Research
Data-Driven and Context-Aware Systems
Context is defined as the information used to characterize the situation. The main components of driving context are driver (e.g., age, driving experience, and fatigue, and drowsiness), vehicle (e.g., equipped ADAS), and environment (e.g., environment, and road network characteristics). The context-aware driver assistance system monitors, detects, interprets, and reacts to changes in any of the components of the driving context. Currently, powerful sensor technologies record naturalistic driving and near-missincidence data. A near-miss-incidence database has been constructed and managed by Smart Mobility Research Center of Tokyo University of Agriculture and Technology in Japan since 2004. This study proposes a context-aware driver model for determining the recommended speed in blind intersection situations based on data-driven approach.
Yuichi Saito, Fumio Sugaya, Shintaro Inoue, Pongsathorn Raksincharoensak, Hideo Inoue, A Context-Aware Driver Model for Determining Recommended Speed in Blind Intersection Situations, Accident Analysis and Prevention, Vol. 163, 106447, 2021, https://doi.org/10.1016/j.aap.2021.106447
Yuichi Saito, Fuma Kochi, Makoto Itoh, Takesato Fushima, Takashi Sugano, and Yasunori Yamamoto, Influence of Road Environmental Elements on Pedestrian and Cyclist Road Crossing Behavior. SICE Journal of Control, Measurement, and System Integration, Vol. 14, No.2, pp. 50-58, 2021, https://doi.org/10.1080/18824889.2021.1894900 [Open access]
Dual Control Theoretic Human Machine Systems
We have proposed a haptic interaction method with a dual-control scheme that attempts to execute vehicle control and driver state identification in the context of manual driving/partial driving automation. In the control theory, dual control deals with the control of uncertain systems whose characteristics are unknown and considers the dual role of control signals for control and real-time estimation. The control signals of adaptive systems have the following features: 1) the controlled variable cautiously tracks the target; 2) the signals excite the controlled object to enhance the process of parameter identification so that the control performance improves in future intervals. The controller has two objectives: action—to perform safety control based on current information—and investigation—to experiment with the system to know and/or learn its behavior. This dual-control scheme was applied to human–machine systems.
Yuichi Saito, Makoto Itoh, Toshiyuki Inagaki, Bringing a Vehicle to a Controlled Stop: Effectiveness of a Dual Control Scheme for Identifying Driver Drowsiness and Preventing Lane Departures under Partial Driving Automation Requiring Hands-on-Wheel, IEEE Transactions on Human-Machine Systems, Early Access, 13 pages, 2021 https://doi.org/10.1109/THMS.2021.3123171
Yuichi Saito, Makoto Itoh, Toshiyuki Inagaki, Driver Assistance System with a Dual Control Scheme: Effectiveness of Identifying Driver Drowsiness and Preventing Lane Departure Accidents, IEEE Transactions on Human-Machine Systems, Vol.46, No.5, pp.660-671, 2016, https://doi.org/10.1109/THMS.2016.2549032 [Postprint, Tsukuba Repository]
MRM for safely guiding a vehicle to the roadside zone
in preparation
Monitoring Requests
In conditionally automated driving, drivers are relieved of steering (hands-off), accelerating, and braking actions as well as of continuous monitoring of driving situations and the system operation status (eyes off). This enables continuously engagement in non-driving-related activities. Managing the allocation of a driver’s attention to the surrounding environment and automation status presents a major challenge in human–machine system design. In this study, we propose a verbal message with a reminder (monitoring request) to divert the driver’s attention from non-drivingrelated activities to peripheral monitoring under conditionally automated driving.
Yuichi Saito, Yuta Watahiki, Chokiu Leung, Huiping Zhou, Makoto Itoh, Effect of Verbal Messages with Reminders to Communicate Driving Situations to Alter Driver Behavior in Conditional Driving Automation, Transportation Research Part F: Traffic Psychology and Behaviour, Volume 85, 69-82, 2022 https://doi.org/10.1016/j.trf.2022.01.003
Proactive Safety
Pedestrians and vehicles share common road spaces. Speed is a major factor in the lethality of these collisions: mortality rates increase sharply when the collision velocity is greater than 30 km/h. Therefore, it is important to predict the ‘‘hidden risk” in driving situations and control the vehicle velocity to prepare for any unexpected upcoming hazardous events. When expert drivers are confronted with uncertainty, they naturally seek to reduce it by obtaining more information and attempting to fit their current driving situation into a pre-existing category from their previous experience. When necessary, expert drivers adopt preventive measures, socalled ‘‘hazard-anticipatory driving behaviors,” such as decreasing their velocity and ensuring that a sufficient braking distance is available in case an unexpected event occurs. We propose the driver assistance system to perform a proactive braking intervention in achieving a referenced terminal speed in uncertain situations, such as one in which an unobserved pedestrian might initiate a road crossing.
Yuichi Saito, Ryoma Yoshimi, Shinichi Kume, Xun Shen, Akito Yamasaki, Ryosuke Matsumi, Takuma Ito, Toshiki Kinoshita, Shintaro Inoue, Tsukasa Shimizu, Masao Nagai, Hideo Inoue, Pongsathorn Raksincharoensak, Effectiveness of a Driver Assistance System With Deceleration Control and Brake Hold Functions in Stop Sign Intersection Scenarios, IEEE Transactions on Intelligent Transportation Systems, early access, 2021, https://doi.org/10.1109/TITS.2021.3085847
Yuichi Saito, Ryoma Yoshimi, Shinichi Kume, Masahiro Imai, Akito Yamasaki, Takuma Ito, Shintaro Inoue, Tsukasa Shimizu, Masao Nagai, Hideo Inoue, Pongsathorn Raksincharoensak, Effects of a Driver Assistance System with Foresighted Deceleration Control on the Driving Performance of Elderly and Younger Drivers, Transportation Research Part F: Traffic Psychology and Behaviour, Volume 77, pp. 221-235, 2021, https://doi.org/10.1016/j.trf.2020.12.017
Takuma Ito, Masatsugu Soya, Kyoichi Tohriyama, Yuichi Saito, Tsukasa Shimizu, Akito Yamasaki, Masao Nagai, Hideo Inoue, Minoru Kamata, Evaluation of Acceptability of Adaptive Proactive Braking Intervention System Based on Risk Map for Elderly Drivers, International Journal of Automotive Engineering, Vol.11, No. 2, pp. 40-48, 2020, https://doi.org/10.20485/jsaeijae.11.2_40 [Open access]
Hazard Prediction Test
in preparation
Shared and Cooperative Guidance Control
The use of automation and opportunities of humanmachine interaction are expanding in driving system domain. Contrary to the idea of replacing humans by technology, the benefits of haptic shared control, in which the driver and automation shares control authority of the vehicle and communicates continuously, have been widely discussed in both theoretical and experimental studies: The drivers benefit from increased performance and/or reduced workload by keeping in the control loop. Shared control links the operational actions of the driver and automation on input terminal (i.e., steering, and pedal) at the force level, and it has a direct impact on the mutual task of both agents. Flemisch et al. stated shared control is where cooperation comes sharply into effect at the operational (or control) level. Haptic and continuous information would enable the driver to virtually touch their potentially hazardous environment. The design philosophy is to communicate to drivers about potentially hazardous situations by manipulating the steering and pedal stiffness with an adaptive level of authority while maintaining the driver in the control loop.
Yuichi Saito, Pongsathorn Raksincharoensak, Effect of Risk-Predictive Haptic Guidance in One-Pedal Driving Mode, Cognition, Technology & Work, 21 (4), 671-684, 2019, https://doi.org/10.1007/s10111-019-00558-3 [Postprint, Tsukuba Repository]
Yuichi Saito, Hussam Muslim, Makoto Itoh, Design of Haptic Protection with an Adaptive Level of Authority Based on Risk Indicators under Hands-on Partial Driving Automation, Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, pp.1613-1618, October 17-20, 2021, https://doi.org/10.1109/SMC52423.2021.9658933