CitationDownload as .RIS
Emerald Group Publishing Limited
Copyright © 2004, Emerald Group Publishing Limited
Walking in the right direction
Walking in the right direction
Our theme for this issue is “Climbing and Walking Robots” and the majority of the research articles are greatly expanded and edited versions of papers presented at last year's CLAWAR conference (17-19 September 2003, Catania, Italy). I have been assisted in my selection by Professor Giovanni Muscato (our Guest Specialist for this issue and chairman of CLAWAR) Professor Gurvinder Virk of Leeds University, UK and leader of the Robovolc project.
Our aim was to select papers that best encompassed the full range of climbing and walking robots, and which demonstrated a high level of ingenuity and practicality.
There is one aspect of walking robot research that I have always found highly frustrating and which I am delighted to see finally being addressed. Over the years I have put my hand up at numerous conferences after a presentation on a particular walking robot, and asked the questions “How does the energy consumption compare with that of a person and what can be done to improve it?”.
Nine times out of ten it became immediately apparent that the speaker had never considered such a question and furthermore had no idea how much power a person does consume walking. From my own research it seems that a person consumes about 250 W, or 20 A at 12 V. An old fashioned lead acid battery weighing about 10 kg will provide about 1 h of operation. More modern cell technologies will do the same job at half the weight or less.
The most worrying thing about the speaker's revelation was not so much that they had not considered the question but that they did not see it as relevant.
In my view research needs to have a purpose and have a challenging goal. It is a bit like system design which needs both attention to detail and a solid overview plan – the one is useless without the other.
It seems to me that one very good goal, at least for biped walking robots, is to make them at least as good as people. If you are working in this area and know for sure that you will never achieve this objective then it is useful to ask “why not?”.
Another frustration for me has been to see different research groups working in effective isolation, busily reinventing the work of others.
I am absolutely delighted to see that all of the above concerns are being addressed both by the CLAWAR organisation itself and by papers included in this issue. The idea of a modular approach to robot design where one group might only be responsible for one part of a robot as opposed to the whole thing, is one of the stated aims of CLAWAR (refer to The CLAWAR project on mobile robotics, pp. 130–138).
The drive for energy efficiency is being addressed by our paper on optimising power consumption in walking robots (refer Building an energetic model to evaluate and optimize power consumption in walking robots, pp. 201–208) and also by the paper on the SMART actuator (refer Adding extra sensitivity to the SMART non-linear actuator using sensor fusion, pp. 179–188). Both show that significant gains can be made by careful research and innovation.
Walking robots cannot yet compete with us in performance and efficiency – but they are heading in the right direction.