车辆队列在保证安全性的基础上采用几何构型紧凑的跟驰策略，可以提高交通效率，减少能源消耗，是智能交通的重要发展方向。现有研究涉及的通信拓扑结构单一，未考虑建模不确定性和通信扰动的影响，难适用于复杂多变的交通环境。针对这些问题，本文基于四元素构架车辆队列模型，研究了多型通信拓扑结构下车辆队列的内稳定性和鲁棒性，提出了计算量独立于队列规模的控制器设计方法，为多型通信拓扑结构下车辆队列的分布式控制系统设计和稳定性分析奠定了基础。首先对具有非线性节点动力学的车辆队列进行动力学建模。将车辆系统分为上、下层，下层系统通过反馈线性化技术以获得带惯性延迟的线性车辆模型，上层系统采用分布式控制律以保持期望的队列几何构型。将通信拓扑建模为有向图，以拉普拉斯阵描述队列成员车之间的信息交互关系，从而建立起包含节点动力学、队列几何拓扑、通信拓扑和分布式控制律的四元素高维车辆队列模型。然后对多型通信拓扑动力学耦合下的四元素高维车辆队列模型进行闭环内稳定性分析。将高维度车辆队列的稳定性问题转化为低维度子模态的稳定性问题，在实数域中给出了一般通信拓扑下车辆队列系统的低维度稳定性充要条件。通过Riccati不等式将通信拓扑结构与控制器的设计相解耦，以通信拓扑阵特征值对线性矩阵不等式可行域的影响表征通信拓扑结构对车辆队列内稳定性的影响，使得控制器求解计算量独立于队列规模。其次分析了匀质/异质参数摄动下车辆队列的内稳定性条件。基于所提出的控制器设计方法，证明了时变匀质参数摄动下车辆队列的稳定性取决于时间常数取最大值时队列系统的稳定性。通过将异质参数摄动表达为范数有界形式，结合车辆模型的结构特点，给出了控制器所能镇定的异质参数摄动区间。为适应复杂的交通环境，进一步讨论了异质通信时延和随机通信拓扑切换对车辆队列内稳定性的影响。给出了异质通信时延上界，所设计控制器能够保证不大于该上界的异质通信时延下车辆队列系统的内稳定性。一般通信拓扑切换条件下，若子通信拓扑均具有有向生成树，且平均驻留时间不小于本文所提出的下界，则队列系统内稳定性得到保证。而对称通信拓扑切换条件下，车辆队列系统的内稳定性需要通信拓扑在有限时间内具有联合生成树。最后，开展了基于动态模拟试验台的车辆队列试验研究。试验结果表明，所设计的控制器在匀质/异质参数摄动下、异质时延下以及通信拓扑切换情况下均能保证车辆队列系统的鲁棒稳定性。

Platooning has the potential to increase traffic efficiency and decrease energy consumption without sacrificing safety due to the empoloyment of wireless communication, thus has been a promising research topic in the field of ITS (Intelligent Transportation Systems) since last century. Existing works on platoon control usually focus on simple topologies without considering uncertainty, which is difficult for real implementation. To address the issue, a high-order linear platoon model based on four-component framework is established. Internal stability and robust stability are investigated, resulting in Riccati inequality based controller synthesis algorithm whose computation complexity is independent of platoon size.A hierarchical structure is employed for the vehicle control system with the lower layer using feedback linearization technique aimed at achieving a linear node dynamic, and the upper layer using distributed control law aimed at achieving a desired platoon formation. Communication topology is modeled by algebraic graph theory in which the informnation exchange between different vehicles are described by the Laplacian matrix, leading to the high-order linear platoon model including four-component: node dynamics, platoon formation, communication topology and distributed control law.Based on the proposed platoon model, internal stability analysis is applied. The stability analysis of high-order platoon system is transformed into the stability analysis of low-order modes, based on which a sufficient and necessary condition of internal stability is proposed in real number field. Combining the Riccati inequality and the structure of the vehicle model, a controller synthesis algorithm whose computation complexity is independent of platoon size is developed, in which the influence of communication topology on platoon internal stability is represented by the influence of topology eigenvalue on the feasibility of linear matrix inequality.Robust stability subjected to parameter uncertainty is further investigated. As far as homogeneous uncertainty is concerned, the platoon is robustly stable if the closed loop internal stability is guaranteed when the time constant of the vehicle model reaches its maximum value. The robust stability analysis concerning heterogeneous uncertainty is transformed into a H_∞ problem in which the heterogeneous uncertainty is bouded by a H_2 norm. Then the heterogeneous uncertain region is provided within which the robust stability can be guaranteed.To be more applicable, robust stability analysis with respect to communication disturbance is carried out. An upper bound is provided for the communication delay, within which the platoon can be stabilized by the proposed controller. For symmetric topology switching, joint spanning tree is required in order to stabilize the platoon system. And for generic communication topology, a minimum average dwell time is demanded to guarantee the platoon internal stability.At last, experiments based on a driving simulator are carried out to verify the developed methods. Experiment results demonstrate that the proposed controller can stabilize the platoon system under parameter uncertainty and communication disturbance.