Abstract

In this paper a novel Polar Scan Matching (PSM) approach is de-scribed that works in the laser scanner’s polar coordinate system, therefore taking advantage of the structure of the laser measurements and eliminating the need for an expensive search for corresponding points in other scan match approaches. PSM belongs to the family of point to point scan matching approaches with its matching bearing association rule. The performance of PSM is thoroughly evaluated in a simulated experiment, in experiments using ground truth, in exper-iments aimed at determining the area of convergence and in a SLAM experiment. All results are compared to results obtained using an it-erated closest point (ICP) scan matching algorithm implementation. It is found that PSM is superior to the ICP implementation in process-ing speed and that PSM converges to a correct solution from a larger range of initial positions.