Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Modelling and Hovering Stabilisation of a Free-Rotating Wing UAV

Florian Sansou, Gautier Hattenberger, Luca Zaccarian, Fabrice Demourant, Thomas Loquen

TL;DR

This work tackles the challenge of stabilizing a tail-sitter freewing UAV across hover and forward flight by introducing a pivoted-wing architecture that keeps sensor orientation fixed. It develops a constrained multi-body model using the Udwadia-Phohomsiri equations, and couples this with state estimation via EKF and a high-gain observer to recover wing and fuselage states. Hover stability is achieved through Incremental Nonlinear Dynamic Inversion (INDI) for the wing and a PD controller for the fuselage, validated on a Colibri prototype with experimental hovering in a controlled environment. The results demonstrate the feasibility of maintaining a stable fuselage attitude while the wing handles lift during hover, and point to future work on centralized control to span the full flight envelope and better mitigate coupling and disturbances.

Abstract

We propose a multibody model of a freewing UAV. This model allows obtaining simulations of the UAV's behaviour and, in the future, to design a control law stabilising the entire flight envelope (hovering and forward flight). We also describe the realisation of a prototype and a comparison of possible methods for estimating the UAV's states. With this prototype, we report on experimental hovering flights with a non-linear incremental dynamic inversion controller to stabilise the wing and a proportional derivative controller for the fuselage stabilization.

Modelling and Hovering Stabilisation of a Free-Rotating Wing UAV

TL;DR

This work tackles the challenge of stabilizing a tail-sitter freewing UAV across hover and forward flight by introducing a pivoted-wing architecture that keeps sensor orientation fixed. It develops a constrained multi-body model using the Udwadia-Phohomsiri equations, and couples this with state estimation via EKF and a high-gain observer to recover wing and fuselage states. Hover stability is achieved through Incremental Nonlinear Dynamic Inversion (INDI) for the wing and a PD controller for the fuselage, validated on a Colibri prototype with experimental hovering in a controlled environment. The results demonstrate the feasibility of maintaining a stable fuselage attitude while the wing handles lift during hover, and point to future work on centralized control to span the full flight envelope and better mitigate coupling and disturbances.

Abstract

We propose a multibody model of a freewing UAV. This model allows obtaining simulations of the UAV's behaviour and, in the future, to design a control law stabilising the entire flight envelope (hovering and forward flight). We also describe the realisation of a prototype and a comparison of possible methods for estimating the UAV's states. With this prototype, we report on experimental hovering flights with a non-linear incremental dynamic inversion controller to stabilise the wing and a proportional derivative controller for the fuselage stabilization.

Paper Structure

This paper contains 10 sections, 19 equations, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. Design and modelling of Colibri UAV
  3. State estimation
  4. Sensors placement
  5. Angular speed estimation
  6. Wing state estimation
  7. Fuselage orientation estimation
  8. Incremental Nonlinear Dynamic Inversion combined with PD pendulum control
  9. Experimentation
  10. Conclusions and Future Work

Figures (6)

  • Figure 1: Inertial (I) and wing (W) reference frames and the Colibri architecture.
  • Figure 2: Inertial (I), fuselage (F) and wing (W) reference frames and forces acting on the Colibri UAV.
  • Figure 3: Position and orientation simulation of the multi-body UAV Colibri in closed loop with a simple double-loop controller.
  • Figure 4: Angular position measurement (black,top plot), wing gyro velocity measurement (blue,bottom plot), finite difference velocity estimation (green, bottom plot) and high-gain estimates (red curves)
  • Figure 5: Colibri experimental prototype.
  • ...and 1 more figures