Smoothing framework for automatic track initiation in clutter

Automatic track initiation in clutter within the smoothing framework is considered in this paper. We derive a new smoothing algorithm based on integrated probabilistic data association (IPDA) filter and augmented state smoothing framework. We demonstrate the benefits of smoothing based approaches in multi-target tracking in clutter scenarios using tracker performance measures like true/false track discrimination statistics, track initiation/termination times and RMS errors of confirmed true tracks.