Mathematical moves

An Australian researcher is helping find the best way to tap into the body’s electrical signals, so that a robotic prosthetic device can be operated like a real limb.

Rami Khushaba, a PhD student at the University of Technology Sydney’s Faculty of Engineering and Information Technology, is helping find the best way to tap into the body’s electrical signals, so that a robotic prosthetic device can be operated like a real limb.

It has been known for some time that human muscle activity, known as the Electromyogram (EMG), carries the distinct signature of the voluntary intent of the central nervous system.

These so-called myoelectric signals are already being used to control prosthetic devices.

But a lot of work still needs to be done before a robotic arm will respond instantaneously and accurately to the intention to move.

Khushaba said: ‘Right now the best that can be done is a few simple tasks with rather unsatisfactory performance, due to poor signal recognition and the high computational cost that leads to extra time delays.

‘Improvement in analysing the myoelectric signals will spur improvement of the hardware, and that’s where our work is directed.’

Supervised by Dr Adel Al-Jumaily and Dr Ahmed Al-Ani of the School of Electrical, Mechanical and Mechatronic Systems, Khushaba is developing a mathematical means of identifying what biosignals relate to particular arm movements and where electrodes should be placed to achieve the optimum result.

He added: ‘This project uses novel ‘swarm intelligence’-based algorithms to tackle many of the problems associated with the current myoelectric control strategies.

‘The way the members of a colony of ants will interact to achieve goals like finding food is metaphor that can be expressed in algorithms that are powerful tools for pattern recognition.’

Current methods for capturing biosignals on the forearm can involve mounting up to 16 electrodes on the skin, generating a vast quantity of data to be processed.

But Khushaba has already demonstrated that applying swarm logic will simplify that set-up and achieve significantly better results.

Applying the algorithms on 16-channel EMG datasets from six people found patterns that made it clear only three surface electrodes were actually needed.

Khushaba said: ‘These few electrode positions achieved 97 per cent accuracy in capturing the crucial biosignals for movement.

‘This significantly reduced the number of channels to be used for a real-time problem, thus reducing the computational cost and enhancing the system’s performance.’

Khushaba hopes that in the near future his work will allow amputees, who can still imagine moving a lost limb, to make use of a prosthetic device that will respond to their intentions.