Autonomous Mobile Robot Exploration in Unknown Indoor Environments Based on Rapidly-exploring Random Tree

In recent years, robots have been quickly integrated into people's daily lives. To allow the robots to navigate autonomously, accurate maps have to be provided. Therefore, it is important to make robots to obtain maps automatically and improve the efficiency of autonomous exploration. In this paper, we propose a method based on the rapidly-exploring random tree (RRT) and frontier 2D-SLAM exploration techniques. The proposed system is divided into three parts. First, we construct an initial map with laser range data, and use RRT and frontier detector to identify the frontier points of the initial map. The frontier points are then filtered and clustered to reduce the total number and the computation load. Finally, the score of each frontier point is calculated and the mobile robot is directed to the unknown areas until the map is constructed. In the experiments, the performance is evaluated in various synthetic scenes and real indoor environments. The results show that our system is able to successfully complete the autonomous exploration task in a reasonable time.