Abstract

Abstract — We present a novel terrain classification technique both for effective, autonomous locomotion over natural, unknown terrains and for the qualitative analysis of terrains for exploration and mapping. Our straight-forward approach requires a single camera with little processing of visual information. Specifically, we derived gait bounce and gait roll measures from visual servoing errors that result from vehicle-terrain interactions during normal locomotion. Characteristics of the terrain, such as roughness and compliance, manifest themselves in the spatial patterns of these signals and can be extracted using pattern classification techniques. For legged robots, different limb-terrain interactions generate gait bounce signals with different information content, thus deliberate limb motions can effect higher information content (i.e. the robot is an active sensor of terrain class). Segmentation of the gait cycle based on the limb-terrain interaction isolates portions of the gait bounce signal with high information content. The decoding of, then sequencing of, this content from each cycle segment yields a robust classification of terrain type from known benchmarks. This is analogous to word recognition in which the spatial pattern of speech encodes phonemes and the sequence of the phonemes encodes the word. To extract the spatiotemporal pattern of the gait bounce signal, we developed a meta-classifier using discriminant analysis and hidden Markov models. In this paper, we present the gait bounce and gait roll derivation, terrain classification using both spatial discriminants and our meta-classifier, and we describe how terrain classification can be used for gait adaptation, particularly in relation to an efficiency metric. We also demonstrate that our technique is generally applicable to other locomotion mechanisms such as wheels and treads. I.

Citations