Dynamic Performance Enhancement of a High‐Agility Hypersonic Mechanical System via a Learning‐Augmented Robust Control Architecture

The design of control systems for high‐agility mechanical systems operating in extreme environments is a significant challenge in nonlinear dynamics. This paper addresses the complex dynamic control problem of a hypersonic interceptor, modeled as a multi‐input multi‐output (MIMO) mechanical system characterized by strong aerodynamic cross‐couplings, rapid parametric variations, and severe external disturbances. We propose a novel, adaptive and resilient integrated guidance and control (IGC) architecture designed to guarantee robust dynamic performance. The architecture integrates three core modules: (1) an adaptive‐gain continuous higher‐order sliding mode controller (ACHOSMC) that acts as the robust dynamic stabilizer, effectively suppressing chattering and ensuring stability; (2) a long short‐term memory (LSTM) network for predictive rejection of unstructured dynamic disturbances, such as those induced by electronic attacks; and (3) an Online Aerodynamic Coefficient Calibration (OACC) module for real‐time adaptation of the system's internal dynamic model. The proposed framework's impact on the system's dynamic response is rigorously validated through extensive simulations and processor‐in‐the‐loop (PIL) experiments. Quantitative results demonstrate a 91.2% reduction in terminal dynamic error (miss distance) and a 78% improvement in dynamic response time compared to a benchmark controller. Furthermore, PIL tests confirm the entire control algorithm executes in 885 µs, proving its feasibility for the real‐time control of fast mechanical systems. This work provides a validated, modular framework for enhancing the dynamic performance and resilience of next‐generation mechanical systems.