Calibration of a multi-sensor system laser rangefinder/camera

Abstract

This paper proposes a new calibration method for a perception system designed for a mobile robot. The perception system we use is made of a gray-level CCD camera and a laser rangefinder. The problem is to match data provided by those two sensors and therefore to identify the laser rangefinder/camera transformation. A calibration process based upon a specific calibration pattern is used to identify such a transformation. Adapted procedures define a geometric relationship between these two sensors. We have tested in simulation several models in order to identify the transformation. The most efficient one has been chosen and implemented on the perception system of an experimental mobile robot. Experimental results are presented.