Principal Investigator: Mingxi Zhou

Co-PI: Chengzhi Yuan (URI College of Engineering)

Funding Agency: Foundational Research in Robotics, National Science Foundation (NSF) The project is recommended for award and waiting for the final announcement

Project Summary: Complex underwater environments, such as underwater caves, flooded mines, offshore wind farms and under-ice waters have significant societal and science impacts. For example, surveying underwater cities and exploring underwater caves provide evidence about our history, and the health of underwater infrastructure is extremely important for reliable energy and the blue economy. While these complex environments are challenging and risky for human divers to perform repeated surveys to establish baseline measurements, underwater robots have been considered as a potential candidate for surveying these environments. This award supports fundamental research in underwater robot design, control, and perception to overcome the current challenges that will be encountered when operating underwater robots in these complex environments. Specifically, a highly maneuverable underwater robot will be designed, constructed and tested for moving in a tight complex space. The robot will be made accessible and customizable aiming to grow and broaden the underwater robotics research community. Moreover, a learning-based control system will be developed to detect, identify and respond to the uncertain influences posed by the environment disturbance for an improved robot motion control performance. Lastly, a new sonar-based environmental reconstruction algorithm will be designed and validated to aid robot localization, navigation, and path planning in the challenging underwater environments. Experiments in simulation, indoor tanks and outdoor environment will be conducted to evaluate the methods. 

The overarching goal of the project is to increase the functionality and reliability of underwater robots in complex environments. To this end, the project will make contributions to the fundamental research in robotics in three aspects. First, the novel hydrobatic (a term derived from “hydro and acrobatic) robot will advance the state-of-the-art underwater robot design, allowing it to perform maneuvers (turn-in-place, hovering, and vertical descent) that are impossible for existing underwater robots. Second, a novel adaptive neural-network learning control scheme will be designed to enable the robot to efficiently track a desired trajectory precisely under dynamic uncertainty and time-varying environmental disturbances. Meanwhile, it will accurately identify/learn the nonlinear uncertain robot dynamics during online closed-loop control with provable guarantees in both control and learning performance. Finally, a learning-based terrain reconstruction algorithm will be created for wide aperture sonars. This new algorithm could overcome the elevation ambiguity problem which causes errors when reconstructing underwater environments from a single sonar image. The improved terrain reconstruction result is expected to advance other fundamental research in robot localization, navigation and path planning.