In this study, a robust driver inattention detection system based on a driver model and dynamic relational network (DRN) using sensor data such as the quantity of vehicle speed and pedal operation is proposed. The system can be divided into three parts: the sensor diagnosis module, the driver model and inattentive driving detection. The sensor diagnosis module using the DRN is developed to analyze and identify faulty sensors (if any) through its measurement data. The driver inattention detection system was tested with actual data from a car driving on a highway and the system's diagnosis performance was evaluated. The results obtained indicate the effectiveness of the proposed system for inattentive driving detection. © ICIC International 2014.