Comparison of GPS and MEMS Support for Smartphone-Based Driver Behavior Monitoring

Smartphone-based driver monitoring and other forms of smartphone-based telematics are fast becoming a prevalent trend, especially in the vehicle insurance industry. Choosing the most appropriate sensor system is a pivotal aspect of any driver behavior monitoring system. This paper evaluates the performance of the two sensor systems for smartphone-based driver behavior monitoring and telematics -- GPS and MEMS -- in terms of the usefulness of data that can be collected for detecting and classifying driving maneuvers as well as the convenience of use and the effect on battery life. Comparisons are done in terms of selected Figures of Merit (FoM) that are commonly employed as metrics for classifying driving style. The results of comprehensive testing show that the MEMS inertial sensors outperform a GPS platform in terms of sampling rate, battery life and the accuracy with which acceleration, braking, swerving and cornering can be detected.