Robust control and state estimation of micro aerial vehicle using sliding mode technique

Abstract

Micro Aerial Vehicles (MAVs) are smaller version of UAVs and operate at very small velocities, which makes it very vulnerable to external disturbances. Also, MAV being a small vehicle, obtaining accurate aerodynamic data is difficult. The vulnerability of the vehicle towards external disturbances and the model inaccuracies calls for a robust control. Sliding Mode Control with its invariance property against external disturbances and the model inaccuracies is seen as a potential solution. The Sliding Mode Control algorithm needs all the states, which may not be measurable always. This calls for the design of an observer in order to estimate the unmeasurable states. The application of Sliding Mode Observer is one way to overcome this hurdle. In this paper, the various steps involved in estimating states with a Sliding Mode Observer using Linear Matrix Inequalities (LMI) approach for Longitudinal and Lateral channels of an MAV is discussed. During aggressive maneuvering requirements, it is useful to have a tighter control of Body Accelerations. However, measurement of Angular Acceleration or numerical differentiation of Body Rate measurements is not practical. A Super Twisting Algorithm (STA) based Differentiator to estimate Angular Accelerations from Body Rates is proposed in this paper, which may be useful for tight Body Acceleration control.