PUB - Publikationen an der Universität Bielefeld

We investigate a new approach towards maneuver prediction that is based on personalization and incremental learning. The prediction accuracy is continuously improved by incorporating only the individual driving history. The study is based on a collection of commuting drivers who recorded their daily routes with a standard smart phone and GPS receiver. Prediction target is the expected maneuver on the next intersection with three classes: stop, turn, or go straight. We show that a personalized prediction based on at least one experience of a certain intersection already improves the prediction performance over an average prediction model trained on all test driver commute routes. This performance gain increases further with more personal training data.