Data association in clutter with an adaptive filter

This paper presents an adaptive filter for tracking targets in clutter. The filter employs a scale factor, which accounts for the target unpredictability at any time as estimated front the available data. The adaptive approach, in which the gain is adapted according to changing target dynamics, is used in conjunction with a widely accepted data association routine called probabilistic data association (PDA) to form the adaptive probabilistic data association filter (APDAF). Performance comparison between the PDA and APDA filters is demonstrated through simulations.