This paper presents an internal model-Kalman filtering-based optimal hybrid feedforward/feedback control strategy for traffic platoon control coordination enabled by SAE Automation Level 2 or Level 3 vehicles, i.e., partially automated vehicles (PAVs). Based on the Helly linear car-following model, a PAV platoon is established. Taking each vehicle’s characteristic polynomial as the dominant internal model polynomial, an augmented system state model with filtered inputs is formed, and then both the system states and the external disturbance/internal perturbation from previous vehicle can be estimated by a delicately designed Kalman filter. By using a linear quadratic regulation approach, a distributed hybrid optimal feedforward/feedback controller utilizing the local estimated states is constructed to achieve a practically decoupled platoon regulation around an equilibrium state. String stability of the control system thus obtained is further analyzed. Finally, extensive numerical simulations verified the theoretical analysis in decoupling between successive vehicles, in suppressing the influence of various external disturbances, and in maintaining string stability.

本文提出了一种基于内部模型卡尔曼滤波的最佳混合前馈/反馈控制策略，用于由 SAE 自动化 2 级或 3 级车辆（即部分自动驾驶车辆（PAV）实现的交通队列控制协调）。基于Helly线性跟驰模型，建立PAV排。以每辆车的特征多项式作为主导内部模型多项式，形成带有滤波输入的增强系统状态模型，然后可以通过精心设计的卡尔曼滤波器来估计系统状态和来自前一车辆的外部扰动/内部扰动。通过使用线性二次调节方法，构建了利用局部估计状态的分布式混合最优前馈/反馈控制器，以实现围绕平衡状态的实际上解耦的排调节。对由此获得的控制系统的串稳定性进行进一步分析。最后，大量的数值模拟验证了连续飞行器之间的解耦、抑制各种外部扰动的影响以及保持串稳定性的理论分析。