基于主车和目标车辆动力学的前车转向角估计

Steering Angle Estimation for Preceding Vehicles Considering Host-target Vehicle Dynamics

转向角是表征目标车辆侧向运动的基本要素,可以被用于辨识车辆换道等行为。文中提出了一种利用由相机、激光雷达、毫米波雷达等传感器获取的信息估计目标车辆转向角的新方法,基于车辆动力学和道路约束建立了一个目标车辆侧向运动模型,该模型可以有效表征目标车辆侧向运动。基于该模型,本文采用卡尔曼滤波器完成了对目标车辆转向角的估计,并利用仿真分析验证了模型在直道和弯道情况下的准确性。结果显示,基于该模型可以准确地对目标车辆转向角进行估计,并可以在将来应用于对目标车辆驾驶行为的辨识。

Steering angle is an essential parameter tightly linked to the lateral motion of the target vehicle that could be used to characterize the vehicle behaviors such as lane changing.This paper presents a new method for estimating the target vehicle steering angle using the information obtained from sensors such as the camera,light detection and ranging(LiDAR)and radar.A complete lateral motion model based on vehicle dynamics and road constraints is constructed and can be applied to interpret the target vehicle lateral motion.Kalman filter is adopted to estimate the steering angle of the target vehicle and simulations are carried out both in straight and curve roads.The results show that this model established in this article can be used to realize accurate estimate of the target vehicle steering angle.The proposed model can be applied for better recognition of target vehicle behaviors.