Ethan Chan (V), Keira Chen (VI), Matt Dicks (V), Connor Francis (V), James Kotsen (V), Nick Meng (VI), Alexander Recce (V), Victoria Xie (IV)
Project Summary
Drover is a low-cost integrated system of robots composed of an unmanned aerial vehicle (UAV, or drone) and unmanned ground vehicle (UGV, or rover) that autonomously cooperate and pathfind. Our rover is able to follow drone-calculated paths with centimeter accuracy.
In the past, we focused on spotting small fires with aerial photos from the drone for wildfire prevention. We used techniques like correcting image distortion, pathfinding algorithms, and mapping pixels to GPS coordinates to help the drone and rover navigate autonomously and as efficiently as possible.
Recently, we’ve been using photogrammetry, which is a method of turning images into a 3d model. By analyzing 3D models of roofs, we can estimate the annual solar power output with the roof’s gps coordinates, slant, and other geometry that we can only get by getting a detailed scan with a drone.
Our autonomous system is safe and requires less manpower than conventional methods; drones are less expensive than large aerial vehicles, more accurate than satellite imagery, and can provide real time data.
Our members have written two papers detailing our methods:
UAV and UGV Autonomous Cooperation for Wildfire Hotspot Surveillance: https://ieeexplore.ieee.org/document/10002208 Autonomous Roof Solar Potential Estimation Using UAV Photogrammetry: awaiting publication
Project Goals
- Rover navigation using drone photogrammetry (applications include trash pickup)
- Long Range Rover Driving, LIDAR
- Expand applications for drover system
Faculty Advisor
Dr. Jolly
