Drone-Based Aerial Security System with 5G and AI Edge Computing
NSERC / IDEaS-funded project (C$1.54M): flight control and energy-optimal path planning for drones carrying 5G base stations, genetic-algorithm area surveillance, and YOLOv8 detection of ground intruders.
Period: 2023/05 – 2026/03 · University of Toronto Institute for Aerospace Studies
- The project was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Canada’s Innovation for Defence Excellence and Security (IDEaS) program, with total funding of C$1.54 million (approx. RMB 7.5 million), conducting innovative theoretical and engineering research for intelligent air transportation, aerial security, and the 5G Internet of Things in Canada.
- Proposed flight control algorithms and energy-optimal path planning methods for drones carrying onboard 5G base stations; the results were published at the WF-IoT 2023 and WF-IoT 2024 international conferences: “Robust Stabilization of a Quadrotor as a 5G Access Point Using Acceleration Feedback” and “Energy Aware Planning of A Quadrotor With 5G Access Point in Disaster Response”.
- Proposed a genetic-algorithm-based flight path planning algorithm for drone scanning surveillance of designated areas, achieving the maximum scanned area with the minimum battery consumption. The result was published in Science China Information Sciences, a CAS Zone 1 top journal: “A path planning algorithm for a crop monitoring fixed-wing unmanned aerial system” (2024).
- Proposed a YOLOv8-based algorithm for detecting ground intruders / illegal drone operators and planning flight trajectories; combined with “Autonomous Drone Operator Localization Using UAVs With Reinforcement Learning” (IEEE Smart World Congress 2025), achieved automatic detection of ground intruders on a quadrotor flight platform.