From Interplanetary Cruise to the Surface of Mars The Challenges of infusing AI in Space
In 1999, NASA flew the first AI based closed-loop control system in space, the Remote Agent Experiment (RAX), on the New Millennium Deep Space One (DS1) spacecraft. This deployment fundamentally changed the perception of AI research within (and without) the agency, and helped institute a robust funding program for Autonomy and Robotics research for NASA. While the promise of more applied AI and Robotics systems within the space domain has yet to be fully realized, the agency is taking small and incremental steps towards realizing the ambitions of scores of researchers in the field. This talk is the story of why progress has been incremental and what it took to get to deploy a mission-critical AI application on a NASA science mission.
The Mixed-initiative Activity Plan GENerato(MAPGEN) is a system that combines a rich formalism of a flexible temporal constraint network with a familiar front end used by mission operations personnel at JPL for mission planning. It provides a ground-based decision-support system in the critical part of the uplink command cycle for the 2003 Mars Exploration Rovers (see http://marsrovers.jpl.nasa.gov & http://ic.arc.nasa.gov/story.php?sid=106&sec=space).
Mission operators with the help of science personnel, have continued to use MAPGEN, twice daily to build a complex conflict free plan that is packaged and uplinked to command the Spirit and Opportunity rovers on the surface of Mars. This generative planner, automatically enforces mission and flight rules encased in a declarative model, as well as constraints imposed to encode the scientific intent of observations for that Sol(Martian day) to be executed onboard the two rovers.
I will briefly discuss the results of the deployment so far (which have far exceeded our own expectations) and will take you along on the wild ride we went thru to get to be a critical part of NASA’s most complex scientific mission to date.
Kanna is a Senior Research Scientist and a member of the management team of the the Autonomy and Robotics Area at NASA Ames Research Center Moffett Field, California. He is one of the principals of the Remote Agent Experiment (RAX) which designed, built, tested and flew the first AI based closed loop control system on a spacecraft. The RA was the co-winner of NASA's 1999 Software of the Year, the agency's highest technical award (http://ic.arc.nasa.gov/projects/remote-agent/).
His interests are in Planning/Scheduling, modeling and representation for real world planners and agent architectures for Distributed Control applications. Prior to joining Ames, he was in the doctoral program at the Courant Institute of Math Sciences at New York Univ. Prior to that he was at the Knowledge Systems group at American Airlines, helping build a Maintenance Routing scheduler(MOCA) which continues to be used by the airline 365 days of the year. The MAPGEN team at Ames and JPL has been awarded the 2004 Turning Goals into Reality award under the Administrators Award category. He is currently the Principal Investigator for MAPGEN and is looking forward to returning to leading a normal human existence soon.
He is the recipient of the 2002 NASA Public Service Medal and the First NASA Ames Information Directorate Infusion Award also in 2002. He is the Co-chair of the 2005 Intnl. Conference on Automated Planning and Scheduling (ICAPS), to be held in Monterey California (http://icaps05.icaps-conference.org/) and the chair of the Executive Board of the International Workshop on Planning and Scheduling for Space (http://www.congrex.nl/04c06/).
Our work is centered around a series of Focus Areas that we believe are the future of science and technology.
We’re continually developing new technologies, many of which are available for Commercialization.