A Two-Step Particle Filter for SLAM of Corridor Environment

Abstract

The simultaneous localization and map building (SLAM) problem is critical to the research of mobile robot. Enabling robots to build the map of corridor environment is the basis for them to accomplish the tasks independently in the corridor environment. In this paper, the characteristic of the corridor environment is analyzed, the reason why it is hard to apply the general particle filter to solving the SLAM problem in the corridor environment is explored, and a novel two-step particle filter SLAM approach is presented. The experimental results demonstrate this approach can solve the correlation among states without the need of appending any artificial landmarks beforehand in the corridor environment in the process of SLAM