Fault-Tolerant Control for Quadcopters Under Actuator and Sensor Faults

Area of Science:

• Aerospace Engineering
• Control Systems
• Robotics

Background:

• Unmanned aerial vehicles (UAVs), especially quadcopters, face operational risks from sensor and actuator faults due to complex dynamics and environmental factors.
• Ensuring the safety, reliability, and cost-effectiveness of UAV operations necessitates robust fault detection and diagnosis (FDD) and fault-tolerant control (FTC) strategies.

Purpose of the Study:

• To implement and evaluate different Kalman filter (KF)-based FDD approaches for fault estimation in quadcopters.
• To achieve effective fault-tolerant control (FTC) for quadcopters experiencing nonlinear actuator and sensor faults, including simultaneous occurrences.

Main Methods:

• Implementation of three KF variants: linear KF, extended KF (EKF), and unscented KF (UKF).
• Inclusion of three-stage and adaptive variations of the KF to enhance fault estimation.
• Integration of FDD methods within an FTC architecture for quadcopter stabilization.

Main Results:

• Adaptive KF-based FDD methods demonstrated superior fault estimation performance in complex scenarios.
• The FTC architecture successfully maintained quadcopter stability despite sensor and actuator faults.
• Significant improvements in quadcopter safety and reliability were observed with the proposed FDD/FTC system.

Conclusions:

• Kalman filter-based FDD approaches, particularly adaptive variants, are effective for enhancing quadcopter fault tolerance.
• The developed FTC system provides a robust solution for maintaining stable UAV operation under fault conditions.
• This research contributes to safer and more reliable autonomous aerial vehicle systems.