Coordinated Robotics
Observing ocean processes is dominated by three paradigms: (1) satellite remote sensing, (2) in-situ measurements from ships; and (3) in-situ measurements from autonomous platforms such as gliders, drifters, autonomous underwater vehicles (AUVs), and moorings. Satellite remote sensing methods provide long-term measurements over large horizontal regions of the ocean but lack the spatial resolution and cadence of in-situ measurements. The cost, availability, and ability of vessels precludes using them for persistent observations especially in harsh remote areas, such as the Southern Ocean, where the greatest paucity of observations exist.
I research approaches that use mobile heterogeneous robotic platforms — i.e., robots with different but complementary range, endurance, sensing, and speed characteristics — in a coordinated fashion to provide long-term ocean observations. For example, gliders, long-range AUVs (LRAUVs), and autonomous surface vehicles (ASVs) could work cooperatively. Multiple gliders could survey large areas with basic sensing payloads and a LRAUV could carry more advanced sensors and, because of its additional speed, respond to events while ASVs obtain measurements, provide external navigation aiding to submerged platforms, serve as a communication relay mode, and, as the computing power of ASVs increases, coordinate subsea observation efforts. Data obtained with this fleet of robots could be communicated ashore for data assimilation, human interpretation, and subsequent definition of new observation goals for the fleet to achieve.
Coordinated, Long Duration, Robotics Missions
The high operating costs of the ships required to deploy, communicate with, and externally aid the navigation of deep-water AUV motivates short, rapid AUV missions rather than extended duration missions. An operations paradigm that reduces the use of ships would facilitate multi-vehicle operations and provide a mobile persistent observational presence: potentially transforming how we use AUVs. In pursuit of this goal, collaborators and I are developing methods for coordinated autonomous surface vehicle (ASV) and AUV operations (German et al., 2012). I co-led experiments in 2012 that showed we could use an ASV to simultaneously externally aid navigation and communicate with the AUV (Kinsey et al., 2013). In the short-term this allows us to leave AUVs unattended on 24-48 hour missions without compromising AUV navigation or telemetry; however, in the future, it will enable longer missions in which an ASV and fleet of AUVs autonomously survey large areas of the ocean. To achieve this, I plan to research a number of robotic science topics such as optimized trajectory planning of ASVs and AUVs, coordinated control, and autonomy. Results of this research will contribute to future research on coordinated operations with AUVs, ASVs, gliders, and satellites focused on obtaining measurements at differing resolutions and spatial scales over entire ocean basins.
In 2012, I collaborated with Mike Jakuba, Chris German, Dana Yoerger, and Lee Frietag to demonstrate the concept of using an ASV to tend a shallow water AUV.
Satellites to the Seafloor: Autonomous Science to Forge a Breakthrough in Quantifying the Global Ocean Carbon Budget
Cooperative robotic technologies are a crucial technology for understanding large scale oceanographic observation challenges such as the fate of carbon in the ocean. In collaboration with PIs at the California Institute of Technology and NASA’s Jet Propulsion Laboratory, I co-lead a two-part workshop in 2013-2014 that will bring together oceanographic, space, and robotics researchers to develop a roadmap for the robotics research necessary for achieving large scale, coordinated oceanographic surveying. More info on the workshop and its results are here.
