Research On The Fusion Path Planning Algorithm For Mobile Robot

Mobile robot is an important branch in the field of robotics,with research directions mainly focused on localization,path planning,and formation collaborative control of mobile robot.The increasingly complex working environment poses higher requirements for mobile robot.Traditional path planning algorithms often focus on ideal static environments,while a single algorithm is difficult to meet practical needs.Therefore,the fusion of multiple algorithms is an important means to solve the application of mobile robot.The D*Lite algorithm is commonly used for robot global path planning,and its advantage is that when the environment changes,such as new obstacles affecting the existing path,it can be re-planned to obtain a new global path.In response to the problem of the path being too close to obstacles and the path being not smooth planned by the algorithm,safety detection rules and node deletion strategies are introduced,and an improved D*Lite algorithm is proposed to significantly reduce the path turning angle and safety risk.The dynamic window approach(DWA)is commonly used in robot local path planning,which allows robot to plan smooth and safe routes near obstacles,but there is a local extremum problem.This article proposes a fusion path planning algorithm that considers both the improved D*Lite algorithm and DWA.The improved D*Lite algorithm is used for robot global path planning and re-planning,while the DWA algorithm takes local nodes extracted from the global path as target points,providing the robot with local obstacle avoidance capability.The DWA algorithm compensates for the insufficient control of robot states by the D*Lite algorithm.The D*Lite algorithm provides a safer local target point for the DWA algorithm and solves the local extremum problem.Through MATLAB simulation and ROS robot path planning experiments,the fusion algorithm has been verified to have strong planning and obstacle avoidance abilities in complex environments.On this basis,the above fusion path planning algorithm is applied to multi robot formation.By using fusion algorithms to plan the path for the formation system,the follower robot runs the DWA to autonomously avoid obstacles,while maintaining the stability of the formation using the coordinates of the virtual follower as the target point.To verify the application effect of the algorithm,a triangular formation obstacle avoidance experiment of ROS robots was carried out,and the dynamic obstacle avoidance effect of the multi robot formation system was achieved,achieving the expected goal.