Abstract
An autonomous driving system using GNSS and a high definition map can continue to drive autonomously even when the road condition are bad. However, under slippery road conditions such as on a snow-covered road, it is expected that the performance of tracking a target path deteriorates or in the worst case the vehicle could start spinning. Therefore, vehicle control to accurately follow the target path is important. In this study, integrated longitudinal and lateral vehicle control to follow the target path is formulated as a nonlinear model predictive control scheme using a nonlinear tire model and a single-track vehicle model. We prepared two types of tire models for asphalt and a snow-covered road, and simulated a double lane change on the snow-covered road. Furthermore, we implemented this controller in GNSS based autonomous driving system and its control performance is validated based on tests with an actual vehicle.
