Feedback Estimators for Identifying Headway Distance and Velocity of Longitudinal Platooned Vehicles
Feedback Estimators for Identifying Headway Distance and Velocity of Longitudinal Platooned Vehicles
复制标题
用于识别纵向队列车辆车头距离和速度的反馈估计器
DOI：
10.11175/easts.10.1631
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
Hironori Suzuki and Takashi Nakatsuji
中科院分区：
文献类型：
--
作者：
公文俊平;大橋正和 編著;Hironori Suzuki and Takashi Nakatsuji
开发了一种动态反馈系统，用于估计纵向三车排的车头距离和速度。估计系统使用粒子滤波器（PF）和无迹卡尔曼滤波器（UKF）进行建模，通过测量排中探测车辆的加速度和/或速度来估计它们。状态方程被定义为车头距离和速度的离散守恒方程，而测量方程则基于传统的跟车模型。 UKF 和 PF 的优点是在实现滤波过程时避免一阶近似以提高估计精度。使用人工模拟数据以及真实跟车数据进行的数值分析表明，与扩展卡尔曼滤波器（EKF）或神经卡尔曼滤波器（NKF）等传统方法相比，PF 和 UKF 在大多数情况下减少了估计误差。这在车头时距估计中尤其重要，因为 EKF 估计的准确性较低。
A dynamic feedback system is developed for estimating the headway distance and velocity in a longitudinal three-vehicle platoon. The estimation system is modeled using a particle filter (PF) and an unscented Kalman filter (UKF) that estimate them by measuring the acceleration rate and/or velocity of probe vehicle (s) in the platoon. State equations are defined as a discrete conservation equation of headway distance and velocity, whereas the measurement equation is based on a conventional car-following model. The UKF and PF have the advantage of avoiding first-order approximation when implementing a filtering process to increase the estimation accuracy. Numerical analyses using artificial simulated data as well as real car-following data showed that the PF and UKF reduce the estimation errors in most cases compared to conventional approaches such as an extended Kalman filter (EKF) or neural Kalman filter (NKF). This was significant especially in the headway estimation, where the accuracy of the EKF estimates was low.