Contemporary systems and cybernetics

Citation

Rudall, B.H. (2002), "Contemporary systems and cybernetics", Kybernetes, Vol. 31 No. 6. https://doi.org/10.1108/k.2002.06731faa.001

Contemporary systems and cybernetics

Keywords: Automation, Cybernetics, Research, Technological developments

Abstract Provides reports and surveys of selected research and development in the interdisciplinary fields of systems and cybernetics. They include: Biocybernetics – Brain Scanners and Emotions; Wearable Computers; Decision Support-Systems; Mind Machines; Innovations; Automation and Cybernetics.

Biocybernetics

Brain scanner discovers emotions

It is reported that the activity of the amgydala in the brain can now be monitored by scanners to reveal the true nature of suppressed emotions and feelings even those that we may have been concealing. Science, it seems may have discovered the ultimate “truth machine” with these new scanners that could, perhaps, be used to find out our beliefs, involvements with crime or marital indiscretions. These scanners are being developed by neurologists who are taking seriously the possible breakthroughs they have made into the functions of the amygdala, the almond-shaped region of the forebrain. The report that during a recent trial, scientists had read the electrical patterns of amygdalas in a series of trials concerning the cognitivity of humans and then succeeded in decoding their emotions from their scanners has caused a great deal of hope in the field. Dr Stephen Lawri of Edinburgh University, Scotland, U.K. Psychiatry Department says that:

The whole field has become immensely exciting. It has immense potential, not just for understanding how we think and feel, but for helping people with severe emotional problems, including those with schizophrenia.

Biocyberneticians will already know that the amygdala has been described as the “mind's emotional engine” and research into its operation has attracted scientists for a long time. It is, we are told, currently regarded as one of the most important areas of research by scientists studying the functions of the brain.

So much so that a recent international conference in Galveston, Texas USA, devoted three days dedicated to amygdala research.

Readers will be aware that so far most researchers on the brain have concentrated on “thinking patterns” so that understanding how we feel or express our feelings has taken second place. Now it would about seem we may have the means to obtain more information about why we feel emotions and their effects.

Animal experiments have already given us much information so that we believe the amygdala controls the expression of fear and anxiety. Whether the human brain also works in this way is the subject of much research. This is obviously going to be aided by the new magnetic resonance scanners that can measure brain activity by analysing changes in oxgenated blood flow.

US University research studies

Researchers from the Universities of Yale and New York have been carrying out new studies in this field. Participants in an experiment were shown pictures of black and white people. The white volunteers being checked in advance that they had no overt anti-black feelings. The experimental results indicated that when these participants were shown pictures of black people, images of some of them revealed intense amygdala activity which suggested that they were feeling fearful or anxious. Other tests with pen and paper, we are told, were psychological exercises designed to reveal inate racism. These tests uncovered in some participants a suppression of their anti-black feelings. These were the same individuals whose amygdalas were most active during tests. It could be deduced therefore that although they claimed not to be racist their brains indicated otherwise. The results of these and other tests give rise to the future prospect that the feelings and attitudes we attempt to conceal could be discovered by such scanning methods. This may or maynot be an attractive or a desirable state. The prospects, however, of such a breakthrough still remain rather distant. Meanwhile the advances reported here can be used for formulating new methods for dealing with psychiatric conditions. Dr Lawrie of Edinburgh University believes that:

In schizophrenics, and in those who are at risk of developing the condition, we find amygdalas that are much smaller than those in most people. In other words, the emotional barometers they use to measure the outside world are damaged. They may feel anxious or fearful about innocent objects or passers-by, which can trigger all sorts of paranoid responses.

We can now think about how to develop drugs that could alleviate the problem or design cognitive therapies to improve their condition.

The ability to scan the amgydalas and to monitor has also produced new information about people who have stress conditions. It has been shown for example, says Dr Rashid Shaikh of the New York Academy of Sciences, that when you scan patients with post-traumatic stress, such as that brought on by bereavement, or witnessing a tragedy you discover that their amygdalas are switched on to far higher levels than you would find in the average person.

Although some of the more dramatic results of successful brain scanning may be achieved in the longer term and lead to applications such as truth testing the prospects of amygdala research transforming treatments of psychiatric conditions is the more likely immediate benefit.

UK Research studies using magnetoencephalography (MEG)

The findings of Professor Steven Rose's research team from the Open University and the London Business School, UK, are being published in the journal Neural Plasticity (2002). They outline the progress being made using the MEG technique which is regarded as the fastest of all scanner methods. It was used to measure brain activity in four men and four women for the project that involved shopping. This it was said is regarded as an activity that keeps the right parietal cortex of the brain busy. The scanner was used during the second or so that it takes a person to select a product. The team, which was also led by Professor Steven Swithenby, were then able to link the measurements taken with what the shoppers said about their preferences.

The scanner itself and the planned experiment were also described by Dr Roger Highfield (Science, Daily Telegraph, 20/03/02), who writes:

The scanner looks like a glorified salon hair-drier and is too bulky to be portable. Instead, the experienced supermarket shoppers – including the team leaders – were given a virtual tour of Sainsbury's and, as the MEG scanner watched what was going on in their heads, they were invited to make a choice between different brands by pressing a button.

The team found that the brain was hugely active during the 2.5 seconds it took for the button to be pressed, revealing a flow of activity, starting at the back of the brain and moving to the front.

Professor Rose describes the action being monitored. He says that:

Within 80 milliseconds (thousandths of a second) the visual cortex, where information from the eye is processed, responds as we perceive the choice items.

A little later, regions of the brain associated with memory and speech become active, when we begin to interpret the image and vocalise it silently – saying the name of the brand in our heads – especially if we don't have a strong preference.

Finally, after about 800 milliseconds – and this was the surprising thing – if and only if they really do prefer one of the choice items then a region called the right parietal cortex becomes active. This lights up if and only if you actually do prefer Coke to Pepsi or Pepsi to Coke.

This then is the region of the brain involved in making conscious decisions – in this experiment about shopping choices, but maybe for more important life choices too, holiday destinations and choice of husband as examples.

MEG is reported to use an array of 344 devices known as SQUIDS – superconducting quantum interference devices – that work at very low temperatures and are in fact bathed in helium, to measure minute fluctuations in the magnetic fields that are generated by electric currents in the brain.

It was pointed out that the UK needed to use the scanner at the Low Temperature Laboratory at Helsinki University of Technology, Finland, and that the UK does not have MEG machines that would be sufficiently powerful.

The application of this new technology will bring in reports of other projects and will no doubt amaze the general public by analysing many of the “brain activities” made during the lifetime of a human. The results will obviously not only be of interest to biocyberneticians and others but also manufacturers and organisations which could well benefit by knowing how our decisions are made.

Wearable computers

What is a wearable computer?

The initial concept of having intelligent walls soon moved on to clothes. On the last occasion in these columns when we described the latest “wearable computers” it was a matter of envisaging something the size of a backpack now, however we have the problem of defining what is meant by such a device. The trouble is that computer systems vary so much in size that, for example, a wearable computer could be the size of the smallest chip or indeed the size of suitcase. The other question to be considered in framing any definition of such devices is whether they are to be worn outside or inside the body. We all know that for many applications computing devices have been implanted in the body or fixed onto the skin. With the current rate of advance in this form of technology it is better to take an extremely loose description of such computers rather than aim at a precise definition. Consider then some of the possible categories recently reported to us.

Microchip implanted in the arm

Professor Kevin Warwick, the cybernetics expert of Reading University UK and a past contributor to this journal recently had a microchip implanted into his arm. By anyone's definition he became, as far as we know, the world's first cyborg. There is no doubt that he is a pioneer in this research and we have described his endeavours on many occasions. He is investigating whether data can be transmitted between human and machine. He wants to know the extent to which it is possible to capture data signals which on the one hand would include commands for movement say, and on the other physical emotions such as pain of pleasure and in each case be able to transmit them to other humans or, maybe devices. The operation was recently televised and screened worldwide so that even if it produces no positive results it has publicised cybernetics and Professor Warwick has to be admired for his pioneering endeavour.

More conventional wearable computers

We are not sure who initiated the concept of wearing and not carrying computers, but Dr. Mark Weiser of Xerox's Palo Alto Research Centre USA, has been given the credit for the idea. In 1991 he is said to have proposed that computers could be made so small they could be worn as jewellery or as clothing. Other computer researchers have made claims for producing the idea and, of course, there are now many manufacturers and research agencies who are exploiting the concept. We now have the hightech backpack in many different forms so that it can be worn by cameraman, production worker or warehouse operator. You can wear a web-cam and microphone for real-time interactions that can happen online. Cyberhelmets are popular so that you can view information of your choice at any time. There are watches that can be worn to give you full computing power or just as or in addition to serve as a cellphone. All the devices that are wearable come with the possibility of having the latest interface systems whether it be voice-dialing, or any other advanced input/output facility.

New systems being marketed

Many new systems are being introduced to the target user, who is usually the user who has got everything or is merely technologically hungry. In most cases the devices on offer have been produced with a very serious purpose in view and have very real applications in science, industry and business.

Many new systems are being launched this year. The WIA-I00NB Wearable Internet Appliance is being launched by Hitachi in Japan and by Xybernaut in the US. This machine is slim, silver and palm-sized unit that hooks on to your belt or sits in a pocket. Most innovative in its design is that it does not possess the conventional screen, but instead a tiny display sits just in front of your eye, attached to a headband. It is claimed that the laser optics used make a virtual notebook-sized screen appear to float in front of you. Xybernaut has been involved with developing devices of this sort since the beginning: of the 1990s. This new machine, which is called Poma, is the first to enter the market.

This machine and others that will no doubt follow in its wake are aimed at the new breed of wearable computer users, this is someone, they believe, who is keen to compute whilst on the move, surfing the internet, retrieving personal information, playing online games all “anywhere and at anytime”.

MIT has decided that wearable computers should in fact be clothes, just as Weiser's original concept declaimed. Their latest project called MIThril literally wraps one in computer power. It consists of putting input/output systems, wireless networking and all that goes with it in a sleeveless fleece. No doubt other researchers will follow this trend and we will soon see some exotic garments implanted with the latest computer and communications facilities.

Future wearable computers

A recent report suggests that by the end of this decade 40 per cent of adults and 75 per cent of teenagers will be wearing computers. This is bound to happen as manufacturers produce miniaturised systems that carry more powerful processors. They would contain full voice recognition systems, advanced wireless communications and all the new fashionable devices that are currently too large to wear. Many of these systems will have tracking facilities using global positioning, others will allow you to use them without you having to take your eyes off the task in hand. Already Xybernaut have a wearable product called Mobile Assistant which is in use in Tennessee the company say, where the Tenessee Department of Transport has equipped its staff. It allows their workers to view and to update information, even when they are driving.

Most cyberneticians who have researched into the information age that is already with us will not be surprised at this trend. The popularity of mobile telephones indicates how quickly the applications of advanced technology are absorbed by the public at large. It could well be that carrying any sort of technological device will become a thing of the past and wearing our technology will be the reality.

Wearables in the real world

Unfortunately being in the real world of wearable computers can create personal problems. It is reported in Technology News (March 2002) that Professor Steve Mann of the University of Toronto, Canada, was stopped boarding a flight at St. John's Newfoundland after a strip search revealed he was wearing computerised glasses, headgear and an electronic body suit. He is reported as saying that he needed the gear for medical reasons. Professor Mann is a specialist in wearable computers and has worn full cyborg equipment for some 20 years we are told. More stringent searches by security conscious airlines may well create problems for passengers with wearable computers. Professor Warwick with his implanted chip may well cause a rethink of current security vetting at airports, particularly if they become more commonplace in the future.

Advances in mobile technology

Reports tell us that the next generation of mobile technology that will allow users to be always connected to the internet without the need for wires will be the General Packet Radio Service (GPRS). It is claimed that GPRS is faster and a more reliable method of sending data, since it is transmitted in “packets” over different phone lines as opposed to the slower alternative of the General (GSM) which uses one line to both send and receive e-mail and online information.

GPRS we are told can be seen as a way of bridging technology in order to get consumers used to mobile internet services before the introduction of the coming even faster third-generation (3G) services. Companies like BT Cellnet, Vodaphone and Nokia in the UK are already encouraging businessmen and women to “work on the move”. Vodaphone and Cellnet already offer GPRS in the UK wherever there is a GSM network. Similar developments are occurring worldwide as companies seem willing to finance their employees to work on the move and download the information they require. Even so the cost may well influence how quickly mobile technology of the next generation is taken up, and many corporates are monitoring the market very carefully.

Decision support systems

Model management has been one of the main topics in the model oriented approach to decision support systems (DSS). There have been many contributions in the literature and one from Naoki Shiba from the Department of Management Information Science, Chiba Institute of Technology, Japan, was particularly illuminating. Published in the International Journal of General Systems (Vol. 31, No. l, 2002, pp 53-72), Dr Shiba asks the question “What is model” management? which in itself is a very meaningful query. Asking and attempting to answer such questions, particularly in this case where the consensus to the answers is certainly neither obvious nor agreed, forms an excellent starting point to a discussion paper.

We all appreciate that in so many information style technologies there is an accepted understanding about the field of study or the topic in hand. The author of this paper rightly cites database management as a typical example of such an understanding. He then asserts that:

model management is management of the structure of models. Two questions must be answered in order to make this assertion comprehensive. The first one is “what is the structure of models?” and the other one is “what is management?”.

We can all ask questions and obviously we do so to obtain illuminating answers. This paper goes a long way to satisfy this criterion. With well- referenced discussions in its introductory part and a well formulated mathematical approach to hierarchical structures “and to managing structures” together with a demonstration of the applicability of his definitions the author helps us understand more about DSS and model management. The example given belongs to the real world and is a simplified version of a long-term climate model developed by the UNESCO Global Education Network Initiative (GENIe).

What we are told here is that a mathematical formulation of the concepts in model management supports the assertion that model management is management of the structure of models. It also illustrates that this is a topic that can be safely discussed using mathematical techniques and that many systemists who work in this area without using them cannot contribute fully to its development.

It is interesting to note that because of its mathematical approach to these concepts the author can naturally move into the development of a software system which he calls DIALOGO. He believes that the presented formal definition approach helps the implementation of the development of software tools and intends to apply them to real problems.

Mind machines

Background

For many decades since the advances in computer technology gave rise to numerous effective medical uses scientists have toyed with the idea of producing a “mind machine”. In these columns over the years we have described many such endeavours some not to be heard of again whilst others have been transformed in useful marketable products. Basically a mind machine was to be designed around the principle that a computer could be programmed to interpret signals received from the brain and act upon them. In other words if we think “on” then it would be possible to switch some device on and think “off” and it is turned off. Such commands were simple and binary in concept but they did form the basis of a system that could, as we have seen, be used by a handicapped system-user to control actions by devices that would be capable of carrying out his commands. Such systems now exist, many are still simple in function whilst others are becoming more and more sophisticated. Not a fantasy by any means is the concept such machines can be built. Recently in the media another claim has been made that “the final frontier” has been broken and that a revolutionary device, which for the first time, allows humans to control objects simply by the power of thought has been built. Called the “Thought Translation Device (TTD)” it has been created by Professor Niels Birbaumer of Tuebingen University, in Germany.

Thought Translation Device (TTD)

It is admitted at once that the device is still in its infancy, but, even so enables its users to focus their minds on to moving icons on a computer screen. As a result it is predicted that when it becomes more refined it will be possible for the brain of the human user to carry out some of the tasks described as “opening a window”, “flicking a light switch” and even, which seems more ambitious than most other similar projects, order the food supplies and initiate other such operations.

At present, however, the basic version is a device that consists of six electrodes that are linked to a standard desktop computer via a small control box.

The research and development is being carried out at the Institute of Medical Psychology and Behavioural Sciences at the university, where the researchers described the “box” linking the electrodes to the computer as the standard electroencephalograph (EEG) which is able to record the signals from the brain. These electrical signals are transmitted to the computer which has been programmed to respond to certain subconscious thoughts when they are detected The device in its present state of development is said to cost some £3,000 to £5,000, although the developers predict that it could sell for as little as £150 if it was marketed as a mass -produced unit.

Experimentation with the machine

Demonstrations of the machine have been held for the media. One such exercise involved electrodes being connected to the brain so that the frontal lobe, which is used for planning and many high-level cognitive activities, could be used to control a ball that was displayed on a computer screen. It formed the basis of an experiment that simulated a football match and the screen was setup with two goals and a marked-out pitch, the aim being to control the movement of the ball with your mind so that goals could be scored. Currently we are informed a “thought” instruction is obeyed correctly in eight attempts out of ten.

When the basic operations are mastered by the machine's user higher levels can be reached, as in so many computer games. One remarkable demonstration is that of the use of the machine by a handicapped user who is a patient with “locked-in syndrome”. He has learned, we are told, how to use the machine at this higher level so that he can operate a word processing system, giving him, once again, the ability to write.

Many other applications, particularly in the medical field are being carried out or initiated at the project-level. They involve Professor Birbaumer says, work with, stroke victims, epileptics, hyperactive children and many other conditions.

Future developments

Research in this area is, of course, being carried out at other institutions, and several companies are marketing simple devices. Researchers at Tuebingen are, however, planning more advanced versions. One using a magnetoencephalograph (MEG) is a device that costs £1.2 million and is capable of providing more detailed readings by using magnetic fields instead of electric currents. The software used in the system is also being further enhanced so that when the word processing system is controlled by TTD words can be completed after only a few letters have been selected by the user, One main thrust of their development is to produce a device that will allow patients to control medical aids and household appliances. Professor Birbaumer sums up the state of the research by saying that:

The technology is still slow but not that we have crossed the border and been able to interpret thoughts we have the ability to use TTD for many other operations-switching on the oven, ordering groceries or making phone calls. Some patients can already open doors or switch on the TV.

Finally, he believes that it is only a matter of time before thought can be interpreted as words. This will then lead to many more applications that will cover a wide range of activities. The main criterion, however, is that a truly effective system can be produced which has a high reliability and an acceptable “hit rate” for “thought collection” and “thought evaluation”. Then the device can be commanded to carry out efficiently the system user's intention. Most cyberneticians and systemists will agree that this is a daunting task which will not be completed in the short term.

Innovations

1. Computer program to translate Latin

The online journal of the British Computer Society (2002) has given details of a new computer program called Brutus which can translate Latin into English. It has been developed by Dr Paul Bowden of the Department of Computing at Nottingham Trent University, UK. It has been designed intially to help children learn the classical language and for teachers to mark translations. Language translation has become very sophisticated since the early days of 1968 when language professionals became interested in applying computers for teaching purposes and to assist in the translation of languages for business, industrial and social purposes. There are a number of differing techniques that can be applied to the problem, and in this project of Latin to English we are told that the simplest one has been utilised. This involves taking one word at a time, with the first translations being refined in a series of subsequent operations. Dr Bowden says that:

As far as I am aware there is no other program like it. I used my old school textbook to program Brutus and it turns out that it can translate about ten Latin sentences per second from the schoolbook exercises. I remember it took rather longer when I did those very same exercises at the age of 12.

It may be that this approach to language translation is particularly suitable for a structured language such as Latin but the success of the basic algorithm should encourage others.

2. UK Curling team apply computers and science

Everyone now knows that there is more to winning a sports award than simply turning up for the competition and presumably a degree of talent for the sport is also essential for success but that, it would appear, is not sufficient in today's international games. With national pride at stake large financial assistance is found by states to back their representatives. This enables a sports person to have the full backing for facilities for training, and for specialist professional coaching. It now also means that scientists can also be involved as well as the most up-to-date computers. Indeed cybernetics because of its interdisciplinary nature would provide the ideal backup for a training program if its researchers and proponents could be encouraged to take part.

Researchers at Edinburgh University's Department of Materials Science and Engineering, led by Dr Jane Blackford were only too glad to assist in preparing the British Olympic Curling Team for the recent Winter Olympic Games. Central to the success of the team, which, of course won Britains's first gold medal for 18 years, was a pioneering “sweep ergometer” developed by the Edinburgh scientists. The device, which uses sensors embedded in the curlers brushes to send data to a computer, was used to perfect the British team's sweeping technique, which is one of the most important aspects of the game. During their preparations for the games the team took part in a series of experiments to simulate match play, when players sweep in front of the moving stone with various degrees of force to melt the ice and make the curling stone travel straighter and faster. The experiments involved having the brushing strokes analysed by the computer program to create the perfect sweep.

In addition the team was backed by another project that used computer systems for a “notational analysis” to dissect shots and to improve tactics Every stone played by the women's team was filmed and relayed to the computers to give them an instant breakdown of their play.

The researchers at Edinburgh University who invented the sweep ergometer say that they:

... dismantled a conventional curling brush and fitted an accelerometer in the brush head to measure the speed and the direction in which the brush was moving. Between the brush head and the shaft they fitted a strain gauge load cell, which measured the horizontal and vertical force the sweeper was applying.

The measurements were downloaded electronically through a wire connected to the broom shaft and relayed to a computer. Over two days' work with the women's team, the data was converted into graphs and the scientist curlers and the team coach, went over the results.

They looked at the consistency with which each sweeper kept her brush head moving for the 25 seconds the stone was in motion. On average, a sweeper could keep her broom moving at a frequency of roughly four times a second. They looked at the pressure each sweeper applied and whether it varied. The results allowed each member to build up a picture of her technique.

This research was assisted by a machine developed by a Canadian University that can fire a curling stone with the same force each time. Researchers now plan to continue their innovative programme to measure which type of sweeping is the most effective. Using the Canadian machine they can vary the direction of travel and the speed of the stone over the ice.

It has been suggested that at the next Olympic games, and indeed at lesser events, there should be awards for the scientists who provide the expertise needed to win.

3. Personal flying machine

The company Millennium Jet, Santa Clara, California, USA is developing a personal flying machine called Solotrek. The device is a strap-on helicoper which, it is claimed takes off and lands vertically. It is powered by two fans each protected by casings. The machine, the developers say has a range of some 150 miles on one tankful of petrol. The user of the device stands in the machine controlling it with hand controls and body weight, whilst the fans are able to provide lift and when tilted forward will provide thrust.

The machine is being developed for the police, military and the rescue services and NASA is testing it in a wind tunnel.

The project has been previously introduced when it was made clear that it remains to be seen whether it is a viable one. There have been a number of previous attempts worldwide to produce such machines. The main problems being encountered were stability and control as well as practicality. Much is learnt from the failures that have been reported in this columns. We are told that even if Solitrek is found not to be feasible it is a step towards a more workable system in the future. A system of “personal propulsion” recently tested in Russia has also provided more test data for these innovative experimental projects.

Automation and cybernetics

Robots in the home – no longer a cyber fantasy

New robots for the domestic scene. In November 2001 Electrolux claimed it had launched the world's first “autonomous floor cleaner” about a hundred years after Hubert Cecil Booth invented the first successful vacuum cleaner. In March 2002, however, Matsushita announced its rival product also billed as “the world's first robotic vacuum”. No doubt there are other manufacturers with similar claims. But what is of most importance is that robots designed for the domestic market are being developed and marketed. Of those recently publicised in these reports it is of significance to note that they are of the “single function” variety. These are robots that are designed to perform a simple and necessary operation. These are the ones that carry out one task and do it well. The range now available for the home market includes robots for tasks in the garden which are single-tasked such as the robot lawn mower. The robot engineer Hans Moravec is reported as believing that:

There was a lot of optimism in the beginning when robots first appeared on the scene. But some of us began to realise there was a major miscalculation in building robots and that had to do with the belief that complicated programming was involved.

The best robots today are those that are the simplest in what they do. These are the ones that have one task and do it well. Any time you try and build one that does a multitude of things, like a robot that can do everything at home, you inevitably fail because there are too many ways in which something can go wrong.

We report on two robot cleaners in the next sections.

The Trilobite – first robot cleaner?

Electrolux's Trilobite is claimed to be the first vacuum cleaner to go on sale with the promise that it will perform the most tedious of domestic tasks better than humans. It immediately was criticised by its rivals as being “little more than a motorised carpet sweeper”. The company on the other hand believe their device is the first autonomous floor cleaner to be brought onto the market.

It is 13 inches wide and it is said to be able to vacuum some 95 per cent of the area to be cleaned and could slip under such low furniture as coffee tables much more easily than the conventional ones. It has already been on sale in Sweden with the United Kingdom next in line as a marketing venue. The company say it will cost about £1,000, which in the UK is about six times the cost of an ordinary vacuum cleaner. We are told that the device does not negotiate steps and that it requires a magnetic strip hidden under the carpet to prevent it leaving the target area. Electrolux research has indicated, it is reported, that with the 3.6 million vacuum cleaners sold in the UK in a year some 180,000 homes could be prepared to pay more to make the chore a little easier. No technical details are currently available but presumably the designers have harnessed a suction motor and sensors to a carriage that is sufficiently mobile to navigate walls, furniture and other obstacles. Many of the developers rivals who have been attempting to produce robot vacuum cleaners suggest that it only uses 90 watts of power when conventional cleaners need 800-1,600 watts. In developing such devices much of the available power has to be utilised to beat the carpet, suck up debris and manoeuvre the machine around the room. Any shortage of power would result in the vacuum having reduced suction the rival developers say. They even claim that a robot vacuum with an adequate suction power would required either heavy batteries or run for only short periods. We will see how the competing developers of such robot devices fare in the coming months.

Matsushita's robotic cleaner

Developed by the Japanese electronics giant Matsushita the robot cleaner is slightly bigger than a football and apparently has the intelligence to avoid falling down stairs, or hitting walls.

Not a great deal of technical data is available, for obvious commercial reasons, We are told, however, that it is fitted with 50 sonic and infra-red sensors and that it can work for up to an hour on a single battery charge after using its sensors to measure the size of each room that needs to be cleaned. Some £1 million has been spent on its development and when the current trials are completed a full market launch will be made. This has been provisionally set for mid 2002.

The company who also own Panasonic have been researching automation in relation to domestic chores for some time and believe that robots can be developed for many household tasks. They cite guarding the house against fire or burglary in homes whilst the occupants are asleep or away as a likely project. The renewed interest in robots in domestic environments will undoubtedly see new ranges of personal robots being made available for the consumer market.

Brian H. RudallNorbert Wiener Institute and the University of Wales UK