Advanced vehicular sensing of road artifacts and driver behavior

The use of smartphone-only systems has relatively low accuracy, high computational complexity, and high battery consumption. Utilizing built-in vehicle sensors via OBD dongles can enable a more comprehensive road conditions monitoring system with lower computational cost and higher accuracy. This paper presents a system that utilizes the vehicle´s CAN-Bus, as a source of sensory data, and a smartphone, as the processing unit of the mentioned data, to detect road artifacts and monitor driver behavior. Preliminary results of the proposed system reveal a maximum of 92%, with an average of 84%, in detection of road artifacts. Similarly, in a well-defined environment, driver behavior detection approaches 100%.