MBARI is developing onboard intelligence for autonomous underwater vehicles (AUVs) and autonomous surface vehicles (ASVs) to recognize and react to oceanic features of interest by using onboard sensors, water sample acquisition and analysis systems, as well as input from remote sensing and collaborating robots.

In the vast ocean, many ecologically important processes occur episodically in time and are localized in space. We therefore need to develop methods to direct limited and precious resources to the most relevant and informative locations to reliably and repeatedly achieve observation and sampling at the appropriate place and time. Traditional ship-based methods for detecting and sampling dynamic ocean features, such as fronts, are often laborious and very difficult, and long-term tracking of such features is practically impossible.

In this research, we design onboard intelligence for autonomous underwater vehicles (AUVs) and autonomous surface vehicles (ASVs) to recognize oceanic features of interest and accordingly adapt behavior to optimally characterize the feature by using onboard sensors, water sample acquisition and analysis systems, as well as input from remote sensing and collaborating robots.

These intelligent robots enable targeted sampling—quickly detecting particular ocean features and intensively sampling them as dictated by the science question at hand. A targeted sampling capability allows researchers to direct a limited number of marine assets to find features of interest and collect desired measurements/samples absent a human presence, thus significantly enhancing accuracy and efficiency of ocean studies. We seek to maximize targeted sampling gain by emphasizing collaboration between robots.

Current Projects

Adaptive Sampling by Autonomous Vehicles