Comparative Analysis of Three Kinds of Laser SLAM Algorithms

Abstract

With the development of artificial intelligence, the application of robots is also rapidly increasing. How to autonomously navigate and complete complex tasks for robots in an unknown environment is a hot spot in the research domain of simultaneous positioning and map construction (SLAM) algorithms. To better study and apply three common laser SLAM algorithms, by building a SLAM environment on the ROS robot platform, Hector SLAM, Gmapping, and Cartographer algorithms were used to conduct actual indoor mapping experiments. All three algorithms can achieve effective indoor two-bit mapping construction. By comparing and analyzing the three SLAM algorithms, the mapping accuracy of the Cartographer algorithm is significantly better than Hector SLAM and Gmapping algorithms. Meantime, the Cartographer algorithm has better robustness.