Energy-Efficient, Multi-Scale, Biologically-Inspired Mobile Sensor Networks with Real-Time Observation Adaptability

点击翻译按钮获取中文摘要

ECS-0501407Energy-Efficient, Multi-Scale, Biologically-Inspired Mobile Sensor Networks with Real-Time Observation AdaptabilityThis integrative systems proposal focuses on the cooperative propulsion of hydrodynamically-coupled biomimetic underwater vehicles. The superiority of the proposed platform hinges on the integration of flow sensing into the closed-loop control of vehicle-wake interactions. A fish-like robot will be constructed that responds in real time to ambient flows estimated using micro-machined hair cells distributed along the vehicle's sides. Computational multi-scale algorithms will be developed to enable the rapid extraction of relevant flow information from sensor data. Analytical and computational models will be derived to clarify the principles by which fish exploit ambient flows for efficient swimming, focusing on schooling behavior. These principles will be used to design feedback control laws for efficient robotic locomotion through unsteady flows. Fast approximate dynamic programming algorithms will then be developed for the redirection and reshaping of vehicle schools to maximize the collection of information from time-varying sources. The intellectual merit of the research proposed is in creating analytical and computational tools and technology that will enable schools of vehicles to exploit principles of unsteady flow to achieve energy efficient locomotion and maneuvering for adaptive sensing.The broad impact of the research is in providing the basis for autonomous mobile sensor arrays to be deployed over unprecedented distances for underwater applications ranging from environmental sampling to the collection of military intelligence. This project will also enhance the interdisciplinary engineering curriculum at UIUC and directly train one PhD student and numerous undergraduates.