Uncertainty modelling and computational aspects of data association

TL;DR

This paper introduces a new hierarchical statistical model and an efficient MCMC-based method for data association in multi-object dynamical systems with uncertain and noisy observations, improving performance over existing algorithms.

Contribution

It presents a novel hierarchical model and an approximate MCMC method tailored for complex multi-object systems with uncertain data association.

Findings

Outperforms existing algorithms in various scenarios

Efficient handling of unknown and varying number of objects

Improved uncertainty representation in data association

Abstract

A novel solution to the smoothing problem for multi-object dynamical systems is proposed and evaluated. The systems of interest contain an unknown and varying number of dynamical objects that are partially observed under noisy and corrupted observations. An alternative representation of uncertainty is considered in order to account for the lack of information about the different aspects of this type of complex system. The corresponding statistical model can be formulated as a hierarchical model consisting of conditionally-independent hidden Markov models. This particular structure is leveraged to propose an efficient method in the context of Markov chain Monte Carlo (MCMC) by relying on an approximate solution to the corresponding filtering problem, in a similar fashion to particle MCMC. This approach is shown to outperform existing algorithms in a range of scenarios.