Controller Design of an Airship

TL;DR

This paper presents a robust control strategy for vectored-thrust airships using an Extended Incremental Nonlinear Dynamic Inversion approach, effectively managing hover, high-speed flight, and environmental disturbances.

Contribution

It introduces a novel control framework combining E-INDI with a high-level outer loop and a control allocation method for tiltable propellers, tailored for airships.

Findings

Achieves excellent tracking performance in simulations.

Demonstrates high robustness against disturbances and parameter uncertainties.

Outperforms an alternative control method in various scenarios.

Abstract

Airships offer unique operational advantages due to their ability to generate lift via buoyancy, enabling low-speed flight and stationary hovering. These capabilities make them ideal for missions requiring endurance and precision positioning. However, they also present significant control challenges: their large, lightweight structures are highly sensitive to environmental disturbances, and conventional aerodynamic control surfaces lose effectiveness during low-speed or hover flight. The objective of this thesis is to develop a robust control strategy tailored to a vectored-thrust airship equipped with tiltable propellers. The proposed approach is based on an Extended Incremental Nonlinear Dynamic Inversion inner loop in combination with a high level outer loop, controlling the attitude and velocity of the airship. The proposed method is able to effectively control the airship over the whole envelope, including hover and high speed flight. For this, effective use of available actuators is key. This includes especially the tilt rotors, for which a control allocation method is presented. The controller's performance is validated through a series of simulation-based test scenarios, including aggressive maneuvering, gust rejection, atmospheric turbulence, and significant parameter mismatches. The controller is compared against an alternative controller developed at the institute, offering insight into the trade-offs between direct inversion and incremental control approaches. Results demonstrate that the proposed E-INDI controller achieves very good tracking performance and high high robustness against parameter uncertainties.