The HURON task allocation and scheduling system: Planning human and robot activities for lunar missions

摘要

NASApsilas new lunar exploration program encompasses a broad set of objectives, to be implemented through a sequence of missions. Lunar surface activities to be conducted as part of these missions include ground exploration, characterization of the lunar environment, construction or deployment of permanent structures and systems, engineering experiments and science surveys, among others. Astronauts will have access to tools and vehicles, and these vehicles in turn may also be capable of functioning independently of an astronaut, being teleoperated from Earth or operating with some level of onboard autonomy. Astronauts and vehicles will also be constrained by available consumables, maximum allowable astronaut working times, and other concerns. In this paper, we describe HURON (human-robot task network optimization), an optimal multi-agent task allocation and scheduling system developed for lunar mission planning. HURON assigns tasks to the various agents on the surface, plans their cooperation in the execution of tasks, and computes activity schedules while taking into account the appropriate and optimizing the mission plan based on a given objective function. We present the overall architecture of HURON and discuss its application to lunar mission exploration scenarios.