Over the last few years, there has been an increasing interest in monitoring driving activities. Information about drivers and the way they drive is useful for several organizations, including car-insurers, fleet managers and traffic safety administrations. Moreover, new smartphones and mobile devices had enabled advance monitoring by embedding different motion sensors, GPS and data connectivity. Due to the increasing market penetration of smartphones, there is a potential for a reliable and distributed sensing platform. In particular, they can be used to identify risky driving maneuvers and compute accurate driver profiles. Such profiles can then be used by insurance companies to provide tailored premiums to their customers. This concept, known as Pay As You Drive (PAYD) or Usage Based Insurance (UBI), is currently undergoing a paradigm shift as the traditional black boxes telematics systems are being replaced, respectively extended by mobile devices.
To this end, we developed SenseFleet, a novel driver monitoring system that makes use of embedded smartphone sensors (e.g., GPS, motion sensors) to compute a set of metrics including an overall score that reflects the driving characteristics of a user and thus can be used to identify risky driving maneuvers. The system can optionally be augmented by an additional car-to-device interface that allows retrieving vehicle specific information that can be used as additional input by the system.
As opposed to existing systems, SenseFleet is able to provide reliable scores independently from the smartphone model and vehicle used by implementing a novel dynamic profiling algorithm.
After a small-scale evaluation study that showed promising results, we aim to validate our system on a much larger scale to demonstrate the reliability of the scoring technique under real and continuous constraints. The result of this validation project will provide evidence on the reliability of the SenseFleet system and open new perspectives and business opportunities in the telematics market.
Spin-Off Website: Motion-S