- This event has passed.
AggieSat Laboratory Autonomous Rovers Showcase
April 3, 2024 @ 4:00 pm – 5:00 pm
AggieSat Laboratory is a student-run space program that trains students in systems engineering through hands-on experience in the design, building, testing, and operation of space systems. Two projects will discuss how their autonomous vehicles will help explore other planets. Join us to see what new adventures await you.
Location: Blocker 129 (Inside Computing Lab) & Zoom (registration link)
Time: April 3 at 4:00 pm – 5:00 pm
Autonomous Rover Navigation using LiDAR and Position Sensors
Rovers on other planets need to be capable of operating autonomously since communication delay makes direct human control impractical. To autonomously navigate, the rover must know what the environment looks like, and where the rover itself lies in that environment. AggieSat Laboratory’s AggieRover1 (AR1) rover uses Light Detection Ranging and Light (LiDAR), which measures the time-of-flight of reflected light to gauge distances, to scan the environment. While moving, the rover utilizes an inertial measurement unit and motor encoders to track its location. Together, these allow the rover to move freely and autonomously towards its destination.
Presentation By: George Mersiowsky, Amanda Macha
Image Segmentation and Stereo Vision for Autonomous Planetary Vehicles
Vehicles operating on the surfaces of other planets need to do so autonomously so that communication delays and disruptions do not interfere with their missions. To be autonomous, these vehicles must be able to sense and analyze the surrounding terrain, identify obstacles, and path-find to their objectives. One approach, which is being implemented by AggieSat Laboratory’s Geographical Radio Guided Exploration (GeoRGE) rover, uses convolutional neural networks such as UNet to segment recorded images and stereo vision to measure distances to obstacles. Once obstacles are identified and a viable path is found, the vehicle can traverse the path and repeat the sense-analyze-identify-path process.
Presentation By: Arjun Madana, Austin Salmon