A constrained optimal data association for multiple target tracking

One of the major problems in multiple target tracking is to obtain an accurate association between targets and noisy measurements. We introduce a new scheme, called constrained optimal data association (CODA), that finds the optimal data association by a MAP estimation method and uses a new energy function. In this scheme, the natural constraints between targets and measurements are defined so that they may contain missed detection and false alarm errors. Most current algorithms involve many heuristic adjustments of the parameters. Instead, this paper suggests an adaptive mechanism for such parameter updating. In this manner, the system automatically adapts to the clutter environment as it continuously changes in time and space, resulting in better association. Experimental results, using PDA, NNF, and CODA, show that the new approach reduces position errors in crossing trajectories by 13.9% on average compared to NNF.