2026-05-21T08:10:03Zhttps://keep.lib.asu.edu/oai/request

oai:keep.lib.asu.edu:node-2013102025-05-06T22:43:47Zoai_pmh:repo_items

201310 https://hdl.handle.net/2286/R.2.N.201310 http://rightsstatements.org/vocab/InC/1.0/ All Rights Reserved 2025 2025-11-01T17:47:00 80 pages Masters Thesis Academic theses en Lok, Johnathan Berman, Spring Tsakalis, Konstantinos Nguyen, Duong Lewis, John Arizona State University Partial requirement for: M.S., Arizona State University, 2025 Field of study: Electrical Engineering Every year, thousands of lives are lost globally due to delayed emergency response in floods, boating accidents, and even rip tide incidents at public beaches. Traditional search-and-rescue operations are slow, human-dependent, and often ineffective in low-visibility conditions. With the aim of preventing these oceanic drownings, this thesis presents a multi-phase control strategy that guides a swarm of torpedo-like underwater robots to rapidly search for and cooperatively rescue an overboard victim in complex environments that may contain convex and concave obstacles. In the search phases of the robots’ mission, potential field-based controllers are employed to guide the robots to a uniformly-spaced circular formation around the victim’s last known location, and then the robots execute an Archimedean spiral search pattern until the victim is detected. The Archimedean spiral is demonstrated to have a substantially higher detection rate than other deterministic search methods and random-walk searches in Monte Carlo simulations, and it is shown through geometric analysis to be the most efficient 2D search pattern for finding a survivor from their last known location. While conducting the spiral search, the robots avoid obstacles using a novel motion planning algorithm that incorporates dynamic identification of the widest navigable space (“best gap”) between obstacles, a perimeter-following bug algorithm, and attraction-repulsion control inputs. All phases of the proposed search-and-rescue control strategy are validated in two-dimensional simulations and experiments with real robots on the Robotarium, a remotely-accessible swarm robotics testbed. Robotics Aerospace Engineering Electrical Engineering drone search mine sweeping Rescue Robotics Search and Rescue swarm Design of a Multi-Phase Controller for Search-and-Rescue Operations by Swarms of Autonomous Underwater Vehicles