Seminar On Advancing Biological Oceanography with Inference-based Robotic Exploration of the Coastal Ocean

11 November 2013

Kanna Rajan, Monterey Bay Aquarium Research Institute (MBARI), US

Persistent under-sampling of complex coastal ocean processes has resulted in calls for new
methods to approach the sampling problem. Autonomous Underwater Vehicles (AUVs) have been
used over the last few years to observe dynamic events such as blooms, anoxic zones and ocean
fronts. However they had till recently, been stymied with the use of simple reactive approaches
which depended on a priori “plans” which prevented any substantiative adaptation of mission
I will motivate the use of Artificial Intelligence based Planning/Execution using generative planning
techniques in-situ for such robots and highlight the range and diversity of applications which have
been impacted to enable novel observations. The talk will explore how the Teleo-Reactive EXecutive
(T-REX) has impacted MBARI collaborative scientific missions used within a very inter-disciplinary
environment at the institute.
T-REX is an open source software framework which implements a variation of the sense-plan-act
paradigm by using a partitioned approach to onboard problem solving synthesizing and executing
plans onboard. The system is general enough to be used on a terrestrial platform for a personal robot
(Silicon Valley’s WillowGarage PR2) as well as on a planetary rover testbed at the European Space
Agency and dual use for civil, military and homeland security applications. T-REX is the only
operational AI system in use on an AUV anywhere and has a rich legacy from two NASA space
missions, including the 2004 Mars Exploration Rovers mission which is still ongoing.
In this talk I will highlight the overall architecture, its divide-and-conquer strategy to solve complex
mission planning and synthesis problems and the use of Machine Learning techniques which inform
Sampling and adaptation of robotic vehicles in the water-column. I will endeavor to articulate how
biological oceanography in particular can reap benefits of such advanced capabilities derived from
decades of work in Artificial Intelligence and Robotics and do with with actual examples of upper
water-column sampling primarily in Monterey Bay, California, the densest patch of the ocean studied
any where in the world.

Kanna is the Principal Researcher in Autonomy at the Monterey Bay Aquarium Research Institute a
privately funded non-profit Oceanographic institute which he joined in October 2005. Prior to
that he was a Senior Research Scientist for the Autonomous Systems and Robotics Area at NASA
Ames Research Center Moffett Field, California. At MBARI he heads the only AI group in an
operational oceanography setting anywhere.
At Ames, he balanced programmatic and technical responsibilities. He was the Principal Investigator
of the MAPGEN Mixed-Initiative Planning effort to command and control the Spirit and
Opportunity Mars Exploration rovers on the surface of the Red Planet. MAPGEN continues to be
used to this day, twice daily in the mission-critical uplink process at the Jet Propulsion Laboratory in
Pasadena. Kanna was one of the six principals of the Remote Agent Experiment (RAX) team, which
designed, built, tested and flew the first closed-loop AI based control system on a spacecraft. RAX
was the co-winner of NASA's 1999 Software of the Year, the agency's highest technical
award (
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, JPL
awarded him the NASA Exceptional Service Medal for his role on the Mars Exploration Rovers
mission. He was the Co-chair of the 2005 International Conference on Automated Planning and
Scheduling (ICAPS), Monterey California and till recently the chair of the Executive Board of the
International Workshop on Planning and Scheduling for Space. He continues to serve on review
panels for NSF, NASA, the Italian Space Agency and European Space Agency in addition to a
number of academic conferences and journals in AI and Robotics.
His academic interests are in automated Planning/Scheduling, modeling and representation for real
world planners and agent architectures for Distributed Control applications. Prior to joining NASA,
he was in the doctoral program at the NYU Courant Institute of Math Sciences. 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