This Article Full Text (PDF) References Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Web of Science (2) Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Biesiadecki, J. J. Articles by Maimone, M. W. Search for Related Content Social Bookmarking                         What's this?

Tradeoffs Between Directed and Autonomous Driving on the Mars Exploration Rovers

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA USA

P. Chris Leger

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA USA

Mark W. Maimone

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA USA, mwm{at}helios.jpl.nasa.gov

NASA’s Mars Exploration Rovers (MER) have collected a great diversity of geological science results, thanks in large part to their surface mobility capabilities. The six wheel rocker/bogie mobility system provides driving capabilities in a range of terrain types, while the onboard IMU measures actual rover attitude changes (roll, pitch and yaw, but not position) quickly and accurately. Four stereo camera pairs provide accurate position knowledge and/or terrain assessment. Solar panels generally provide enough energy to drive the vehicle for at most four hours each day, but drive time is often restricted by other planned activities. Driving along slopes in nonhomogeneous terrain injects unpredictable amounts of slip into each drive. These restrictions led to the creation of driving strategies that alternately use more or less onboard autonomy, to maximize drive speed and distance at the cost of increased complexity in the sequences of commands built by human Rover Planners each day.

Commands to the MER vehicles are typically transmitted at most once per day, so mobility operations are encoded as event-driven sequences of individual motion commands. Motions may be commanded using quickly-executing Directed commands which perform only reactive motion safety checks (e.g., real-time current limits, maximum instantaneous vehicle tilt limit), slowly-executing position measuring Visual Odometry (VisOdom) commands, which use images to accurately update the onboard position estimate, or slow-to-medium speed Autonomous Navigation (AutoNav) commands, which use onboard image processing to perform predictive terrain safety checks and optional autonomous Path Selection.

In total, the MER rovers have driven more than 10 kilometers over Martian terrain during their first 21 months of operation using these basic modes. In this paper we describe the strategies adopted for selecting between human-planned Directed drives versus rover-adaptive Autonomous Navigation, Visual Odometry and Path Selection drives.

Key Words: Mars Rover • MER • space robotics • autonomy • mission planning

The International Journal of Robotics Research, Vol. 26, No. 1, 91-104 (2007)
DOI: 10.1177/0278364907073777


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?