Research Project Title:
Machine Learning Methods for Predictive Mapping of Sea Floor Features
abstract:The goal of this project will be to develop, implement, and evaluate machine learning methods for fast prediction of sea floor feature maps from sparse data collected by an Autonomous Underwater Vehicle (AUV). Although current approaches to this task apply Gaussian Process models to produce accurate results, they typically require a large amount of processing power, making it infeasible for an AUV to adapt its path during a data collection mission without returning to the surface and taking advantage of an external processing device. My aim is to apply modern machine learning techniques to perform the task of habitat map prediction with acceptable accuracy under the speed and memory consumption constraints of an AUV’s on-board computer.
I am participating in SuperUROP because I want to gain experience doing scientific research in robotics. I hope to gain a better understanding of predictive and model-based algorithms and techniques, to learn how to develop, evaluate, and express my ideas, and to contribute towards the research goals of my group.