CAREER: A New Paradigm in Control and Coordination of Robot Teams in Geophysical Flows

University Of Pennsylvania, Philadelphia PA

Investigators

Abstract

Little prior work in unmanned underwater vehicles (UUVs) has addressed the tight coupling that is inherent between the fluid dynamics of the body of water and the vehicle itself. In fact, the fluid dynamics of large, open bodies of water can be quite complex and this proposal addresses large coherent structures that naturally occur in the flow structure of these large bodies of water. The goals of this project are to extend the PI's ONR YIP award into the realm of 3-D, expanding the mathematical and control framework for distributed autonomous sensing and tracking of geophysical fluid dynamics and to understand the long-term impact of geophysical fluid dynamics to improve the autonomy of underwater vehicles. The key idea exploits the capability of the team to cover large regions to increase the spatio-temporal sampling resolution of the flow field. The data will then be processed in a distributed fashion to obtain a global description of the flow dynamics that can be maintained and updated in real-time. The specific objectives that expand prior proposals include not only the 3-D modeling, but the development of an energy efficient stochastic pulse controller for tracking the ridges of the coherent structures and expanding the estimation of the structures through stochastic approaches that allow the fusing of larger teams of sensing entities. The intellectual merit of the proposed work stems from the synthesis of nonlinear dynamical systems theory, transport theory, and robotics to develop a modeling, control, and analysis framework for collaborative unmanned systems operating in dynamic and uncertain environments. The information gleaned from the coherent structures will be used to refine motion control and resource allocation strategies to determine minimum-effort stochastic control policies for long-term operation in GFD environments. To the PI's knowledge, this is the first attempt to use robots to track and map unstable coherent structures in the ocean, and to exploit knowledge of them to improve the autonomy of AUVs/ASVs. Broader Impact: Success of these endeavors will improve the forecast of weather-climate systems, underwater transport dynamics, and the modeling and prediction of various other physical phenomena in geophysical flow environments. Since the proposed methods are very general and developed for continued operation in dynamic and uncertain environments, success of the proposed activities will likely increase the maneuverability and energy-efficiency of existing AUVs/ASVs; enable teams of AUVs/ASVs to continuously adapt to changing environmental conditions as they execute their assigned tasks; and provide greater situational awareness for various scientific, commercial, and military applications in the ocean. In addition, the proposed educational activities include a comprehensive plan to integrate the study of GFD into robotics through online educational modules for general K-12 audiences; an interdisciplinary undergraduate and graduate curriculum; and contributions to the robotics community in the form of open source software and hardware development tools.