Research On Robot Indoor SLAM Technology Based On Lidar

As a key technology in the field of mobile robots,the SLAM has attracted more and more attention from scholars.As an important branch of SLAM,laser SLAM is widely used because of its stability and mature technology.As a mainstream laser SLAM method,the Cartographer SLAM algorithm has been applied in many occasions because of its low cumulative error,suitable for low cost radar and other features.However,in similar environments or environments with fewer feature points,Cartographer SLAM may cause loop closure detection matching errors due to its scan-to-map matching method,resulting in inaccurate mapping and long mapping time.Aiming at the above problems,this paper improves the cartographer SLAM algorithm to improve the mapping accuracy of mobile robot and reduce the mapping time,and verifies its feasibility and effectiveness in the mobile robot platform.The main contents of this paper are as follows:(1)To solve the problem of easy matching errors in Cartographer SLAM loop closure detection,this paper proposes a map-to-map loop closure detection matching method with adaptive submap expansion.This method takes the map-to-map loop closure detection matching algorithm as the basic principle,and then improves it on this basis.First,the similarity between the current submap and the previous submap is calculated according to the number of scanning points in the submap.If the similarity is less than the set threshold,the submap selection stage is entered,and the expansion of two adjacent submaps is further completed according to the number of scanning points in this stage;secondly,according to the map-to-map matching rule,loop closure detection matching is achieved between the selected submap and the map.(2)Software and hardware design of mobile robot platform is carried out.Taking the onboard computer as the core,a mobile robot platform is built through modules such as sensors and driving chassis.At the same time,a software program framework is proposed based on the platform,and the SLAM mapping of the wheeled mobile robot under ROS is realized.(3)The improved algorithm is verified by experiments.In four different indoor environments,Karto,Cartographer and the improved algorithm are used for comparative experiments.The mapping effect and experimental data show that the improved algorithm can improve the mapping accuracy,shorten the mapping time,solve the problem of Cartographer,and verify the feasibility and effectiveness of the algorithm.