Cooperative Control for Automated Air-to-Air Refuelling

Abstract

The proliferation of Unmanned Aircraft Systems has led to demand for Air- to-Air Refuelling capabilities similar to those available to manned aircraft, and there also exist fuel and cost savings which can be realised. Autonomous Air-to-Air Refuelling presents interesting challenges in the areas of control, sensing, and decision making. This thesis focuses primarily on probe-drogue refuelling. Work exists on automating Air-to-Air Refuelling using conventional leader- follower architectures. In these schemes, the refuelling drogue is aerodynam- ically stabilised but subject to disturbances such as gusts, tanker wake, and receiver bow wave, and the receiver is controlled in such a way as to capture the drogue. This method does not effectively utilise all available degrees of freedom – the tanker is usually much larger and less manoeuvrable than the receiver, but there is potential to control and harness the drogue’s motion, for ex- ample by adding aerodynamic control surfaces. This thesis investigates the feasibility of this approach and compares control architectures by which these degrees of freedom may be harnessed. A drogue control model is developed, and two candidate architectures are investigated: Common-target-point Control uses a shared target point approach, which is found to improve capture rate under turbulence, and is ex- tended using a novel scheduled-gain method. Intimate Control optimises the whole drogue-receiver system using Multiple-Input Multiple-Output tech- niques. Verification of the control schemes is conducted via the Univer- sity of Bristol’s Relative Motion Robotic hybrid testing facility. A well- characterised F-16 aircraft model is used as a surrogate for future mid-sized Unmanned Aircraft System. This thesis presents evidence that harnessing the additional degrees of freedom available via drogue control is likely to improve capture performance in Autonomous Air-to-Air Refuelling, and has been tested at a higher Tech- nology Readiness Level than is usual in academic fields. The work presented here forms part of the Autonomous Systems Technol- ogy Related Airborne Evaluation & Assessment (ASTRAEA) programme in the UK, and was funded by Cobham plc.

Date of Award	9 May 2017
Original language	English
Awarding Institution	University of Bristol
Supervisor	Tom S Richardson (Supervisor)