Adaptive Robust Control

Linear Matrix Inequality-based Nonlinear Adaptive Robust Control of a Quadrotor [1]

In this study, we propose a novel linear matrix inequality-based nonlinear adaptive robust control algorithm to design the attitude and position controllers for an X-configuration quadrotor helicopter. 

• The inner loop of the autopilot controls the attitude and altitude of the quadrotor, and 

• the outer loop controls its position in the Earth-fixed coordinate frame. 

By assuming knowledge of the bounds of the quadrotor’s uncertain parameters (e.g., mass and moments of inertia) and predetermined bounds on the external and unstructured disturbances (e.g., wind gusts and unmodeled dynamics), we attain a guaranteed transient and steady-state tracking performance. 

We demonstrate the performance of the controllers via two illustrative examples. 

• In the first example mission, the quadrotor follows an altitude reference trajectory in the presence of wind gusts and delivers a package of uncertain mass midflight. 

• The second example presents an autonomous waypoint flight in the presence of wind disturbances. 

The results of our simulations indicate that our controller design is useful for a variety of quadrotor applications, including precise trajectory tracking, autonomous waypoint navigation in the presence of disturbances, and package delivery without loss of performance.

Related Publications

1. D. Kun and I. Hwang, "Linear Matrix Inequality-based Nonlinear Adaptive Robust Control of a Quadrotor," AIAA Journal of Guidance, Control and Dynamics, Vol.39(5), pp. 996-1008, May 2016, DOI: 10.2514/1.G001439

2. J.S. Nandiganahalli, C. Kwon, and I. Hwang, “Prediction-based Adaptive Robust Tracking Control of an Uncertain First-Order Time-Delay System,” Asian Journal of Control,(submitted on December 22, 2016; 1st revision submitted on June 22, 2017; under review)