Research On Path Planning Of Mobile Robot Based On Improved RRT Algorithm

The motion-planning technology of mobile robots is one of the core parts of mobile robot navigation technology.Path planning will play a more and more important role as mobile robot technology develops.This paper mainly studied the path planning and path smoothing algorithms of mobile robots in acquired obstacle environments.It offered an improved Rapidly-exploring Random Tree(RRT)path planning algorithm for less obstacle environments and one for multi-obstacle environments.Two improved algorithms were proposed based on the in-depth study of the classical RRT algorithm,to improve the search efficiency and shorten the planning path length.The main research contents included the following aspects:Firstly,in less obstacle environment,to lower the randomness of the classical RRT algorithm in the global environment,this paper proposed a bidirectional RRT path planning algorithm based on the target gravity.It used the objective gravity theory and the bidirectional search strategy in the artificial potential field method.The algorithm adopted the objective gravitational idea and only needed local sampling to make the random tree grow towards the target point.The bidirectional search strategy was used to make the random tree grow from the initial point and the target point simultaneously.The improved algorithm effectively solved the problem of randomness in the environment of less obstacles.It improved the path search efficiency and shortens the planned path length.Secondly,in the multi-obstacle environment,the bidirectional RRT algorithm based on target gravity was less effective in obstacle avoidance.It also tended to fall into local minimum value and target unreachable problem.The author introduced variable step size strategy and greedy-choice strategy and proposed a method based on variable the step-by-step gravitational bidirectional smoothing RRT algorithm.This algorithm introduced a variable step size strategy,so that the random tree could adjust the step size to bypass the obstacle when the new node encounters the obstacle.It also used the greedy strategy to bypass the variable steps to avoid the redundancy at the obstacle.Nodes were eliminated to improve the smoothness of the path.The improved algorithm effectively avoided obstacles in a multi-obstacle environment,prevented the path from being partially localized and made the target reachable.It retained the original advantages of small randomness and greatly improved the path search efficiency and the path smoothness,which ensures the feasibility and superiority of the algorithm.Finally,the thesis implemented MATLAB to develop an experimental simulation environment platform that were required.The main function of the platform is to establish the required working environment space.It is convenient to carry out experimental simulation test on the path planning algorithm.This paper used the simulation experiment environment platform to simulate the improved RRT path planning algorithm proposed in different environments.Based on the experimental simulation of the platform,the feasibility and superiority of the improved path planning algorithm are proved.