Title :
A marginalised particle filter for bearings-only tracking
Author :
Morelande, Mark R.
Author_Institution :
Melbourne Syst. Lab., Univ. of Melbourne, Parkville, VIC, Australia
Abstract :
A marginalised particle filter (PF) for bearings-only tracking in modified polar coordinates (MPC) is developed. Using an Euler approximation to the dynamical equation it is shown that the range can be marginalised out so that only the three remaining elements of the state vector need to be sampled in the PF. The marginalised PF in MPC is shown to significantly outperform existing PFs for BOT in a numerical example.
Keywords :
approximation theory; particle filtering (numerical methods); Euler approximation; MPC; PF; bearings-only tracking; dynamical equation; marginalised particle filter; modified polar coordinates; Approximation methods; Atmospheric measurements; Equations; Noise; Particle measurements; Radar tracking; Vectors; Particle filters; bearings-only tracking; nonlinear filters;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Conference_Location :
Vancouver, BC
DOI :
10.1109/ICASSP.2013.6638877
