对未知变化风场下固定翼无人机编队的三维路径跟踪控制问题，基于单机路径跟踪与编队协同双层框架进行了研究。对单机路径跟踪问题，提出了一种基于预定义时间预测器的视线制导律，能够估计由风扰引起的水平与垂向偏流角，并保证路径跟踪误差与偏流角预测误差在设计者预先设定的时间内实现有界收敛。对于编队协同问题，基于规定时间控制方法，设计了分布式观测器与一致性控制律，使得所有跟随者的路径参数严格自规定时刻起与虚拟领导者实现同步，从而为各跟随者提供准确的参考位置与速度信息。此外，针对无人机速度受限的实际情况，引入了抗饱和辅助系统，从而在底层动态饱和时仍然保证系统的预定义时间稳定性。通过在多种轨迹与风场条件下的数值仿真对比实验和多机编队实验，充分验证了所提控制策略的有效性与优越性。

This paper addresses three-dimensional path-following control for fixed-wing unmanned aerial vehicle formations under unknown varying wind with a two-layer guidance and coordination framework. For the guidance layer, a predefined-time predictor-based line-of-sight guidance law is proposed, which estimates horizontal and vertical crab angles induced by wind, and guarantees bounded convergence of both path-following errors and crab angle predictor errors within a designer-chosen time. For the formation coordination layer, based on the novel stipulated-time method, a distributed observer and a consensus law are designed to synchronize the path parameters of all followers with a virtual leader at a precisely prescribed instant, providing each follower with accurate reference position and velocity from that instant onward. Given that the reference speed may exceed physical speed constraints before the convergence instant, an anti-saturation auxiliary system is incorporated to preserve the predefined-time stability guarantee. Comparative simulations against multiple baseline line-of-sight guidance methods and integrated multi-UAV formation tests under various paths and wind conditions validate the proposed approach.