Unknown labyrinth traversal method based on detection and following of gap which is closest to target and is not visited

Abstract

The invention discloses an unknown labyrinth traversal method based on detection and following of an unvisited gap closest to a target. The method comprises the following steps: determining a free space around a robot based on depth information around the robot and the size of the robot, wherein the depth information is scanned by a sensor; scanning the depth points in the free space to search for discontinuity so as to obtain a gap and further obtain an effective gap; establishing a topological map for the effective gaps in a kdtree form; in combination with the position of the robot at the current moment, gaps which can be searched by the robot at present are searched in the topological map, the distances between the midpoints of the gaps and the labyrinth end point are ranked from large to small, and finally the gaps which are closest to the labyrinth end point and are not accessed are selected as sub-targets explored and accessed at the next moment; whether repeated access is carried out or not is checked by using a topological map of the gap, and the robot is helped to be prevented from being stuck in dead beard or circulation; according to the method, the robot can reach any target point in the maze.

Description

technical field [0001] The invention belongs to the technical field of robot control, in particular to an unknown maze traversal method based on detecting and following the gap closest to the target and unvisited Background technique [0002] A complex environment containing different road branches is called a maze. The goal of solving the maze problem is to find a path with the shortest path and the highest efficiency in the shortest possible time, and follow this path to reach the end of the maze (target point). The maze solving system can be applied in many fields such as fire rescue, disaster relief robots, and intelligent traffic control systems. There are many different maze solving schemes today, which can be roughly divided into two categories: solving based on known maze models, the structure of the entire maze model is known at the beginning of solving the problem; solving based on sensor information, the maze structure is known at the beginning of solving the pro...

AI Technical Summary

Problems solved by technology

[0010] However, most of the above traditional solutions not only need to spend a lot of time to calculate, but also easily make the solution process fall into a dead end

Images