Solution for multiple mobile robot path conflict based on dense storage regions

Abstract

The invention relates to a solution for multiple mobile robot path conflict based on dense storage regions. The technical scheme of the solution comprises steps of (1), establishing an initialized mesh path image by dispatching a server; (2), detecting a cargo access task and dividing the task into a plurality of sub-tasks; (3), outputting task paths of current sub-tasks by using an image search method; (4), determining whether target points of sub-tasks are released by a last task; (5), marking all task paths by using task numbers; (6), returning to the step (2) after marking all task paths; (7), seizing resource variables through all tasks in a task executing queue; (8), dividing conflict regions; (9) determining whether conflict regions are released completely; (10), sending a task execution command to a mobile robot; and (11) returning to the step (7) after releasing resource variables to the server. The solution has the advantages of being reasonable in design, capable of distributing tasks at any time, planning tasks at any time and avoiding conflict at any time, high in accuracy, simple in calculation and practical.

Description

technical field [0001] The invention belongs to the field of mobile robots, in particular to a method for solving path conflicts of multiple mobile robots based on dense storage areas. Background technique [0002] The conditions for multi-robot collision avoidance planning in dense storage areas are: (1) the movement of robots is limited to two-dimensional space, which belongs to plane motion; (2) the roadways in dense storage areas are not only narrow but also coherent. Its width only allows a robot to pass through lying down; the so-called coherence is that the combination of several waypoint areas is formed by the interlacing and superimposition of areas with different attitudes: the robot can pass through this area with different attitudes; (3) the collision of the robot Refers to the task deadlock caused by walking towards each other and staggered on the same lane, regardless of the situation of walking in the same direction; (4) The moving robot constitutes a dynamic ...

AI Technical Summary

Problems solved by technology

The traffic rule law is to formulate rules for multiple robots to abide by and maintain the orderliness of the group. The walking channel must ensure the right side of the passage. When two robots walking towards each other are walking on the same channel, only when they occupy their own lanes can deadlocks not occur. If the width of the channel cannot accommodate the width of the two robots, only When letting a robot pass, the traffic law is clearly invalid at this time

The rate adjustment method and the priority method are used together. The rate adjustment includes acceleration, deceleration, stop and even reversing. When two robots moving towards each other have a deadlock, the one with the higher priority has the right to give way to the one with the lower level. The disadvantage of this method is that only relying on local information will generate local poles, which cannot guarantee that the robot can reach the destination smoothly. For example, one of the robots must retreat to an open area to solve the problem, but The environmental conditions for the robot to move backwards cannot be obtained from the current local information: how to ensure that it will not collide with other robots on the way during the backwards process? Where is the most economical and reasonable place for the robot to reverse? How to ensure that the regressing target point is a valid target point and a series of problems, because the distributed conflict resolution method can not really achieve the task only by relying on local information, and it also increases the difficulty. In actual operation, this method is rarely adopted

[0005] To sum up, the centralized management method is the most accurate, but the operation is complicated and the real-time performance is poor. Although the distributed method is simple in operation, strong in real-time and flexible, it often cannot complete the task completely due to the occurrence of local extremes.

Images