Flight control of a rotary wing UAV using backstepping

Bilal Ahmed, H. R. Pota, Matthew Garratt

International Journal of Robust and Nonlinear Control · 2009 · 111 citations · 34 references

Concepts

TL;DR

Backstepping control, originally developed for general rigid‑body velocity regulation, is extended here to cover the full flight envelope of rotary‑wing UAVs. The study introduces and analyzes a backstepping controller for autonomous landing of a rotary‑wing UAV, incorporating a flapping‑angle correction term. Using a nonlinear RUAV model that includes flapping and servo dynamics, the controller decouples translation and rotation, first computing desired thrusts and flapping angles and then determining control inputs, and its performance is evaluated in a high‑fidelity miniature rotorcraft simulation. Simulation results show the UAV reaches the commanded position after a brief transient and the derived control inputs confirm the algorithm’s feasibility on practical rotary‑wing platforms. © 2009 John Wiley & Sons, Ltd.

Abstract

This paper presents a novel application of backstepping controller for autonomous landing of a rotary wing UAV (RUAV). This application, which holds good for the full flight envelope control, is an extension of a backstepping algorithm for general rigid body velocity control. The nonlinear RUAV model used in this paper includes the flapping and servo dynamics. The backstepping-based controller takes advantage of the ‘decoupling’ of the translation and rotation dynamics of the rigid body, resulting in a two-step procedure to obtain the RUAV control inputs. The first step is to compute desired thrusts and flapping angles to achieve the commanded position and the second step is to compute control inputs, which achieve the desired thrusts and flapping angles. This paper presents a detailed analysis of the inclusion of a flapping angle correction term in control. The performance of the proposed algorithm is tested using a high-fidelity RUAV simulation model. The RUAV simulation model is based on miniature rotorcraft parameters. The closed-loop response of the rotorcraft indicates that the desired position is achieved after a short transient. The Eagle RUAV control inputs, obtained using high-fidelity simulation results, clearly demonstrate that this algorithm can be implemented on practical RUAVs. Copyright © 2009 John Wiley & Sons, Ltd.

References

34