An autonomous exploration and obstacle avoidance method based on adaptive frontier exploration target point selection

Abstract

An autonomous exploration and obstacle avoidance method for self-adaptive frontier exploration target point selection. Firstly, the principle of traversable area selection is explained through laser scanner scanning information; secondly, based on the principle of adaptive frontier exploration target point selection, the robot position that adapts to the current moment is selected. According to the exploration threshold and the evaluation function, the optimal exploration target point is selected, and the dynamic obstacle may appear when the robot moves to the exploration target point, and the least square method is used to fit the movement of the dynamic obstacle. trajectory, to judge the relative relationship between the robot and the obstacle, and then select an effective exploration target point according to the principle of adaptive frontier exploration target point selection; finally, by querying the data set of the exploration target point, the possibility of repeated exploration and missing exploration can be effectively avoided. Ensure the integrity of the explored map. The invention can not only meet the precision and real-time requirements of practical application, but also has low cost.

Description

technical field [0001] The invention is applied to the field of simultaneous positioning and map construction of mobile robots, and involves the autonomous exploration of mobile robots, especially when there are obstacles and narrow passages in the scene, how to select a feasible area and how to select an exploration target point, and realizes An efficient approach to autonomous exploration. Background technique [0002] The present invention mainly considers how a mobile robot can avoid static obstacles and dynamic obstacles while using a laser scanner to effectively conduct autonomous exploration and scene map construction. The main autonomous exploration methods include the judgment method of the feasible area, the selection method of the exploration target point and the dynamic obstacle collision avoidance method. Although the use of laser scanners to construct scene maps is very mature, the technology to allow robots to explore autonomously and to construct scene maps ...

AI Technical Summary

Problems solved by technology

The information scanned by the laser sensor can effectively identify the current environment information, so as to obtain local sub-targets and effective travel routes. However, the traditional autonomous exploration method is based on the frontier breakpoint information, and the robot cannot effectively bypass when encountering obstacles. Obstacles, as well as the problem of ignoring some narrow passages and repeated exploration, lead to incomplete and time-consuming problems in the scene map constructed by the robot, and when there are dynamic obstacles in the process of exploring the target point, it is easy to Collision

Images