COORDINATED PATH-FOLLOWING IN THE PRESENCE OF COMMUNICATION LOSSES AND TIME DELAYS
COORDINATED PATH-FOLLOWING IN THE PRESENCE OF COMMUNICATION LOSSES AND TIME DELAYS
复制标题
DOI：
10.1137/060678993
复制
发表时间：
2009-01-01
影响因子：
2.2
通讯作者：
Hespanha, J.
中科院分区：
文献类型：
--
作者：
Ghabcheloo, R.;Aguiar, A. P.;Hespanha, J.
本文讨论的问题，引导一组车辆沿着给定的空间路径，同时保持所需的时变几何形成模式。这个问题的解决方案，此后被称为协调路径跟踪（CPF）问题，展开在两个基本步骤。首先，路径跟踪（PF）控制律的设计，以驱动每个车辆到其指定的路径，与标称速度曲线，可能是路径相关的。这是通过使每个车辆接近一个虚拟目标来完成的，该虚拟目标根据方便定义的动态法则沿路径沿着移动。在第二步骤中，虚拟目标的速度（也称为协调状态）被调整到它们的标称值附近，以便同步它们的位置并间接地实现车辆协调。在问题的制定，它被明确认为是每个车辆传输其协调状态的一个子集的其他车辆只，所采用的通信拓扑结构确定。结果表明，由PF子系统和协调子系统组合而成的系统可以自然地看作是后两者的反馈或级联。利用这一事实和最近的结果从非线性系统和图论，条件下，PF和协调误差被驱动到一个附近的零存在的通信损耗和时间延迟。考虑两种不同的情况。第一个捕获通信图交替连接和断开的情况（短暂的连接丢失）。第二个反映了一个操作的情况下，工会的通信图在统一的时间间隔保持连接（均匀连接的意思）。为了更好地根在一个非平凡的设计实例中的文件，CPF算法推导出多欠驱动自主水下机器人（AUV）。仿真结果进行了介绍和讨论。
This paper addresses the problem of steering a group of vehicles along given spatial paths while holding a desired time-varying geometrical formation pattern. The solution to this problem, henceforth referred to as the coordinated path-following (CPF) problem, unfolds in two basic steps. First, a path-following (PF) control law is designed to drive each vehicle to its assigned path, with a nominal speed profile that may be path dependent. This is done by making each vehicle approach a virtual target that moves along the path according to a conveniently defined dynamic law. In the second step, the speeds of the virtual targets (also called coordination states) are adjusted about their nominal values so as to synchronize their positions and achieve, indirectly, vehicle coordination. In the problem formulation, it is explicitly considered that each vehicle transmits its coordination state to a subset of the other vehicles only, as determined by the communications topology adopted. It is shown that the system that is obtained by putting together the PF and coordination subsystems can be naturally viewed as either the feedback or the cascade connection of the latter two. Using this fact and recent results from nonlinear systems and graph theory, conditions are derived under which the PF and the coordination errors are driven to a neighborhood of zero in the presence of communication losses and time delays. Two different situations are considered. The first captures the case where the communication graph is alternately connected and disconnected (brief connectivity losses). The second reflects an operational scenario where the union of the communication graphs over uniform intervals of time remains connected (uniformly connected in mean). To better root the paper in a nontrivial design example, a CPF algorithm is derived for multiple underactuated autonomous underwater vehicles (AUVs). Simulation results are presented and discussed.