A Novel Map-matching Algorithm to Improve Vehicle Tracking System Accuracy

Satellite-based Vehicle Tracking System accuracy can be improved by augmenting the positional information using road network data, in a process known as map-matching. Point-to-curve matching algorithm is generally not suitable for problem since it ignores any historical data and often gave unstable results. The curve-to-curve matching algorithms measure the similarity between the track and possible road path. However, most of these algorithms cannot handle multiple track hypotheses that have varying degree of similarity. This paper presents a new curve-to-curve map matching technique which picks the most relevant hypothesis, enabling corrections if necessary, therefore providing intelligent guesses with considerable accuracy.