Research On Environment Exploration And Path Planning Of Mobile Robot

Path planning and robot environment exploration strategies have been key research questions in the field of robotics.Currently,path planning algorithms in autonomous robot environment exploration tasks often suffer from problems such as path quality that fails to meet practical applications,slow environment exploration speed,and incomplete compositional regions.In order to solve these problems,this paper proposes an environmental exploration strategy based on fast RRT to improve the efficiency of map construction,and an improved path planning algorithm based on RRT* to improve the speed and quality of robot path planning.Finally,the proposed algorithm is validated by real-world experiments using robots in real-world environments.The main points of the paper are as follows:(1)Environmental exploration Strategy Research: RRT(Rapidly exploring Random Tree)algorithm is an important part of robot environmental exploration,and its path planning efficiency fundamentally affects the robot environmental exploration efficiency.Therefore,in this paper,we study the environment exploration strategy from the perspective of efficiency of the RRT algorithm.First,an improved RRT algorithm is proposed based on the RRT algorithm.The algorithm sets different partition sampling probabilities,restricts the growth direction of the random tree,and increases the expansion efficiency of the random tree by adding the nearest node connection strategy and the collisional re-selection strategy.Simulation experiments on three different difficulty maps show that the running time of the improved RRT algorithm is reduced by about 62 percent and the number of iterations is reduced by about 60 percent compared to the conventional RRT algorithm.Second,by combining the partition sampling idea and the target bias idea of the improved RRT algorithm with the environmental exploration task,we propose the sampling point range limit policy and the heuristic node policy.Experimental results show that the improved environmental exploration strategy can improve the expansion efficiency of the tree structure by restricting the sampling area in real-time and the ergodic efficiency of the map by heuristically node-wise optimization.Exploration speed was increased by 18.7% and 21.2% in two different types of simulated environments,respectively.(2)Research on robot path planning Problem: An improved path planning algorithm ATS-RRT* is proposed based on RRT* algorithm.An alternative path strategy is first proposed in,where multiple initial paths are generated depending on whether the newly generated node is directly connected to the target point or not,and the path with the lowest cost is set as the final initial path,which speeds up the initial path planning.Second,a triangle region sampling strategy is proposed,which generates a set of triangle regions and the corresponding half-triangle regions for every three path nodes,and rapidly reduces the path cost by limiting the sampling range.Finally,the triangle-node direct connection strategy and the tabu-table strategy are proposed.The former tries to connect the generated nodes to the path nodes,while the latter avoids repeated collision detection of existing nodes with the path nodes and improves the running speed of the algorithm.The experimental results show that compared with RRT*,Quick-RRT*,and Informed + Quick-RRT*,the initial path discovery speed of ATS-RRT* algorithm is increased by 2.3 times,and the second-best path discovery rate is increased by 1.45 times.(3)Robot experiment: In the real environment,Turtlebot2 robot is used to carry out real experiments on the improved environmental exploration strategy and ATSRRT* planning algorithm.Firstly,the experimental results show that the modified environmental exploration strategy can quickly discover the information of boundary points in the map in real-world environments,and the modified strategy combined with the map creation algorithm can build a complete map of the environment.Demonstrating the availability of improved environmental exploration strategies.Secondly,experiments show that the ATS-RRT* path planning algorithm can be applied to real robots to complete fixed point navigation in real environment,and the path planned by robots is suitable for robots to walk,which proves the availability of ATS-RRT* algorithm.