Study on Test Method for Takeover Time of L3 Automatic Vehicles



3rd International Forum on Connected Automated Vehicle Highway System through the China Highway & Transportation Society
Authors Abstract
According to SAE’s classification of autonomous driving, L3 systems are conditional autonomous driving. The responsibility of vehicle control and environmental monitoring are systems. But when the system requires the driver to respond its request, the driver must respond soon, and if an accident occurs, the responsibility is the driver. So, it requires the driver have ability to take over the vehicle all the time. Therefore, the length of the driver’s takeover time in different scenarios is becoming more and more important. The system needs to send a takeover signal at a certain time advance according to the driving environment, so that the driver has time to complete the takeover of the system. This article describes the takeover control flow chart of the L3 autopilot systems, and the factors that affect the takeover time, such as the time when the takeover signal is issued, the form of takeover signal, and the state of the driver. A test method for the takeover time at different signal moments is proposed, it includes the test tools, the test scenario construction, and takeover time analysis methods are described in detail, and a detailed test flow chart is given. Based on the driving simulator, the takeover time of eight drivers at different signal timings was tested. The results indicate that the later the alarm signal is sent, the less time the driver takes to take over the vehicle. The test result of the takeover time provides a reference for the design such as the time when the takeover is issued, and the entire test method for the takeover time also provides a reference for testing other factors affecting the takeover.
Meta TagsDetails
Shi, J., Xu, L., Qin, K., Xu, Z. et al., "Study on Test Method for Takeover Time of L3 Automatic Vehicles," SAE Technical Paper 2020-01-5144, 2020,
Additional Details
Dec 30, 2020
Product Code
Content Type
Technical Paper