Decentralized Poisson Multi-Bernoulli Filtering for Vehicle Tracking

TL;DR

A decentralized Poisson multi-Bernoulli filter is proposed to track multiple vehicles using multiple high-resolution sensors and an efficient implementation is achieved by parametric state representation, utilization of single hypothesis tracks, and fusion of vehicle information based on a fusion mapping.

Abstract

A decentralized Poisson multi-Bernoulli filter is proposed to track multiple vehicles using multiple high-resolution sensors. Independent filters estimate the vehicles' presence, state, and shape using a Gaussian process extent model; a decentralized filter is realized through fusion of the filters posterior densities. An efficient implementation is achieved by parametric state representation, utilization of single hypothesis tracks, and fusion of vehicle information based on a fusion mapping. Numerical results demonstrate the performance.

Figures (5)

• Figure 1: Scenario with three et observed by two sensors with partially overlapping fov.
• Figure 2: Simulation scenario with two et observed by two sensors with overlapping fov. The et state is plot every $20$ time steps.
• Figure 3: Estimated shape of one et for different etpmb filters. The true et center and extent is plotted (green dot, blue solid line), as well as the estimated ones (red dot, red solid line for mean extent, red dotted line for one standard deviation).
• Figure 4: The true and estimated measurement rate for each et and sensor/filter is plotted over time.
• Figure 5: The average gospa value is plotted over time. Posterior fusion is performed in every time step.