As modern society relies on intelligent mobile robots deepens,researchers are increasingly focusing on the robots’ adaptability to their surrounding environment.Autonomous navigation includes the main components of SLAM,path planning and intelligent obstacle avoidance,which are the technical foundation for robots to accomplish various tasks in different work scenarios.At present,mobile robots have oscillations in speed and unsmooth trajectories when performing path planning on general roads,and there is limited room for optimization because of the single obstacle avoidance strategy on undulating roads.Therefore,thesis takes a wheeled mobile robot as the research object,optimizes the traditional path planning algorithm through ROS platform simulation and experimental verification.And investigates the obstacle avoidance strategy under undulating road surfaces to determine the optimal technical solution for intelligent obstacle avoidance of the robot.Based on the analysis of existing algorithms,thesis analyzes the required scenario of the mobile robot and the design of autonomous navigation scheme based on the ROS platform,and introduces topology-based parallel trajectory optimization in the traditional TEB algorithm,using trajectory curvature as a new constraint and proposes the CSC-TEB algorithm;Then the 3D point cloud information of the undulating road surface around the robot is obtained by a depth camera to achieve an intelligent obstacle avoidance strategy.The A* algorithm and the CSC-TEB algorithm cooperate with each other throughout the path planning process to avoid unstable speed output during strong turns,safely and quickly avoid the obstacles encountered so as to reach the specified target point smoothly.And joint simulation and experiments verify that implement an intelligent obstacle avoidance strategy in the face of undulating roads to complete the specified path planning task. |