Project number
16026
Organization
UA Department of Electrical and Computer Engineering
Academic year
2016-2017
The team’s goal is to develop a computer vision system that can be mounted on an unmanned aerial vehicle, or UAV, to detect, localize and classify ground targets. The UAV will compete in the Association of Unmanned Vehicle Systems International Student Unmanned Aerial Systems competition and be required to search for objects and identify attributes such as shape, color and letter.
This is often done by an operator monitoring an image feed, which limits the versatility and scalability of the system. The designed system eliminates the human operator and automates the imaging while minimizing hardware requirements through use of a hybrid onboard-offboard processing architecture.
The onboard system uses a single-board computer to detect objects, which reduces the communication bandwidth required, and the offboard ground-based system uses machine-learning algorithms such as convolutional neural networks to classify received images.
This is often done by an operator monitoring an image feed, which limits the versatility and scalability of the system. The designed system eliminates the human operator and automates the imaging while minimizing hardware requirements through use of a hybrid onboard-offboard processing architecture.
The onboard system uses a single-board computer to detect objects, which reduces the communication bandwidth required, and the offboard ground-based system uses machine-learning algorithms such as convolutional neural networks to classify received images.