Toward Symbiosis with Risk from Human-Technology Interaction
The advent of technology has placed people in complex and uncertain situations. It is a critical issue to determine how to effectively and efficiently identify and address such risks and uncertainties. In this lab, we conduct research on risk management, with a focus on human factors, to gain social acceptance. To enhance safety in the face of these risks, we employ statistical models to quantitatively clarify the human cognitive processes and psychological structures that underlie human behavior and to evaluate proposed ideas.
Trust in Automation
Highly developed automation become pervasive. Naive users can easily interact with such automation without professional training like aircraft pilots. However, recent automation still requires human supervisory controllers. Here, trust is one of determinators influecing the usage of automation. The goal of this research is to clarify contributors to trust development for general automation users and apply this to automation design to support appropriate human-machine interaction. The research first approaches human-machine trust with the Muir & Moray (1996)'s trust framework, then newly interpreted the long-accepted view considering the current context. As a next step, the new interpretation has been expanded in another automation domain, driving automation. This research is going to explore the finding to other professional domains and different user perspectives.
Hazard Perception
Drivers' hazard perception is crucial to keep driving safety. The main objective of this study is to clarify when and why drivers encounter failures in information processing and how to support this. In particular, this research has focused on impacts of driver compensatory behaviour on the cognitive process of visually impaired drivers by analyzing driver behavior, such as response to pedals. When drivers are aware of their visual impairment, they take extra actions to overcome the impariment, such as moving their head wider and more frequently or driving at low-speed. The first step was to investigate effects and limits of driver compensation: reducing speed and head movement. The second step was to clarify the relationship between compensation, congnitive process, and collision risk quantitatively. The overreaching goal of this research is to design trustworthy assistance systems for visually-defected drivers.
Evaluation Methods Development
Operators have stress in not only working places but also their dailylives. We aim to develop measures and tools to evaluate the stress from operating machines by simulating the working environment considering social contexts.
Driving Automation
Driving automation is a promising technology, and it will change the conventional role of human drivers and ways of interaction between road users and vehicles. To provide insights for both drivers and vulnerable road users, such as pedestrians and cyclists, this research explores human factors problems related to automated vehicles (AVs) and how to address the problem with the feature of driving automation.
External Human-Machine Interfaces
External Human-Machine Interfaces (eHMI) are considered to be a tool for explicit communication between vulnerable road users and AVs. Given that vulnerable road users cannot communicate with human drivers by eye-contact or hand gestures, a new substitute for road safety and efficiecy becomes essential. The eHMIs are expected to provide AV's intention with road users, resulting in safe crossing. However, one concern related to the eHMI is that vulnerable road users may develop trust toward AVs, leading less allocation of cogtinive resources to traffic situation or increase in collision risk. This research aims to develop eHMIs which mitigates such side effect by clarifying effects and limits of eHMIs.
New Transportation System in Symbiotic Environment
Vehicles have featured several levels of driving automation from SAE L0 to L5. Still, people are not ready to accept the development of driving automation due to several reasons. This study aims to figure out public's perception of driving automation using a text-mining of newspapers from perspectives of both human factors and city planning. Words that imply traffic risk will be categoriezed in both positive and negative aspects, then new research problems tangled with transportation systems will be suggested using the framework of Social-Psychology-Technological Systems.
Driver Trust in Automated Vehicles
Driver trust in automated vehicles are shaped by a myriad of factors. This cross-cultural study aims to define driver trust in driving automation and develop new framework of driver-AV trust by exploring several dominant factors influencing trust in both Korean and Japanese drivers. The framework will address how driver trust changes before/after the use of driving automation.
Driver Situation Awareness
Conservative in-vehicle systems are expected to keep drivers on the loop in partial driving automation. The main research question is this: can the conversative system like human passengers improve driver situation awarenss, resulting in timely and appropriate take-over? The relation between drivers and passengers, frequency, amount, and contents are considered to be contributors to system design. This research has investigated three matters: relation, freqeuncy, and amount excpet for the contents and found that continuous oral communication with in-vehicle system leads decreases in driver fatigue and drowsiness.