Vehicle’s Turning Time Span Estimation Based on Linear and Non-linear Computational Turning Model

Abstract

The time span of a vehicle making turns, which is from the time a turn starts to the time the turn ends,  are the basis of determination of essential data of vehicle motion like slip angles, forces on tires, except the vehicle is in an ideal one-dimensional motion. When the road condition is sensed by sensors on vehicle’s driver assistance systems, how to take a turn is the next step that should be calculated. After the turn starts, the end time of the turn should be adjusted in real time to remedy the imperfection digital simulation of the road sensed by the sensors. This paper compares two turning models and proposes an algorithm to adjust the end time of the turning time span during the turning motion of the vehicle. Subsequently, the algorithm is applied to two classic models of car, BMW M3 and Chevrolet Cavalier in four common motions: U-turn, lane change, sharp U-turn and sharp lane change. The algorithm proposed can be applied to every scenario that should take use of the turning time length, including the mathematical model of crossings, traffic flow of rush hours, or simply estimate the time span in unmanned vehicle control systems.