Big Data Analytics for Vehicle Multisensory Anomalies Detection.

Autonomous driving is assisted by different sensors, each providing information about certain parameters. What we are looking for is an integrated perspective of all these parameters to drive us into better decisions. To achieve this goal, a system that can handle these Big Data issues regarding volume, velocity and variety is needed. This paper aims to design and develop a real-time Big Data Warehouse repository, integrating the data generated by the multiple sensors developed in the context of IVS (In-Vehicle Sensing) systems; the data to be stored in this repository should be merged, which will imply its processing, consolidation and preparation for the analytical mechanisms that will be required. This multisensory fusion is important because it allows the integration of different perspectives in terms of sensor data, since they complement each other. Therefore, it can enrich the entire analysis process at the decision-making level, for instance, understanding what is going on inside the cockpit. [ABSTRACT FROM AUTHOR]