Improved Particle Filters for Vehicle Localisation

TL;DR

This paper introduces improved particle filter algorithms that enhance vehicle localization accuracy and efficiency, especially under conditions of infrequent, low-noise observations, by sampling around the latest observation.

Contribution

The paper proposes a novel particle filtering approach that samples around recent observations, significantly improving performance over traditional methods.

Findings

Order of magnitude accuracy improvement

Enhanced efficiency with infrequent observations

Effective in non-linear, non-Gaussian scenarios

Abstract

The ability to track a moving vehicle is of crucial importance in numerous applications. The task has often been approached by the importance sampling technique of particle filters due to its ability to model non-linear and non-Gaussian dynamics, of which a vehicle travelling on a road network is a good example. Particle filters perform poorly when observations are highly informative. In this paper, we address this problem by proposing particle filters that sample around the most recent observation. The proposal leads to an order of magnitude improvement in accuracy and efficiency over conventional particle filters, especially when observations are infrequent but low-noise.