Research On Mobile Robot Map Building And Path Planning Based On Robot Operating System

With it developing ever faster,mobile robots are becoming increasingly important in people's daily lives.Operating in various areas including entertainment,industry,military and aviation they can be found almost everywhere.After the explosive development in artificial intelligence the study in mobile robotics is now enabled with many new directions of research.Today,there are mature mobile robot devices in various industries,such as cleaning robots,space robots and industrial robots.In the technical research of mobile robots,the key issue is to solve map construction and path planning,so that mobile robots can complete specified tasks in unknown environments.Increasing the operational accuracy and speed of mobile robots,as well as dealing with unexpected situations,are increasingly becoming the focus of research at this stage.This thesis analyzes the real-time localization and mapping technology(SLAM;Simultaneous localization and mapping)of mobile robots and the ROS(Robot Operating System).By studying the mathematical model of the mobile robot,and building the physical model of the mobile robot in the ROS system,the simulation study is carried out.Aiming at the traditional A*algorithm in path planning,an improved A*algorithm based on artificial potential field is proposed.In order to improve the accuracy and speed of the SLAM algorithm,the SLAM algorithm with extended Kalman filtering(EKF-SLAM)is improved and optimized.Through the analysis and research on the algorithm for building maps,we choose the algorithm suitable for the laboratory environment to build the map,so as to obtain the appropriate map information,and then compare the difference between the traditional A*algorithm and the improved algorithm in path planning in the simulation environment.First of all,through the analysis and research on the problem of real-time positioning and map construction,it is concluded that the mobile robot needs to build an updated map in an unknown environment,and it is clear that the SLAM problem itself is a probability estimation problem.At the same time,the ROS simulation platform to be used in this thesis is studied,and the mobile robot model to be used in the experiment is designed and constructed in the ROS system,which plays an effective platform guarantee for algorithm verification.Secondly,analyze and study the traditional A*algorithm used in path planning.In the actual environment,the optimal path planned by the A*algorithm may be too close to the obstacle,resulting in collision and safety problems.This thesis presents an improved A*algorithm based on artificial potential field.By introducing artificial potential field,the mobile robot can choose a sub-optimal safe path relatively far away from obstacles during path planning.The experimental results show that the path planned by the improved A*algorithm based on artificial potential field is more secure and operable.Third,the traditional extended Kalman filter SLAM(EKF-SLAM)algorithm has the disadvantages of low accuracy and low operation efficiency.This thesis proposes an improved adaptive EKF-SLAM algorithm based on Cholesky decomposition.Optimize and improve the observation noise and error covariance matrix in the traditional EKF-SLAM algorithm.Experiments show that the improved algorithm has higher accuracy and computational efficiency.Finally,the traditional map construction algorithm,Gmapping algorithm and Hector-SLAM algorithm are analyzed.Build a laboratory-based simulation environment on the ROS platform,and analyze the applicability of the two algorithms based on experimental results.The algorithm with high applicability is selected as the map construction algorithm in this thesis,and the constructed map information is obtained.The improved A*algorithm based on artificial potential field proposed in this thesis is compared with the traditional A*algorithm for path planning.Experiments show that the path planned by the improved A*algorithm based on artificial potential field has relatively high security.