In recent years,mobile robots have been increasingly applied in various places and are gradually becoming more intelligent due to rapid developments in science and technology.Path planning,which is the main content of mobile robots,has become a hot research topic.The path planning algorithms in this thesis are based on the Robot Operating System(ROS)and have been improved to address existing shortcomings.Specifically,the A* algorithm in global path planning and the Dynamic Window Approach(DWA)in local path planning were improved.The improved A* algorithm and DWA algorithm are then fused and the effectiveness of the improved algorithm is validated using MATLAB simulation software.Finally,the improved algorithm is deployed on a real mobile robot platform to improve the efficiency and quality of mobile robot path planning.The research contents of this thesis are as follows:(1)Regarding global path planning,the principle of the traditional A* algorithm is first explained.Shortcomings of this algorithm such as too many turning points,low search efficiency,and an unsmooth path are revealed in simulation experiments.To address these issues,improvements have been made to the traditional A* algorithm.Improvements to the algorithm were made.Firstly,the evaluation function is improved by dynamically adjusting the heuristic function part to enhance node search efficiency.Secondly,the Floyd algorithm idea is utilized to remove redundant points.Finally,Bezier curves are used to smooth turning points.Experiments are conducted in various simulation environments to compare results.The experiments demonstrate that the improved A* algorithm is more efficient in operation and produces smoother paths that comply with the kinematic constraints of mobile robots.(2)In local path planning,the principle of the traditional DWA algorithm is explained and its shortcomings,such as lengthy paths and a tendency to become trapped in local optima,are revealed through simulation experiments.To address these issues,the evaluation function of the DWA algorithm is improved by considering global information and adding a new evaluation index to the existing evaluation function.Simulation experiments are conducted to improve the smoothness of the path and the local obstacle avoidance capability.(3)To further validate the effectiveness of the improved algorithm,simulation experiments are conducted in the Gazebo physical simulation environment under the ROS system to simulate real-world environments.Then,a two-wheeled differential mobile robot model is used to conduct simulation experiments,and the improved algorithm is deployed on a real mobile robot model.The experimental results show that the improved algorithm eliminates path fluctuations at turning points compared to traditional algorithms,improves the efficiency of path planning,and has better obstacle avoidance performance,facilitating the robot’s movement and allowing it to safely and quickly reach its target point. |