The chink in the cobot armor?

Brian McMorris of Futura Automation linked to this article today and has some comments about it. Brian and I share many of these opinions but I felt I had more to say. And since a LinkedIn post has a character limit (who knew?) and since I am not short on opinions, I decided to post this article...

The article from Asian Robotics Review shows some developing chinks in the cobot armor. They make some very real and very important points. Of one to note:

"A recent 36-page survey Global Robotics Report conducted by TM Robotics, revealed that over 55 percent of distributors “do not believe that collaborative robot technology is advanced enough to deliver the performance required by manufacturers today.”

What's more telling is that only 20.7% answered YES to this question (the remainder saying they did not know).That's not a large number. I hear daily stories of manufacturing engineering buying a cobot only to find out that it cannot do everything they thought (or were told) it could do (or at the speeds they imagined). That often sours management to future robotic applications, often leaving the poor cobot in the corner, gathering dust.

But perhaps the more startling statistic is

"In addition, the survey found that 79 percent of distributors do not believe that “their customers understand the safety requirements of installing a cobot,” while another 10 percent are “unsure of their customers’ level of knowledge surrounding safety.”

That's 89% of distributors that believe their customers do not understand the safety aspects of cobots. All it takes is a quick glance at LinkedIn to see cobot application videos with WILDLY unsafe aspects. Most of these are self integrated and it's obvious that safety is either misunderstood or simply ignored. At RND Automation we play a little game called "spot the safety violation" when we see these cobot videos. To date it's been hard to find a video without at least a minor one. Customers simply do not do, nor do they understand, risk assessments.

Of course this article did get a few things wrong.

Especially at the price points that range from $6,000 to $35,000, with ROIs that are measured in a handful of months.

I have yet to see a $6,000 cobot. Anyone? And the price points go easily up to the $60,000+ range. I also defey ANYONE to show me a few month ROI on a cobot project if you are honest and really calculate all of the costs (robot, end of arm tool, and most importantly, the time it takes to design, program and debug the system.) One manufacturer in particular loves to use this ROI in the "months" phrase but I have never seen any data that backs up these claims.

Cobots are the 3D printers of today's age. Remember how the 3D Printer market claimed (10-12 years ago) that within a decade we'd all have one in our homes; printing broken refrigerator handles, screwdrivers, parts of our kid's broken toys? While there are many hobbyists that have home units (I was one) they are far from the ubiquitous appliance that was promised.

Cobots have their place and there are very good applications for them. But there are also a lot more BAD applications. Applications where a standard industrial robot with some safety fencing or laser scanner will outperform and in many cases cost less than the cobot.

Just because your cobot is "easy to program" or "easy to implement" does not mean it's intrinsically safe, nor does it mean it's the best tool for the job.

Right now the hype is high. Time will tell how long that will last.