|
|
Department of Aerospace Engineering Phone: (301) 405-3282 | E-mail: kdimble (at) umd.edu Research Interests: Electrostatic analysis of underwater objects, fish dynamics and motion modeling, application of boundary element methods to underwater electric fields, wide-field integration processing, control oriented robotic implementation of electrolocation Dissertation Title: Electroreception for Autonomous Underwater Guidance, Navigation and Control Abstract:
|
J. Sean Humbert, Ph.D. |
Kedar D. Dimble
Viable strategies for extracting and fusing uncertain observations of the underwater environment are exemplified by biological systems that use optical and electrical sensing. Thousands of years of evolution has provided some species of weakly electric fish to perform target detection, obstacle avoidance and prey capture by using self-generated electric fields as the principal sensing modality. These fish are capable of detecting tiny subtle perturbations to the fields to extract the distance, location, size, electrical properties and shape of the perturbing object. The perturbations are visible to the fish in the form of Gaussian-like peaks, commonly termed as electric images.
In this project, we investigate the opportunity to apply bio-inspired sensor processing to the problem of low-cost absolute and relative navigation in complex environments. Electro-sensing will be used as the principle sensing modality to obtain the electric images on an underwater vehicle. Wide-field integration processing of these images will be implemented to extract the necessary motion cues for autonomous navigation. The entire sensory apparatus and control algorithm will be designed to be computationally cheaper and easily implementable.
©2010 University of Maryland