Adaptive Linear Quadratic Attitude Tracking Control of a Quadrotor UAV Based on IMU Sensor Data Fusion

Area of Science:

• Robotics
• Control Systems Engineering
• Aerospace Engineering

Background:

• Quadrotor unmanned aerial vehicles (UAVs) require precise attitude control for various applications.
• Real-world operations introduce uncertainties in system parameters and sensor measurements, challenging control performance.
• Existing control methods may struggle with dynamic parameter variations and sensor noise.

Purpose of the Study:

• To design and validate an infinite-horizon adaptive linear quadratic tracking (ALQT) control scheme for optimal quadrotor attitude tracking.
• To enhance the robustness of the control system against parametric uncertainties and sensor measurement noise.
• To compare the effectiveness of Kalman filtering and complementary filtering for sensor fusion in improving controller performance.

Main Methods:

• An infinite-horizon adaptive linear quadratic tracking (ALQT) control scheme was developed.
• An online least squares based parameter identification was integrated to estimate quadrotor inertia in real-time.
• Two sensor fusion techniques, Kalman filtering and complementary filtering, were implemented and compared.
• Experimental validation was performed under conditions with system parameter uncertainties and sensor noise.

Main Results:

• The ALQT control scheme demonstrated asymptotic stability for the closed-loop system.
• The integrated parameter identification effectively estimated instantaneous quadrotor inertia, enhancing robustness.
• Kalman filtering resulted in lower mean-square estimation error compared to complementary filtering.
• The Kalman filter-based approach yielded superior attitude estimation and control performance.

Conclusions:

• The proposed ALQT control scheme provides a robust and stable solution for optimal quadrotor attitude tracking.
• Online parameter identification is crucial for mitigating performance degradation due to system uncertainties.
• Kalman filtering is the preferred sensor fusion technique for improving attitude estimation and control accuracy in noisy environments.
• The study validates the practical applicability of the developed control strategy for quadrotor UAVs.