Driver Takeover

Research Title: Development of Driver Risk Scenario and Response Technologies for Autonomous Vehicles Control Takeover Verification

Sponsor

• Namyang R&D Center of HMG

Research Period

• Total Period: 2021.05.10 - 2022.05.09

• Participation Period: 2021.09.01 - 2022.05.09

Role

• Practitioner (Contribution: 30%)

• Participated as a master’s student at Kookmin University

Keywords

• Take-over request (TOR), availability, alert, eye tracking, misuse

Research Goal

• Development of a mental model framework based on driver availability for proper take-over during Take-over Requests (TOR) of Level 3 autonomous driving systems.

Research Introduction

 This research aims to develop a take-over framework considering the driver’s mental model to ensure safe take-over during Take-over Requests (TOR) of Level 3 autonomous driving systems. In Level 3 autonomous driving, the system fully controls the vehicle, and the driver is not required to hold the steering wheel or keep their eyes on the road.

 However, during driving, the driver must be ready to take over promptly if the system issues a Take-over Request (TOR). Recently, cases of misuse such as over-reliance on the system, misunderstanding its operational limits, headphone use, and sleeping have been reported. These factors can prevent the driver from responding promptly to a Take-over Request (TOR).

 Therefore, in this research, a vehicle simulator equipped with an eye tracker was used to monitor driver postures that hinder quick take-over (e.g., crossing legs, reclining, closed eyes) based on eye-tracking data. The driver’s mental model in response to ADS alerts was analyzed using the STPA method. As a result, scenarios requiring availability warning alerts were derived through the analysis of the driver’s mental model.

Research Achievements

• 1 International Conference Presentation (First Author)

• 1 Domestic Conference Presentation (First Author) 

• 1 Registered Domestic Patent (First Inventor)