Adaptive Second-Order Strictly Negative Imaginary Controllers Based on the Interval Type-2 Fuzzy Self-Tuning Systems for a Hovering Quadrotor With Uncertainties

Abstract

We introduce a new adaptive control technique based on the second-order strictly negative imaginary (SNI) controller, coupled with the interval type-2 fuzzy self-tuning mechanism. We demonstrate the performance of our controllers in the three control loops of the AR.Drone quadrotor to achieve stable and balanced hover performance in flights with a variable Center-of-Gravity (CoG). To facilitate an automatic tuning mechanicsm in our SNI controller, we employ the knowledge-based interval Type-2 Takagi-Sugeno fuzzy system, known for its ability to accommodate the footprint-of-uncertainties (FoU). The proposed adaptive control law is implemented to compensate the dynamic variations in the quadrotor's CoG when the mass of the payload is constantly varied. The robustness and the effectiveness of the proposed control technique are highlighted not only using extensive computer simulations, but also through numerous real-time flight tests. Besides, we also conduct the stability analysis based on the SNI theorem.