A key function of mobile networks is the ability to dynamically reshape itself to any desired geometry. Lacking absolute position awareness, agents often rely on distance-limited inter-agent spatial measurements to maintain state awareness. Methods of formation control must therefore ensure a minimal level of persistent pairwise measurement feedback throughout transition, giving rise to the classic connectivity maintenance problem. To address this problem, we propose a method of structure-preserving assignment, matching agents to desired positions such that persistent global connectivity is naturally and automatically satisfied under smooth transition. Compared to other approaches, this complementary technique reduces reliance on aggressive or costly mid-flight formation control protocols. The technique is shown to scale and even improve with network size.

移动网络的一个关键功能是能够动态地将自身重塑为任何所需的几何形状。由于缺乏绝对位置感知，智能体通常依赖距离受限的智能体间空间测量来维持状态感知。因此，编队控制方法必须确保在整个过渡过程中持续存在最低水平的成对测量反馈，从而引发经典的连接维护问题。为了解决这个问题，我们提出了一种保留结构的分配方法，将智能体匹配到所需的位置，以便在平滑过渡下自然且自动地满足持久的全局连接性。与其他方法相比，这种补充技术减少了对激进或昂贵的飞行中编队控制协议的依赖。该技术被证明可以随着网络规模的扩大而扩展甚至改进。