Research On RRT Path Planning Algorithm For 2D Static Environment

When the working environment is dangerous,robots can replace manpower to perform tasks,which can not only ensure human safety,but also save time and obtain more economic value.The path issue problem has always been the focus of research in the field of robotics.The effectiveness of a path planning algorithm impacts how fast and safely the planning process may be accomplished.Path planning refers to finding a collision-free path from the starting point to the target point in a given workspace,so that the robot can move and perform tasks autonomously.The Rapidly-exploring Random Tree(RRT)algorithm is a sampling-based path planning algorithm.Aiming at the problems of poor search efficiency,low search path quality,and too many path inflection points in the path planning process of the RRT algorithm,the main research content of this thesis is as follows:(1)The RRT algorithm is random,and the path is not optimal.The RRT*algorithm adds linear optimization to the RRT algorithm,but because the sampling is still global random sampling,it takes a long time and has poor real-time performance.Therefore,based on the traditional RRT* algorithm,the traditional RRT* algorithm is optimized and improved for the problems of slow planning speed,high cost of the generated path,and many redundant points of the path.First ly,change the selection method of sampling points in the traditional algorithm,select better sampling points first,accelerate the speed of obtaining the path,and then optimize the generated path,and use the method based on interpolation to make the path keep approaching obstacles,thereby reducing the path.Finally,the simulation experiment and data analysis of the improved algorithm are carried out to verify the effectiveness of the improved algorithm.(2)The Informed-RRT* algorithm is an improvement based on the RRT*algorithm.This algorithm uses a heuristic search strategy,which greatly speeds up the search for the initial path,and narrows the sampling area to the ellipse,which accelerates the convergence of the algorithm.However,the sampling part of the algorithm is still the same as the RRT* algorithm,which is global random sampling,which has poor guidance,blindness in the search path,and poor quality of the obtained path in a limited number of iterations.In view of the above problems,a goal-oriented(GOInformed-RRT*)algorithm is proposed.By changing the generation method of new nodes,the new nodes are constrained by the sampling points and target points at the same time,making the algorithm more oriented,speed up the growth of the path toward the target point,and assign different weights in the two directions respectively,optimize the performance of the algorithm in complex environments,and finally smooth the generated path based on cubic β-splines.Through the experimental simulation verification,the improved algorithm reduces the path cost and planning time,and improves the convergence speed of the algorithm.(3)In order to verify the effect of the improved algorithm in practice,the improved algorithm is tested by the Turtle Bot3 mobile robot car in the real map environment.Build a map environment in the experimental site,then control the car to scan and build the map,save the map and start the navigation module.In the experiment,the car will detect the surrounding map information and carry out autonomous navigation and real-time path planning according to the improved algorithm.Through multiple experiments,the improvement is proved the effectiveness of the algorithm in practical applications.