Research Of Mobile Robot Path Planning In Complex Environment

Path planning is a relatively important part of the field of mobile robot research and a prerequisite for completing various tasks.During the movement from the starting point to the target point,the robot can not only reach safely,but also keep the path as smooth as possible.Good algorithms can help mobile robots adapt to various changes in complex environments and complete tasks to reach specified locations.This article takes individual robots as research objects,and studies path planning in a twodimensional environment.The main research contents of this paper are as follows.1.Global path planning can find the optimal solution,but it is necessary to know the accurate information of the environment in advance.If an unknown dynamic obstacle occurs,this method may not be completed.Local path planning can provide real-time feedback and correction of the planning results,but due to the lack of global environmental information,the planned path may not be optimal.By analyzing the advantages and disadvantages of each algorithm,this paper proposes a hybrid algorithm combining the rapid expansion random tree(RRT)method with local path planning and the artificial moment method,which performs well in global path planning,to perform path planning for mobile robots.Because the artificial moment method may cause the robot to choose the wrong direction of motion,make the path cumbersome,and in line with the new attraction segment,the new attraction segment and attraction point should be recalculated,the calculation is more complex First,the RRT is used for static preplanning globally,and the child nodes and trajectories of the final path are saved,treat each child node as the attraction of the robot.Second,the artificial moment controller is used to make the robot trace the child nodes and trajectories generated by the RRT pre-planning to reach the target point.The robot can avoid the problem that the robot chooses the wrong initial motion direction and path.2.Because the classic RRT chooses to expand in a random direction or a target direction through the probability value P(0 to 1),the final path is relatively tortuous and messy.Therefore,based on the classic RRT method,a smooth processing RRT method is proposed,which increases the step size that is biased toward the target direction.In the free area,the nodes are completely extended toward the target point.When there are obstacles,the step size is increased toward the random point.Make the globally planned path closer to the ideal path.Because the RRT method is not suitable for path planning when there are dynamic obstacles,the robot cannot track the next node when it encounters dynamic obstacles,and then guide the robot through the attraction line segments and points generated by the knowledge obstacle wall of dynamic obstacles.Make the robot avoid dynamic obstacles,and then track the next child node until it reaches the target point.3.Using MATLAB to simulate and analyze the hybrid algorithm of smooth processing of RRT and artificial torque method,whether in complex static environment or dynamic environment,the hybrid method can not only obtain the global optimal path,but also have good local dynamic obstacle avoidance ability.The feasibility and efficiency of the hybrid algorithm are presented.