AUTONOMOUS HYBRID AERIAL-TERRESTRIAL TRANSPORT AND RETRIEVAL FOR IN-SITU COLLECTION

Abstract

This thesis investigates the use of hybrid vehicles, capable of both aerial and terrestrial motion, for sample search and retrieval, enabling autonomy through the customization of existing software for new applications. This research addresses challenges related to autonomous detection of, and navigation with respect to, fiducial markers - in both the aerial and terrestrial domains. A key contribution of this work is that of the hybrid mobility system, which is capable of transitioning autonomously between flying and driving. To assist in sample detection, an existing software package for fiducial marker recognition is utilized and adapted for localization of the target and positioning of the vehicle for sample collection. For sample collection, a simple gripper is designed to grasp the target object before depositing at a home location. This research leverages software-in-the-loop simulation for validating code performance prior to deployment on the actual vehicle, which is tested in a controlled, indoor laboratory environment.