Target Identification and Tracking

Abstract

In this chapter we first recall the notion of contracted GPTs. Then we show that the CGPTs have some nice properties, such as simple rotation and translation formulas, simple relation with shape symmetry, etc. More importantly, we derive new invariants for the CGPTs. Based on those invariants, we develop a dictionary matching algorithm. We suppose that the unknown shape of the target is an exact copy of some element from the dictionary, up to a rigid transform and dilatation. Using the invariants, we identify the target in the dictionary with a low computational cost. We also apply the Extended Kalman Filter to track both the location and the orientation of a mobile target from MSR data.