The Open UniversitySkip to content
 

Network technologies for intelligent modular prostheses

Poulton, A.S.; Hill, S. and Binnie, D. (2004). Network technologies for intelligent modular prostheses. In: 15th Congress of the International Society of Electrophysiology and Kinesiology, 18-21 Jun 2004, Boston, USA, p. 254.

Google Scholar: Look up in Google Scholar

Abstract

Introduction: A controllable multiple degree of freedom prosthesis must allow for distributed intelligence across multiple joints. CAN (Controller Area Network) and LonWorks are two widely available technologies that can facilitate this, by allowing multiple modules to be plugged into one central network. The development of modular control electronics has been encouraged by recent interest in designing a prosthesis as a set of interchangeable components. This simplifies servicing and reduces stock keeping, while retaining the flexibility to produce prosthetic devices that will fit all or most users. A modular approach to the electronics can also result in reliability gains, and makes it easier to add and test new intelligent features.

Methods: We have investigated LonWorks and CAN for use in modular prostheses in separate projects. They are mature technologies with wide application bases, and offer cost effective solutions to distributed applications. CAN was developed by Robert Bosch for critical automotive applications such as anti-lock braking systems (ABS) and engine management, which like prostheses require high reliability. LonWorks is a standard that was developed by Echelon originally for intelligent building applications such as lighting, heating, air conditioning and security, and is also used in other areas such as industrial automation and transportation. Here we compare LonWorks and CAN as technologies for prosthetics, and we also consider other emerging technologies that could be candidates for similar applications in the future.

Results: LonWorks and CAN both have features which are needed in a modular prosthesis. Each offers good immunity to electrical noise, adequate real-time performance, mechanisms for recovering from data errors, and widely available components which are energy efficient and have low weight and cost. They have different methods for dealing with network addressing, network configuration, data priority, error recovery, network robustness and a number of other issues.

Discussion: A modular prosthesis requires the following from a network: predictable real-time performance, low power consumption, graceful degradation and simple installation, among other things. Both LonWorks and CAN have the real-time capabilities needed. Serial busses that are becoming common for PC use such as USB and Firewire
may also have a place here. However, while these promise higher raw data rates, they were not designed at the outset for real-time control in electrically and physically demanding environments, unlike LonWorks and CAN, which are ideally suited to the practical implementation for reliable control of a multiple degree of freedom prosthesis. This has been demonstrated through limited trials of a LonWorks system, and forthcoming fittings of a CAN based system in Edinburgh.

Item Type: Conference or Workshop Item
Copyright Holders: 2004 International Society of Electrophysiology and Kinesiology
ISBN: 0-87270-136-0, 978-0-87270-136-6
Extra Information: ISBN 0-87270-136-0
p254
Academic Unit/School: Faculty of Science, Technology, Engineering and Mathematics (STEM) > Computing and Communications
Faculty of Science, Technology, Engineering and Mathematics (STEM)
Item ID: 5655
Depositing User: Adrian Poulton
Date Deposited: 30 Oct 2006
Last Modified: 07 Dec 2018 08:59
URI: http://oro.open.ac.uk/id/eprint/5655
Share this page:

Actions (login may be required)

Policies | Disclaimer

© The Open University   contact the OU