Probabilistic map-matching using particle filters

TL;DR

This paper introduces a probabilistic map-matching method using particle filters to improve GPS data accuracy by generating multiple candidate solutions with associated probabilities, validated on diverse GPS datasets.

Contribution

It presents a novel probabilistic approach to map-matching employing particle filters, enhancing accuracy over traditional methods.

Findings

Effective in handling GPS data of varying quality

Produces multiple candidate solutions with probability scores

Validated thoroughly on real GPS datasets

Abstract

Increasing availability of vehicle GPS data has created potentially transformative opportunities for traffic management, route planning and other location-based services. Critical to the utility of the data is their accuracy. Map-matching is the process of improving the accuracy by aligning GPS data with the road network. In this paper, we propose a purely probabilistic approach to map-matching based on a sequential Monte Carlo algorithm known as particle filters. The approach performs map-matching by producing a range of candidate solutions, each with an associated probability score. We outline implementation details and thoroughly validate the technique on GPS data of varied quality.