Swing-free manoeuvre controller for rotorcraft unmanned aerial vehicle slung-load system using echo state networks

Vargas, A., Ireland, M. L. and Anderson, D. (2015) Swing-free manoeuvre controller for rotorcraft unmanned aerial vehicle slung-load system using echo state networks. International Journal of Unmanned Systems Engineering, 3(1), pp. 26-37. (doi: 10.14323/ijuseng.2015.3)

Full text not currently available from Enlighten.


There is a growing interest in developing Rotorcraft Unmanned Aerial Systems (RUAS) with advanced onboard autonomous capabilities. RUAS is a very versatile vehicle and its unique flying characteristics enable it to carry loads, hanging in wires underneath it. This suspended load alters the flight characteristics of the vehicle. 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 Echo State Networks (ESN) that enables simple modeling of the controller and outperforms linear techniques in terms of robustness to unmodelled dynamics and disturbances. The external load behaves like a pendulum; this can change the natural frequencies and mode shapes of the low frequency modes of the RUAS. The technique chosen to solve the problem is to achieve both robust performance and computational efficiency. Reservoir Computing (RC) is an alternative to gradient descent methods for training Recurrent Neural Networks (RNN), which represent a very powerful generic tool, integrating both large dynamical memory and highly adaptable computational capabilities. ESN is a type of reservoir computing; the advantage lies in the ability to overcome the difficulties in RNN training, it is conceptually simple and computationally inexpensive. It has been demonstrated that a model and controller design using ESN may be developed. ESN performs well to control unknown nonlinear systems.

Item Type:Articles
Keywords:autononmous systems, echo state networks, nonlinear systems, slungload multicopter, recurrent neural networks, reservoir computing
Glasgow Author(s) Enlighten ID:Anderson, Dr David and Ireland, Dr Murray
Authors: Vargas, A., Ireland, M. L., and Anderson, D.
College/School:College of Science and Engineering > School of Engineering > Autonomous Systems and Connectivity
Journal Name:International Journal of Unmanned Systems Engineering
Publisher:Marques Aviation Ltd - Press
ISSN (Online):2052-112X

University Staff: Request a correction | Enlighten Editors: Update this record