PMHT with Application of Road Information Based on GMTI

Abstract

Probabilistic Multi-Hypothesis Tracking (PMHT) is an algorithm for multiple target tracking when association of measurement and track is unknown. PMHT is an efficient algorithm for GMTI in cluttered environments of which the computational complexity is linear in the number of targets and measurements. An algorithm is presented in this paper to apply road information to PMHT. Road information is used to adjust estimated target state during the process of Kalman filter. Tracking performance of PMHT is increased apparently with the application of road constraint.