Abstract
In open traffic environments, humans still have to remain in the control loop of vehicle due to the insufficient of the existing technologies and their high costs. For the realization of cooperation between the human and the automatic driving system, the determination of the time when automatic driving is necessary is very important. To avoid unnecessary intervention when the driver has the control authority of vehicle, a new driving capability-based transition strategy was proposed, which comprehensively considers the driver’s correction ability and the driving risk. The transition time from the human driver to the automatic driving system is determined by an unreliable domain (UD), whose boundary is modeled according to the driving data recorded by a driving simulator and statistically described by a log-normal distribution. Furthermore, an adaptive algorithm is designed to update the parameters of UD boundary online to make this strategy suitable for different drivers. This UD-based transition strategy is validated by several tests on the driving simulator. The bench test results show that the individual driving characteristic can be identified by the adaptive algorithm in time, the transition time determined by UD is more accurate, and sufficient time is reserved for the correction carried out by the automatic driving system.
Highlights
Traffic safety has always been an area of social concern, especially with the ever-increasing number of vehicles [1]
Automatic driving is preferred under the condition that the correction ability of driver is not enough and the collision risk is comparatively high
It should be considered that the driver’s behavior is less certain and less consistent especially under critical and abnormal scenarios. This is bad for the convergence and stability of training process, so the data recorded under normal driving conditions by the driving simulator is used to determine the boundary of unreliable domain (UD)
Summary
Traffic safety has always been an area of social concern, especially with the ever-increasing number of vehicles [1]. To enhance the adaptability and robustness of detection for the driver’s distraction, Enache et al fused the information of steering torque with the posture of driver body [25] This cooperative system for steering control is applicable for different vehicle speeds and a variety of roads. Sentouh et al was already aware of this when developing the switching logic and suggested a mandatory three-second delay to prevent such false interventions [27] Another one is that driver state is not entirely equivalent to the driver’s capability of keeping the vehicle in safe driving conditions. Automatic driving is preferred under the condition that the correction ability of driver is not enough and the collision risk is comparatively high Considering these fundamentals, the following switch-based cooperative logic is designed: xd ∈ , Automaticdriving.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
