Fault Tolerant Control of UAV

Report Link:- Report

Code Link:- Github

Position and Altitude Control of Quadrotor with Single Motor Failure

Abstract

This project develops a robust control algorithm for quadrotors to maintain stability during motor failures, crucial for UAV safety in surveillance and mapping applications. It includes a fault detection algorithm based on a quasi-linear Parameter Varying (qLPV) observer and Fault Tolerant Control (FTC) methods. The system ensures stable flight by adjusting the remaining motors and supporting controlled landing, stable hover, and basic navigation.

Challenges

1. Failure Detection: Quick, real-time identification of motor failure.
2. Control Adjustments: Real-time motor adjustments to maintain stability after failure.

Solution

Using the IRIS quadrotor model in Gazebo simulator, a PX4 flight controller was developed to:

• Detect motor failure using qLPV PIO.
• Transition to Fault-Tolerant Control (FTC), using control allocation to adjust motor contributions and maintain stability.

Results

• Fault Detection: The system detected motor failures in 417 ms and maintained control in various conditions.
• Fault-Tolerant Control: Successfully kept the quadrotor stable during a motor failure and guided it to a safe landing without crashes.

Conclusion

This project demonstrates an effective fault detection and fault-tolerant control system for quadrotors, ensuring stability and safe operation despite motor failures. Further testing is planned to enhance the system’s robustness under diverse conditions.