Research On Mobile Robot Lidar Map Construction And Pathplanning Under ROS

With the rapid update of robotics technology,more and more smart robots enter different fields to help people accomplish diverse tasks.How an intelligent robot can determine its own position in a complex and uncertain terrain environment,provide a high-quality map,and plan the optimal routes in a complex environment has important significance.Based on the Robot Operation System(ROS),in this paper,we design a program for laser simultaneous localization and mapping(SLAM)and path planning.Aiming at the problems of inadequate positioning accuracy and inaccurate mapping in traditional mapping algorithms,a mobile robot improved laser SLAM algorithm is proposed.At the same time,inadequate optimization of the planning path of algorithm A*,low search efficiency and frequent inflection points are improved,and the improved A*algorithm is combined with the local path planning algorithm.The specific research content is as follows:1.Aiming at the problems of the traditional RBPF(Rao-Blackwillised Particle Filter)algorithm with low positioning accuracy and lack of diversity of particles,an improved laser SLAM algorithm for mobile robots is proposed.Based on the principal component analysis method,the point cloud of adjacent frames are coarsely registered,and the improved point-to-line iterative closest point registration algorithm is used to correct the coarse registration result to complete the accurate registration.In the improved resampling algorithm,the low-weight particle set is introduced when the high-weight particle set is replicated multiple times to improve the lack of particle diversity.The original algorithm,the algorithm proposed in this paper and the minimum variance sampling method(MSV)proposed in the literature[75]are simulated and compared,it is guaranteed that the improved algorithm in this paper has better performance in improving the accuracy of lidar registration and state estimation.2.Aiming at the problems of insufficient path optimization,low search efficiency and many inflection points in the traditional A*algorithm,an improved path planning algorithm is proposed.The node distance information is integrated into the heuristic function with the exponential decay function as the coefficient to speed up the search speed of the global optimization.At the same time,the influence of the parent node of the previous generation is taken into account,which reduces the probability of node round-trip search.After obtaining the initial path,the gradient descent method is used to smooth the path.Finally,the improved A*algorithm and the local path planning algorithm are merged,and the evaluation function is changed to consider the global optimum.compared with the original A*algorithm,the improved A*algorithm significantly reduces the search time,the number of traversal nodes and the number of inflection points,in environments with different characteristics,the fusion algorithm can succeed on the basis of global optimization.The collision-free path is planned,and the global optimal performance and real-time obstacle avoidance ability of the fusion algorithm are verified.3.In different actual scenarios,turtlebot mobile robots are used to carry out SLAM and path planning experiments.The traditional SLAM algorithm,MSV-SLAM algorithm and the improved SLAM algorithm in this paper are used to conduct mapping comparison experiments,and the improved SLAM algorithm can obtain maps with higher accuracy.The efficiency and smoothness of the path planned by the improved A*algorithm are verified on the constructed map.At the same time,in the environment of known prior information and unknown prior information,the global and local path planning fusion algorithm proposed in this paper is tested.The fusion algorithm can not only maintain the global optimality of the path,but also avoid unknown obstacles in real time.The experimental results verify the effectiveness and reliability of the improved mapping and path planning algorithms in this paper.