| 摘要 |
A method and system for closed-loop vehicle lateral control, using image data from front and rear cameras and information about a leading vehicle's position as input. A host vehicle includes cameras at the front and rear, which can be used to detect lane boundaries such as curbs and lane stripes, among other purposes. The host vehicle also includes a digital map system and a system for sensing the location of a vehicle travelling ahead of the host vehicle. A control strategy is developed which steers the host vehicle to minimize the deviation of the host vehicle's path from a lane reference path, where the lane reference path is computed from the lane boundaries extracted from the front and rear camera images and from the other inputs. The control strategy employs feed-forward and feedback elements, and uses a Kalman filter to estimate the host vehicle's state variables. |
