Lane tracking control in vehicle-following collision situations

This paper describes work on the subject of vehicle-following collisions and the implementation of a controller to maintain the vehicles in the desired trajectory. The goal of the study is to evaluate the feasibility of lane-tracking control systems in vehicle-following collisions. The effectiveness of such systems implies the potential for an automated or a driver-assistance function that can help maintain the paths of vehicles in collisions. This study was conducted with a combined model of collision and vehicle dynamics, SMAC (simulation model of automobile collisions). SMAC allows the longitudinal and lateral movements as well as the yaw motion about the vertical axis of vehicles on a horizontal plane. A controller is incorporated into the model to allow steering inputs. A feedback controller was selected to perform the lane tracking function. The control law corrects the current and foreseen deviation from the desired position. The scenarios evaluated in this paper include straight and curved road cases. Besides the basic controller, a gain compensator was introduced to reduce steady state tracking errors in curved roads. Simulations were also conducted to evaluate wheel offset caused by collision damage and time delay in sensing or actuation cycle. The controller was found to be effective. The feedback controller utilized illustrates a simple yet robust control law that can be applied to a wide range of conditions. The result is significant because a robust controller is essential to such applications when a system are open to uncertainties and variations in operation parameters