Ambler: Performance of a Six-Legged Planetary Rover

Abstract

Exploration of planetary surfaces by mobile robots is now within technical reach. On the Moon, robots could be used to explore lunar resources, to conduct scientific observations and to carry out a variety of simple construction tasks. On Mars, robots could be employed to survey the planet's composition, monitor its weather and return samples for analysis on Earth. To plan such missions, a host of technical questions must be answered. What degree of mobility must be achieved to accomplish different missions? What rates of power consumption are required for different terrains? What levels of precision and accuracy are necessary? What are the proven capabilities of autonomous machines in terms of a long-duration mission? Planetary rover researchers around the world are attempting to answer these questions. Significant research efforts are underway in Europe [6], Japan [4], North America(2,l I] and Russia[3]. Many of the research programs are in the early stages, and have not yet produced extensive experimental performance results. In the spirit of providing data to mission planners, in this paper we attempt to provide quantitative answers to some of the questions posed earlier. The answers are based on our practical experience with the performance of an autonomous, six-legged robot for an exploration mission in Mars-like terrain. In Section 2, we briefly describe the configuration and operation of the Ambler walking robot. In the next three sections, we concentrate on performance metrics in power consumption, positioning accuracy and autonomous walking. For each of these topics, we present our approach, describe the experimental methods, analyze the results and outline future research directions.