Conflict search path optimization method based on warehouse automatic guiding robot

Abstract

The invention discloses a conflict search path optimization method based on a warehouse automatic guiding robot, and relates to the technical field of artificial intelligence and computer science, and the method comprises the following steps: S100, determining the input and output of path optimization; s200, path optimization preparation is carried out; s300, defining an action space; s400, defining a target of path optimization; s500, designing and searching a binary constraint tree; s600, expanding nodes of the binary constraint tree; s700, segmenting a key conflict; s800, bypassing the conflict; s900, searching for the maximum matching of the conflict graph; and S1000, establishing an automatic guiding robot path set.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and computer science technology, and in particular to a conflict search path optimization method based on a warehouse automatic guidance robot. Background Art

[0002] The path planning problem for multiple automated guided robots in a warehouse means that in a given graph, multiple automated guided robots need to move from their respective starting points to their respective target points, and the paths cannot conflict. Time is discretized into time steps, and each automated guided robot either moves to an adjacent empty position or stays at the current position at each time step. The goal of path planning is to find a set of conflict-free paths that minimize the total path cost of all automated guided robots. This problem is an NP (Non-deterministic Polynomial) problem with a wide range of practical applications.

[0003] The conflict-based search (CBS) algorithm is an algorithm based on a two-layer search structure to solve the path planning problem for multiple automated guided robots. The core idea is divided into two layers: the upper layer is conflict detection, and the lower layer is path search. In the conflict detection stage, the algorithm detects whether there are conflicts in the paths between the automated guided robots. If there are conflicts, they need to be resolved to ensure that all automated guided robots can reach their destination safely. In the path search stage, based on the results of conflict detection, a path search is performed on each automated guided robot to find an optimal path so that all automated guided robots can reach their destination according to the planned path. However, the conflict search variants of the prior art only use the path cost as the priority of the node, without utilizing heuristic information.

[0004] Improved conflict search improves search efficiency by selecting critical conflicts and bypassing non-critical conflicts. Improved conflict search divides conflicts into three categories: critical conflicts, semi-critical conflicts, and non-critical conflicts, and prioritizes critical conflicts for branching. In addition, some conflicts can be avoided by modifying the path, thereby reducing the size of the constraint tree. At the same time, for different conflict scenarios, corresponding constraints can also be added to avoid multiple node branches, such as rectangular symmetric scenarios, target symmetric scenarios, and corridor symmetric scenarios. Recently, some studies have classified critical conflicts, semi-critical conflicts, and non-critical conflicts through multi-valued decision graphs and mutually exclusive propagation techniques, and added corresponding constraints to reduce the depth of the binary constraint tree. Despite this, the improved conflict search method still only uses path cost as the priority of the node, and fails to make full use of heuristic information.

[0005] Therefore, technicians in this field are committed to developing a conflict search path optimization method based on a warehouse automatic guidance robot. Summary of the invention

[0006] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is how to improve the optimization efficiency and conflict reduction efficiency of the path optimization method based on conflict search.

[0007] The path optimization method aims to solve the problem of establishing an efficient motion path for a group of automated guided robots in a warehouse map after a given target position. It is necessary to solve the conflict problem that may occur during the movement of multiple automated guided robots and avoid conflicts. The search-based method is used to design the optimal path, grid all the passable areas of the automated guided robots in the warehouse, and adopt the conflict avoidance method to reduce the possible conflicts that may occur among multiple automated guided robots during the movement. The heuristic enhancement method is further used to improve the optimization efficiency and form an efficient automatic guided robot optimization path.

[0008] In one embodiment of the present invention, a conflict search path optimization method based on a warehouse automatic guidance robot is provided, comprising the following steps: S100, clarifying the input and output of path optimization, where the input of path optimization includes path planning map information, the number of automated guided robots, the starting position and target position of each automated guided robot, and the output of path optimization is a conflict-free optimal path for each automated guided robot; S200, path optimization preparation, selecting an automatic guided robot, generating a graph according to the path planning graph information, and discretizing the time of the automatic guided robot path optimization into time steps; S300, defining an action space, defining the action space of the automatic guiding robot and the action of the automatic guiding robot at each time step; S400, defining a path optimization goal, where the action cost of automatically guiding the robot to be the lowest under the condition that there is no conflict in the paths is the path optimization goal; S500, designing and searching a binary constraint tree, designing and constructing a binary constraint tree, traversing each node in the binary constraint tree, and sorting the nodes in ascending order according to the action cost of the automatic guiding robot; S600, expanding the binary constraint tree node, adopting the best first search strategy, selecting the node with the smallest action cost for expansion, branching the node for the conflicting path, and generating two child nodes; S700, split the key conflict and determine the conflict type. If it is a key conflict, branch to generate two child nodes and add corresponding constraints in each child node. Otherwise, execute S800; S800, bypass conflicts. For non-critical conflicts, re-plan the path of the automatic guidance robot, use heuristic information to guide the lower-level search, and find a conflict-free path; S900, finding a maximum match of a conflict graph, constructing a conflict graph according to non-intersecting key conflicts between the automatic guidance robots, and determining heuristic information through a maximum match of the conflict graph; S1000, establishing a path set of automatic guided robots, and repeating steps S200-S900 until the conflict-free optimal paths of all automatic guided robots are completed.

[0009] Optionally, in the conflict search path optimization method based on the warehouse automatic guided robot in the above embodiment, the path planning map information includes a grid node set after gridding all the traversable areas of the automatic guided robot in the warehouse. The set of edges connecting adjacent nodes , the number of automatic guided robots is The automatic guided robot is represented as , automatic guided robot For the Automatic guided robot , automatic guided robot Starting position and target location ,in , are positive integers, , The upper limit is the ratio of the area of ​​all traversable areas to the area occupied by a single automated guided robot, rounded down and minus 1.

[0010] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, step S200 includes: S210, generate a map, grid the traversable area of ​​the automatic guided robot, and record it as ; S220, time discretization, discretize the time of automatic guidance robot path optimization into time steps, automatically guide the robot At time step At the starting position .

[0011] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, step S300 includes: S310, action space definition, defining the action space A of the automatic guidance robot = {up, down, left, right, wait}; S320, action selection, selecting the action for automatically guiding the robot at each time step, including moving to an adjacent idle node or waiting in the current grid.

[0012] Optionally, in any of the above embodiments, in the conflict search path optimization method based on the warehouse automatic guided robot, the path in step S400 is the automatic guided robot From the starting position To the target location The route passed, each path corresponds to an automatic guided robot, and the conflict is the automatic guided robot and automated guided robots At the same time step Occupies the same node at the same time , expressed as ,The action cost of the automatic guided robot is the action path cost of the automatic guided robot, that is the time consumed by the action.

[0013] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, step S500 includes: S510, design and construct binary constraint tree, binary constraint tree Each node A set of constraints containing two autonomous robots that are in conflict , the set of paths for all automatically guided robots and the total action cost of all automated guided robot paths , where the constraint set of the root node is empty; S520, traverse the binary constraint tree, search the binary constraint tree Each node in , imposes a set of constraints on the two autonomous guided robots that are in conflict , among which, for the automatic guided robot The constraints imposed are a tuple , indicating automatic guidance of the robot At time step Banned from occupying nodes , and obtain a single path that satisfies all constraints and the corresponding action cost ; S530, node sorting, upper conflict detection, search binary constraint tree Perform a best-first search, sorting the nodes in ascending order based on action cost, and in case of ties, select the node with the least conflicting solutions.

[0014] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, step S600 includes: S610, node expansion, adopting the best first search strategy to select the node with the smallest action cost for expansion; S620: Conflict check: check the path set that satisfies the constraint set Is there a conflict? If not, return to the path. Otherwise, execute S630. S630, conflict handling, selecting a conflicting path, splitting the node into two sub-nodes, each sub-node having a set of constraints.

[0015] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, the conflict type in step S700 includes: Critical conflict: Both autonomous robots in conflict need to increase their action costs to avoid the current conflict; Semi-critical conflict: The conflict can be avoided if only one of the two conflicting automated guided robots increases the action cost. Non-critical conflict: Both autonomous robots in conflict can avoid the current conflict without increasing action costs.

[0016] Furthermore, in the conflict search path optimization method based on the warehouse automatic guidance robot in the above embodiment, step S700 also includes giving priority to key conflicts, branching the node into two sub-nodes, and adding corresponding constraints in each sub-node.

[0017] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, step S800 includes: S810, replanning the path, for semi-critical conflicts and non-critical conflicts, replanning the path of the automatic guided robot where the conflict occurs, and resolving the conflict without increasing the action cost; S820, find a conflict-free path, use the maximum number of matching edges contained in all matches of the conflict graph as heuristic information, use the heuristic information, combined with the Manhattan distance between the current position and the target position, to guide the lower-level search and find the conflict-free path faster.

[0018] Optionally, in the conflict search path optimization method based on a warehouse automatic guidance robot in any of the above embodiments, step S900 includes: S910, constructing a conflict graph, where a node in the conflict graph represents each automatic guided robot, and if there is a key conflict between two automatic guided robots, there is an edge between the corresponding nodes; S920: Determine heuristic information, and calculate the maximum match of the conflict graph using a flower tree algorithm.

[0019] This application is based on the conflict search path optimization method of warehouse automatic guided robots. By rationally utilizing heuristic information and improved conflict handling strategies, it significantly improves the efficiency and effect of path planning for multiple automatic guided robots and reduces the calculation time. Compared with traditional conflict search algorithms, the introduction of heuristics enables the algorithm to maintain good performance in the case of large-scale and multiple automatic guided robots.

[0020] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flow chart illustrating a conflict search path optimization method based on a warehouse automatic guidance robot according to an exemplary embodiment. DETAILED DESCRIPTION

[0022] The following describes several preferred embodiments of the present invention with reference to the drawings in the specification, so that the technical content is clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the protection scope of the present invention is not limited to the embodiments mentioned in the text.

[0023] In the drawings, components with the same structure are indicated by the same numerical reference numerals, and components with similar structures or functions are indicated by similar numerical reference numerals. The size and thickness of each component shown in the drawings are arbitrarily shown, and the present invention does not limit the size and thickness of each component. In order to make the illustration clearer, the thickness of the components is schematically exaggerated in some places in the drawings.

[0024] The present invention designs a conflict search path optimization method based on a warehouse automatic guidance robot, such as Figure 1 As shown, the following steps are included: S100, clarify the input and output of path optimization, the input of path optimization includes path planning map information, the number of automatic guided robots, the starting position and target position of each automatic guided robot, and the path planning map information includes the grid node set after all the traversable areas of the automatic guided robots in the warehouse are gridded The set of edges connecting adjacent nodes , the number of automatic guided robots is The automatic guided robot is represented as , automatic guided robot For the Automatically guided robots, automatic guided robots Starting position and target location ,in , is a positive integer, , The upper limit is the ratio of the area of ​​all traversable areas to the area occupied by a single automated guided robot, rounded down and minus 1; the output of path optimization is the conflict-free optimal path of each automated guided robot.

[0025] S200, path optimization preparation, selecting an automatic guided robot, generating a graph according to the path planning graph information, and discretizing the time of the automatic guided robot path optimization into time steps; specifically including: S210, generate a map, grid the traversable area of ​​the automatic guided robot, and record it as ; S220, time discretization, discretize the time of automatic guidance robot path optimization into time steps, automatically guide the robot At time step At the starting position .

[0026] S300, defining an action space, defining the action space of the automatic guiding robot and the action of the automatic guiding robot at each time step; specifically including: S310, action space definition, defining the action space A of the automatic guidance robot = {up, down, left, right, wait}; S320, action selection, selecting the action for automatically guiding the robot at each time step, including moving to an adjacent idle node or waiting in the current grid.

[0027] S400, define the goal of path optimization, in the case of no conflict in the path, the action cost of the automatic guidance robot is the lowest as the goal of path optimization, the path is the automatic guidance robot From the starting position To the target location The route passed, each path corresponds to an automatic guided robot, and the conflict is the automatic guided robot and automated guided robots At the same time step Occupies the same node at the same time , expressed as ,The action cost of the automatic guided robot is the action path cost of the automatic guided robot, that is the time consumed by the action.

[0028] S500, designing and searching a binary constraint tree, designing and constructing a binary constraint tree, traversing each node in the binary constraint tree, and sorting the nodes in ascending order according to the action cost of the automatic guiding robot; specifically including: S510, design and construct binary constraint tree, binary constraint tree Each node A set of constraints containing two autonomous robots that are in conflict , the set of paths for all automatically guided robots and the total action cost of all automated guided robot paths , where the constraint set of the root node is empty; S520, traverse the binary constraint tree, search the binary constraint tree Each node in , imposes a set of constraints on the two autonomous guided robots that are in conflict , among which, for the automatic guided robot The constraints imposed are a tuple , indicating automatic guidance of the robot At time step Banned from occupying nodes , and obtain a single path that satisfies all constraints and the corresponding action cost ; S530, node sorting, upper conflict detection, search binary constraint tree Perform a best-first search, sorting the nodes in ascending order based on action cost, and in case of ties, select the node with the least conflicting solutions.

[0029] S600, expanding the binary constraint tree node, adopting the best first search strategy, selecting the node with the smallest action cost for expansion, branching the node for the conflicting path, and generating two child nodes; specifically including: S610, node expansion, adopting the best first search strategy to select the node with the smallest action cost for expansion; S620: Conflict check: check the path set that satisfies the constraint set Is there a conflict? If not, return to the path. Otherwise, execute S630. S630, conflict handling, selecting a conflicting path, splitting the node into two sub-nodes, each sub-node having a set of constraints.

[0030] S700, split key conflicts and determine conflict types, which include: Critical conflict: Both autonomous robots in conflict need to increase their action costs to avoid the current conflict; Semi-critical conflict: The conflict can be avoided if only one of the two conflicting automated guided robots increases the action cost. Non-critical conflict: Both autonomous robots in conflict can avoid the current conflict without increasing action costs; If it is a critical conflict, the critical conflict is processed first, branching is performed to generate two child nodes, and a corresponding constraint is added to each child node. Otherwise, S800 is executed.

[0031] S800, bypass conflicts. For non-critical conflicts, re-plan the path of the automatic guidance robot, use heuristic information to guide the lower-level search, and find a conflict-free path; specifically, it includes: S810, replanning the path, for semi-critical conflicts and non-critical conflicts, replanning the path of the automatic guided robot where the conflict occurs, and resolving the conflict without increasing the action cost; S820, find a conflict-free path, use the maximum number of matching edges contained in all matches of the conflict graph as heuristic information, use the heuristic information, combined with the Manhattan distance between the current position and the target position, to guide the lower-level search and find the conflict-free path faster.

[0032] S900, finding the maximum match of the conflict graph, constructing the conflict graph according to the non-intersecting key conflicts between the automatic guidance robots, and determining the heuristic information through the maximum match of the conflict graph; specifically including: S910, constructing a conflict graph, where a node in the conflict graph represents each automatic guided robot, and if there is a key conflict between two automatic guided robots, there is an edge between the corresponding nodes; S920: Determine heuristic information, and calculate the maximum match of the conflict graph using a flower tree algorithm.

[0033] S1000, establishing a path set of automatic guided robots, and repeating steps S200-S900 until the conflict-free optimal paths of all automatic guided robots are completed.

[0034] In order to verify the technical effect of the present invention, an experiment was conducted in an automated warehouse. The specific situation is as follows: the traversable area of ​​the automated warehouse is gridded and divided into a 20×20 square grid map. Each grid on the map represents the robot's traversable area. There are 10 automated guided robots (AGVs), represented as , transport goods from their respective starting locations to the specified target locations. Find the optimal path for all automatic guided robots, that is, the path with the shortest possible total driving time, while avoiding conflicts during driving.

[0035] The automatic guided robot number, starting point and terminal information are as follows: Automatic guided robot A: starting point (2,2), end point (18,18); Automatic guided robot B: starting point (2,18), end point (18,2); Automatic guided robot C: starting point (10,2), end point (10,18); Automatic guided robot D: starting point (2,10), end point (18,10); Automatic guided robot E: starting point (10,18), end point (10,2); Automatic guided robot F: starting point (18,2), end point (2,18); Automatic guided robot G: starting point (18,10), end point (2,10); Automatic guided robot H: starting point (10,10), end point (15,15); Automatic guided robot I: starting point (15,15), end point (10,10); Automatic guided robot J: starting point (5,5), end point (15,5).

[0036] Therefore, the input and output of the path optimization are: Inputs for path optimization: warehouse map (20×20 grid), number of automated guided robots (10 AGVs), starting position and target position of each automated guided robot; Output of path optimization: conflict-free optimal path for each automatic guided robot.

[0037] According to the above steps S200-S1000, the path optimization is performed, and the specific results (i.e., the path set of the automatic guidance robot) are as follows: Automatic guided robot A: Moves diagonally from (2,2) to (18,18); Automatic guided robot B: moves along another diagonal line from (2,18) to (18,2); AGRO C: moves along the longitudinal straight line from (10,2) to (10,18), choosing to wait at certain time steps to avoid conflicts with AGRO A and B. AGV D: Moves along a horizontal straight line from (2,10) to (18,10), waiting appropriately to avoid collision; AGRO E: From (10,18) to (10,2), in the opposite direction of AGRO C, using the waiting action to avoid conflict; Automatic guided robot F: From (18,2) to (2,18), the path intersects with automatic guided robots A and B, and avoids conflict by adjusting the time step and waiting; AGV G: From (18,10) to (2,10), in the opposite direction of AGV D, waiting appropriately to avoid conflict; AGRO H and AGRO I: exchange positions between (10,10) and (15,15) and make minor adjustments to the paths to avoid collision in the middle area. AGRO J: Moves from (5,5) to (15,5), traveling in a straight line horizontally, avoiding other AGROs.

[0038] It can be seen that the present invention successfully plans the optimal path without conflict for 10 automatic guided robots, and the total cost is minimized. The present invention improves the search efficiency of the algorithm and reduces the calculation time through heuristic enhancement. Compared with the traditional conflict search algorithm, the introduction of heuristics enables the algorithm to maintain good performance in the case of large-scale and multi-automatic guided robots.

[0039] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A conflict search path optimization method based on a warehouse automatic guidance robot, characterized in that: The steps include: S100, clarifying the input and output of path optimization, where the input of path optimization includes path planning map information, the number of automated guided robots, the starting position and target position of each automated guided robot, and the output of path optimization is a conflict-free optimal path for each automated guided robot; S200, path optimization preparation, selecting an automatic guided robot, generating a graph according to the path planning graph information, and discretizing the time of the automatic guided robot path optimization into time steps; S300, defining an action space, defining the action space of the automatic guided robot and the action of the automatic guided robot at each time step; S400, defining a path optimization goal, where the action cost of the automatic guided robot is the lowest when there is no conflict in the paths, as the path optimization goal; S500, designing and searching a binary constraint tree, designing and constructing a binary constraint tree, traversing each node in the binary constraint tree, and sorting the nodes in ascending order according to the action cost of the automatic guiding robot; S600, expanding the binary constraint tree node, adopting the best first search strategy, selecting the node with the smallest action cost for expansion, branching the node for the conflicting path, and generating two child nodes; S700, split the key conflict and determine the conflict type. If it is a key conflict, branch to generate two child nodes and add corresponding constraints in each child node. Otherwise, execute S800; S800, bypass conflicts. For non-critical conflicts, re-plan the path of the automatic guidance robot, use heuristic information to guide the lower-level search, and find a conflict-free path; S900, finding a maximum match of a conflict graph, constructing a conflict graph according to non-overlapping key conflicts between the automatic guiding robots, and determining heuristic information through a maximum match of the conflict graph; S1000, establishing a path set of automatic guided robots, and repeating steps S200-S900 until the conflict-free optimal paths of all automatic guided robots are completed.

2. The conflict search path optimization method based on the warehouse automatic guidance robot according to claim 1, characterized in that: The path planning map information includes a grid node set after all traversable areas of the automatic guided robot in the warehouse are gridded The set of edges connecting adjacent nodes , the number of automatic guided robots is The automatic guided robot is represented as , automatic guided robot For the Automatically guided robots, automatic guided robots Starting position and target location ,in , is a positive integer, , The upper limit is the ratio of the area of ​​all traversable areas to the area occupied by a single automated guided robot, rounded down and minus 1.

3. The conflict search path optimization method based on the warehouse automatic guidance robot according to claim 2, characterized in that: The step S200 includes: S210, generating a map, generating a grid map of the traversable area of ​​the automatic guided robot, denoted as ; S220, time discretization, discretizing the time of the automatic guided robot path optimization into time steps, and automatically guiding the robot At time step At the starting position .

4. The conflict search path optimization method based on the warehouse automatic guidance robot according to claim 3 is characterized in that: The step S300 includes: S310, action space definition, defining the action space A of the automatic guided robot = {up, down, left, right, wait}; S320, action selection, selecting the action of the automatic guiding robot at each time step, including moving to an adjacent idle node or waiting in the current grid.

5. The conflict search path optimization method based on warehouse automatic guidance robot according to claim 4, characterized in that: The path is for an automatically guided robot From the starting position To the target location The route passed, the conflict is the automatic guidance robot and automated guided robots At the same time step Occupies the same node at the same time , expressed as ,The action cost of the automatic guided robot is the action path cost of the automatic guided robot, that is the time consumed by the action.

6. The conflict search path optimization method based on warehouse automatic guidance robot according to claim 5, characterized in that: The step S500 includes: S510, design and construct a binary constraint tree, the binary constraint tree Each node Contains the constraint set of the two conflicting automated guided robots and the path set of all automated guided robots and the total action cost of all automated guided robot paths , where the constraint set of the root node is empty; S520, traverse the binary constraint tree and search the binary constraint tree Each node in , imposes a set of constraints on the two autonomous guided robots that are in conflict , among which, for the automatic guided robot The constraints imposed are a tuple , indicating automatic guidance of the robot At time step Banned from occupying nodes , and obtain a single path that satisfies all constraints and the corresponding action cost ; S530, node sorting, upper conflict detection, search binary constraint tree Perform a best-first search, sorting the nodes in ascending order based on action cost, and in case of ties, select the node with the least conflicting solutions.

7. The conflict search path optimization method based on warehouse automatic guidance robot according to claim 6, characterized in that: The step S600 includes: S610, node expansion, adopting the best first search strategy to select the node with the smallest action cost for expansion; S620: Conflict check: check the path set that satisfies the constraint set Is there a conflict? If not, return to the path. Otherwise, execute S630. S630, conflict handling, selecting a conflicting path, splitting the node into two sub-nodes, each sub-node having a set of constraints.

8. The conflict search path optimization method based on warehouse automatic guidance robot according to claim 7, characterized in that: The step conflict types include: Critical conflict: Both autonomous robots in conflict need to increase their action costs to avoid the current conflict; Semi-critical conflict: If only one of the two conflicting automated guided robots increases the action cost, the current conflict can be avoided. Non-critical conflict: Both autonomous robots in conflict can avoid the current conflict without increasing action costs.

9. The conflict search path optimization method based on warehouse automatic guidance robot according to claim 8, characterized in that: The step S800 includes: S810, replanning the path, for the semi-critical conflict and the non-critical conflict, replanning the path of the automatic guided robot where the conflict occurs, and resolving the conflict without increasing the action cost; S820, find a conflict-free path, use the maximum number of matching edges contained in all matches of the conflict graph as heuristic information, use the heuristic information, combined with the Manhattan distance between the current position and the target position, to guide the lower-level search and find the conflict-free path faster.

10. The conflict search path optimization method based on warehouse automatic guidance robot according to claim 9, characterized in that: The step S900 includes: S910, constructing a conflict graph, wherein a node in the conflict graph represents each automatic guided robot, and if there is a key conflict between two automatic guided robots, there is an edge between the corresponding nodes; S920: Determine heuristic information, and calculate the maximum match of the conflict graph by using a flower tree algorithm.

Images