Relative Position Estimation Using Monocular Camera and Ultra-Wide Band Ranging

This paper presents a methodology for estimating the relative position between a micro aerial vehicle (MAV) and a base-station using a monocular camera and an ultra-wide band (UWB) ranging module. The pinhole camera model is used to derive a relationship that relates the 3D position of the vehicle to its location in the image frame and the estimated range. A tracking algorithm was implemented to track the location of a light emitting diode (LED) array mounted to the MAV with a low cost webcam. The range was determined by using a two-way ranging algorithm with the UWB modules. The presented methodology is applied to experimental data collected from a quad-rotor. The results are compared to a motion capture system, and it is shown that this method is able to track the position of the MAV to within the range accuracy of the UWB modules. This relative position estimation methodology has the potential to be a viable component of a navigation system for cooperative vehicles.