Research On Path Planning Of Wheeled Mobile Robots Based On RRT Algorithm

Wheeled mobile robots are playing an increasingly important role in contemporary industrial development.Path planning can provide feasible paths for wheeled mobile robots in a variety of environments,which is the premise to ensure that robots can complete tasks.For all kinds of path planning algorithms,Rapidly-exploring Random Tree(RRT)path planning algorithm does not depend on the model and does not need to model the space,so it can solve all kinds of key path planning problems.However,in engineering applications,due to the randomness of nodes and paths generated by RRT algorithm,although it can effectively solve the problem of path planning in different environments in different application scenarios,it is difficult to show stable performance.Aiming at the defects of the above RRT algorithm,the path planning problem of RRT algorithm in static,dynamic and complex maze environment is studied.The main research work of this paper is as follows:The rapidity and optimality of Informed RRT* algorithm in static environment are studied.An improved path planning algorithm based on Informed RRT* is designed to generate sampling steps according to changes in the environment.The sampling steps and nodes generated by the Informed RRT* algorithm will influence the planning efficiency and path optimality.Enables it to quickly find Informed subsets,reduce time spent searching for initial solutions,remove nodes outside the ellipse,reduce unnecessary calculations,improve planning efficiency,and finally optimize the generated paths repeatedly to the best possible result.Then,the optimality and complexity of the proposed algorithm are analyzed.Finally,the effectiveness,rapidity and optimality of the proposed method are verified by simulation and experimental results.Path planning for wheeled mobile robots based on dynamic Informed sampling in dynamic environments is studied.In the dynamic environment,moving obstacles may affect the robot movement at any time,and the planned path is difficult to ensure the safety of the robot movement and the effectiveness of the path.In this paper,an algorithm is proposed to ensure the global asymptotic optimization,and the locally reprogrammed path is also asymptotic optimal.When it is observed that moving obstacles are about to affect the robot’s movement,only the part of the path that may be affected is replanned,and this part of the path is optimized within the safe area,which greatly reduces the complexity of planning and improves the safety of the replanned path.Finally,the asymptotic optimality of the reprogramming path is analyzed,and the feasibility,optimality and robustness of the proposed method are verified by comparing various algorithms in experiments.The path planning problem of wheeled mobile robot in complex maze environment is studied.For the complex maze environment,the existing algorithms can not overcome the complexity,tortuous,irregularity and other problems of the maze environment in the planning process,so a two-tree and omnidirectional search sampling algorithm is designed.The algorithm overcomes the disadvantages of the original double RRT search tree for path planning in the maze environment,and does not use "greedy strategy" or "heuristic method".The sampling and expansion method of starting and ending double tree omnidirectional search is adopted to delete redundant paths in infeasible areas and improve planning efficiency.After all feasible paths are found,the shortest path is optimized and the final path is generated.Finally,the effectiveness and superiority of the proposed algorithm are verified by simulation and experiment.