Dp-slam

We present a novel, laser range finder based algorithm for simultaneous localization and mapping (SLAM) for mobile robots. SLAM addresses the problem of constructing an accurate map in real time despite imperfect information about the robot's trajectory through the environment. Unlike other approaches that assume predetermined landmarks (and must deal with a resulting data-association problem) our algorithm uses the sensor range data directly to build metric occupancy maps. Our algorithm uses a particle filter to represent both robot poses and possible map configurations. By using a new map representation, which we call distributed particle (DP) mapping, we are able to maintain and update hundreds of candidate maps and robot poses efficiently. Through careful design, we are able to achieve a time complexity which is linear in both the area observed and the number of particles used. This thesis also presents an autonomous method for learning an appropriate motion model for the robot, providing a more accurate posterior distribution, and thus requiring fewer resources. Furthermore, we develop a hierarchical framework for SLAM which allows DP-SLAM to maintain appropriate amounts of uncertainty for significantly longer than would otherwise be possible. Our technique makes essentially no assumptions about the environment yet it is accurate enough to close loops over 250m in length with crisp, perpendicular edges on corridors and minimal or no misalignment errors, despite significant noise and ambiguity in the environment. We also provide methods for extending DP-SLAM into three dimensional maps.