Multi-agent control of high redundancy actuation

  • Authors:

  • Jessica Davies;Roger Dixon;Roger M. Goodall;Thomas Steffen

  • Affiliations:

  • Lancaster Environment Centre, Lancaster University, Lancaster, UK LE11 3TU;School of Electrical, Electronic & Systems Engineering, Loughborough University, Loughborough, UK LA1 4YQ;School of Electrical, Electronic & Systems Engineering, Loughborough University, Loughborough, UK LA1 4YQ;Department of Aeronautical and Automotive Engineering, Loughborough University, Loughborough, UK LE11 3TU

  • Venue:
  • International Journal of Automation and Computing
  • Year:
  • 2014

Quantified Score

Hi-index 0.00

Visualization

Abstract

The high redundancy actuator (HRA) concept is a novel approach to fault tolerant actuation that uses a high number of small actuation elements, assembled in series and parallel in order to form a single actuator which has intrinsic fault tolerance. Whilst this structure affords resilience under passive control methods alone, active control approaches are likely to provide higher levels of performance. A multiple-model control scheme for an HRA applied through the framework of multi-agent control is presented here. The application of this approach to a 10脳10 HRA is discussed and consideration of reconfiguration delays and fault detection errors are made. The example shows that multi-agent control can provide tangible performance improvements and increase fault tolerance in comparison to a passive fault tolerant approach. Reconfiguration delays are shown to be tolerable, and a strategy for handling false fault detections is detailed.