engleski

A framework for collection, contextual enrichment and advanced analytics of automotive data

The thesis addresses the challenges related to collection, contextual enrichment and analytics of automotive data. Current trends that focus on vehicle connectivity, i.e. vehicle-2 infrastructure and vehicle-2-vehicle, are seen as beneficial for data collection and analysis. However, collecting data from connected vehicles is not a simple task because connected vehicle hardware and software are a closed system meant to be used by vehicle manufacturers only. Additionally, when automotive data is combined with various heterogeneous data sources, it can provide valuable additional insights in data analysis. The key component for impactful research in this domain is a high-quality source of contextually enriched automotive data, enabling interdisciplinary study from environmental sustainability, automotive engineering, behavioural science, telecommunications and transportation science perspectives. To the best of the author's knowledge, there is no open source data set of contextually enriched automotive data available. To facilitate the creation of such a data set, this thesis presents a framework for the collection and contextual enrichment of automotive data. The framework utilizes smartphones to collect the vehicle's On-Board Diagnostics (OBD) or Controller Area Network (CAN) data and enriches it with Internet-based and built-in smartphone sensor data. The framework was evaluated on a case study of a data collection experiment. In this experiment, 9 drivers collected more than 90 hours of driving data, forming a contextually enriched automotive data set. A special emphasis in the analysis of the collected data was placed on deriving and calculating metrics for ranking the drivers and trips according to their eco-efficient driving patterns, which is a valuable insight that can be used to improve the transportation sustainability.

automotive data, data collection, contextual enrichment, data analytics

nije evidentirano