Distributed Platoon Control Under Topologies With Complex Eigenvalues: Stability Analysis and Controller Synthesis
Distributed Platoon Control Under Topologies With Complex Eigenvalues: Stability Analysis and Controller Synthesis
复制标题
DOI：
10.1109/tcst.2017.2768041
复制
发表时间：
2019-01-01
影响因子：
4.8
通讯作者：
Zhang, Hongwei
中科院分区：
文献类型：
--
作者：
Li, Shengbo Eben;Qin, Xiaohui;Zhang, Hongwei
自动驾驶车辆的队列行驶可以极大地改善道路交通。以前大多数关于排控制的研究只关注特定的通信拓扑，特别是那些具有真实特征值的拓扑。本文将分布式排控制的现有研究扩展到具有复杂特征值的更通用的拓扑，包括内部稳定性分析和线性控制器综合。使用逆车辆模型补偿导出线性队列动力学，并采用图论对通信拓扑进行建模，从而形成闭环队列动力学的集成高维线性模型。利用相似变换，推导了内部稳定性的充分必要条件，该条件完全定义在实数域中。然后，我们提出了一种基于 Riccati 不等式的算法来计算可行的静态控制增益。此外，扰动传播被公式化为 H 无穷大性能，并且使用 Lyapunov 分析明确导出间距误差的上限。非线性车辆模型的数值模拟验证了所提出方法的有效性。
The platooning of autonomous vehicles can significantly benefit road traffic. Most previous studies on platoon control have only focused on specific communication topologies, especially those with real eigenvalues. This paper extends existing studies on distributed platoon control to more generic topologies with complex eigenvalues, including both internal stability analysis and linear controller synthesis. Linear platoon dynamics are derived using an inverse vehicle model compensation, and graph theory is employed to model the communication topology, leading to an integrated high-dimension linear model of the closed-loop platoon dynamics. Using the similarity transformation, a sufficient and necessary condition is derived for the internal stability, which is completely defined in real number field. Then, we propose a Riccati inequality based algorithm to calculate the feasible static control gain. Further, disturbance propagation is formulated as an H-infinity performance, and the upper bound of spacing errors is explicitly derived using Lyapunov analysis. Numerical simulations with a nonlinear vehicle model validate the effectiveness of the proposed methods.