In the automotive industry, understanding and mapping of customer needs into functional requirements is usually done using engineering knowledge obtained through subjective understanding of the manufacturer. Data analysis of real customer vehicles enables objective understanding of customer behavior, vehicle usage and insights into the driving environment. The current approaches for big automotive data collection and analysis on customer vehicles entails challenges with data transfer, latency, storage and management, data analysis algorithms and methods. Nevertheless, the integrity and security aspects also need to be addressed in order to meet the legal requirements and to establish a trustworthy relationship with the users and fleet owners. The current approaches require the big data computing infrastructure to be either in cloud or on premises of the automotive manufacturer. Currently it is challenging to centrally store and process the entire collection of vehicle data either on premises or cloud. However, it is more efficient and safe to complement the off-board big data analytics with flexible on-board data analysis capabilities. In addition, and advantage of the on-board data analysis is that sensitive raw data of the customer need not be transmitted to the vehicle manufacturer. Thus there is a need for research and development of a flexible on-board and off-board data processing and analysis platform in the automotive industry. The aim of this project is to develop these methods and demonstrate on specific use cases.
Docent at Computer Science and Engineering, Networks and Systems (Chalmers)
Funding years 2016–2018