Depth Camera based Localization and Navigation for Indoor Mobile Robots

Abstract

We present here the Fast Sampling Plane Filtering (FSPF) algorithm, which reduces the volume of the 3D point cloud by sampling points from the depth image, and classifying local grouped sets of points as belonging to planes in 3D (called the plane filtered points) or points that do not correspond to planes (the outlier points). We present a localization algorithm based on an observation model that down-projects the plane filtered points on to 2D, and assigns correspondences for each point to lines on the 2D map. The full sampled point cloud (consisting of both plane filtered as well as outlier points) is processed for obstacle avoidance for autonomous navigation. We provide experimental results demonstrating the effectiveness of our approach for indoor mobile robot autonomy. We further compare the accuracy in localization using 2D laser rangefinders vs. using 3D depth cameras.