Prescribed Performance Path-Following Control for a Parafoil System Under Varying-Curvature Paths: Theory and Experimental Validation
This paper investigates the path-following control problem for flexible parafoil systems operating in complex wind-disturbed environments. The work aims to achieve accurate tracking of varying-curvature paths, thereby strengthening the potential for collision avoidance. In this paper, a high-fidelity 9-degree-of-freedom equivalent model is established, incorporating both external wind disturbances and internal multi-body coupling torques. Based on this model, an observer-based controller is designed to achieve asymptotic convergence of lateral tracking errors within predefined performance bounds. The primary contribution of this work is its capacity to explicitly prescribe both transient and steady-state performance of the lateral tracking error in a 9-DOF parafoil system—a critical capability for ensuring safe navigation in cluttered airspace. Unlike conventional parafoil control methods, the proposed strategy ensures predictable and constrained error behavior throughout the entire flight phase. The effectiveness and practicality of the approach are demonstrated through comprehensive simulations and hardware-in-the-loop experiments, confirming its viability for real-world parafoil mechatronic systems. Note to Practitioners—This paper was motivated by the problem of tracking desired varying-curvature paths for multi-body flexible parafoil systems under wind disturbances, while the proposed method is also applicable to path tracking control for 6-DOF rigid-body UAVs. Existing control methods for parafoil systems mostly cannot predefine the performance bounds of tracking errors during the initial control design phase and often rely heavily on extensive parameter tuning to improve control performance. To overcome this limitation, our solution combines a predefined performance controller, which explicitly enforces pre-defined error constraints, with two high-precision observers that actively compensate for wind effects and disturbance torques. This means practitioners can specify the acceptable tracking error bounds beforehand, and the controller will strive to maintain them. We have validated this method through ground-based, hardware-in-the-loop simulations. The results confirm that our approach is not only theoretically sound but also feasible for real-world applications. It simplifies the tuning process and enhances system reliability in the presence of disturbances. Future work will focus on flight tests to further demonstrate its practicality. Engineers interested in systems requiring guaranteed performance under environmental uncertainties may find this method particularly useful.