Year: 2023
Type: dissertation
Abstract: This PhD thesis explores how to make robots and autonomous systems smarter and more capable in uncharted spaces. The author combines techniques from the established field of Simultaneous Localization and Mapping (SLAM) with advancements in Machine Learning and Object Detection. The focus is on creating semantically meaningful maps without relying on pre-known information about the environment. The research delves into the effects of partially observed objects and evaluates solutions for this issue. Lastly, the work illustrates a method for identifying and tracking objects based on visual data, making the overall process more efficient and applicable to real-world scenarios.
Author: Lachlan Nicholson
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
Award: G252308265
OpenAlex ID: W4386546345