Driving Style Recognition at First Impression for Online Trajectory Prediction

TL;DR

This paper introduces a hybrid offline-online driving style recognition method that improves autonomous vehicle trajectory prediction accuracy by up to 37.7% using minimal data and real-time style recognition techniques.

Contribution

It presents a novel hybrid approach combining PCA, K-means, and Maximum-Likelihood for efficient online driving style recognition in trajectory prediction.

Findings

Reduces trajectory prediction error by up to 37.7%.

Outperforms existing methods in RMSE on real driving data.

Effective with minimal data for real-time applications.

Abstract

This paper proposes a new driving style recognition approach that allows autonomous vehicles (AVs) to perform trajectory predictions for surrounding vehicles with minimal data. Toward that end, we use a hybrid of offline and online methods in the proposed approach. We first learn typical driving styles with PCA and K-means algorithms in the offline part. After that, local Maximum-Likelihood techniques are used to perform online driving style recognition. We benchmarked our method on a real driving dataset against other methods in terms of the RMSE value of the predicted trajectory and the observed trajectory over a 5s duration. The proposed approach can reduce trajectory prediction error by up to 37.7\% compared to using the parameters from other literature and up to 24.4\% compared to not performing driving style recognition.