No abstract

Purpose: Drowsy driving accounts for many accidents and has attracted substantial research attention in recent years. Electroencephalography (EEG) signals are shown to be a reliable measure for the early detection of drowsiness. Unfortunately, there is no comprehensive study showing the applicability of drowsiness detection systems with EEG signals. In this research, we targeted the studies under the category of drowsiness detection, which adopted an EEG-based approach, to inspect the applicability of these systems from different aspects. Materials and Methods: We included documented studies that utilized clinical devices and consumer-grade EEG headsets for detection of drowsiness and investigated the selected studies from different aspects such as the number of EEG channels, sampling frequency, extracted features, type of classifiers, and accuracy of detection. Among available headsets, we focused on the most popular ones, namely Muse, NeuroSky, and EMOTIV brands. Results: Considerable number of studies have used EEG headsets, and their reports showed that the highest average accuracy belongs to EMOTIV, and the highest maximum detection accuracy, 98.8%, was achieved by the Muse headset. Spectral features extracted from short periods of 1, 2, or 10 secs are the most popular features, and the support vector machine is the most commonly used classifier in drowsiness detection systems. Therefore, implementing a reliable detection system does not necessarily include complicated features and classifiers. Conclusion: It is shown that, despite their few electrodes, commercial headsets have gained decent detection accuracy. This study sheds light on the current status of drowsiness detection systems and paves the way for future industrial designs of such systems.

Driver drowsiness causes fatal driving accidents. Thermal imaging is a suitable drowsiness detection method as it is non-invasive and robust against changes in the ambient light. In this paper, driver drowsiness is detected by measuring the forehead temperature at the region covering the supratrochlear artery and also the cheek temperature. About 30 subjects drove on a highway in a driving simulator in two sessions. A thermal camera was used to monitor the facial temperature pattern. The subjects’ drowsiness levels were estimated by three human observers. The forehead and the cheek regions were located and tracked in each frame. The forehead and the cheek skin temperatures were obtained at three levels of drowsiness. The Support Vector Machine, the K-Nearest Neighbor, and the regression tree classifiers were used. From wakefulness to extreme drowsiness, the forehead skin temperature and the absolute cheek-forehead skin temperature gradient decreased by 0.46°C and 0.81°C, respectively. But the cheek skin temperature increased by 0.35°C in two sessions. The gradient difference is on average 50% higher than the forehead or the cheek temperature change alone. The results indicate that drowsiness can be detected with an accuracy of 82%, sensitivity of 85%, specificity of 90%, and precision of 84%. Driver drowsiness can be detected by monitoring changes in the forehead and the cheek temperature signal. Also, the temperature gradient can be used as a more robust and sensitive indicator of drowsiness.

This research deals with the autonomous landing maneuver of a quadrotor unmanned aerial vehicle (UAV) on an unmanned ground vehicle (UGV). It is assumed that the UGV moves independently, and there is no communication and collaboration between the two vehicles. This paper aims at the design of a closed-loop vision-based control system for quadrotor UAV to perform autonomous landing maneuvers in the possible minimum time despite the wind-induced disturbance force. In this way, a fractional-order fuzzy proportional-integral-derivative controller is introduced for the nonlinear under-actuated system of a quadrotor. Also, a feedback linearization term is included in the control law to compensate model nonlinearities. A supervisory control algorithm is proposed as an autonomous landing path generator to perform fast, smooth, and accurate landings. On the other hand, a compound AprilTag fiducial marker is employed as the target of a vision positioning system, enabling high precision relative positioning in the range between 10 and 350 cm height. A software-in-the-loop simulation testbed is realized on the windows platform. Numerical simulations with the proposed control system are carried out, while the quadrotor system is exposed to different disturbance conditions and actuator dynamics with saturated thrust output are considered.

The high number of fatal crashes caused by driver drowsiness highlights the need for developing reliable drowsiness detection methods. An ideal driver drowsiness detection system should estimate multiple levels of drowsiness accurately without intervening in the driving task. This paper proposes a multi-level drowsiness detection system by a deep neural network-based classification system using a combination of electrocardiogram and respiration signals. The proposed method is based on a combination of convolutional neural networks (CNNs) and long short-term memory (LSTM) networks for classifying drowsiness by concurrently using heart rate variability (HRV), power spectral density of HRV, and respiration rate signal as inputs. Two models, a CNN-based model and a hybrid CNN-LSTM-based model were used for multi-level classifications. The performance of the proposed method was evaluated on experimental data collected from 30 subjects in a simulated driving environment. The performance and the results of both models are presented and compared. The best performance for both three-level and five-level drowsiness classifications was achieved by the CNN-LSTM model. The results indicate that the three-level and five-level classifications of drowsiness can be achieved with 91 and 67% accuracy, respectively.