Invited Talks

Larry Matthies
Jet Propulsion Laboratory
University of Southern California

http://www-robotics.jpl.nasa.gov/people/Larry_Matthies/

Biography: Larry Matthies obtained a PhD in Computer Science from Carnegie Mellon University in 1989, then moved to the Jet Propulsion Laboratory, where he has supervised the Computer Vision Group since 1997. His group developed the stereo vision and visual odometry algorithms used by the Mars Exploration Rovers (MER), the Descent Image Motion Estimation System used for terrain-relative velocity estimation by the MER landers, and algorithms for detecting dust devils and clouds onboard the MER rovers. He initiated several threads of research on vision algorithms for safe and precise landing on Mars and for landmark-based orbit estimation around asteroids, comets, and moons. He has also been very active in vision system research for defense robotics for three decades. He is an Adjunct Professor of Computer Science at the University of Southern California and a member of the editorial boards of Autonomous Robots and the Journal of Field Robotics.

Robotic Solar System Exploration: Progress and Challenges

Robotic solar system exploration is difficult for many reasons, including temperature extremes, radiation, limited power and energy budgets, very limited computational power, limited communication with the spacecraft, and the need for ultra-reliable operation over many years. Nevertheless, there have been several notable successes in robotic exploration of Mars, including semi-autonomous rover navigation in the Mars Pathfinder and Mars Exploration Rover (MER) missions, a vision system for terrain-relative velocity estimation for MER landers, and autonomous science modules onboard the MER rovers to detect rare events like dust devils and clouds. Technology development is fairly advanced for more sophisticated vision and guidance systems for landers for Mars and the Moon, including precision landing and landing hazard avoidance capabilities. Closely-related visual navigation algorithms have already been used to support sampling an asteroid and are undergoing further development for comet sampling. Novel robotic mobility systems are in development for our Moon (wheels-on-legs) and for Titan and Venus (balloons and airships). In these and other missions, the key problems for robotic systems are often different from the problems that most occupy the international robotics research community. I will survey the missions, core problems, progress to date, and major open challenges in these areas.