The Autonomous Flocking u-Sub (AFuS) Project

KARAUL, JACOB The Autonomous Flocking µ-Sub (AFµS) Project. Department of Computer Engineering, June 2020. ADVISOR: Walter Dixon Underwater data collection is a field that concerns the tracking, monitoring, or gathering of underwater data and has significant financial, social, and environmental implications. Technologies used to attain this data are vast, but suffer from key shortcomings that prevent their adoption at mass in society. Autonomous Underwater Vehicles (AUV’s) provide a cheap and autonomous option, however current industry products suffer either from a high price point or lack in ease of use (i.e. meters long and several tons heavy). Recently, AUV’s have been emerging on the micro-scale (a few feet in length), easily operable by one person. This project is the Autonomous Flocking µ-Sub project, or AFµS project. The goal of this project is to develop a micro-scale AUV at a cost of no more than $500, thus both handlable and affordable by a much larger audience. Flocking capabilities will increase the applicability of the product, and is achievable by utilizing a low latency optical communication system. Sonar is used for obstacle avoidance, and a combination of inertial measurement and global positioning allows for navigation and exploration. Results of this project show that decent performance in communication, locomotion and obstacle avoidance can still be achieved with off-the-shelf electrical components, however dead-reckoning using an off-the-shelf inertial measurement unit is not feasible due to unbounded error accumulation