University of Glasgow student receives best presentation award at the 2014 World Congress on Unmanned Systems Engineering

Issued: Thu, 17 Oct 2013 16:33:00 BST

Aldo Vargas, one of our PhD students at the Aerospace Sciences Division working with Dr. David Anderson head of the Micro Air Systems Technology (MAST) laboratory, has won the best presentation award at the 2014 World Congress on Unmanned Systems Engineering held at St. Anne's College at Oxford University, 30th July to 1st August.

Abstract is given below.


Abstract

In this paper, an anti-swing manoeuvre controller for a rotorcraft unmanned aerial system with an attached suspended load (slung-load) is proposed. The presented architecture is powered by an Echo State Network that enables simple modelling of the controller and outperforms linear techniques in terms of robustness to unmodelled dynamics and disturbances. The external load behaves like a pendulum, and when excited the system coupling can change the natural frequencies and mode shapes of the low frequency modes of the RUAS during flight. In choosing a controller technique to solve the problem, the primary requirements were to achieve robust performance in a computationally efficient algorithm.  Reservoir computing is an alternative to gradient descent methods for training Recurrent Neural Networks and Echo State Networks are a specific type of reservoir computing algorithm. In this paper, it has been demonstrated that a robust, functioning flight controller may be designed using Echo State Networks to stabilise the coupled rotorcraft/load system subject to unknown parameters. By extension, Echo State Networks are shown to be a powerful method for the control of nonlinear coupled systems.