The academic community has extensively studied fault management in dynamical and cyber-physical systems, leading to the development of various model-based and data-driven/learning-enabled methods. Although these advanced designs show promise for improving conventional practices in aircraft systems, there is a noticeable disparity between academic methodologies and the specific needs of the aviation industry. The paper begins with an examination of the current practices within the aviation industry alongside the academic state of the art. It highlights commonly overlooked issues that hinder the transition from laboratory development to practical flight applications. Looking ahead, the paper anticipates evolving needs driven by the transition towards greater autonomy and intelligence within connected and distributed cyber-physical flight environments. This includes the emerging trend towards the introduction of Single Pilot Operations (SPO). The paper presents an outline of a combined model-based/data-driven vision, under human oversight, to navigate this complex transition.

中文翻译：

---

飞机系统故障管理问题综述：现状和未来方向

学术界广泛研究了动态和网络物理系统中的故障管理，导致了各种基于模型和数据驱动/学习支持的方法的发展。尽管这些先进的设计有望改善飞机系统的传统实践，但学术方法与航空业的具体需求之间存在明显的差异。本文首先考察了航空业当前的实践以及最新的学术水平。它强调了阻碍从实验室开发到实际飞行应用过渡的常见被忽视的问题。展望未来，本文预计，在互联和分布式网络物理飞行环境中向更大程度的自主性和智能化过渡所驱动的不断变化的需求。这包括引入单一飞行员操作（SPO）的新兴趋势。本文概述了在人类监督下基于模型/数据驱动的组合愿景，以应对这一复杂的转变。