Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Gain-Scheduled Passive Fault-Tolerant Control Design for Dual-System UAV Transition Flight

Junfeng Cai, Marco Lovera

TL;DR

This work tackles fault-tolerant control for dual-system UAVs during transition flight by introducing a gain-scheduled passive fault-tolerant control (GS SHIF PFTC) that treats actuator faults as multiplicative input uncertainties within a structured $H_{\infty}$ framework. A multi-model approach with six discrete, airspeed-dependent design points is employed, enabling robust synthesis via fixed-structure $H_{\infty}$ optimization and reducing the design burden compared to prior self-scheduled methods. The resulting controllers are validated in a nonlinear 6-DOF simulator under multiple fault scenarios, where GS SHIF provides superior attitude tracking and robustness relative to LQR and SHIF approaches, and does not rely on fault-detection and diagnosis. Overall, the method enhances safety and reliability for transitioning dual-system UAVs under actuator faults and parametric uncertainties, with potential applicability to other over-actuated, uncertain aerospace systems.

Abstract

Dual-system UAVs with vertical take-off and landing capabilities have become increasingly popular in recent years. As a safety-critical system, it is important that a dual-system UAV can maintain safe flight after faults/failures occur. This paper proposes a gain-scheduled passive fault-tolerant control (PFTC) method for the transition flight of dual-system UAVs. In this novel FTC design method, the model uncertainties arising from the loss of control effectiveness caused by actuator faults/failures, for the first time, are treated as model input uncertainty, allowing us to use multiplicative uncertainty descriptions to represent it. The advantages of the proposed method consist in significantly reducing the number of design points, thereby simplifying the control synthesis process and improving the efficiency of designing the FTC system for dual-system UAV transition flight compared with the existing FTC design methods. As a general method, it can be applied to the design of FTC systems with multiple uncertain parameters and multiple channels. The developed passive FTC system is validated on a nonlinear six-degree-of-freedom simulator. The simulation results demonstrate that the gain-scheduled structured H infinity (GS SHIF) PFTC system provides superior fault tolerance performance compared with the LQR and structured H infinity control systems, thereby showcasing the effectiveness and the advantages of the proposed GS SHIF PFTC approach.

Gain-Scheduled Passive Fault-Tolerant Control Design for Dual-System UAV Transition Flight

TL;DR

This work tackles fault-tolerant control for dual-system UAVs during transition flight by introducing a gain-scheduled passive fault-tolerant control (GS SHIF PFTC) that treats actuator faults as multiplicative input uncertainties within a structured framework. A multi-model approach with six discrete, airspeed-dependent design points is employed, enabling robust synthesis via fixed-structure optimization and reducing the design burden compared to prior self-scheduled methods. The resulting controllers are validated in a nonlinear 6-DOF simulator under multiple fault scenarios, where GS SHIF provides superior attitude tracking and robustness relative to LQR and SHIF approaches, and does not rely on fault-detection and diagnosis. Overall, the method enhances safety and reliability for transitioning dual-system UAVs under actuator faults and parametric uncertainties, with potential applicability to other over-actuated, uncertain aerospace systems.

Abstract

Dual-system UAVs with vertical take-off and landing capabilities have become increasingly popular in recent years. As a safety-critical system, it is important that a dual-system UAV can maintain safe flight after faults/failures occur. This paper proposes a gain-scheduled passive fault-tolerant control (PFTC) method for the transition flight of dual-system UAVs. In this novel FTC design method, the model uncertainties arising from the loss of control effectiveness caused by actuator faults/failures, for the first time, are treated as model input uncertainty, allowing us to use multiplicative uncertainty descriptions to represent it. The advantages of the proposed method consist in significantly reducing the number of design points, thereby simplifying the control synthesis process and improving the efficiency of designing the FTC system for dual-system UAV transition flight compared with the existing FTC design methods. As a general method, it can be applied to the design of FTC systems with multiple uncertain parameters and multiple channels. The developed passive FTC system is validated on a nonlinear six-degree-of-freedom simulator. The simulation results demonstrate that the gain-scheduled structured H infinity (GS SHIF) PFTC system provides superior fault tolerance performance compared with the LQR and structured H infinity control systems, thereby showcasing the effectiveness and the advantages of the proposed GS SHIF PFTC approach.

Paper Structure

This paper contains 19 sections, 12 equations, 14 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. The Dual-system VTOL UAV
  4. Control-oriented Model
  5. Roll Dynamics
  6. Pitch Dynamics
  7. Yaw Dynamics
  8. Actuator Redundancy Management
  9. Fault-tolerant Control Design Using a Multi-model Approach
  10. Mixed Sensitivity Optimization Based on Structured $H_{\infty}$
  11. Descriptions of the Perturbed Models
  12. Control Synthesis
  13. Robustness Analysis
  14. Stability of the Nominal Systems
  15. Robust stability and performance
  16. ...and 4 more sections

Figures (14)

  • Figure 1: The top view of the dual-system UAV
  • Figure 2: The general tuning block diagram of structured $H_{\infty}$
  • Figure 3: The multiplicative uncertainty weight, $\bar{V_{3}}$ = 4 m/s
  • Figure 4: The perturbed model with multiplicative uncertainty
  • Figure 5:
  • ...and 9 more figures