Global Path Planning for Fire-Fighting Robot Based on Advanced Bi-RRT Algorithm

Successful and efficient path planning is essential for autonomous mobile robot. Traditional Rapidly-Exploring Random Tree(RRT) algorithm usually suffer from long path planning time and strong randomness in specific environment such as depression trap and narrow channel. In order to tackle such problem, in this paper a bidirectional fast search algorithm based on violent matching and regression analysis is proposed for fire-fighting robot. Specifically, by using violent matching strategy, the path can be searched directly when there are few obstacles. In addition, the regression analysis reduces the probability of sampling nodes appearing in the same area. Meanwhile, a bidirectional RRT algorithm is employed to improve the path search efficiency. Finally, the redundant nodes in the random tree are discarded by the method of node backtracking, which reduces the robot control complexity. Several comparison experiment case studies are carried out on Rviz, and the results validate the effectiveness and advantages of our proposed approach.