Research Of Robot SLAM Technology Based On ROS

In order to improve human life better, the robots are invented to help people do what robots can do, especially some work requiring tedious processes, high precision in harsh environment. Since mobile robot needs to achieve autonomous localization and path planning under such conditions, the technology of robot simultaneous localization and mapping (SLAM) becomes particularly important. As the most fundamental and significant skill of robot, SLAM technology equips robot with the ability of autonomous walking, obticle avoidance and object recognition, which means the extensive application in the fields of space exploration, resource exploration, and technology development of indoor, outdoor or underwater robots.This paper presents the research background and significance of SLAM technology, introduced the research status, and states the basic theory and mainstream algorithms of SLAM. It respectively describes principles of Kalman filter (KF), extended Kalman filter (EKF) and particle filter (PF), and complete simulation experiments in the Matlab environment. What’s more, this paper mainly introduces EKF-SLAM algorithm based on EKF and FastSLAM based on PF, simulates under Matlab and analyzes the experiment outcomes. According the simulation results, we have carried out refinement strategy of FastSLAM algorithm. Under the robot operating system (ROS), we apply the improved algorithm in the SLAM experiment of indoor robot. During the process of experiment, the improved FastSLAM algorithm needs less memory capacity and computing time, exhibiting better outcome in real-time process.Finally, this paper summarizes three aspects of characteristics of improved FastSLAM algorithm in stability, precision and rapidity. We also analyze the deficiencies and development direction of our SLAM experiments, and make the outlook of future work at...