Aardvark Daily

Please visit the sponsor!

Can REX deliver on its promise?

Yesterday's media was filled with another new Kiwi invention, the robotic exoskeleton known as REX.

At first glance, this looks like an astonishing breakthrough for those who have previously been confined to a wheelchair. At last, a way for these people to walk again and enjoy the freedom/independence that ambulatory mobility delivers.

Look a bit closer however, and I fear that there's a lot more sizzle than steak involved.

Now I should say right from the outset that it's not my desire to knock REX but I do think it's not a product worth the hefty pricetag (US$150,000) being asked for it.

For a start, it's slow -- very, very slow.

While it looks great in a demo, where most observers will be simply gobsmacked by the fact that it effectively allows disabled people to walk, it would not take long for its snail-like pace to become very frustrating to anyone looking for an effective mobility device.

It also looks to have only very limited stability and from the videos I saw no sign of any intelligent controller handling the balance issues involved. This would be no fun if, when you picked up a heavy object, the result was a fall caused by the centre of gravity falling outside the area of the feet.

What's more, if you did fall while wearing this device, there would appear to be zero capacity for returning to a vertical position without assistance.

Now I can fully understand that the real feature of REX is that it simulates actual walking, thus providing its wearer with an experience that is much closer to normality than could be experienced in a conventional wheelchair. However, this experience would seem to come at the cost of huge sacrifices in practicality (and a big price tag).

From a practicality perspective, Dean Kamen's iBot has REX beaten hands-down. iBot has the ability to negotiate stairs and other obstructions that normally impede the progress of traditional wheelchairs, it also elevates the user to the same height as when standing but it does all this without the speed penalty that REX seems to suffer.

Now I'm not saying that REX couldn't be improved and improved significantly, perhaps to the point where it really does deliver on its promise, however I really do think that this machine is a prototype that is far from ready for "the big time".

So how would I improve it?

First of all I'd add an intelligent balance system using accelerometers and gyros to compensate for any tilt or weight-shifts as they occurred. This would significantly improve the safety of the device as well as allowing it to automatically negotiate inclines and unexpected external forces.

Secondly I'd increase the speed at which it operates. To be truly practical, REX must allow its wearer to keep up with a normally-abled person with a normal walking pace. Of course the ability to operate at such speed will also require the automated stability mentioned above.

As implemented, REX seems to be a very "dumb" device and it would appear that the user is responsible for activating each leg using hand-controls. The inclusion of automated stability would also allow for the provision of programmed "walk" routines that would remove the need for operators to guide each leg through each step. Such software could deliver far simpler control -- perhaps just a single lever that would be pushed forwards to walk, from side-to-side to turn and up and down to stand/sit.

This would further open up the market for REX to those who had also lost a degree of arm/hand function in one or both limbs.

Then there's the issue of recovery from a fall. I'm pretty sure that it would be possible to build in a software controlled self-righting capability that would allow recovery from most "downed" positions in the unlikely event this ever occurred.

The provision of an adequate energy source would be a difficult but not insurmountable problem. Modern lithium batteries have very good power to weight/volume ratios and once again, that smart software would allow for far more efficient use of that energy in much the same way as you and I establish a walking rhythm that effectively minimises the amount of exertion required to travel long distances with minimum effort.

My big fear for the designers of REX is that they've shown the world the potential of this exoskeleton concept and someone else (with more money) will take that idea, add the necessary intelligent systems, and come up with a vastly superior product which will steal the market from them.

Rex Bionics claim that they've invested $10m in developing the product to this stage -- something that staggers me, given the apparently low-tech simplicity of the system. I sure hope they've got enough money left to take it from "proof of concept" to an actual, practical, marketable product.

I hate to say it but I would have expected a far superior product could have been designed, developed, tested and commercialised for a tenth the amount these guys have spent. What's more I think they should be aiming for a sell-price of US$20K or less, something which is entirely achievable, even once the improvements I've suggested have been added.

For almost as long as I can remember, I've always been very interested in ways of improving the lot of the disabled through technology. Unfortunately, there is virtually no money around for funding the development of technology-based systems for the disabled and I've never had the money to go it alone. Let's hope that Rex Bionics are able to run with the ball and deliver the on the promise they've made this week.

Please visit the sponsor!

Have your say on this...

PERMALINK to this column

Oh, and don't forget today's sci/tech news headlines

Rank This Aardvark Page

Change Font