Path following of the autonomous airship with compensation of unknown wind and modeling uncertainties

Jie Wang, Xiuyun Meng*, Guanghui Wu

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

16 Citations (Scopus)

Abstract

The motion control of the airship has been an attractive topic in recent years. This paper investigates the problem of the path following of the autonomous airship. Due to the characteristics of the large volume, slow speed, and low-maneuverability, the airship is sensitive to the wind. Based on the design idea of the line-of-sight (LOS) guidance law, a parameter-observer-based LOS (POBLOS) guidance law is proposed. The unknown and time-varying influence from the wind is estimated by the parameter observer and then compensated in the design of the guidance loop. In order to overcome the disturbances of the modeling uncertainties and drift forces induced by the wind, the method of the command filtered backstepping control (CFBC) combined with the disturbance observer is employed in the control loop. The disturbance observer is constructed based on the formulation of the nonlinear disturbance observer (NDO) and improved by adding the status estimation error feedback, which results in a more robust performance when modeling uncertainties and time-varying disturbances exist. The stability analysis shows that the path following error is ultimately bounded. The results of the simulation indicate that the path following system has a good performance and is robust against the influence of the wind and modeling uncertainties.

Original languageEnglish
Article number105349
JournalAerospace Science and Technology
Volume93
DOIs
Publication statusPublished - Oct 2019

Keywords

  • Autonomous airship
  • Backstepping control
  • Disturbance observer
  • LOS guidance law
  • Path following

Fingerprint

Dive into the research topics of 'Path following of the autonomous airship with compensation of unknown wind and modeling uncertainties'. Together they form a unique fingerprint.

Cite this