Driver Physiological Signals Research Articles

EEG and ECG-Based Multi-Sensor Fusion Computing for Real-Time Fatigue Driving Recognition Based on Feedback Mechanism.

A variety of technologies that could enhance driving safety are being actively explored, with the aim of reducing traffic accidents by accurately recognizing the driver's state. In this field, three mainstream detection methods have been widely applied, namely visual monitoring, physiological indicator monitoring and vehicle behavior analysis. In order to achieve more accurate driver state recognition, we adopted a multi-sensor fusion approach. We monitored driver physiological signals, electroencephalogram (EEG) signals and electrocardiogram (ECG) signals to determine fatigue state, while an in-vehicle camera observed driver behavior and provided more information for driver state assessment. In addition, an outside camera was used to monitor vehicle position to determine whether there were any driving deviations due to distraction or fatigue. After a series of experimental validations, our research results showed that our multi-sensor approach exhibited good performance for driver state recognition. This study could provide a solid foundation and development direction for future in-depth driver state recognition research, which is expected to further improve road safety.

Unsupervised Scalable Multimodal Driving Anomaly Detection

Driving anomaly detection aims to identify objects, events or actions that can increase the risk of accidents, reducing road safety. While supervised approaches can effectively identify aspects related to driving anomalies, it is unfeasible to tabulate and address all potential driving anomalies. Instead, it is appealing to design unsupervised approaches that can automatically identify unexpected driving scenarios. This study formulates the detection of driving anomalies as a binary-discrimination task between expected and unexpected driving behaviors. We propose an unsupervised contrastive method using conditional generative adversarial networks (GANs) implemented with the attention model and the triplet loss function. A feature of our framework is its scalability, where it is easy to add new modalities. We consider five different modalities: the vehicle's CAN-Bus signals, driver's physiological signals, distance to nearby pedestrians, distance to nearby vehicles and distance to nearby bicycles. Our approach trains a conditional GAN to extract latent features from each of the five modalities. An attention model combines the latent representations from the modalities. The entire framework is trained with the triplet loss function to generate effective representations to discriminate normal and abnormal driving segments. We conduct experimental evaluations on the driving anomaly dataset (DAD), achieving improved performance over alternative approaches.

A review of driver fatigue detection and its advances on the use of RGB-D camera and deep learning

Driver fatigue is an essential reason for traffic accidents, which poses a severe threat to people’s lives and property. In this review, we summarize the latest research findings and analyze the developmental trends of driver fatigue detection. Firstly, we analyze and discuss four types of different fatigue detection technologies based on driver physiological signals, behavior features, vehicle running features, and information fusion, respectively. Then, we focus on RGB-D camera and deep learning which are two state-of-the-art solutions in this field. Finally, we present the work on integration of RGB-D camera and deep learning, where Generative Adversarial Networks and multi-channel schemes are utilized to enhance the performance. We conducted experiments to show that the fatigue features extracted by Convolutional Neural Networks are superior to traditional handcrafted ones while single features cannot guarantee robustness. Moreover, the latent fatigue features extracted by deep learning methods have been demonstrated to be effective for fatigue detection.

Joint learning of video images and physiological signals for lane-changing behavior prediction

Understanding and predicting human driving behavior play an important role in the development of intelligent vehicle systems, particularly for Advanced Driver Assistance System (ADAS) to estimate dangerous operations and take appropriate actions. However, well-predicted lane-changing (LC) behavior is still challenging on account of the complexity and uncertainty of traffic status, and labeled data are required. To address this problem, we propose a novel framework, denoted as LCNet, for lane-changing behavior prediction via joint learning of the front view video images and driver physiological signals. Firstly, with a temporal consistency module, both labeled and unlabeled video frames can be utilized in the training phase, while no extra computation is required during inference. Secondly, a new penalty term is introduced for learning sequential physiological signals, which is sensitive to local continuity property. Finally, a new loss function is designed for LCNet to learn co-occurrence features from the video scene-optical flow branch and physiology branch efficiently. Moreover, the experiments are conducted on a real-world driving data set. The experimental results demonstrate that the LCNet can learn the underlying features of upcoming lane-changing behavior and significantly outperform the other advanced models.

Experimental Study of Thermal Comfort Based on Driver Physiological Signals in Cooling Mode under Summer Conditions

In this study, electroencephalogram (EEG), photo-plethysmography (PPG), and surface temperature measurements of subjects were taken while performing a driving simulation when the cabin and vent discharge air temperature in summer were changed from discomfort to comfort conditions. Additionally, subjective questionnaires were used to analyze the subject’s thermal comfort under the various driving environments. As a result, the surface temperatures of the forehead, left hand, right hand, and abdomen of the subject during driving were reduced by 2, 0.97, 2.18, and 5.86 °C, respectively, by operating a 12.5 °C vent cooling function at a cabin temperature of 35 °C. As a comprehensive analysis of the subjective survey, PPG, and EEG results, total power (TP), the standard deviation of N-N interval (SDNN), and the root mean square of successive differences (RMSSD) of subjects increased and stress index decreased at cabin and vent discharge air temperatures of 30–27.5 °C and 16.5–18.5 °C, respectively. Furthermore, the relative sensory motor rhythm (SMR) wave and concentration index (CI) of the frontal lobe tended to increase under the same temperature conditions. Accordingly, it was confirmed that these temperature conditions provided a pleasant driving environment for the driver and increased concentration on driving.

Driver Drowsiness Measurement Technologies: Current Research, Market Solutions, and Challenges

Significant advances in embedded technologies hold promise to characterize and monitor driver’s state of alertness and detect critical levels of driver drowsiness in real-time. While some enduring solutions have been available as prototypes for a while, many of these technologies are now in the development, validation testing, or even commercialization stages. Several studies have reviewed some available fatigue and/or drowsiness detection solutions. This paper builds on previous studies and aims to provide up-to-date yet complete review on emerging driver drowsiness and alertness monitoring technologies. The contribution of this paper is to identify and review the key objective driver state monitoring technologies namely (i) driving behavioural (vehicle-based), (ii) driver behavioural (video-based), and (iii) driver physiological signals measure based technologies. Each technology is presented with a detailed description of associated detection methods and measuring metrics along with the current research activities and market products in this era. An evaluation of these technologies is also provided in term of intrusiveness, accuracy of detection, and practical use point of view. Therefore, the paper highlights open issues with these emerging systems which need further investigation in the future. We think that this study contribute to a better understanding of sleepiness at the wheel, and will help promote the implementation of accurate crash prevention technologies.

How the Autonomic Nervous System and Driving Style Change With Incremental Stressing Conditions During Simulated Driving

This paper reports on the autonomic nervous system (ANS) changes and driving style modifications as a response to incremental stressing level stimulation during simulated driving. Fifteen subjects performed a driving simulation experiment consisting of three driving sessions. Starting from a first session where participants performed a steady motorway driving, the experimental protocol includes two additional driving sessions with incremental stress load. More specifically, the first stressing load consists of randomly administering mechanical stimuli to the vehicle during a steady motorway driving by means of a series of sudden and unexpected skids, such as those produced by a strong wind gust. These skids were supposed to produce in the driver a given level of stress. In order to assess this mental workload, dedicated psychological tests were performed. The second stressing load implied an incremental psychological load, consisting of a battery of time pressing arithmetical questions, added to the mechanical stimuli. For the whole experimental session, the driver's physiological signals and the vehicle's mechanical parameters were recorded and analyzed. In this paper, the ANS changes were investigated in terms of heart rate variability, respiration activity, and electrodermal response along with mechanical information such as that coming from steering wheel angle corrections, velocity changes, and time responses. Results are satisfactory and promising. In particular, significant statistical differences were found among the three driving sessions with an increasing stress level both in ANS responses and mechanical parameter changes. In addition, a good recognition of these sessions was carried out by pattern classification algorithms achieving an accuracy greater than 90%.

차량시뮬레이터 환경에서 운전 중 주의분산에 따른 생체신호 변화 연구

Driver distraction is a major cause of traffic accidents in Korea. Various measures are being introduced to detect and warn driver distraction. The objective of this research is to invcstigate changes in driver's physiological signals due to distraction during driving. Driving simulator experiments have been canicd out to investigate discrepancy in EEG signals among normal driving, DMB watching during driving, and cellular phone use during driving. Based on the discrepancy, combination of EEG signals have been identfied as candidate variables for detecting driver distraction. Statistical analysis has been carried out to verify their statistical significance.