Abstract:

A gimbaled camera can be used in vision-based target tracking and geolocation on small unmanned aerial vehicles (UAV). The effect of camera rotation and translation on target’s image movement on an image plane is analyzed, the transfer matrix between UAV’s body rates and gimbal attitude rates is inferred, a gimbal controller is designed which uses UAV’s velocity and body rates as feedback to compensate camera’s line-of-sight change, and the target’s position on the image plane is used to correct the gimbal angle error. Simulation results show that the controller can improve tracking precision and eliminate gimbal vibration.