A domestic research team has developed guidelines to help drivers safely take control of self-driving cars. This will greatly assist government agencies and automakers in establishing standards for autonomous driving control and incorporating safety features.

The Electronics and Telecommunications Research Institute (ETRI) announced on the 6th that it has developed guidelines and three related technologies to guide the safe transfer of control in autonomous vehicles by considering the driver's condition, and released a related dataset.

The Autonomous Vehicle Human Factors Guidelines present scenarios that could arise in various control transfer situations from the driver, vehicle, and environmental perspectives, making them easily understandable for both experts and the general public.

For example, if the car is in autonomous driving mode and is about to switch to manual driving, but an important phone call comes in, a decision must be made about what to do. The guidelines advise that even in situations where a transfer of control is expected, the driver's condition should be monitored and autonomous driving should be maintained through warnings to ensure preparation time.

These guidelines can help automakers and parts suppliers implement features for safe autonomous driving and help users correctly understand the control transfer process.

In particular, government ministries, including the Ministry of Land, Infrastructure and Transport, are expected to play a major role in increasing the acceptance of autonomous vehicles by referring to these guidelines and thereby increasing their utilization and effectiveness.

ETRI stated that it was able to develop detailed and practical guidelines by incorporating a high-quality dataset built through experiments involving 458 participants and over 1,500 control transfers, and also by referencing foreign guidance from the US, Europe, and Japan. Thanks to this research, recommendations developed based on this research were adopted as standards at the ITS General Assembly.

The research team also developed three different technologies to assist in the transfer of control in autonomous vehicles. First, the DVE (Driver, Vehicle, Environment) monitoring system monitors and analyzes driver, vehicle, and driving environment information in real time to predict the driver's readiness to take control.

Additionally, the Co-Pilot Agent (CPA) is a technology that helps drivers and autonomous vehicles communicate by providing biometric information or warnings to enable drivers to quickly and safely take over driving control.

We also developed a system to evaluate the AI control transfer capability of autonomous vehicles. This is a technology that utilizes AI to monitor and analyze driving readiness, driving load, and control transfer reaction time to evaluate the driver's situational awareness, perception, physical response, driving skills, and situational judgment ability in real time.

ETRI Cognitive and Transportation ICT Research Lab Director Dae-Seop Yoon said, “The guidelines, technology, and data developed this time will not only help ensure safe autonomous driving, but will also greatly contribute to increasing trust in autonomous vehicles and revitalizing the related industrial ecosystem.”

The guidelines developed by the research team will be distributed to relevant organizations and manufacturers, and can also be accessed through the ETRI Knowledge Sharing Platform (https://ksp.etri.re.kr/ksp/plan-report/read?id=828). Details on the three developed technologies and datasets can be found on the ETRI Technology Transfer website.

ETRI was able to achieve this result through in-depth research on human factors in the transfer of control of autonomous vehicles over a four-year period as part of a transportation and logistics research project supported by the Ministry of Land, Infrastructure and Transport and the Korea Agency for Infrastructure Technology Advancement (KAIT) since 2017.

Through related research, the research team published 53 papers, including four SCI/SCIE-level papers, and filed 16 patents. Two technology transfers have been completed, and one additional technology transfer is in progress.