However, deep-sea missions offer an excellent environment. Like space missions, deep-sea missions are often science driven, have severe communication constraints, and operate within a highly uncertain environment. The cost of unmanned deep-sea vehicles, however, are priced at the level of hundreds of thousands to millions of dollars, rather than hundreds of millions, are easier to deploy, and are easier to retrieve upon failure. To cross the technology validation gap, a team comprising MIT, Caltech, JPL, and the Woods Hole Oceanographic Institution (WHOI) is conducting a two-year KISS-funded study to demonstrate risk-aware execution (MIT, JPL) combined with correctness-by-construction controller synthesis (Caltech), applied to a simulated rover and a fielded deep-sea mission. In March 2015, the team tested goal-directed execution using Enterprise during a research cruise off the northwest coast of Australia, supported by Schmidt Ocean Institute. Over three weeks, the MIT MERS team, along with groups from WHOI, the Australian Center for Field Robotics, the University of Rhode Island, and elsewhere, tested several classes of AUVs, and their ability to work cooperatively to map the ocean environment. Enterprise was tested on an autonomous underwater glider, in collaboration with Rich Camilli at WHOI, and demonstrated that the glider could operate safely among a number of other autonomous vehicles, while receiving high-level goals. The glider, guided by Enterprise, was able to adapt its mission plan to avoid getting in the way of other vehicles, while still achieving its most important scientific objectives. If another vehicle was taking longer than expected to explore a particular area, Enterprise would help the glider reshuffle its priorities, and, for example, choose to stay in its current location longer, in order to avoid potential collisions. The team will soon test Enterpriseâ€™s risk-aware capabilities on a deep-sea glider off the coast of Santa Barbara, California, and the MIT-WHOI team is working with Exxon to extend risk-aware execution to the coordination of teams of exploration vehicles.
Goal-directed risk-aware autonomous explorers
Published on Nov 18, 2016
Annual magazine review of MIT Aeronautics and Astronautics Department research and educational initiatives.