When men first imagined robots, they visualized automatons that looked and moved like humans and who served their mortal masters by undertaking the tedious work of the day.
For example, Czech author Karel Čapek (who is credited for coining the word “robot” from the Slavic language root “robota,” meaning forced laborer) pictured robots that mimicked people and were happy to slavishly perform chores around the house and garden, at least at first (things turned nasty a little later in Čapek’s 1921 play, Rossumovi Univerzální Roboti (Rossum’s Universal Robots)).
But it’s only now that anthropomorphic “service” robots are starting to fulfill the potential that was imagined by Čapek. Recent advances in computing power, artificial intelligence (AI), and electromechanics have introduced commercial service robots within applications (e.g., to assist the developed world’s aging population).
Prior to the emergence of service robots, the major revolution in robotics had been driven by industry. Once the electronics and software matured enough to make the technology practical and cheap, robots were introduced to assembly lines, in the 1970s, to grind and polish pipe joints, weld together automobiles, paint fridges, or assemble furniture. In these high-volume applications where repetitive, precise, mistake-free operations are essential, robots put humans firmly in the shade. Manufacturers embraced a technology that didn’t need comfort breaks, sick leave, pay rises, or union representation. And while industrial robots weren’t cheap, they did provide the years of reliable service required to justify the initial capital investment.
Today’s industrial robots are far removed from the metal men of science fiction. The International Organization for Standardization (ISO) defines an industrial robot as “an automatically controlled, reprogrammable, multipurpose manipulator programmable in three or more axes, which can be either fixed in place or mobile for use in industrial automation applications.” In other words, an industrial robot is a powerful, giant arm that is at home in a factory and can move inside a large operational envelope to where it needs to be to complete its work quickly, repeatedly, and precisely.
But humans still have a major role to play in manufacturing. What they lack in stamina, speed, and precision, they make up for in dexterity, flexibility, and problem-solving skills. And in many parts of the world, their labor is cheap. That makes people ideal for finicky assembly where the product mix changes frequently and a wide range of skills is necessary.
Between the robot- and human-domains lies a zone where the talents of both could dramatically ramp up manufacturing productivity. This zone is characterized by relatively low-volume production and relatively high-value products. Bringing in the robots would dramatically enhance productivity by automating the picking of parts, lifting and fetching, and repetitive, routine elements of the assembly process.
This is not a job for the metal monsters of car assembly lines. For one, they are too expensive to buy and run, and more importantly, they’re as dangerous to humans as apex predators. Making workers toil alongside industrial robots would be akin to using a zoo’s lions for kiddie rides—it’s a disaster waiting to happen.
That’s why a new breed of “friendly” robots are filling this niche. Dubbed collaborative robots (or “cobots” for short), the machines are lightweight, human-scaled, inexpensive, and empathetic to their co-workers.
Designing cobots is tricky: The challenge is not so much in making them perform various assembly line tasks—that’s fairly straightforward—but more in ensuring workers don’t get hurt. Engineers must combine operational parameters such as strength, speed, and repeatability with safety elements such as motion sensors, force limiters, and streamlined design (to eliminate things like pinch points). It’s relatively simple to build in sensors to bring a cobot to a dead stop, if a human worker enters the operational envelope, but there will be occasions when the human and robot have to interact—demanding a degree of slack in the cobot’s joints, such that it doesn’t knock against flesh and bone too hard. Today’s industrial robots feature joints built to high-specifications precisely to eliminate slack, because slack compromises accuracy; maintaining this accuracy while easing tolerances requires new design techniques.
And that’s just the hardware. Industrial robots require experienced technicians to program them. That’s not such a big deal when a machine needs a very infrequent update to cope with a new car body, after years spent welding together the previous type. In contrast, cobots will need to be easily “programmable” by the human co-worker to cope with frequent introduction of new products. The programming operation should be as simple as, for example, manually guiding the robot’s arm through a gluing operation while the machine “learns” the sequence of movements. But the underlying complexity that will support such simple programming is yet to be fully developed.
Designing cobots is a nascent discipline, and as such, there is little guidance to draw upon. International safety standards for collaborative robots are being developed in parallel with the introduction of the first models for the workplace. The ISO 10218 standard provides a few specific guidelines for collaborative robots, while ISO 15066 outlines some rules for collaborative operation. A technical specification (TS), a document one level below an international standard, that is currently being drafted by ISO technical committee (TC) 184/SC 2 (for robots and robotic devices) promises to add much more to the body of knowledge for cobot design.
And this information will come not a moment too soon. The cobot market is set to be huge among capital goods—as analysts at Barclays (a UK-based bank) estimate that the segment will likely grow from $116 million in 2015 to $11.5 billion by 2025. This is roughly equal to the size of the entire industrial robotics market today.
Steven Keeping gained a BEng (Hons.) degree at Brighton University, U.K., before working in the electronics divisions of Eurotherm and BOC for seven years. He then joined Electronic Production magazine and subsequently spent 13 years in senior editorial and publishing roles on electronics manufacturing, test, and design titles including What’s New in Electronics and Australian Electronics Engineering for Trinity Mirror, CMP and RBI in the U.K. and Australia. In 2006, Steven became a freelance journalist specializing in electronics. He is based in Sydney.