Review of model predictive control and its applications in aircraft systems

Advanced and efficient aircraft typically refer to a category of vehicles that exhibit significantly enhanced flight performance, high operational efficiency, and environmental friendliness, aided by the latest technologies and design philosophies. Such aircraft integrate innovations from fields including advanced aerodynamic design, novel materials, and intelligent control systems, aiming to improve overall efficiency, safety, and sustainability. Among the various technical components, the control system serves as the central command unit. It is capable of regulating the aircraft’s attitude and trajectory in real time, ensuring stable and accurate tracking of desired paths even in complex environments, thus constituting an indispensable technological element of advanced efficient aircraft. Sophisticated control systems can optimize the control processes, leading to comprehensive improvements in energy consumption, flight efficiency, and safety.

This paper conducts a systematic investigation into four types of flight platforms: quadrotors, helicopters, fixed-wing aircraft, and high-speed vehicles. Each category operates in distinct flight environments and presents specific control requirements, necessitating the adoption of corresponding model predictive control strategies. Specifically, quadrotors face control challenges primarily due to strong system nonlinearities and external disturbances. Robust model predictive control can be employed to enhance disturbance rejection, while Lyapunov-based model predictive control helps ensure flight stability. Fixed-wing aircraft, often deployed in long-endurance missions, must cope with dynamically changing environmental conditions. Robust model predictive control is suitable for handling such perturbations, and explicit model predictive control can be applied to optimize flight trajectories. Helicopter systems encounter challenges related to multi-mode flight transitions. Switched model predictive control offers an effective approach to achieve smooth switching between different operational modes. High-speed vehicles operate in the most complex flight environments, where control involves multiple coupled factors such as trajectory planning, external disturbances, multi-mode switching, and stability assurance. Consequently, a combined design integrating various model predictive control methods is generally required.

Model predictive control and its applications in aircraft systems remain a prominent research focus. With the ongoing development of novel aircraft, research interest in model predictive control for flight systems is expected to persist, likely giving rise to further investigative topics. Currently, model predictive control techniques for low-speed aircraft are relatively mature. In contrast, model predictive control methods suitable for high-speed vehicles still require further research and development. High-speed vehicles represent a typical class of coupled hybrid systems, operating across flight regimes ranging from subsonic to supersonic speeds. They must transit smoothly through multiple flight phases—such as boost, acceleration, cruise, and re-entry—across a wide range of Mach. Throughout different mission stages, the engine operates in multiple thrust modes, inevitably involving discrete switching between operational states. Moreover, as the vehicle traverses varying Mach, the aerodynamic environment changes drastically. The interplay between aerodynamic characteristics and engine thrust becomes significant, making the coupling between aerodynamics and propulsion a critical factor that cannot be overlooked.

For high-speed vehicle control, a significant gap persists between model predictive control theory and practical implementation. Consequently, there is a pressing need to develop a unified model predictive control framework. This requires deeper integration of various algorithms—including robust model predictive control, Lyapunov-based model predictive control, switched model predictive control, and explicit model predictive control—to propose innovative solutions that bridge the gap between ideal performance targets and practically achievable limits. Research on control problems during transonic and supersonic flight of high-speed vehicles must address the following key scientific questions: how to construct a multi-physics coupled model that accurately represents cross-regime flight conditions; how to design a performance-guaranteed control strategy that ensures—in a computationally tractable form—the vehicle remains within a safe operational envelope while maintaining stable operation; and how to achieve safe switching between different flight modes, providing sufficient conditions for switching stability to guarantee recursive feasibility and stable convergence throughout the transition process. These areas constitute critical directions for future investigation.