Adaptive Control for Autonomous Underwater Vehicles
Speaker: Kanna Rajan, Principal Researcher for Autonomy Monterey Bay Aquarium Research Institute Moss Landing, California
Date: 15 July 2008 Time: 11:00-13:00
Location: "Mediterranean Studies" Seminar Room, FORTH, Heraklion, Crete
Host: Yannis Tsamardinos


With global climate change in the news, there is an increasing focus of robotics research in the ocean sciences where deep ocean robots are few and substantially less capable than those built for terrestrial and space research. Energy and communications (or lack there of) are two drivers of how robotic devices in the sea are challenged compared to their terrestrial and space cousins. These constraints coupled with lack of detailed understanding of ocean processes within the water column as well as the benthos, have made ocean exploration in its harsh environment an extremely challenging domain for robotics research. Yet recent shifts in ocean exploration with a move from the Expeditionary to the Observatory mode of doing science have only pushed for the use of robotic platforms with substantial onboard intelligence so as to enable a cost-effective way to observe, characterize, map and sample what lies under 70% of the Earths surface.

At the Monterey Bay Aquarium Research Institute (MBARI), a private non-profit ocean science inter-disciplinary institution, Autonomous Underwater Vehicles (AUVs) untethered mobile robots, are used for routine scientific missions for water-column time series measurements as well as for generating bathymetric relief maps of mid-ocean ridges. These vehicles carry advanced scientific payloads yet are commanded using pre-scripted plans by highly skilled mission operators. Such a mode of operation precludes these very capable vehicles from being adaptive to observe and sample dynamic episodic events in our oceans while relying on such error prone modes of operation. The Autonomous Systems group at MBARI is building an onboard advanced adaptive control system for our AUVs that integrates Artificial Intelligence (AI) Planning and Probabilistic State Estimation in a goal-directed hybrid executive, enabling scientists to detect, survey and sample events of an unpredictable nature. This effort is informed from our decades worth of experience building autonomous systems for NASA including the Remote Agent (1999) the first AI based closed-loop control system to command a spacecraft (NASA's Deep Space One), when it was 65 Million miles from Earth and MAPGEN the first AI based system to command a vehicle on the surface of another planet (Mars) as part of the Mars Exploration Rovers mission.

I will discuss the continuum of our efforts in autonomy and attempt to weave together the challenges and opportunities in this emergent field of intelligent robotics in our deep oceans.


Kanna Rajan is the Principal Researcher in Autonomy at the Monterey Bay Aquarium Research Institute (MBARI) (http://www.mbari.org) a small privately funded Oceanographic institute. At MBARI, where he leads the Autonomous Systems Group, his current interests are focused in fully autonomous operations of unmanned Autonomous Underwater Vehicles (AUV's) for ocean observation and sampling. His current research is focused on adaptive sampling and control for deep ocean exploration.

Prior to joining MBARI in 2005, he was a Senior Research Scientist and a member of the management team of the the 95 member Autonomous Systems and Robotics Area at NASA Ames Research Center Moffett Field, California. As the Program Manager for Autonomy & Robotics for a $5M FY05 program at Ames he was tasked with putting together a credible demonstration of Human/Robotic collaboration on a simulated planetary surface. The field demonstration at the Ames Marscape in September 2005 end, showcased how autonomous systems and EVA astronauts could "work" together towards exploration tasks. Before this programmatic role, he was the Principal Investigator on the MAPGEN Mixed-Initiative Planning effort to command the Spirit and Opportunity rovers on the surface of the Red Planet. MAPGEN continues to be used to this day, twice daily in the mission-critical uplink process. Kanna was one of the six principals of the Remote Agent Experiment (RAX) which designed, built, tested and flew the first closed-loop AI based control system on a spacecraft. The RAX 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/).

Prior to joining NASA Ames, he was in the doctoral program at the Courant Institute of Math Sciences at New York University. Prior to that he was at the Knowledge Systems group at American Airlines (the largest US airline), helping build a Maintenance Routing scheduler (MOCA) which continues to be used by the airline 365 days of the year.

MAPGEN has been awarded NASA's 2004 Turning Goals into Reality award under the Administrators Award category, a NASA Space Act Award, a NASA Group Achievement Award and a NASA Ames Honor Award. Kanna is the recipient of the 2002 NASA Public Service Medal and the First NASA Ames Information Directorate Infusion Award also in 2002. In Oct 2004, Jet Propulsion Laboratory awarded him the NASA Exceptional Service Medal for his role on the Mars Exploration Rovers misson.

He was the Co-chair of the 2005 Intnl. Conference on Automated Planning and Scheduling (ICAPS), Monterey California (http://icaps05.icaps-conference.org/) and till recently the chair of the Executive Board of the International Workshop on Planning and Scheduling for Space. He continues to serve in review boards for NASA, the Italian Space Agency and the European Space Agency. As part of his academic collaboration, he advices PhD students at MIT, Carnegie Mellon, Strathclyde (UK), Birmingham (UK) universities and at LAAS (Toulouse, France).

© Copyright 2007 FOUNDATION FOR RESEARCH & TECHNOLOGY - HELLAS, All rights reserved.