Citation
Loughlin, C. (2005), "Toys Rn't Us", Industrial Robot, Vol. 32 No. 3. https://doi.org/10.1108/ir.2005.04932caa.001
Publisher
:Emerald Group Publishing Limited
Copyright © 2005, Emerald Group Publishing Limited
Toys Rn't Us
Editorial
Toys Rn't Us
I am always pleased to receive a Christmas card from Joe Engelberger and this (last) year was no exception. Joe's cards normally include an open letter to the robotics community. This year's message highlighted the wasted financial resources and technical effort that goes into the design and manufacture of robotic toys.
One of Joe's recent examples concerned a robotic muscle man that could dance a bit and (with luck) pick up a specially designed beaker. It sold for about 100 US$ and had been designed by a highly respected, and now wealthy, robot designer whose CV includes NASA consultancies.
Newton's long established laws of motion make it pretty clear that such systems will never be able to do anything useful. If you want to lift a kilogram by 1 m then the minimum amount of energy you will need to expend will be 9.81 J and you simply cannot achieve that with a few pen cell batteries and bendy plastic appendages.
In my view the problem with these toy robots is not with the toys themselves, but with our expectations. If you buy an exercise video expecting it to change your life then you will mostly be disappointed, but if you buy it for a bit of fun then the chances are you will be satisfied, and even if you didn't manage to shed those pounds you can at least seek consolation in the fact you tried. Toys are designed to entertain and therefore should be judged on their merit to do so and not on over hyped expectations.
This journal is not concerned with entertainment robots, but it is interested in useful technical developments. I would expect that about 98 per cent of the effort that goes into the combined mass of robotic toys will be of absolutely no interest to us. However, I do believe that such systems can occasionally have spin off benefits, and if nothing else they can sometimes provide the low cost platform for preliminary useful research and experimentation, as our Aibo based contribution from Golubovic and Hu demonstrates (Evolving locomotion gaits for quadruped walking robots, pp. 259-67).
As our regular readers will know, Joe's cause célèbre is the development of a two-armed sensate robot for assisting the elderly or infirm. So far however, no major manufacturer has come forward, that is prepared to put in the technical and financial investment to make such a system a commercial reality – despite an overwhelmingly good case in its favour.
One area where Joe and I differ is that I do not consider that existing industrial robot manufacturers are necessarily best placed to develop his home helper; and I certainly would not blame them for not trying. I am not sure who is best placed but perhaps it is the same organisations that make the subject of this issue – the white goods industry.
Domestic appliances are technically quite complex and they are made to exacting standards for low prices and in large volumes. They are also designed to be used by members of the general public. In my book this ticks most of the right boxes.
Of course, washing machines have yet to include stereo vision and a host of the other technologies that would be required, but at least they are not toys.
Clive Loughlin