Abstract:
To solve the problem of slow speed and low efficiency of the soccer robots using reactive obstacle avoidance strategy at soccer game, an improved rapidly-exploring random tree (RRT) algorithm is adopted, and a path planner based on this algorithm is designed to adapt to the dynamic moving obstacles environment in the robot soccer field. Firstly, the basic RRT algorithm is introduced, and some improvement ways to solve the disadvantages of strong randomness and long path length are proposed, such as probabilistically selecting the target node, adding the gravity component, and smoothing the path. The path planning in dynamic moving obstacles environment is solved by using the methods of path buffer and dynamic-expanding random tree. The simulation in complex obstacle environment shows that the path generated by the improved algorithm is about 20% shorter than the path generated by the basic RRT algorithm. Finally, the strategy is applied to the physical NAO robot to participate in the RoboCup competition and the top eight grade is achieved.