Path Planning Based on RRT-Stream+ Algorithm in Narrow Road Conditions

In order to solve the problems of RRT-Connect algorithm generating a large number of invalid nodes in narrow channels, long planning time, long path length and low pass rate in narrow road conditions, an algorithm named RRT-Stream is proposed, which is an improved RRT-Connect algorithm based on the idea of water flow, which first enhances the orientation of the RRT-Connect algorithm by limiting the sampling interval, so as to converge quickly. Then, when encountering obstacles, the water flow algorithm is used to quickly bypass the obstacles, and the circle drawing method is used to detect whether it has entered the narrow passage. In order to improve the versatility of the algorithm and make the algorithm explore normally with or without narrow channels, the RRT-Stream+ algorithm is proposed, which uses a heuristic search strategy, and the strategy of leaving the obstacle is changed to choosing to leave the obstacle at a lower cost. In order to prove the effectiveness of the proposed algorithm, MATLAB is used to simulate in three different maps, and the simulation results show that compared with the original algorithm, the algorithm has a smaller number of nodes, faster convergence speed, shorter planned path, and higher pass rate in narrow road conditions. Finally, the practicability of the algorithm is confirmed by the path planning experiment of the physical robot.