About this document
Driving Event Recognition via ML Sensors by yueetang05 is a document available to read on EtoBox.
This paper presents a robust machine learning structure for recognizing driving events using smartphone motion sensors, focusing on brake and turn detection. The proposed two-phase method employs Random Forest and Artificial Neural Network classifiers to enhance prediction accuracy, achieving average F1-scores of 71% for brake detection and 82% for turn detection. The study highlights the importance of data filtering techniques and robustness in performance across different drivers and road conditions.
- Author
- yueetang05
- Language
- EN