target path set.

2. A collaborative inspection method based on heterogeneous multi-machine combination as claimed in claim 1, characterized in that: The population exploration operation includes: Obtaining a global optimal solution, a global path pheromone, and a global expected reward table; wherein the global optimal solution is initially empty, the global path pheromone is used to record the pheromone on each valid path in the path set generated by each ant colony, and is initially a preset constant value; the global expected reward table is used to record the reward value of each male ant in each ant colony exploring the corresponding valid path, and is initially empty; Inputting the undirected connectivity graph, the global path pheromone, and the global expected reward table into an ant colony that has not performed a node exploration operation, so that the ant colony, according to the undirected connectivity graph, the global path pheromone, the global expected reward table, the initial position of the male ants, and the initial resources, repeatedly performs the node exploration operation until the ant colony has explored all the collection points and the male ants are all at the supply point; When it is confirmed that a target ant colony has explored all the collection points and the male ants are all at the supply point, a path set of the target ant colony is obtained; Calculate a first performance indicator of the path set, and compare the performance indicator with a second performance indicator of the global optimal solution. When it is determined that the first performance indicator of the path set is greater than the second performance indicator of the current global optimal solution, update the global optimal solution to the path set; wherein the first performance indicator and the second performance indicator are composed of completion time and task balance rate; According to the valid paths in the path set, updating the global path pheromone and the global expected reward table; Determining whether the total number of times the ant colonies perform population exploration operations is not less than a preset number threshold; If yes, re-perform the population exploration operation on the target ant colony according to the current global optimal solution, the current global path pheromone, and the current global expected reward table; If not, the global optimal solution is used as the target path set.

3. A collaborative inspection method based on heterogeneous multi-machine combination as claimed in claim 2, characterized in that: The node exploration operation includes: Obtain the current global path pheromone, the current global expected reward table, the current taboo table, the current position of each male ant, and the current resources; wherein, initially, the current position of each male ant and the current resources are the initial position and initial resources corresponding to each male ant, and the taboo table is used to record the collection points that the ant colony has explored, and is empty initially; Determining a target male ant for exploring an effective path according to the current position and current resources of each male ant; Determine a target node to be explored according to the target male ant, the global path pheromone, and the global expected reward table; wherein the target node is a collection point or a supply point; When the target node is determined to be a collection point, the remaining resources of the target male ant after reaching the target node from the current position are calculated, and the target supply point closest to the target node is determined; After determining that the target node is a supply point, or determining that the target male ant can reach the target node from the current position and can reach the target supply point by relying on the remaining resources, generating a valid path according to the current position and the target node, taking the target node as the new current position of the target male ant, taking the remaining resources as the new current resources of the target male ant, and adding the target node to the taboo table; Determine whether all collection points are recorded in the current taboo table, and whether the male ants are at the supply points; If so, a path set is generated based on several valid paths; If not, the next round of node exploration operation is performed.

4. A collaborative inspection method based on heterogeneous multi-machine combination as claimed in claim 3, characterized in that: After calculating the remaining resources of the target male ant after reaching the target node from the current position and determining the target supply point closest to the target node, the method further includes: After determining that the target male ant reaches the target node from the current position, when it is impossible to reach the target supply point by relying on the remaining resources, the global path pheromone and the global expected reward table are updated according to the current valid paths, the taboo table is cleared, and the node exploration operation is re-executed.

5. A collaborative inspection method based on heterogeneous multi-machine combination as claimed in claim 4, characterized in that: The updating of the global path pheromone and the global expected reward table includes: Obtaining the path pheromone of each valid path in the global path pheromone; Traversing the male ants, and calculating the pheromones left by each male ant on each valid path when the male ant passes through the valid path; When the traversal is completed, the sum of the pheromones on each of the valid paths is calculated; According to the sum of the pheromones of each valid path and the corresponding path pheromone, updating the path pheromone of each valid path in the global path pheromone; The effective path is input into a preset reward prediction model so that the reward prediction model outputs the expected reward value of each male ant exploring the corresponding effective path, and updates the expected reward table according to the expected reward value; wherein the reward prediction model is a model constructed using a Q learning algorithm.

6. A collaborative inspection method based on heterogeneous multi-machine combination as claimed in claim 5, characterized in that: Determining a target node to be explored according to the target male ant, the global path pheromone, and the global expected reward table includes: Take several collection points that are not recorded in the current taboo table, as well as idle supply points, as initial candidate nodes; According to the current resources of the target male ant, determining several initial candidate nodes that the target male ant can reach as candidate nodes; Calculate the probability of the target male ant colliding with other male ants when it arrives at each candidate node from the current position; Randomly generate a probability value, and when the probability value is less than a preset probability, randomly determine a target node from the candidate nodes by using a roulette wheel method; When the probability value is not less than a preset probability, a target node is determined from the candidate nodes according to the conflict probability, the global path pheromone, and the global expected reward table.

7. A collaborative inspection method based on heterogeneous multi-machine combination as claimed in claim 6, characterized in that: The step of determining a target node from the candidate nodes according to the conflict probability, the global path pheromone, and the global expected reward table includes: Determine the working time for the male ant to reach each of the candidate nodes from the current position and perform the inspection task according to the current position of the male ant and the candidate nodes; Calculating the heuristic information of each candidate node according to the working time and the conflict probability; The transition probability of each candidate node is determined according to the following formula: in, is the transition probability of candidate node j, τ ij (t) is the path pheromone between the current position of the male ant and the candidate node j, η ij is the heuristic information of candidate node j, ζ ij is the expected reward of the male ant from the current position to the candidate node j, α is the preset pheromone heuristic factor, β is the preset expectation heuristic factor, γ is the preset reward heuristic factor, is the set of candidate nodes.

8. A collaborative inspection device based on heterogeneous multi-machine combination, characterized in that: include: The data acquisition module is used to obtain a number of collection points to be inspected in the inspection area, a number of supply points for supplying the inspection robot, the initial positions of the inspection robots, the initial resources, and the number of robots; A connectivity graph construction module, used to determine connectivity paths between the collection points and the replenishment points according to the collection points and the replenishment points, and to construct an undirected connectivity graph of the inspection area according to the connectivity paths; wherein the collection points and the replenishment points are nodes of the undirected connectivity graph; An ant colony construction module, used to construct a plurality of ant colonies, and initialize a plurality of male ants in the ant colonies according to the number of the robots, the initial positions, and the initial resources; wherein each of the ant colonies contains a plurality of male ants equal to the number of the robots, and in each of the ant colonies, the male ants correspond one to one with the inspection robots; A population exploration module is used to use an ant colony algorithm to repeatedly perform population exploration operations on a number of ant colonies according to the undirected connected graph, initial positions of a number of male ants in the number of ant colonies, and initial resources, until a target path set is determined from a number of path sets generated by the ant colonies in a number of population exploration operations; wherein the path set is composed of a number of valid paths that a number of male ants in an ant colony traverse through all collection points in the undirected connected graph, and the valid path is a path that each male ant in the ant colony moves from one node to another node; the target path set is a path set with the shortest completion time and the highest task balancing rate among a number of path sets; The task allocation module is used to determine the inspection task of the inspection robot corresponding to each male ant according to the effective path explored by each male ant in the target path set.

9. A terminal device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a collaborative inspection method based on heterogeneous multi-machine combination as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute a collaborative inspection method based on heterogeneous multi-machine combination as described in any one of claims 1 to 7.

Images