An auto-piloted drone which observes objects as they move, be it people or cars. The test case scenario with which we begin will be tracking cars on race circuits. This system enables the shooting of a whole circuit continuously and automatically, without the need for professional drone handlers.
The drone works by utilizing 3 key systems of technology: GPS, iBeacon and Computer vision. Computer vision technology is based upon a neural network, trained on more than 150k worth of pictures. The network runs on a Nvidia Jetson video graphic adapter, the world’s leading visual computing platform for GPU-accelerated parallel processing in the mobile embedded systems market.
Taking into account that the Computer vision component is the most important part of the system, there is a licensed PhD specialist in computer vision and neural networks, working on this project. Our business partners in this endeavor are the owners of the largest race track circuit in Romania and national racecar champions.