| 研究実績の概要 |
This research focuses on modeling the lane change behavior of individual drivers and designing a guidance-as-needed steering system through haptic interface. At the beginning of this research project, a novel lane change model that takes account of different driving styles was developed by analyzing driver behavior data collected from driving simulator experiments. After that, an intention-based haptic guidance steering system was designed by real-time measuring vehicle sensory data and driver gaze behavior, which shows its effectiveness for both lane keeping and lane changing tasks. The experimental results indicate that the adaptive shared control system reduced the driver-automation conflict and decreased the lane departure risk in the lane keeping tasks and could support a fast and stable lane changing maneuver. Furthermore, an adaptive haptic guidance steering system by real-time monitoring driver arm muscle activity was designed and evaluated in lane change tasks. From the experimental results, the proposed system yielded lower driver workload and reduced the lane departure risks compared to manual driving and shared control with fixed authority. Due to COVID-19, the project was extended for one year, in which a driving simulator experiment with higher demand driving tasks and a real-vehicle experiment were conducted. The experimental results show that, compared to a one-size-fits-all interface, the developed guidance-as-needed interface by taking account of individualization is capable to improve driving safety and comfort as well as driver acceptance.
|
