夜行车辆的主动防撞预警模型仿真研究

Simulation of Active Collision Avoidance Warning Model for Nocturnal Vehicle

在对夜行车辆主动防撞预警进行研究时,由于周围环境及路面不可预测,使得夜行车辆采集的信息存在模糊性,传统的建模算法,主要采用感知周围事物的方法进行建模,没有考虑到在实际的夜间行驶时车辆的动态制动减速信息对建模造成的干扰,存在建模误差大的问题,提出采用改进卡尔曼滤波算法的夜行车辆的车辆主动防撞预警建模方法。对夜行环境及路面类型进行动态识别,确定所得到的动态制动减速度,分析前后两辆车的精确制动时间和车辆制动的过程,建立不同状态下的夜间临界跟车距离模型,获取模型中关键的夜行车辆行驶信息并进行预处理,再将信息传输至融合中心,利用卡尔曼滤波器对信息进行融合,获取安全距离值,建立精确的夜行车辆的车辆主动防撞预警模型。仿真结果证明,改进卡尔曼滤波的夜行车辆的车辆主动防撞预警建模方法能够准确预报夜间车辆行驶的安全状态。

A modeling method of active anti - collision warning for nocturnal vehicle is proposed based on the im- proved Kalman filter. Dynamic identification of the environment at night and road types is carried out, and the ob- tained dynamic braking deceleration is determined. The precise braking time and vehicle braking process of the two vehicles in the front and back are analyzed, the critical following distance model of nocturnal vehicle in different state is established, and key traveling information of nocturnal vehicle in this model is obtained and pretreatment is made. Then, the information is transmitted to the fusion center and is fused based on Kalman filter, a safe distance value is obtained, and precise active anti - collision warning model for nocturnal vehicle is established. Simulation results verify that the proposed method can accurately forecast the safety status of night driving.