Unsupervised Driving Event Discovery Based on Vehicle CAN-data
This work addresses the problem of analyzing large-scale, unlabeled vehicle data for fleet management, though it is incremental as it compares existing SSL methods rather than introducing a new paradigm.
The paper tackles unsupervised discovery of driving events from vehicle CAN-data by proposing a simultaneous clustering and segmentation approach using self-supervised learning (SSL) for multivariate time series. It finds that contrastive learning methods achieve performance similar to state-of-the-art generative SSL techniques on real Tesla Model 3 data.
The data collected from a vehicle's Controller Area Network (CAN) can quickly exceed human analysis or annotation capabilities when considering fleets of vehicles, which stresses the importance of unsupervised machine learning methods. This work presents a simultaneous clustering and segmentation approach for vehicle CAN-data that identifies common driving events in an unsupervised manner. The approach builds on self-supervised learning (SSL) for multivariate time series to distinguish different driving events in the learned latent space. We evaluate our approach with a dataset of real Tesla Model 3 vehicle CAN-data and a two-hour driving session that we annotated with different driving events. With our approach, we evaluate the applicability of recent time series-related contrastive and generative SSL techniques to learn representations that distinguish driving events. Compared to state-of-the-art (SOTA) generative SSL methods for driving event discovery, we find that contrastive learning approaches reach similar performance.