Memorial University of Newfoundland

Published purpose

Vertical take-off and landing (VTOL) aircraft are often used for transporting goods to and from inaccessible areas, surveillance, ground support, etc. which can be enhanced and de-risked by increasing the autonomous capabilities of the vehicles. The objective of this research work is to develop an Artificial Intelligence (AI) powered multi-sensor based navigation module, which can be integrated into VTOL vehicles in real-time to improve its autonomous capabilities and to reduce pilot workload. The system requires a combination of LiDAR, vision, and IMU data for generating map and location information. The particular sensor suite enables autonomous operation independent of GPS availability for the safety-critical take-off and last-mile goods delivery portions of a VTOL mission. The recent advances in deep learning research and AI computing power will be used to develop a real-time capable hardware-implemented solution that enhances the robustness and accuracy of mapping and localization modules of the proposed navigation system.