CAREER: Autonomous underwater vehicle-manipulator robots adaptable to the unforeseen

Marine robots can be central tools in reaching and exploring the depths of the ocean, but due to the high costs associated with these technologies, scientists have limited access to such systems. Beyond their use in observing the environment, underwater robots can be used in marine resource exploration or maintenance and construction of marine engineering structures, through the integration of underwater vehicles with robotic manipulators. Lightweight underwater vehicle-manipulator systems (UVMS) are characterized by small-sized vehicles and electric manipulators, with the weight of the manipulator comparable with the weight of the vehicle. These types of underwater robots are essential to lowering the barriers of entry to ocean exploration and science, representing a radical change from the current marine research and intervention practices. The main challenges of lightweight UVMS that hinder their widespread applicability are: 1) the stability issues encountered due to the interaction (coupling) effects between the manipulator and vehicle, and also due to the contact with the environment; 2) the uncertainties in the hydrodynamics; and 3) their lack of robustness due to low underwater visibility or unforeseen events caused by environmental conditions. The overarching goal of this CAREER project is to design autonomous UVMS capabilities robust and reliable to unforeseen events observed by Dynamic Vision Sensors (DVS), advancing the autonomy capabilities of lightweight UVMS and proving their value for ocean exploration. The work proposed in this project aims to showcase that: 1) marine multi-body robotic systems, like lightweight underwater-vehicle manipulator systems that operate in uncertain environments, are robust, reliable, and efficient in the context of unforeseen events if event- and model-based reactive control strategies are created; and 2) marine robotics concepts can be used to build self-esteem of students from low-income socio-economic backgrounds. To achieve these, the project will: (O.1) Create novel physics-based and data-driven dynamic models with uncertainty quantification using DVS to analyze lightweight multi-body systems in multi-scale ocean flows under uncertain conditions; (O.2) Establish new probabilistic model-based control theories for lightweight UVMS governed by non-vanishing uncertainties, unknown disturbances, and multi-physics characteristics; and (O.3) Develop education and outreach activities, centered around underwater robotics and specifically UVMS, for college students, low-income K-12 students, and the general public to demonstrate the impact marine robotics can have in their lives. This project is jointly funded by the Electrical, Communications and Cyber Systems Division (ECCS) and the Established Program to Stimulate Competitive Research (EPSCoR). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.