Prior Knowledge Input to Improve LSTM Auto-encoder-based Characterization of Vehicular Sensing Data

TL;DR

This paper introduces a prior-knowledge input method combined with a recurrent denoising auto-encoder to enhance the accuracy of vehicular event characterization systems, addressing data imbalance and infrequency issues.

Contribution

It proposes a novel integration of prior knowledge and a recurrent auto-encoder to improve event detection accuracy in connected vehicle data.

Findings

Achieved a 14.7% accuracy improvement over baseline.

Enhanced F1-score with knowledge-based modeling.

Demonstrated robustness in training with augmented data.

Abstract

Precision in event characterization in connected vehicles has become increasingly important with the responsive connectivity that is available to the modern vehicles. Event characterization via vehicular sensors are utilized in safety and autonomous driving applications in vehicles. While characterization systems have been shown to be capable of predicting the risky driving patterns, precision of such systems still remains an open issue. The major issues against the driving event characterization systems need to be addressed in connected vehicle settings, which are the heavy imbalance and the event infrequency of the driving data and the existence of the time-series detection systems that are optimized for vehicular settings. To overcome the problems, we introduce the application of the prior-knowledge input method to the characterization systems. Furthermore, we propose a recurrent-based denoising auto-encoder network to populate the existing data for a more robust training process. The results of the conducted experiments show that the introduction of knowledge-based modelling enables the existing systems to reach significantly higher accuracy and F1-score levels. Ultimately, the combination of the two methods enables the proposed model to attain 14.7\% accuracy boost over the baseline by achieving an accuracy of 0.96.