GOBS: an intelligent system architecture for planetary robotic sampling

Abstract

An architecture of intelligent robotic systems, referred to here as GOBS: goal-oriented behavioral synthesis, is presented. Connecting sensing and action in real-time, PAN automatically synthesizes goal oriented behaviors under uncertainties, errors, and faults, through task monitoring and replanning. GOBS is significant for autonomous robotic tasks in unstructured environments, including planetary robotic sampling, which require robust and fault tolerant behaviors under uncertainties and errors in sensing and actuation as well as in environmental constraints. The application of GOBS to planetary robotic science is shown.