Range Imagery Algorithms for the Detection of Obstacles by Autonomous Vehicles

摘要

Algorithms are presented which segment range images and classify regions as being navigable or unnavigable by a land vehicle. The algorithms are applied to data collected from an active laser range sensor mounted on an autonomous land vehicle and their comparative results are analyzed. The sensitivity of various algorithms to uncertainty in the orientation of the range sensor is studied. Experiments on sensor calibration and image enhancement are also presented. A computer model of an autonomous land vehicle and its environment is described which provides a valuable tool for investigating may issues of navigation with range sensors. Obstacle detection algorithms are used in conjunction with the model to demonstrate a vehicle navigating itself through an obstacle strewn world to a goal location.