PF-HMap: A Target Track Maintenance Approach for Mobile Sensor Platforms with Intermittent and Regional Measurements
Zhijun Tang, Ümi̇t Özgüner · 2006
A new approach, the particle-filter-and-hospitability-map algorithm (PF-HMap), is proposed to deal with the general target track maintenance problem for mobile sensor platforms with intermittent and regional measurements. The posterior probability density function (PDF) of the target state is maintained on-line through a particle filter. The hospitability map (HMap), which describes the terrain-dependent motion capability of ground targets, is encoded into the prediction stage of the particle filter. The PF-HMap approach can be considered as a generic way to incorporate non-linear, non-Gaussian target/sensor model and non-analytic environmental information into the tracking scheme. Simulation results are presented to demonstrate the capability of the PF-HMap algorithm in maintaining track information of maneuvering targets in various application scenarios such as target search and target following