Robotics technology trends

By : Jim Pinto,
San Diego, CA.
USA

When it comes to robots, reality still lags science fiction. But, just because robots have not lived up to their promise in past decades does not mean that they will not arrive sooner or later. Indeed, the confluence of several advanced technologies is bringing the age of robotics ever nearer - smaller, cheaper, more practical and cost-effective.

This article was published by:

Automation.com, October 2003

When it comes to robots, reality still lags science fiction. But, just because robots have not lived up to their promise in past decades does not mean that they will not arrive sooner or later. Indeed, the confluence of several advanced technologies is bringing the age of robotics ever nearer – smaller, cheaper, more practical and cost-effective.

Brawn, Bone & Brain

There are 3 aspects of any robot:

Brawn – strength relating to physical payload that a robot can move.
Bone – the physical structure of a robot relative to the work it does; this determines the size and weight of the robot in relation to its physical payload.
Brain – robotic intelligence; what it can think and do independently; how much manual interaction is required.

Because of the way robots have been pictured in science fiction, many people expect robots to be human-like in appearance. But in fact what a robot looks like is more related to the tasks or functions it performs. A lot of machines that look nothing like humans can clearly be classified as robots. And similarly, some human-looking robots are not much beyond mechanical mechanisms, or toys.

Many early robots were big machines, with significant brawn and little else. Old hydraulically powered robots were relegated to tasks in the 3-D category – dull, dirty and dangerous. The technological advances since the first industry implementation have completely revised the capability, performance and strategic benefits of robots. For example, by the 1980s robots transitioned from being hydraulically powered to become electrically driven units. Accuracy and performance improved.

Industrial robots already at work

The number of robots in the world today is approaching 1,000,000, with almost half that number in Japan and just 15% in the US. A couple of decades ago, 90% of robots were used in car manufacturing, typically on assembly lines doing a variety of repetitive tasks. Today only 50% are in automobile plants, with the other half spread out among other factories, laboratories, warehouses, energy plants, hospitals, and many other industries.

Robots are used for assembling products, handling dangerous materials, spray-painting, cutting and polishing, inspection of products. The number of robots used in tasks as diverse as cleaning sewers, detecting bombs and performing intricate surgery is increasing steadily, and will continue to grow in coming years.

Robot intelligence

Even with primitive intelligence, robots have demonstrated ability to generate good gains in factory productivity, efficiency and quality. Beyond that, some of the "smartest" robots are not in manufacturing; they are used as space explorers, remotely operated surgeons and even pets – like Sony's AIBO mechanical dog. In some ways, some of these other applications show what might be possible on production floors if manufacturers realize that industrial robots don't have to be bolted to the floor, or constrained by the limitations of yesterday's machinery concepts.

With the rapidly increasing power of the microprocessor and artificial intelligence techniques, robots have dramatically increased their potential as flexible automation tools. The new surge of robotics is in applications demanding advanced intelligence. Robotic technology is converging with a wide variety of complementary technologies – machine vision, force sensing (touch), speech recognition and advanced mechanics. This results in exciting new levels of functionality for jobs that were never before considered practical for robots.

The introduction of robots with integrated vision and touch dramatically changes the speed and efficiency of new production and delivery systems. Robots have become so accurate that they can be applied where manual operations are no longer a viable option. Semiconductor manufacturing is one example, where a consistent high level of throughput and quality cannot be achieved with humans and simple mechanization. In addition, significant gains are achieved through enabling rapid product changeover and evolution that can't be matched with conventional hard tooling.

Boosting Competitiveness

As mentioned, robotic applications originated in the automotive industry. General Motors, with some 40-50,000 robots, continues to utilize and develop new approaches. The ability to bring more intelligence to robots is now providing significant new strategic options. Automobile prices have actually declined over the last two to three years, so the only way that manufacturers can continue to generate profits is to cut structural and production costs.

When plants are converted to new automobile models, hundreds of millions of dollars are typically put into the facility. The focus of robotic manufacturing technology is to minimize the capital investment by increasing flexibility. New robot applications are being found for operations that are already automated with dedicated equipment. Robot flexibility allows those same automated operations to be performed more consistently, with inexpensive equipment and with significant cost advantages.

Robotic Assistance

A key robotics growth arena is Intelligent Assist Devices (IAD) – operators manipulate a robot as though it were a bionic extension of their own limbs with increased reach and strength. This is robotics technology – not replacements for humans or robots, but rather a new class of ergonomic assist products that helps human partners in a wide variety of ways, including power assist, motion guidance, line tracking and process automation.

IAD’s use robotics technology to help production people to handle parts and payloads – more, heavier, better, faster, with less strain. Using a human-machine interface, the operator and IAD work in tandem to optimize lifting, guiding and positioning movements. Sensors, computer power and control algorithms translate the operator's hand movements into super human lifting power.

New robot configurations

As the technology and economic implications of Moore's law continue to shift computing power and price, we should expect more innovations, more cost-effective robot configurations, more applications beyond the traditional “dumb-waiter” service emphasis.

The biggest change in industrial robots is that they will evolve into a broader variety of structures and mechanisms. In many cases, configurations that evolve into new automation systems won't be immediately recognizable as robots. For example, robots that automate semiconductor manufacturing already look quite different from those used in automotive plants.

We will see the day when there are more of these programmable tooling kinds of robots than all of the traditional robots that exist in the world today. There is an enormous sea change coming; the potential is significant because soon robots will offer not only improved cost-effectiveness, but also advantages and operations that have never been possible before.

Envisioning Vision

Despite the wishes of robot researchers to emulate human appearance and intelligence, that simply hasn't happened. Most robots still can't see – versatile and rapid object recognition is still not quite attainable. And there are very few examples of bipedal, upright walking robots such as Honda’s P3, mostly used for research or sample demonstrations.

A relatively small number of industrial robots are integrated with machine vision systems – which is why it's called machine vision rather than robot vision. The early machine vision adopters paid very high prices, because of the technical expertise needed to “tweak” such systems. For example, in the mid-1980s, a flexible manufacturing system from Cincinnati Milacron included a $900,000 vision guidance system. By 1998 average prices had fallen to $40,000, and prices continued to decline.

Today, simple pattern matching vision sensors can be purchased for under $2,000 from Cognex, Omron and others. The price reductions reflect today's reduced computing costs, and the focused development of vision systems for specific jobs such as inspection.

Robots already in use everywhere

Sales of industrial robots have risen to record levels and they have huge, untapped potential for domestic chores like mowing the lawn and vacuuming the carpet. Last year 3,000 underwater robots, 2,300 demolition robots and 1,600 surgical robots were in operation. A big increase is predicted for domestic robots for vacuum cleaning and lawn mowing, increasing from 12,500 in 2000 to almost 500,000 by the end of 2004. IBot’s Roomba floor cleaning robot is now available at under $200.00.

In the wake of recent anthrax scares, robots are increasingly used in postal sorting applications. Indeed, there is huge potential to mechanize the US postal service. Some 1,000 robots were installed last year to sort parcels and the US postal service has estimated that it has the potential to use up to 80,000 robots for sorting.

Look around at the “robots” around us today: automated gas pumps, bank ATMs, self-service checkout lanes – machines that are already replacing many service jobs.

Fast-forward another few decades. It doesn't require a great leap of faith to envision how advances in image processing, microprocessor speed and human-simulation could lead to the automation of most boring, low-intelligence, low-paying jobs.

Marshall Brain (yes, that's his name) founder of HowStuffWorks.com has written a couple of interesting essays about robotics in the future, well worth reading. He feels that it is quite plausible that over the next 40 years robots will displace most human jobs. According to Brain's projections, in his essay "Robotic Nation", humanoid robots will be widely available by 2030. They will replace jobs currently filled by people for work such as fast-food service, housecleaning and retail sales. Unless ways are found to compensate for these lost jobs, Brain estimates that more than 50% of Americans could be unemployed by 2055 – replaced by robots.