Adaptive Motion Planning for Autonomous Rough Terrain Traversal with a Walking Robot

• DOI: 10.1002/rob.21610
• Publication Date: 2015-06-24T11:37:28Z
• Authors: Dominik Belter; Przemysław Łabęcki; Piotr Skrzypczyński
• Keywords: 0209 industrial biotechnology; 02 engineering and technology

Achieving full autonomy in a mobile robot requires combining robust environment perception with onboard sensors, efficient mapping, and real‐time motion planning. All these tasks become more challenging when we consider natural, outdoor that has many degrees of freedom (DOF). In this paper, address the issues planning legged walking over rough terrain, using only its sensors to gather necessary model. The proposed solution takes limited perceptual capabilities into account. A multisensor system is considered for perception. key idea planner use dual representation concept map: (i) higher‐level applies A* algorithm coarse path on low‐resolution elevation grid, (ii) lower‐level guided‐RRT (rapidly exploring random tree) find sequence feasible motions precise but smaller map. This paper contributes new method can identify terrain traversability cost benefit algorithm. probabilistic regression technique applied assessment typical RRT‐based used explore space values. efficiency our approach demonstrated simulations provide ground truth data unavailable field tests. However, simulation‐verified then thoroughly tested under real‐world conditions experiments two six‐legged robots having different systems.