Road Constrained Labeled Multi Bernoulli Filter based on PDF Truncation for Multi-Target Tracking

摘要

In this paper, road constrained filtering is applied to labeled multi Bernoulli (LMB) filter using PDF truncation in multi-road environment. In target tracking systems with road map information, road constraints can effectively improve the estimation performance. To apply multiple road constraints information to the tracking filter, all constraints should not be applied simultaneously and only one should be selected for each estimated trajectory. Then, probability density function (PDF) truncation is conducted which is a constrained filtering technique for inequality constraints. To verify the constrained filtering technique to LMB filter, simulations for multi-target tracking in cluttered environments are carried out. The simulation result shows that the proposed method bounded estimated trajectories on the road effectively and reduced OSPA error.