Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data.

Drowsy driving is a major threat to the safety of drivers and road traffic. Accurate and reliable drowsy driving detection technology can reduce accidents caused by drowsy driving. In this study, we present a new method to detect drowsy driving with vehicle sensor data obtained from the steering wheel and pedal pressure. From our empirical study, we categorized drowsy driving into long-duration drowsy driving and short-duration drowsy driving. Furthermore, we propose an ensemble network model composed of convolution neural networks that can detect each type of drowsy driving. Each subnetwork is specialized to detect long- or short-duration drowsy driving using a fusion of features, obtained through time series analysis. To efficiently train the proposed network, we propose an imbalanced data-handling method that adjusts the ratio of normal driving data and drowsy driving data in the dataset by partially removing normal driving data. A dataset comprising 198.3 h of in-vehicle sensor data was acquired through a driving simulation that includes a variety of road environments such as urban environments and highways. The performance of the proposed model was evaluated with a dataset. This study achieved the detection of drowsy driving with an accuracy of up to 94.2%.

Soft Bimodal Sensor Array Based on Conductive Hydrogel for Driving Status Monitoring.

：Driving status monitoring is important to safety driving which could be adopted to improve driving behaviors through hand gesture detection by wearable electronics. The soft bimodal sensor array (SBSA) composed of strain sensor array based on ionic conductive hydrogels and capacitive pressure sensor array based on ionic hydrogel electrodes is designed to monitor drivers' hand gesture. SBSA is fabricated and assembled by the stretchable functional and structural materials through a sol-gel process for guaranteeing the overall softness of SBSA. The piezoresistive strain and capacitive pressure sensing abilities of SBSA are evaluated by the data acquisition system and signal analyzer with the external physical stimuli. The gauge factor (GF) of the strain sensor is 1.638 under stretched format, and -0.726 under compressed format; sensitivity of the pressure sensor is 0.267 kPa-1 below 3.45 and 0.0757 kPa-1 in the range of 3.45-12 kPa, which are sensitive enough to hand gesture detection and driving status monitoring. The simple recognition method for the driver's status behavior is proposed to identify the driver's behaviors with the piezoresistive properties of conductive polymers, and the turning angles are computed by the strain and pressure values from SBSA. This work demonstrates an effective approach to integrate SBSA seamlessly into an existing driving environment for driving status monitoring, expanding the applications of SBSA in wearable electronics.

Real-Time Monitoring System of Automobile Driver Status and Intelligent Fatigue Warning Based on Triboelectric Nanogenerator.

To improve automobile safety, identifying driver fatigue is considerably crucial because it is one of the main causes of traffic accidents. In this research, smart systems based on a triboelectric nanogenerator are designed, which can provide driver status monitoring and fatigue warning in real time. The smart system consists of a self-powered steering-wheel angle sensor (SSAS) and a signal processing unit. The SSAS, which comprises a stator, a rotor, and a sleeve, is mounted on the steering rod, and the electrodes are designed with a phase difference to improve the resolution of the sensor. The turning angle of the steering wheel operated by the driver is recorded by the SSAS; meanwhile, the number of rotations, the average angle, and other parameters in the driver's recorded data are analyzed by the signal processing unit from which a warning threshold for each parameter is determined. The system assesses the status of the driver in real-time by comparing these parameters and threshold values, and experimental results demonstrate that driver status is accurately judged. This work has important potential applications in the fields of traffic safety and intelligent driving.