In the last two decades, with the introduction of new robotics and automation technology, we have seen many changes in the rehabilitation of neurological patients, where various forms of robotics tools have allowed users to engage in extensive physiotherapy training in much more efficient manners. While the predecessors have shown great success in their own rights, they have received mixed reviews from users because the device behaviour is commonly generic and does not reflect the needs of an individual user unlike what a physiotherapist would be able to do. In addition, these devices are often not portable and only available in clinics and rehabilitation centres, which sets a limit to training opportunities.
Therefore, our aim is to develop an upper-body exoskeleton for hand/arm-motion support that is versatile and yet specific to the user. In order to make it lightweight and usable at home, we design our exoskeleton to be fully integrated with a technology called functional electrical stimulation (FES) which triggers muscle contraction through electrical stimulation. In contrast to motorised exoskeletons, FES is lightweight and comes with additional clinical benefits as it inherently promotes muscle use. However, there are many challenges to overcome. With FES, it is very difficult to control the response behaviour, especially on the upper-body motion, since the underlying muscular system is very complex. Motorised exoskeletons are not ideal for small muscles in the forearm and the hand as they can be very bulky. Consequently, we combine the exoskeleton and FES to get the best of the two worlds to make the exoskeleton lightweight and still suitable for upper-body support. To make the interaction fun and long-lasting, this hybrid exoskeleton will be offered with a variety of games and interaction scenarios that are sensitised to particular rehabilitation routines recommended from clinical evidence. From the interpretation of how the exoskeleton is used and how the user performs tasks, the system will be able to interpret cognitive, physiological, and neuromechanical states in order to autonomously offer the right training for each and every user.
