CN-122015837-A - Multi-robot path planning method, device, system and storage medium

CN122015837ACN 122015837 ACN122015837 ACN 122015837ACN-122015837-A

Abstract

The application discloses a multi-robot path planning method, device and system and a storage medium, and belongs to the technical field of robots. The method comprises the steps of obtaining traffic information of each robot in a target area, determining traffic cost of each robot according to the traffic information, establishing a probability model of predicting traffic flow of each robot by using the traffic information to obtain a traffic flow prediction result, and feeding back attribute information of the robot, corresponding traffic cost and traffic flow prediction result to the robot in response to request information of the robot so that the robot can calculate future traffic cost passing through different positions of the target area according to the traffic flow prediction result, and planning a traffic path from a starting point to a target point according to the future traffic cost. According to the application, by predicting traffic flow, dependence on real-time communication is reduced, so that the robot can reduce path conflict or repeated planning caused by limited communication, reduce blockage and improve the task execution efficiency of the robot.

Inventors

LIU JUNBIN
YU KUNLIN

Assignees

炬星科技（深圳）有限公司

Dates

Publication Date: 20260512
Application Date: 20241112

Claims (10)

1. A multi-robot path planning method, comprising: Acquiring the passing information of each robot in a target area; Determining the passing cost of each robot according to the passing information; Establishing a probability model of the predicted traffic flow of each robot by utilizing the traffic information so as to obtain a traffic flow prediction result; Responding to request information of a robot, and feeding back attribute information of the robot, corresponding traffic cost and traffic flow prediction results to the robot so that the robot can calculate future traffic cost passing through different positions of the target area according to the traffic flow prediction results, and planning a traffic path from a starting point to a target point according to the future traffic cost; The traffic information comprises at least one of identification, position, direction and path of each robot.
2. The method of claim 1, wherein said determining the traffic cost of the respective robot based on the traffic information comprises: constructing a graph structure according to the target area, wherein the graph structure comprises a plurality of nodes representing intersection points between channels and a plurality of edges representing channels through which the robot can pass; calculating the basic travel cost passing through each side according to the length of the side and the moving speed of the robot; according to the congestion state of each side in a preset historical time period, calculating the historical passing cost of each side; identifying the current congestion degree of each side according to the traffic information, and calculating the real-time traffic cost of each side; And updating the real-time passing cost according to the updating time of the robot edge.
3. The method according to claim 2, wherein the creating a probability model of the predicted traffic flow of each robot using the traffic information to obtain a traffic flow prediction result includes: Establishing the probability model according to the probability distribution of normal advancing of different nodes of each robot at different moments and the probability distribution of collision of different nodes of each robot at different moments; determining the running condition of each robot at each node at the current moment according to the position, the path and the direction of each robot in the traffic information, wherein the running condition comprises the probability of collision of each robot at each node in the running process of the robot and the probability of normal running of the robot at each node; And predicting probability distribution of each robot at different positions of the target area in a preset time period in the future according to the running condition and the probability model to obtain a traffic flow prediction result of the target area.
4. A multi-robot path planning method, comprising: Receiving attribute information of the robot, corresponding passing cost and traffic flow prediction results fed back by the central manager, and acquiring a target task; establishing a path plan aiming at the target task according to the target task, the traffic cost and the traffic flow prediction result; acquiring information of other robots in the target area and corresponding traffic information; Estimating future traffic cost by utilizing the path planning and the traffic information of the other robots; Calculating the comprehensive cost of the passing path of the robot from the starting point to the target point according to the future passing cost and the passing cost fed back by the central manager; And determining an optimal passing path according to the comprehensive cost.
5. The method according to claim 4, wherein the method further comprises: the robot meeting the preset conditions in the passing path is used as a dynamic obstacle to be added into a dynamic obstacle list, wherein the preset conditions are determined according to task priorities of the robots; and re-planning a passing path from the starting point to the target point according to the future passing cost and the dynamic obstacle list.
6. The method according to claim 4, wherein the method further comprises: and under the condition of travelling to the target point according to the passing path, avoiding the target robot passing through the target point.
7. A multi-robot path planning apparatus, comprising: The first acquisition module is used for acquiring the traffic information of each robot in the target area; the first determining module is used for determining the passing cost of each robot according to the passing information; The first establishing module is used for establishing a probability model of predicting traffic flow of each robot by utilizing the traffic information so as to obtain a traffic flow prediction result; The feedback module is used for responding to the request information of the robot, feeding back the attribute information of the robot, the corresponding passing cost and the traffic flow prediction result to the robot, so that the robot can calculate future passing cost passing through different positions of the target area according to the traffic flow prediction result, and planning a passing path from a starting point to a target point according to the future passing cost; The traffic information comprises at least one of identification, position, direction and path of each robot.
8. A multi-robot path planning apparatus, comprising: the receiving module is used for receiving attribute information of the robot, corresponding passing cost and traffic flow prediction results and obtaining target tasks, wherein the attribute information of the robot is fed back by the central manager; the second establishing module is used for establishing a path plan aiming at the target task according to the target task, the traffic cost and the traffic flow prediction result; The second acquisition module is used for acquiring information of other robots in the target area and corresponding traffic information; The estimating module is used for estimating future passing cost by utilizing the path planning and the traffic information of the other robots; the calculation module is used for calculating the comprehensive cost of the passing path of the robot from the starting point to the target point according to the future passing cost and the passing cost fed back by the central manager; and the second determining module is used for determining the optimal passing path according to the comprehensive cost.
9. A multi-robot path planning system is characterized by comprising a central manager and a plurality of robots; The central manager for performing the method of any of claims 1-3; the robot for performing the method of any of claims 4-6.
10. A non-transitory computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method according to any of claims 1-6.

Description

Multi-robot path planning method, device, system and storage medium Technical Field The application belongs to the technical field of robots, and particularly relates to a multi-robot path planning method, device and system and a storage medium. Background In the intelligent storage field, the intelligent robot greatly improves the efficiency of logistics operation in a warehouse by virtue of the autonomous navigation and task execution capacity of the intelligent robot. However, as warehouse sizes increase and task complexity increases, more and more automated warehouses begin to employ multi-robot systems to cooperatively perform tasks. In the related art, a multi-robot path planning study is generally adjusted based on whether conflicts exist between paths of different robots, so that the planned path points of the robots are not in a collision range in the same time. However, in an actual storage environment, the communication capability of the robot is difficult to guarantee, and in a highly dynamic environment, once the robot does not make adjustment in time, collision and even blockage can be caused, so that the efficiency of the robot to execute tasks is reduced. Disclosure of Invention The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a multi-robot path planning method, device and system and a storage medium, so that the blockage of a robot is reduced, and the task execution efficiency of the robot is improved. In a first aspect, the present application provides a method for planning paths of multiple robots, including: Acquiring the passing information of each robot in a target area; Determining the passing cost of each robot according to the passing information; Establishing a probability model of the predicted traffic flow of each robot by utilizing the traffic information so as to obtain a traffic flow prediction result; Responding to request information of robots, feeding back attribute information of the robots and corresponding passing cost and traffic flow prediction results to the robots, so that the robots can calculate future passing cost passing through different positions of the target area according to the traffic flow prediction results, and planning a passing path from a starting point to a target point according to the future passing cost, wherein the passing information comprises at least one of identification, position, direction and path of each robot. According to the multi-robot path planning method of the application, the central manager collects the traffic information of each robot in the target area, so that the prediction process takes the current traffic condition into consideration to generate more accurate traffic flow prediction results, the prediction results are sent to each robot, and the robots evaluate the traffic cost of different paths according to the prediction results, thereby planning the optimal traffic path, and as the central manager can independently complete the prediction of traffic flow, the dependence on real-time communication is reduced, and even under the condition that communication is limited or unstable, the robot can still conduct path planning according to the prediction result received in advance instead of relying on real-time communication feedback, so that the robot can avoid path selection errors caused by communication delay or interruption, reduce blockage caused by path conflict or repeated planning, and improve the efficiency of the robot to execute tasks. According to one embodiment of the present application, the determining the traffic cost of each robot according to the traffic information includes: constructing a graph structure according to the target area, wherein the graph structure comprises a plurality of nodes representing intersection points between channels and a plurality of edges representing channels through which the robot can pass; calculating the basic travel cost passing through each side according to the length of the side and the moving speed of the robot; according to the congestion state of each side in a preset historical time period, calculating the historical passing cost of each side; identifying the current congestion degree of each side according to the traffic information, and calculating the real-time traffic cost of each side; And updating the real-time passing cost according to the updating time of the robot edge. In the embodiment, the target area is abstracted into a graph structure, the complex space environment is simplified into a series of computable nodes and edges, the attributes of the channels and the intersection points can be clearly identified and quantified, and the passing cost of each robot is determined by combining the basic travelling cost, the historical passing cost and the real-time passing cost, so that when the robot plans a path, the traffic condition is considered, the understanding of