Off-road obstacle classification and traversability analysis in the presence of negative obstacles

In order for an autonomous unmanned ground vehicle (UGV) to drive in off-road terrain at high speeds, it must analyze and understand its surrounding terrain in real- time; it must know where it intends to go, where are the obstacles, and many details of the topography of the terrain. Much research has been done in the way of obstacle avoidance, terrain classification, and path planning, Yet few UGV systems can effectively traverse off -road environments at high speeds autonomously. This paper presents algorithms that analyze off-road terrain using a point cloud produced by a 3D laser rangefinder, determine potential obstacles both above ground and those where the ground cover has a negative slope (negative obstacles), then plan safe routes around those obstacles. To classify negative obstacles, this research uses a combination of a geometry-based method called the Negative Obstacle DetectoR (NODR) and a support vector machine (SVM) algorithm. The terrain is analyzed with respect to a large UGV with the sensor mounted up high as well as a small UGV with the sensor mounted low to the groun