Cybernetics in Industry

R.J 1990

Robots have been talked about and tinkered with ever since Czech writer Karel Capek's 1921 play, Rossum's Universal Robots, which brought the word into the English Language. But it was the advance of the microchip in the 1970s that made possible enormous advances in robotics. The first generation of computer controlled robots in the 1970s were called universal devices (UTDs) and were little more than mechanical arms. They were also "deaf, dumb and blind" and very inflexible. They were used for tasks such as spot-welding, paint-spraying, loading and stacking.

But now, improvements in computer software and the development of vision systems (robots that see) are transforming the prospects for these "steel collar workers". Robots are being used increasingly in general manufacturing, especially in sectors like the automobile industry. Robots enjoyed something of a boom in the early 1980s in the USA and Europe, as firms flocked to join the business. In Japan, Fujitsu already has a plant where robots are helping make other robots. In another japanese factory a robot recently crushed a human worker to death, causing the worlds first high-tech murder.

Robots have come to symbolise high level industrialisation of society. It appears that industrial robots are the wave of the future when it comes to manufacturing, and the perceived demand for greater industrial productivity. A flood of articles, magazines, and conferences on production automation, and, in particular robotics, shows this. So do trade shows and meetings.

Improved productivity, reduced costs, and better manufacturing quality are the objectives on which most agree. How to achieve them is not so clear. Politicians and Labour leaders want the governments' intervention to rebuild conventional, old plants to put workers back on the job. Academics want new approaches to the manufacturing processes. Economic planners call for a shift to manufacturing industries of the future, like aerospace electronics and computers. To increase industrial productivity and to alleviate some of the social problems, changes in methods have been and are expected to be introduced at all levels of industrial production through the use of computers. An overall trend stands out, namely, the development and implementation of computer automated manufacturing leading toward the realisation of the computer integrated automatic factory as a complete entity in the future. This trend is impeded, however, by a loss of flexibility to change products quickly and maintain cost effective production. This is particularly true for small volume and mid-volume production. The production in companies manufacturing in the batch-type mode is therefore in most case still done by human operated machines organized according to conventional workshop or assembly line principles. In contrast, high-volume producers employ highly automated transfer lines where possible. These usually become more cost effective with increasing production volume.

The advent of numerical control and the development of numerically controlled machining centres improve the situation somewhat for the low-volume producers of parts. Nevertheless, most mid-volume manufacturers continue to search for production systems that can significantly reduce per-piece part costs below those of the job-shop with traditional technologies. Computerized manufacturing systems, flexible manufacturing systems, higher level direct numerical control systems, etc. are some of the more promising answers offered by builders of machine tools to the mid-volume manufacturers. Although the approach varies with each machine-tool builder, the ultimate objectives are the same-

1. Minimise in-process inventory; ie parts bins needed to perform the task.

2. Minimise lead time; ie idle time when nothing is happening.

3. Minimise direct labour.

4. Minimise tool changing and setups.

5. Maximise equipment utilization.

6. Maximise flexibility.

In working towards these objectives, robots quietly take their placed alongside humans on the production line to raise productivity and to do the 'dirty work'. They can do most of the work still performed by humans, even in plants filled with automated heavy machinery. They can handle materials, load and unload, sort, stack and do assembly operations. They can position workpieces on machines, weld, spray, rivet, rout, sand and grind. They can do many of the monotonous, hot, disagreeable, dangerous tasks formerly assigned to humans, as well as new tasks that humans can not do. They can work for thousands of hours with, typically, less than two percent downtime. Robots can be re-programmed to do different tasks, and the digital electronics of their control systems places them squarely within the computer-aided manufacturing (CAM) arena.

The purpose of this report is to examine the use and application of robotics in industry with a view to evaluating their impact on the company, its workforce and society as a whole.

The History of Cybernetics

In Capek's 1921 play, hero engineer Rossum creates a new breed of robots to do the world's dirty work. He took the name for his human-like creatures from the Czech word robota, meaning forced or slave labour. After taking on all the humans' unpleasant jobs and fighting their wars, Rossums robots eventually rise up and take over the world, like so many Frankensteins monsters.

Unfortunately, this has given "robots" a bad name and an unjustified reputation, which has been perpetuated down through the years- for example by Isaac Asimov's science fiction writings of the 1940's and films like Star Wars, which featured the robot, R2-D2. Todays industrial robots are really nothing like Capeks humanoids. Most cannot see, none can hear, smell or think. They are simply machine tools, typically no more than a mechanical arm controlled by a computer which can be programmed to carry out different movements. In fact, the Robot Institute of America offers the following definition of a robot: "A reprogrammable, multifunctional manipulator designed to move material, parts tools or specialised devices, through variable programmed motions for the performance of a variety of tasks."

The age of the industrial robot really began in 1946, when US inventor George Devol developed a memory device for controlling machines. The first patent for a programmable arm was filed by Devol in 1954 and in 1960 Devol teamed up with the American engineer Joe Engelberger to form Unimation, the first company to make and sell industrial robots. By using the word "robot" to generate interest in their product, they actually succeeded in confusing everyone!

But the marketing ploy worked in that Unimation's "robots" generated piles of press cuttings- if not revenue. In fact, industrial robot sales were very slow to take off: the world's first industrial robot was installed in 1961 at a General Motors factory in New Jersy USA, but the high costs and teething troubles delayed subsequent installations to such an extent that Unimation failed to show a profit until 1975. But the potential of robots to greatly increase productivity and reduce manpower was steadily growing more apparent.

In the book The Robotics Revolution (Peter B. Scott, 1984), it is pointed out that these so-called "first-generation" robots were ideal for dirty and repetitive jobs such as welding, paint-spraying, grinding, moulding and casting. Of the 4700 robots in use in the USA in 1981, for instance, 1500 were used for welding, 850 in foundries, 840 for loading and 540 for paint-spraying. Only 100 were used for assembly tasks. Scott's "second-generation" machines are able to "see" and "touch", while the coming "third generation" will have intelligence and massive computing power, enabling them, for example, to "infer logically"- that is, to work out for themselves how to do something.

Three key developments sparked off the robotics boom in the early 1980's. The first, of course, was the arrival of the microchip, which meant that the computer brains of robots were much cheaper and took up less space. Second, the fear of foreign competition- particularly from the japanese- and the disclosure that US productivity gains in the 1970's were miniscule and were lagging behind the rest of the world stirred US manufacturers into action. For instance, between 1947 and 1965 US manufacturing productivity grew by 3.4 percent a year on average, but this dropped to 2.3 percent in 1965-75 and below 1 percent in the late 1970's. Third, wage inflation and extra fringe benefits were dramatically increasing the costs of human assembly-line workers relative to robots. The incentive to invest in new technology was never greater.

Furthermore, a number of systems up-and-running - particularly in the vehicle industry - had clearly demonstrated the productivity gains to be obtained from using robots in certain situations. Workforce representatives were generally impressed by their successful application in hazardous environments such as furnaces, chemical factories, nuclear power stations, in outer space and on the ocean floor. Besides, there was no great merit in having people work in dirty, noisy, dusty and dangerous jobs.

By the end of 1984, according to British Robot Association figures, there were about 100,000 robot devices at work in the world, of which 64,600 were in Japan, 13,000 in the USA, 6600 in West Germany, 3380 in France, 2850 in Italy and 2600 in the UK. On a per capita basis, the European order was Sweden first, with about 19 robots per 10,000 production workers, followed by Belgium and West Germany, with only 3 per 10,000 workers, about the same as the USA and far less than Japan. By the end of 1985, the USA had 20,000, West Germany 8800 and the UK 3200.

In the mad scramble to get into robotics in the USA in the early 1980's, computer giant IBM entered the market with two home-developed robots, while General Electric (partly through its acquisition of Calma and Westinghouse which bought Unimation in 1982) aimed to offer "complete solutions" to factory automation, making available a whole range of equipment. But some people alleged that the robotics revolution had been overblown and there were now too many players (perhaps 250 worldwide) competing in what was still a fairly small market in relative terms. As Joe Engelberger once quipped, "I sold the idea of robots to my competitors, not my customers." Indeed, the Japanese were quick to introduce a whole range of robots from small pick and place units to whole systems onto the market. A shakeout of some kind seems inevitable- the suggestion is that many of the smaller, specialist robot companies will be forced out of business or into merger with larger companies which have the resources needed to compete in the Factory-of-the-Future market. For example, American Robot Corporation sold out to Ford in 1985.

The robotics industry in the UK has largely been influenced by the US, and as such, has not been out in the front in the design stages. Relatively few companies use robots as yet, and they are only starting to be used in the car industry especially the Japanese owned Nissan plants. Other applications are found in, for example, R.A.Listers in Dursley, GLOS where small robots are used to remove flash from die castings.

Arguments For and Against Robots

The arguments for and against robots and robotic systems in the production industry are closely related to those of industrial automation in general. Production automation became a national issue in the late 1950s and early 1960s, when labour leaders and government officials had debated the pros and cons of automation technology. Even business leaders, who see themselves as advocates of technological progress, have, on occasion questioned whether automation was really worth its high investment cost.

Some of the motivating factors for introducing robotic systems into the production industry can be subdivided into technical, economic and social categories-

1. Technical Factors: Human capabilities are in many cases not sufficient to satisfy modern requirements of precision, speed, endurance, strength, uniformity etc. Robotic systems offer such capabilities. They provide in addition a link between the rigidity of fixed and direct numerical control automation and the flexibility of humans. They offer-

A. High flexibility of product type and variation, and smaller losses for preparation time than conventional automation.

B. Better product quality, fewer rejects and less waste than human intensive production.

2. Economic Factors: An overriding consideration in the application of robotic systems is the associated economic picture. Ever increasing competition and at the same time increasing compensation costs for labour call for increased productivity. Robotic systems offer to contribute to productivity increases through items such as:

A. Providing maximum use of capital intensive production facilities, possibly in three shifts around the clock.

B. Reducing production losses due to interruptions, absenteeism, and labour shortages.

C. Reducing in-process inventory.

D. Reducing manufacturing lead time.

3. Sociological Factors: The impact on the human factors of production can be beneficial in various respects. Many low level or otherwise undesirable tasks can be done by robots. This includes work in dangerous or unhealthy environments, monotonous activities with short recurring cycles and heavy physical tasks. The benefits of robotic systems include:

A. Reduction of accidents (safety).

B. Removal of conditions dangerous to human health.

C. Shorter working hours.

D. Increased living standards.

To these should be added that the growth of the robotic systems industry will itself provide employment opportunities. This has been especially illustrated in the computer industry. As the companies in this industry have grown (IBM, Burroughs, Digital Equipment Corp., Honeywell etc.) new jobs have been created. These new jobs include not only workers directly employed by these companies, but also computer programmers, systems engineers, and others needed to use and operate the computers. Similar arguments could be made for robotic systems.

The arguments against robotic systems are also closely related to those against automation in general. Such arguments include:

1. The subdue of humanity by the machine.

2. Reduction of the labour force with resulting unemployment.

A resolution of these question will not easily be found. In fact, the impact of robotics on industrialized nations and society in general must be viewed in the light of new evidence each year.

The Anatomy of Robots

Robot capabilities range from very simple repetitive point-to-point motions to extremely versatile movements that can be controlled and sequenced by a computer as a part of a complete, integrated manufacturing system.

All robots consist of two major component systems. First, there are the moving parts, chiefly comprising the arm, wrist and hand elements. This is the most obvious part of the robot to an outside observer. Complementary to the moving robot system is the control system. At its very simplest, this might consist only of a series of adjustable mechanical stops or limit switches. At the other extreme are the computer type controls, which give the robot a programmable memory, which allows the robot to follow a path that is accurately defined all along its length by a series of continuous coordinates, and which can also be coupled with another computer or machine control system to synchronize the robot for the most efficient and safe production operation possible. We can place all industrial robots into one of three classifications-

Limited Sequence robots

As its name implies, a limited sequence robot is at the least sophisticated end of the robot scale. Typically, these robots use a system of mechanical stops and limit switches to control the movements of arm and hand. Operation sequences can often be set up by means of adjustable plugboards, which are themselves associated with electromechanical switching. By electromechanical switching is meant a combination of relays and rotary or stepping switches. As a result of this kind of control, only the end positions of the robot limbs can be specified and controlled. The arm, for instance, can be taken from point A to point B, but the path in between is not specified. Thus, the controls simply switch the drives on and off at the ends of travel. This mode of operation has earned such machines the nicknames of 'pick and place' robots.

The arm drive mechanism could be electrical, pneumatic or hydraulic. Most robots of this type are small, and tend to move faster than their bigger, more complicated brothers. The use of mechanical stops and limit switches gives good positional accuracy, which is typically repeatable to better than ⁺-0.5mm. Purchase price is around 25%-50% of that required for bigger robots. They have been used successfully in a variety of applications, including die-casting, press loading, plastics moulding and as part of special purpose automation. Disadvantages, other than the obvious control limitations, are that the number of limb articulations is likely to be few, and setting the machine up is more time consuming and tedious than for those with better control systems. The number of movements possible in a total production sequence must be limited to the number of limit switches, stops, and programmable switches contained by the robot. Such robots are not 'taught' to perform their job, but have to be set up in the same way as an automatic machine would be adjusted. There is no memory unit as such, other than that embodied in the settings of the plugboard and all the mechanical stops.

Programming limited sequence robots is usually accomplished by setting up all the end stops in their appropriate positions, and by adjusting contacts in the sequencing unit so that all the steps take place in the correct order. A peg board os often used to make the task of sequencing quicker.

Here is a description of a limited sequence device in operation. When a sequence is to be started, the controller has to switch power to the relevant drive motor. If the drives are electric, then the controller will probably close a relay to switch the current through. Where the drives are hydraulic or pneumatic, then appropriate solenoid valves are operated. The motion generated by the drive normally continues until the moving limb is physically restrained by hitting an end stop, the physical shock being cushioned by some form of shock absorbing device. Thus there are only two positions at which the moving part can come to rest, one at the beginning and the other at the end of a programmed move. Obviously, the system is arranged so that a limit switch cuts off the motive power as soon as the end stop is reached. When the initial movement has been finished, the limit switch not only cuts off the drive power, but it also signals the controller that the particular movement has been finished, so that the next movement can start.

The controller is a sequencing or stepping switch. It could be a set of contacts operated by cams on a spindle which can be rotated in steps of a few degrees at a time by a small electric motor. Each time the sequencer receives its own drive signal, it steps to its next position, and so switches drive power on or off to a particular part of the robot. This performance is repeated, step by step, until the whole program has been carried out, the manufacturing cycle is complete, and the robot is ready to start all over again on another cycle.

So how does the controller ensure that the robot doesn't put its arm into the closing jaws of a press, or try to load a workpiece into a spinning chuck? The robot cannot see the machine it is trying to operate. There are no robot senses equivalent to those of a human operator. Some method has to be found to make the robot aware of the real world around it. This is accomplished by providing additional limit switches or other electrical sensing devices on the machine to be operated. These are connected to the controller to provide additional signals to the sequencer, complementary to those obtained from the switches mounted on the robot itself. Robot limb movements are therefore carefully interlocked with the machine being operated. This prevents the robot from trying to commit suicide, avoids collision damage to associated plant, and enables the robot to carry out its operations not only in the correct sequence, but also at the appropriate moments in time.

The next couple of pages show details of a small pick and place robot designed and made by M.Collins of Collins Tooling, Cheltenham. The robot was designed to remove sprues from plastic injection mouldings and involves four articulations, including the gripper, is powered pneumatically and is controlled by a computer. Shown on several photographs are white nylon blocks which house the limit microswitches. The activators for these are adjustable and can be moved to alter the movement permitted for each articulation.

Sprue removing robot-

Sprue removing robot-

The next eight pages show simple pick and place robots that are available commercially. They are all essentially computer controlled and are limited in their abilities due to lack of articulations and programmable manoeuvre.

Playback Robots- with point-to-point control

One characteristic of limited sequence robots is that they are generally difficult to reprogram. This arises from the nature of the control system and memory, which are all embodied in a complex and interdependent set of limit switches, interlocks, end stops and electrical connections. Not only does this kind of electromechanical arrangement prove tedious to change, but it also limits the number of different sequence steps that can be accommodated practicably within the control system. Another method for achieving positional control of each limb relies on the provision of some form of servo mechanism. Each movable robot limb is fitted with a device that produces an electrical signal, the value of which is proportional to the limb position. The system is arranged so that the direction of drive travel is such as to reduce the positional error, and as the limb moves closer to the desired position, the error signal automatically reduces until it reaches zero, and the limb stops in the correct position. This is analogue control, and in practice calls for a high degree of engineering skill in design to achieve satisfactory positional accuracy, and freedom from oscillation.

If a knob is provided on a control panel which can vary the command signal for a particular limb, then that limb will move as the knob is moved. Thus, a form of remote control is achieved, and the control panel can be given as many knobs as there are limbs to provide a man with the means for operating the robot.

The device described so far is a manipulator, and not a robot. A memory unit has to be added to complete the control unit before it can properly be called a robot. Once a memory unit exists, a very flexible robot results. The position of the limbs at each operational step, and the total operational sequence are all recorded in the memory. The memory is then used to stimulate all the servo systems. The procedure for setting up such a robot is far easier than for a limited sequence robot. It is only necessary to use the controls to drive the robot limbs to the required position for each operational step, and then to record the exact condition of the robot in the memory by the simple act of pushing a button before proceeding to the next step in the sequence. In other words, the robot can be taught, by simply driving it through all stages of the operation.

For obvious reasons, a robot which can be taught in this manner is sometimes known as a playback robot. It is still essential to provide safety and control interlocks between the robot control unit and the machinery being operated, to prevent collisions and other problems. Not only is the robot far easier to set up than a limited sequence robot, but the memory unit is able to take advantage of modern technology by digitizing all of the command position data. This means that the robot is able to remember a large number of steps. Point-to-point robots are obviously capable of doing any job performed by a limited sequence robot. Presuming that their memory capacity is sufficient, they are also capable of more sophisticated jobs such as palletising, stacking, spot welding and the like.

Playback Robots- with continuous path control

There are applications in manufacturing industry where it is necessary to control not only the start and finish points of each robotized step, but also the path traced by the robot hand as it travels between these two extremes. A good example of this requirement is provided by seam welding, where a robot is asked to wield a welding gun, and move it along some complex contour at the correct speed to produce a strong and neat weld. One way of looking at this problem is to regard continuous path control as a logical extension of point-to-point control. It is feasible to provide a robot with a memory that is sufficiently large to allow path control, that is, to all intents and purposes, continuous.

Alternatively, the continuous path robot may be taught in real time. The operator takes hold of the robot by its hand, and leads it through the motions that it is going to have to perform by itself. The operator tries to copy the speed of travel required. During this teaching process, the robot has to record the movement and hand attitudes continuously, or approximately continuously, in its memory. This is achieved by giving the robot and internal timing system, which, for example, could be synchronised with the 50Hz main supply frequency. Using this time reference, the robot's movements are sampled at the rate of 50 times each second, with the results being committed to memory. Even at this sampling speed, a large amount of data has to be accumulated in the memory. Consequently, the storage systems for continuous path control robots of this type often are magnetic tape units.

To increase the operational usefulness of continuous path robots, provision is usually made for the playback speed of operation to be different from the teaching speed. This is a great help to the operator, because in some applications the actual speed of travel is very slow, and the operator finds it easier to teach the robot at a faster speed, where he can make a smoother run, free from handshake. The converse is obviously also true. For higher speed continuous path programs as in applying a bead of sealant, it may be preferable to program at a lower than playback speed.

The next two pages contain a report written by FIAT, praising their robot assembly facilities in Italy. These are playback robots with continuous path control but some, as shown, combine their activities with the ability to see- they have small CCD cameras mounted on arms and can tell the outline of objects.

Present Applications of Robots

Since robotics was introduced to industry in the 1960's, the number of uses found for robots has increased dramatically. The next few pages describe some of these.

Die Casting Applications

In the die casting process, parts are formed by forcing molten non-ferrous metals under pressure into metal moulds called dies. Alloys of lead, aluminium, zinc, magnesium, copper and brass are commonly used. The die casting machine consists mainly of two heavy platens, one fixed and the other moving, which accommodate the dies, these normally being fabricated in two halves. The whole design is massive enough to withstand the very high pressures used, typically reaching thousands of pounds per square inch.

In operation, the die halves are closed and locked together automatically under pressure generated either from air cylinders or by hydraulic means. Molten metal is then delivered to a pump which may be cold or heated to the temperature of the molten metal according to the method used. The plunger of the pump is advanced to drive metal quickly through the feeding system while air in the dies escapes through vents provided for this purpose. Sufficient metal is introduced in each slot to overflow wells built into the dies and to produce some surplus metal or 'flash'. The pressure is maintained long enough to allow the metal to solidify, after which the die opens to permit the casting to be ejected.

After removal from the machine, the casting is often quenched by transferring it to a bath of water. In the simplest case the casting simply falls from the machine into a water bath on ejection. From the quench tank, the casting is placed in a trim-press which removes all excess metal.

The die casting industry has been a pioneer in the use of robots, the first being introduced as early as 1961, Since then more than two million hours of operation have been accumulated in workshops throughout Europe and the USA. In many instances these machines work 7 days a week, 24 hours a day on long and short runs.

More and more die casters are using robots to keep up with modern foundry practice and competition. Unimate robots at work in die casting far outnumber all other robots combined, and they work in both custom and captive shops, round the clock and single shift operations, manipulating light and heavy loads.

While robots have been used in a variety of modes in die casting they have without exception provided high net yields, better and more consistent quality, reduced die wear and lower costs. Downtime due to malfunctions average less than 3%. On short-run jobs where changes in the robots' program are required frequently, die-setters have found that they can teach the robot new routines after only a short period of instruction. In the more aggressive shops, up to 12 die casting machines can be serviced by six robots, all under the supervision of one operator. Here are some of the operations which robots are now successfully handling in die casting shops.

1. Unloading one machine, quenching the part and disposing it.

This is the simplest and least sophisticated usage, and as a result it probably offers the lowest economic return.

2. Unloading one machine and rough trimming.

This operation is similar to the previous except that a trimming operation is included, usually to separate the rough casting into several different parts. Rough trimming is required when multi-cavity dies are used to cast several parts simultaneously, after which they must be separated, sorted and finally trimmed and finished. Quenching is often unnecessary, and careful positioning in the separating tool is seldom required, so high productivity can be maintained. In addition there are no pieces of trim and flash to dispose of since the operation merely separates the parts from the sprue and the runner.

3. Alternate unloading of two die casting machines.

The long reach capability of robots makes them adaptable to unloading two die casting machines alternately. This mode of operation is generally employed where larger castings and longer machine cycle times are involved. Depending upon the length of these cycle times it may be possible to include trimming or separating in the robots sequence.This mode of operation achieves maximum utilisation of the robot, particularly in the independent shop where short runs may place a premium on set-up time. Once the robot has been taught a basic program for each of the two die casting machines, a die change requires, at most, a slight adjustment of the pick-up point to cater for the changed sprue-location provided by the new die.

4. Unloading one die casting machine, quenching the part and trimming it.

A dynamically superior robot can unload a die casting machine and quench the part at a rate of about 500 shots per hour. When larger castings are being produced, the production rate is much lower than this, so the robot has plenty of idle time which can be put to productive use. Thus if a trim press is moved to within reach of the robot, the secondary operation of trimming the casting can be accomplished well within the cycle time of the die casting machine.

Long run jobs are especially suited to this operation since then initial set-up time is a minor factor both in time and cost. The sequence has particular attractions when the part requires no further work after trimming, such as plating, inspection or hand work, for then the part can be packaged for delivery to the customer.

5. Unloading a die casting machine with die care performed by robot.

The robot's versatility and repeatability enables it to blow off flash and apply lubricant more effectively than the manual operator or any other automatic unloading technique. This is particularly useful in aluminium die casting. Proper die care accounts for an appreciable percentage of the total casting cycle time and therefore has a great influence on the production rate. Also, the higher temperature alloys require more meticulous die care. These factors coupled with short runs make it difficult and sometimes impractical to employ automatic die care methods which require set-up time themselves. The ease with which the robot can be taught permits the proper air cleaning and lubrication patterns for a particular die to be achieved readily.

6. Load insert into die casting machine and unload casting.

Inserts of materials that are different to the casting alloy can be joined to die castings during the casting operation. For example a steel shaft required for strength and hardness can be inserted in a die cast gear wheel. Similarly, when bearing requirements are too severe for the casting alloy, a sleeve of bearing material may required to be inserted.

Spot Welding Applications

Welding is the process of joining metals by fusing them together. This is unlike soldering or brazing where joints are made by adhesion between two surfaces alloyed with a metal of a lower melting point, such as lead, tin or silver.

For hundreds of years blacksmiths have welded wrought iron by heating the pieces to be joined almost to melting point and then hammering them together on an anvil until they become virtually one piece. In modern welding the heat required to cause the metal to fuse is provided by gas torches, electric arcs or by the passage of an electric current through the metals at the point where they are to be joined.

In spot welding, as its name implies, metal pieces are joined at a number of small localized areas or spots. This is accomplished by passing a large electric current at low voltage through the metal at each point to be welded. Sheets, rather than large metal structures, are suitable for spot welding since more massive pieces will conduct the heat away much too rapidly, making it impossible to raise the temperature of the part sufficiently for fusion to occur.

The part in spot welding arises from work done by the electric current in overcoming the small, but finite, resistance offered by the metals to be welded. In fact, the process is sometimes referred to as resistance welding. For a given level of current, the higher the resistance encountered, the greater the heat generated. In practice, the materials being joined are clamped between copper or copper-alloy electrodes which conduct the welding current to the site of the weld. As current flows through the workpieces from one electrode to the other, localised heat is produced in the column of material directly underneath the contacting area of the electrodes. If this heat is sufficient to fuse the materials, then welding takes place at this 'spot'.

The welding sequence ie. squeeze, weld, hold, off is so short that the bulk of the time taken to complete a job involving several welds is in the movement of the welding gun between spots. Because of the difficulty in handling the unwieldy gun for long periods, and the fact that the operator needs to use judgement in deciding where to position a weld, spot welding is not a very precise operation. This leads to the question whether the process can be automated to remove some of the drudgery from the job and to improve accuracy by removing the human element. When the product is unlikely to change significantly over a long period, then the use of a special purpose automation system would be justified. This might be the case, for example, in a box for a refrigerator which might change very little over a period of years. However, when the line is subject to frequent change, this solution is not economical. This situation prevails in the automotive industry where models traditionally change every year and where typically three or four different body styles mat be under construction simultaneously on the same line. In these situations the robot welder comes into its own.

Over the past decade the main application of industrial robots throughout the world has been in the field of automotive spot welding. It is by no means a simple process; on the contrary it is very complex. Yet in may ways the characteristics of a robot are ideally suited to this application. The entry of the robot into this domain came about as the result of painstaking development work on the part of the robot manufacturers and cooperation from the car makers- the most profit conscious industry of all. It happened in the following way. About 15 years ago in the car industry, body parts, some of them large sub-assemblies, were held in clamping jigs and tacked together by manual workers using multiple welding guns. The unit would then be finally spot welded together by operators who would position the welds using their own judgement. In 1966 the first steps were taken to use a robot to guide the welding tongs, linking its control unit with external signals which commanded the robot when to squeeze, weld, hold etc. By 1969 General Motors had installed a line of 26 Unimates for spot welding car bodies, and this was followed in 1970 by Daimler Benz in Europe who adopted a Unimate for a body side spot welding operation.

From those beginnings, the use of robots in this application has mushroomed, and so important is this process in the industry that it has been necessary for the car manufactures to consider the capabilities and limitations of the robot in deigning the production line itself. Step by step the introduction of robots into the car body spot weld process has continues until more than 1200 such machines are now in use.

To take its pace on an automotive production line the robot must be able to remember several different body styles. This is a difficult task, since cars are deigned from a stylistic viewpoint, with the result that the geometry often gives little help to the robot in establishing reference points such as it might encounter in more functional shapes like appliance housings. Having absorbed the necessary information in its memory system,the robot must, on command, initiate the appropriate program to suit the type of body to be built and then carry out a series of complicated manipulations using a heavy spot welding gun. It must perform these tasks with speed, accuracy and reliability in an industry producing up to 80 car bodies per hour on the assembly line.

This is a job for a high technology robot such as those available from ASEA, Cincinnati, Kuka, Comau, Kawasaki and Unimation.

All of these models are to be found in this application today. They often work a three-shift program, although the third shift is usually a shortened one during which maintenance can be carried out on the line as a whole, not just on the robots. The repeatability and positional accuracy which the robots can achieve provides a much more consistent performance than is obtained with human operators. In fact body strength can be obtained by specifying fewer welds (in the correct locations) than would be the case with a human operator who could not be relied upon to produce the best pattern of welds.

Much of the work currently done by robots is in the area of respotting. This is a technique where an initially tacked together body is finally welded by a group of robots on the production line. The group will be under the control of a single supervising computer which will signal the arrival on-line of a particular body style and cause the robots to switch to the appropriate program in their memory systems. Various interlocks and sensing devices can easily be incorporated to ensure that the robots know what they are expected to make, when to start, stop, etc. Robots thrive on the sort of commands which are inherent in spot welding cycles.

All major car manufacturers are now actively using robots on their production lines. In fact some manufacturers, such as Volkswagen, Renault and Fiat have developed and built their own robots with the specific task of doing spot welding and little else.

In Gothenburg, Sweden, Volvo is using the largest and most sophisticated automatic welding line in the world. A line which previously required a workforce of 67 has been replaced, at a cost of 6 million US dollars, by a robotized line serviced by only a handful of key staff. The line is scheduled to produce 50 car bodies an hour in a mix of 2-door and 4-door models, and the aim is to improve the working environment and produce a more consistent product. The installation was designed by an Italian company which already had designed and installed robot welders in the Fiat works in Italy, Fiat being Europes' biggest user of robots among the car manufacturing fraternity.

This new line introduces several novel features and is the most advanced yet to be built. It consists of ten work stations, connected by a conveyor system. The car bodies in various stages of completion are at all times mounted on pallets. The line is on two levels, the upper one being for welding and the lower for returning the pallets to the starting point. An interesting mechanical feature is that the welding line is indexed whereas the return line moves constantly at about 50 feet per minute. This allows pallets to congregate around the starting point, making the available as required. However the subject of most relevance to this application is the role of the robots. This is best described by outlining what occurs at each of the ten stations along the line.

Station 1. This is the loading station at which the body sections are loaded on to a pallet to be conveyed along the line. No robots are used.

station 2. At this station the body parts are held together in position by swinging 'gates' while they are tacked together by sis robots operating horizontally (Unimate 4000B) and one operating vertically (Unimate 2100). A total of 128 spot welds are carried out at this station.

Station 3. This is the first of two respotting stations at which the tacked body is finally welded. Here, three robots perform 82 spot welds, two operating horizontally (Unimate 4000B) and one vertically (Unimate 2100C).

Station 4. At this, the second of the respotting stations, 138 spot welds are carried out by five robots. (Four are horizontal Unimate 4000B models and the other is a vertical Unimate 2100C).

Station 5. This station illustrates the way in which conventional automation and robots complement each other on this line. The controller here selects a roof from two available types, normal and hatch. It places the roof an the car body shell with sufficient precision to permit the robot installed here (Unimate 2100C) to fix the roof using 14 spot welds.

Station 6. This station is reserved for final roof fixing plus a general check of the body at this stage. A robot finalises the roof position by spot welding in the drip channel, while a manual operator loads tie-plates into an automatic transfer machine which then places them into the correct position for them to be welded into the body. Six robots perform this operation by making 176 welds. (Four are Unimate 4000B and two are Unimate 2000C).

Station 7. No robots are used at this station. Instead, welding guns guided by templates which match the roof configuration produce overlapping spot welds to make a seam. There are four of these guns which each make 60 welds while moving at about 7 feet per minute.

Station 8. This is the point where the final respot is carried out to complete the welding operation on the body. Five robots are employed here (four Unimate 4000B and one Unimate 2100C) and a total of 156 welds are made.

Station 9. This station is purely for back-up and checking. If any robot trouble develops, manual labour can be provided at this point to complete the body.

Station 10. At this position the body is unloaded and the pallet returns by the lower path to the starting point.

The above is typical of the use of robots in spot welding, yet the application is relatively simple. Only two body styles are involved so that the robots being employed are by no means operating at the limit of their memory capabilities.

Arc Welding Applications

The spot welding process described previously is not suited to all welds, especially long-path joints needing a gas tight seal between the two surfaces to be joined. However an alternative electric welding technique exists for such situations in the form of arc welding. In this process the heat required to fuse the metal surfaces together is derived from an electric arc. This is no more than a sustained 'spark' or electrical discharge between two terminals, which in this case are the work and a metal welding electrode.

When the arc is struck, the temperature in the vicinity rises rapidly to as much as 6500 degrees Fahrenheit. At such temperature, a small pool of molten metal forms in the work, and the end of the electrode also melts to contribute additional metal to the pool. Obviously the electrode material must be electrically conducting and compatible with the metal forming the part being welded. A typical electrode is in the form of a metal wire, continuously fed at the correct rate to replace electrode material consumed by the welding process.

The electric arc process creates an unpleasant working environment. Considerable ultra-violet radiation is present in the intense glare from the arc itself, requiring the operator to wear almost opaque goggles or a protective mask to filter out the dangerous radiations and prevent them from damaging the eyes. Sparks tend to fly around the shop, and smoke is generated, so it is little surprise that it is difficult to find welders who are prepared to work a full shift under such conditions and who are at the same time capable of consistent, repeatable results. Once more, enter the robot to take over, unaffected by the environmental problems and virtually incapable of any deviation from a well defined task that it has once been taught.

Most arc welding jobs require a robot with about five articulations because the weld gun is a symmetrical tool. An attractive feature of robot welding lies in the fact that it represents true human replacement, and as such handles perfectly standard arc-welding equipment to do the job and does not demand much in the way of specialised equipment. All that is needed is a fitment to mount the welding gun, attached to its long gas lines and wire feed mechanisms, on to the robots' hand. All other equipment associated with the arc welder will then operate in the normal fashion.

The robots' job is mainly to guide the gun around the programmed path and to signal when it is on station and ready to proceed. The welding unit controller does the rest.

Spray Painting Applications

Today spray painting even in the home environment is common place through the use of aerosols or simple home workshop types of compressor and spray gun. It is perhaps the car industry though, which has set the standards for a high quality, even finished surface, and much research has been done by car makers to develop paints which flow easily, minimising runs and drying rapidly before dust can settle on the wet surfaces.

It is of interest to investigate how a paint-spray operator goes about his business. Holding a gun pressurised by air and fed by paint either from a small tank on the gun itself or a central reservoir, the painter depresses the trigger on the gun to release a fine spray of paint carried in the air stream. The viscosity of the paint must be just right to form a fine spray and not be ejected as lumps or globules to mar the surface finish. The distance from the work at which the spray is projected is also critical in order to produce a coating which is thick enough to cover the surface but not so heavy that unsightly runs occur. The operator can control he process by constantly moving the gun, applying thin rather than thick coats of paint, going back over areas already thinly coated to build up a layer of even thicknesses and good finish.

The paint spraying environment has always had the reputation of being one of the worst which human operators have to encounter. To maintain a dust-free painting area at the right temperature means that the 'shop' should be as restricted in size as possible. Some early paint spraying shops were real death-traps, but nowadays legislation has produced codes aimed at insuring the health and safety of the operator.

Solvents used in painting are toxic and produce a polluted atmosphere, so ventilation must be provided which gives an abundant supply of fresh, clean air to the operator. The optimum paint shop layout to provide this may be not at all the best arrangement for production line painting however, so compromises are necessary. The wearing of masks by painters has been required since the inception of the technique.

More recently, attention has been paid to noise, and the noise levels generated in a busy spray shop, arising from the air discharge through fine nozzles can, in an eight hour shift, cause irreversible damage to the ears. As a result, current standards require operators to wear ear-plugs. Yet another problem is the fire hazard, arising from the highly flammable nature of the materials used.

Finally, as if that were not enough, certain pigments used in the technique are suspected of being carcinogenic agents. This, together with proposed new regulations governing hydrocarbon emissions from manufacturing facilities, will probably result in major changes of practice in the industry, such as the use of waterbase and urethane materials. The whole field is therefore seen as one for further research and development.

The hostility of the paintshop environment has always made it a prime candidate for the adoption of techniques which might take the human operator out of the process so that the implementation of many of the costly protective measures now being demanded could be avoided.

Car production with its competitive approach has been a fertile area for such developments, and many spray painting machines have been introduced. Generally speaking, however, they are limited in their use so that they almost always have to be backed up by human operators who can touch up areas missed by the machine. The machines also tend to be more wasteful of paint; typically they are designed to paint with horizontal and vertical paths on a reciprocator system, that is a back and forth motion. Batteries of spray guns fed from large capacity centralised paint reservoirs move according to a pre-determined program and manage to paint some 70% to 80% of the exterior surface to be covered. Less accessible areas such as the wheel arches, inside the body and engine compartment must be painted by operators who look for unpainted areas as the car body leaves the automatic painter. Colour changes have to be properly scheduled, since this involves changing the paint in the reservoir or switching to a new reservoir, and making sure that the old colour has been cleared out of the paint lines and guns to avoid contamination.

The method is well-established in the industry yet recognized as being far from perfect. The energy requirements are enormous. It takes about 25 million BTU's to build a car. Much of this demand comes from the need to supply fresh air to the spraying booth during finishing. Since governments will be demanding that this be cut, it is clear that there is considerable incentive for the car manufacturers to get human operators out of the paintshop wherever possible, quite apart from the economic advantages.

There is little doubt that the robot offers the best chance for success for the car builder, faced as he is with constant design changes, new colour and finish fads and a buyer who seeks high quality at low cost from a product that must look good as well as perform adequately.

The following justification for robot painting gives some idea of what they expect from their massive investment in this process.

1. Robots will allow us to deal with a hostile environment-

noise

carcinogenic materials

particulate matter

2. Robots will allow us to process with less energy-

reduced fresh air requirements

reduced exhaust

reduced energy cost

3. Robots will allow us to improve paint quality-

less dirt

uniform build

consistent quality level

cope with specialised spray techniques

4. Quality improvements will result in-

reduced warranty

reduced in-house repairs

5. Reduced material costs will follow

6. Reduced direct labour costs will result

To meet the demands of what will clearly be a lucrative market, several robot builders are preparing their robots for this application, and competition may be fierce in the years ahead.

Robot Economics

The success of any commercial industrial undertaking has to be measured in terms of financial performance. The most brilliant technical innovation is a failure if it results in money lost by the entrepreneur or his shareholders. Robots are no exception to this rule. No matter what the social benefits are, no matter how clever the technology, no matter how pretty the robot is to watch, every proposed investment in robotics has to pass the test of a critical financial appraisal.

The following headings provide a framework for management analysis of the costs and benefits of the robotics installation.

1. Purchase price of the robot. The purchase price of a robot is

highly variable, particularly if one definition of robot includes simple pick and place devices with few articulations. The range might extend from ,5,000 to ,100,000 depending upon the number of articulations, operating area, weight handling capacity and control sophistication. Generally speaking the higher priced robots are capable of more demanding jobs and their control sophistication assures that they can be adapted to new jobs when original assignments are completed. So too, the more expensive and more sophisticated robot will normally require less special tooling and lower installation cost. Some of the pick and place robots are no more than adjustable components of automation systems. One popular model, for example, rarely contributes over 20% of the total system cost.

2. Special tooling. Special tooling can be no more than a ,300 gripper for a robot installed at a die casting machine. On the other hand, the tooling may include an indexing conveyor, weld guns, transformers, clamps and a supervisory computer for a complex of robots involved in spot welding of car bodies. For assembly automation the special parts may cost well in excess of the robot equipment costs.

3. Installation. Installation cost is sometimes charged fully to a robot project, but is often carried as overhead because plant layout changes were afoot anyway. At a model change there are usually installation costs to be absorbed even if equipment is to be manually operated. There is no logic to penalising the robot installation for any more than a differential cost which is inherent in the robotising process.

4. Maintenance and periodic overhaul. To keep a robot functioning properly, there is a need for regular maintenance, a periodic need for more sweeping overhaul and a random need to correct unscheduled downtime incidents. A rule of thumb for well-designed production equipment operated for two shifts continually is a total annual cost of 10% of the purchase price. This has been borne out for thousands of Unimates many of which have enjoyed several overhauls whilst accumulating as much as 10,000 hours of field usage each.

There is variability, of course, depending upon the demands of the job and the environment. Maintenance costs in a foundry are greater than those experienced in plastic moulding.

5. Operating power. Operating power is easily calculated as the product of overage power drain and the hours worked. Even with increasing energy costs, this is not a major robot cost.

6. Finance. In some cost justification formulae one takes into account the current cost of money. In others one uses an expected return on investment to establish economic viability.

7. Depreciation. Robots, like any other equipment will exhibit a useful life and it is ordinary practice to depreciate the investment over this useful life. Since a robot tends to be general purpose equipment, there is ample evidence that an 8 to 10 year life running multi-shift is a conservative treatment.

The following observations are offered on potential benefits.

1. Labour displaced. The prime issue in justifying a robot is labour displacement. Industrials are mildly interested in shielding workers from hazardous working conditions, but the key motivator is the saving of labour cost by replacing a human worker with a robot. It is so much better if a single robot can operate for more than one shift and thereby multiply the labour saving potential.

2. Quality improvement. If a job is in a hazardous environment, or is physically demanding, or is simply mind-numbing, there is a good chance that product quality will suffer according to the mood of a human worker. A robot may well be more consistent on the job and therefore it may produce a higher quality output.

3. Increase in throughput. Higher quality naturally means more net output when a robot woks fast enough to just match a human workers' output. However there often are circumstance where a robot can work faster to increase gross output as well. The increased throughput is valuable in its own right, but improved usage of capital assets may greatly supplement the economic benefit of one-for-one displacement of a worker.

Project appraisal by the payback method

Payback calculation is the simplest form of project appraisal. It depends on providing answers to the two questions 'How much is it going to cost?' and 'How soon shall we recover the investment?'

A simple example of using the payback appraisal method is shown. A robot is to be considered as a replacement at a workstation where 250 days are worked in a full calendar year, and where the robot would replace one human worker whose wages and fringe benefit amount to ,12 per hour. Against this saving, the robot would cost ,1.30 per hour to run and maintain. Capital investment, for the robot and its accessories would be ,55,000. The company normally operates one eight hour shift per day, but has the option of increasing this to two shifts per day when production demands are sufficient.

The following lines contain the calculations, and illustrate the method.

simple payback formula-

E = annual expense of maintaining the robot

In this example,

Case 1 Single shift operation

Case 2 Two shift operation

If only one eight hour shift is to be operated on each of the 250 working days, it is seen that the payback period amounts to about 2.6 years. This result would possibly satisfy the companies accountants. If, however, sales forecasts and production plans indicate that two shift working could be maintained, then the payback period is reduced to only 1.3 years or so, which must justify going ahead with the project with no room for doubt at all- or is there?

It seems, at first glance, that there should be no hesitation about investing into the new technology. There are however, two points which can depress the optimism shown by the payback calculations.

The first word of caution concerns the level of robot upkeep costs. The figures used in the above example are applicable to a company which already employs some robots, and which has therefore developed the capability and organisation for maintaining and reprogramming robots. If the robot is going to be a lone newcomer, the first robot in a total population of one, then the company is going to have to set up a maintenance facility from scratch. All costs of that maintenance operation are going to be lumped against the single robot, adding both to the initial investment required and the annual upkeep figures. If the new robot is the second to be acquired, then the situation looks a lot rosier, with the initial investment virtually nil for the maintenance facility (since it already exists), and with the total upkeep costs for both robots amounting to something far less than double the costs for the existing single robot. It has to be said that, for practicable and economic purposes, robots are a far better proposition when the robot workforce numbers greater than three or four.

The second counter argument against the apparently favourable economic incentives shown in the example only becomes apparent which expenditure or returns are considered over timescales measured in two, three or more years. For very short payback periods, there is no hidden snag, and no corrections have to be applied. When payback periods are long, or when return on investment predictions are based on writing the investment off over several years, then discount factors have to be taken into account. Discounting recognizes the fact that a pound spent today is more expensive than a dollar spend next year, since by delaying the expenditure the money can be kept in an interest paying account. Conversely, money earned next year is worth less than the same value of money earned this year. These effects obviously increase with the number of years involved.

Sociological Impact of Robots

Quality of working life

No one will dispute that robots offer unique advantages to those workers who must otherwise spend all of their working day in conditions that include noise, vibration, smells, smoke, excessive heat ir intensive cold, oil spray, flying chips, monotony, or risk of serious personal injury. If, for example, a robot is used to replace a human working in a die casting workshop, the human worker can be transferred to another job in the plant, where he no longer has to breath die lubricant fumes, where hot molten zinc no longer spurts at him, where asbestos gloves are not the standard uniform, and where he can work as a man and not as an eight hour per day zombie.

Social improvements usually conflict with lack of available funds. Robotics is often the exception to the rule, so that moving workers into jobs with better conditions is associated with increased productivity at the robotized work stations. Die casting and press operation are just two of the many examples where moving machine parts pose real physical danger to those who have to tend the machines. There must be many other jobs, not only in the manufacturing industries, where people are at risk, and where robots could take over their duties. It could be argued that those working to produce and market industrial robots have a moral duty to give priority to those specific areas in industry where men and women are being degraded by the nature of their work, or put into actual physical danger.

Attitudes to robots

People, in general, are suspicious of change. Whether a plan is for a new road, a new airport, or the introduction of automated plant in a factory, someone, somewhere is going t object. Even a new employee is going to be regarded sceptically, until he has proven that he can fit in with the rest of the team. Much opposition is the result of fear, and fear is born of ignorance. Every worker wants to be assured that change will not threaten his working conditions adversely, reduce his earnings, or even make his job redundant. If management fails to inform its workers of forthcoming changes, keeps people unaware of the reasons for those changes, and does not discuss the effects of proposed changes on individual livelihoods, then of course people become afraid, resentful, upset, worried and hostile.

Robots perhaps are a special case in this respect. They bear some resemblance to human beings, because they can act autonomously and simulate some human actions. Through fiction, men have long been taught to treat robots with a wary eye. When a robot is introduced into a factory, there would appear to be every reason for a mass walk out.

Since robots were introduced in the early 1970s, the case histories show that industrial robots have qualities that never occurred to market survey analysts. The currently relatively small robot population has already interacted with human co-workers and the union leaderships. Initial curiosity was superseded by tolerance, and finally there has developed something very much akin to rapport in many factories.

As the evidence mounts of man's remarkably benign reaction to robot workers, it is interesting to speculate on the underlying causes. An industrial robot is pathetically sub-human. It is deaf, dumb, and until very recently, blind. The emotion most clearly aroused in human colleagues is one of amused affection. It is interesting to note that a study by the General Electric company into the implementation of robots recognised that management conservation or hesitancy were more serious blocks to a robot installation than worker rejection.

Effect on employment

While quality improvement and increased throughput produce some of the savings, labour displacement is the central benefit. Robots contribute to productivity primarily by displacing human workers. The benefits are clear; but what about any sociological costs?

Even if no individual can point to a particular robot and say that this machine has taken over his job, it is surely true to say that jobs have been eliminated by robots. The impact is miniscule to date. By the end of 1985, there were some 15,000 robots installed worldwide. On average, considering that many robots work multi-shift, the displacement may total 20,000. The net employment impact of the formation of a robotics industry may actually be positive in creating jobs- if one considers the numbers working to develop and produce robots and the numbers working as suppliers to this new industry. This perhaps suggests that unskilled or semi-skilled workers are being taken out of the companies, the workforce being supplemented with new personnel to tend, maintain and program the robots.

So far, robotics is such a new technology that it has no history to predict a sociological cost of job displacement by robots. But, all the technology and specifically all of automation over the past century can be evaluated on a cost/benefit basis.

Consider production gains in U.S. farming. In 1870 47% of the U.S. population was engaged in food production; by 1970, 4% was all that was required to produce all of the food needed for the nation as well as to create a vast surplus. There were dislocations during that 100 years as people left farms for the cities. Still, the benefits have greatly outweighed the costs.

What can be reasonably expected for the robotic industry for the balance of this century?. Many independent surveys seem to be arriving at an average annual growth rate of 35%. At a 35% p.a. growth rate the population will be increasing by some 40,000 per year by 1991. The current technology will not support such a growth rate, but robots with rudimentary vision are already appearing.

Forty thousand robots entering the workforce per year is certainly a large number, but not a disruptive one. It represents about 0.06% of the blue collar work force in the industrialised countries where these machines would be employed. Of course, there will be other automation influences on productivity, as well as robots. These will probably eclipse the impact of robotics since automation is already so much a part of the manufacturing scene. And, automation that is unwaveringly addressed to specific segments of manufacturing is more likely than robotics to have a disruptive influence on employment.

Robots will enter the work force gently. There is no job activity concentration. Productivity that can be achieved without massive displacement is the most sociologically acceptable. Even though all productivity is good, a distributed influence that impacts no faster than natural decline can be the most readily accommodated socially.

Labour leaders in the country will ordinarily support productivity improvements so long as it does not come from some disguised form of speed up or does not result in massive displacement. One other caution is that the union membership shares in the economic benefit.

In short, robotics will contribute importantly to the material well-being of mankind, without painful dislocation of individual workers. If 50 years from now the work week is three days, the air and water are clean again and the industrial life is ever so desirable, we shall be at least partially beholden to robotics.

Future Capabilities

In the following listing of robot qualities sought for the future, there is some reason for explanation. The explanations may reflect upon the technological demands of the problem or they may reflect upon just how urgent is the need for this particular quality. It should be noted that the list is in order, that is, number one, rudimentary vision, is the most likely to have a profound effect in broadening the applications of robotics.

1. Rudimentary vision. Vision allows a robot to recognise things and also to determine where things are. Already vision systems are available on robots and these are pretty good at defining the position of an individual dark part on a white background. This enables the robot to isolate individual parts which may be disorientated. With such rudimentary sight systems, the robot can be helped by maintaining some level of order in the transport of parts throughout the factory. Palletised parts are much easier for a robot to find than are parts dumped randomly into a box.

Research into this field is continuing feverishly, and with cameras becoming cheaper and with greater resolution and computers faster and more powerful, robots with good vision abilities are bound to become available

2. Tactile sensing. A robot with tactile sense could use its capabilities to recognise parts just as humans can do in a darkened room, just by feeling and groping.

For a robot, the more important tactile sense is the sense that tells the robot somehow what is going on during the interaction of parts or tools in its grippers that it is working with. We all know the experience of putting a nut on a bolt in a blind location. We almost instinctively back thread until we feel a click and then we run the nut forward slowly so as to avoid a crossed thread. This is a tactile sense which would greatly enhance the capabilities of a robot.

3. Multiple appendage hand-to-hand coordination. The ability for a robot with more than one arm to move a part or tool between grippers calls for better computing power and software to provide the coordination necessary.

4. Mobility. For the bulk of factory jobs, the robot needs only to stand in one position. There are jobs, however, for roving operators who must tend stations that are widely separated and only need service periodically. For this mobility is needed and so far this has been done by robots mounted and rails to travel between work-stations. What is needed is a robot which can literally stroll.

5. Minimised spatial intrusion. Most robots need substantially more floor space than humans. This can reduce the potential use of the robot because of the cost of laying out new equipment and providing factory floor space may eliminate the economic justification for using a robot in the first place. There is therefore a requirement for a robot which is as small as a human and requires less space than the robot of today.

6. Energy conserving power systems. Robots today use a lot more power than humans ever will. Energy use is not yet an abiding issue in robotics, but it may become so, so there is a requirement for research into more efficient robot muscles.

7. General purpose hands. There may be tasks when one particular type of robot gripper is not enough. There is a requirement to develop a gripper which is as universal as a human task, and can be used for many applications.

8. Man-robot voice communication. With such advances in computers and software is now possible to produce a robot with built in speech capabilities. Such a robot, with speech recognition, could be re-programmed by a series of verbal commands, or the robot could use synthesised speed to explain its view of the work situation or explain internal faults.

9. Total self-diagnostic fault finding. With advances in robot design, so the hardware complexity will increase. If a robot computer is given the ability to analyze itself, giving an indication of the fault area, then the downtime due to system repair will be reduced and production maintained.

It has been seen that the major benefit of robots, especially those employed in the manufacturing industry, will be to improve working conditions by taking over dirty, heavy and repetitive jobs. This is bound to improve the living conditions for those who would otherwise be committed to these tedious occupations. On the other hand, there must be concern for the subsequent lack of jobs resulting from occupation take-over, and especially in the vehicle manufacturing business, even if output increases significantly, no new jobs are going to be created in this department. This is going to be of great concern to the many workers unions because the robot explosion started in a recession, and as the use of robotics grows, so the workforce will diminish and with it, union power.

As a counter to this, more skilled workers are going to be needed. The growth of the number of robots in industry will lead people to spend less time at work. Perhaps the whole idea of work needs to be re-thought; that all employees of all levels should be sent on education courses by their employers to attain the skill needed in modern industry. People need to be motivated to work more conscientiously than ever, yet eventually may need to work far less hours each week. These people will need to be more imaginative in the way they think at work.

In fact, the effect of robots themselves on employment levels are certainly exaggerated. For example, in a nationwide manufacturing industry employing, say, 10 million people, an increase in robot population from 300 to 5,000 will have a negligible effect on employment, so long as there is a commitment among management and employees to remain competitive. As the recent industrial recession has shown, companies that allow themselves to become uncompetitive are suffering much more, and much more quickly than they used to. If each manufacturing company can generate just 1% growth each year, then an extra 10,000 people per million employees will be needed. Thus, with an industry employing 10 million people, 100,000 people or about 50,000 robots are needed to increase output by 1%.

Clearly, a country the size of Britain is unlikely to be able to install 200,000 robots in the next few years- 1,000 a year is more likely. Thus, in the short term at least, the effect on employment levels as a whole will be negligible. If a good proportion of those robots are made where they are used, then their effect on employment will be less. But the pattern of where unemployment crops up will present government and people with major problems- especially where there is no real commitment to remaining competitive.

It is also clear that the nature of employment will change. Whether office workers will actually work at home in large numbers, as forecast by some, is by no means certain. We can say that people will actually do less with their hands than at present.

Forging and machining will be automatic, robots will do almost all handling, and will do assembly, while special tools with some robotic features will spread to the service industries. All this means that people will spend more time watching things, or checking on operations without actually doing them. This will be equally true of engineers and designers who will work with CAD/CAM systems.

While it has been said that jobs will certainly be lost due to robot replacement, there are also great possibilities for engineering companies to produce robots. There are bound to be thousands of opportunities for designers, engineers, computer systems engineers and software writers to go into competition to produce robots for the market. Perhaps one of Britains' major managerial faults is the lack of interest in new technologies. If no research is undertaken by companies into robot design, then, as usual, Britain will lose out to the Japanese and others who always seem to dominate high-tech production.

With a new open European Community opening within only a couple of years, there is an ideal chance for Britain to start manufacturing robots and selling them wholesale to large and prosperous market. Also, with huge changes afoot in the Eastern block countries, Industry should be setting it sights here for prospective consumers of cost effective manufacturing techniques.

If the robotic revolution can be used to increase economic growth, then people will have the time and income to enjoy more leisure or more leisurely work. The three-day week, in which the four holidays a week are well paid, has long seemed a dream. But it will be worth having only if the output per machine is increased dramatically. Otherwise, the three day week will materialise, but not with the necessary living standard to go with it.

With more time on our hands, and higher incomes, there will be a temptation to buy more useless things, and certainly a host of unthought of products, some useful and others useless will come with developments in microelectronics and robotics. Many manufacturers will produce more variations on products with short lives in the hope of remaining profitable. So there is a real danger that we will end up squandering our resources, producing more and more useless products, that people only buy because they have sufficient money to do so. To avoid that problem, a sound education is the best base. It will be necessary to educate engineers and managers how to make the most of robots. It will be necessary to upgrade the technical skills of engineering workers. Then, people must be encouraged to think more positively and expansively about work, and to understand the value of the resources we have. Finally we have to educate the next generations, children, how to cope with these things as well as the forthcoming technology-intensive world.

Bibliography

Industrial Robots H.J.Warnecke,R.D.Schraft

A Textbook of Robots 1. M.Shoham

Robots at Work J.Hartley

Robotics M.Baldwin, G.Hack

An Introduction to Robot Technology P.Coiffet, M.Chirouze

Machine Takeover F.Horge

Microelectronics and Society Freidrichs and Schaff

High Tech Society T.Forrester

Recent Advances in Robotics G.Bebi, S.Hackwood

Robots in Practice J.Engelberger

Robots- Planning and Implementation C.Morgan

[ Home ] [ Up ]

Send mail to webmaster with questions or comments about this web site.
Copyright © 1998 richtel.co.uk
Last modified: October 14, 2012