Moment-based filters, such as the Probability Hypothesis Density (PHD) filter, are an attractive solution to multi-target tracking. However, an underlying assumption for the PHD filter is that each measurement is either caused by a single target or clutter. In this paper, we design a novel moment-based multi-target filter, the Additive Likelihood Moment (ALM) filter, where the measurements are affected by all targets. We focus on the cases where the likelihood can be expressed as a function of the sum of the individual target contributions. As an example, we consider radio tomographic tracking where the attenuation of the signal between a pair of sensors is the sum of attenuations caused by all targets. Our multi-target tracking algorithm is based on a particle approximation of our moment-based filter. Our simulations show that our algorithm has a lower estimation error than MCMC particle methods while achieving 80% savings in terms of computational time.

/pdf/multi-target-tracking-for-measurement-models-with-additive-20skjzm873.pdf

Multi-target tracking for measurement models with additive contributions

Distributed multi-sensor multi-view fusion based on generalized covariance intersection

Asynchronous multi-rate multi-sensor fusion based on random finite set

In this paper, the problem of targets road tracking based on multiple asynchronous sensors is addressed. A method based on Cardinalized Probability Hypothesis Density (CPHD) filter improved with probability of target type is presented. Using the sensor classification, the probability of target type is recursively computed by Bayesian rules. This probability is used to improve the performance of the Multi Target Tracking (MTT) filter and the understanding of the observed area. The tuning of filter parameter depending of sensor is explained and the particular case of non detection due to occlusion with camera is developed. Our system has been validated with real measurements from a smart camera and a radar in real traffic conditions with real time computation.

Multi target tracking with CPHD filter based on asynchronous sensors

A Survey of PHD Filter Based Multi-target Tracking

This paper presents a sequential filter implementation of particle Probability Hypothesis Density (PHD) filter for multisensor multi-target tracking. The tracking system involves potentially nonlinear target dynamics described by Markov state space model and nonlinear measurements. Each sensor reports measurements to the tracking system, which performs sequential estimation of the current state using the particle PHD filter, which propagates only the first order statistical moment of the full posterior of the multi-target state. Simulation results are also given and compared with a single radar multi-target tracking, showing the advantage of the fusion tracking over the single radar multi-target tracking. Keywordsrandom finite sets; particle PHD filter; multitarget tracking; sequential fusion

Taishan OuYang

Papers

A Particle PHD Filter for Multi-Sensor Multi-Target Tracking Based on Sequential Fusion