2012 Poster Sessions : Development of a Modular Unmanned Underwater Robotic System / Design and Feasibility Study on Cooperative Navigation

Student Name : Eduardo Moreno
Advisor : Kenneth Salisbury
Research Areas: Artificial Intelligence
The development of cooperative multivehicle systems has received much attention over recent years, motivated by the fact that multiple robots working together as a team can distribute their resources to perform surveillance and rescue operations more efficiently and robustly than a single robot. Precise navigation of a group of underwater vehicles is a challenging task. In order to make cooperative robotic research and exploration more accessible to the scientific community, we are proposing a new kind of underwater robotic system: Autonomous Cooperative Underwater Robotic Vehicle (A-CURV), which is a modular underwater system of robots that is capable of self-navigation with a first level user interface. This system incorporates a centralized and decentralized control system. In this design, a center platform is used as the system's precision navigation node and several vehicles equipped with lower precision navigation sensors are tethered to the platform. Studies have shown that a centralized system will increase the system's precision when tackling complex tasks and a decentralized control system will improve the system's robustness. Using theories in cooperative modular robotics, an optimization process was formalized to analyze current technology investment strategies that would allow a robust and implementable underwater robotic design.

Eduardo Moreno is a soon to begin his PhD in Mechanical Engineering. He recently received his bachelors degree in Mechanical Engineering from the University of Arizona and is currently a visiting researcher at Stanford in Dr. Kenneth Salisbury Bio Robotics Laboratory. His past research projects include: the mechanical design and test of a robotic cheetah hind limb ( DARPA sponsored project) Advisor: Dr. Anthony Lewis, series of underwater robotic and control projects Advisers: Dr. Ricardo Sanfelice and Dr. John Nafziger, and the design and patent of an active orthotic device for flatfoot. His research interest lie in applied feedback control, novel dynamical systems, sensing with limited localization, mobile sensor networks, human gait analysis, and human-robot interaction. His ultimate goal is to initiate the research and development of inexpensive lower limb robotic prosthetics for developing countries.