An Inertial/Vehicular-Based Driver Situation Monitoring System in Vehicle Driving Assistance
Abstract—Many accidents were happened because of habitual working overtime or tired spirit. This paper presents a methodology of remote warning for driver situation system using inertial and vehicular data. The development system integrates vehicular data and inertial information with embedded system into a research platform using IXP processor in linux system. To sense driver action, the steering angle sensor is adopted via controller area network. An inertial measurement unit is installed on a demonstrated vehicle, and the lateral acceleration is trigger and captured with attitude projection in this study. The processing method comprises of construction of driver model based on autoregressive exogenous method, quaternion and kalman filter using driving data. Driving data with s-cure and lane change motion using demonstrated vehicle was test in highway and express way. From inertial data and vehicular information, this paper provided a warning detection algorithm with learning parameters to do remote warning. The proposed system is carried out with theoretical application and hardware integration, and furthermore the result shows driver monitoring approach applicability.
- IA-00-0037 (585K)