October 16, 2014      

Things have changed

In October of 2012 at RoboBusiness Pittsburgh, Rethink Robotics’ Rod Brooks put a face on the collaborative robot. In a first-ever public appearance, he stood up a real, in-the-flesh collaborative robot named Baxter for all the world to see. The machine was mobbed at RoboBusiness. Overnight, Baxter became international front-page news.

Previous to Baxter, the term collaborative robot was just a part of a growing conversation about new-age robots that was ramping up in labs and bandied about at conferences almost everywhere. It was hard to tell if anyone was really serious about them and, better yet, actually doing something about making a commercial version.

Baxter put an end to the conversations; Baxter made the co-robot a reality.

Fast-forward to RoboBusiness 2014, and collaborative robots singlehandedly dominated a half-day, workshop session on Advanced Manufacturing.

At noon on Wednesday (October 15th), Dan Kara, co-chair of RoboBusiness, opened the four-hour Advanced Manufacturing session to “examine the technology trends and end-user demands driving the creation of smarter, more flexible robots.”

The first hour of the workshop was held down by one of Baxter’s handlers, Carl Palme, an applications product manager at Rethink, who told the audience that the new revolution in robotics is the collaborative robot, especially when it comes to advanced manufacturing.

The reason: flexible manufacturing. “The rising tide of consumer-driven customization in manufacturing,” said Palme “requires flexible robots all along the continuum of production. Flexibility demands that robots quickly reconfigure their activities in order to address the fast-changing, ever-changing needs of factory workflows.”

There is little or no room in future factories for the classic industrial robot that is caged or isolated from workers. The robot must integrate to meet those requirements, or else factories and their supply chains will be unable to keep up with the churn of modern commerce, especially online commerce or e-commerce. Only collaborative robots can meet that call of flexibility.

Then too, there are factors like aging workers, employees out sick, high turnover rates, and even industrial accidents that are conspiring to thin out the ranks of workers and to lower productivity.

Palme added that as consumer electronics makes technologies cheaper, such as sensors, so too do industrial sensors become cheaper, all of which facilitates the emergence of data-driven manufacturing. Collaborative robots, he says, are the outcome of that movement and the answer in moving automation to optimization.

Baxter and its kind have indeed come a long way since Pittsburgh.

Validation of processes

Dovetailing from individual collaborative robots to “fully-integrated, collaborative manufacturing systems which respond in real time to meet changing demands and conditions in the factory” was taken on by Elena Messina, leader of the Manipulation and Mobility Systems Group from NIST, the federal government’s National Institute of Standards and Technology.

She began with questions that she said where necessary to know when building a “smart manufacturing solution.”

“How do we know what a robot needs to do” she asked. “And how do we know that it is working well enough?”

“Are you measuring the right things and are you measuring them correctly?”

“What happens when a gripper and a vision system are put together?”

“How do you validate the processes?”

NIST, she said, asks questions likes these all of the time, and then goes about measuring and securing results.

Such measurement science is necessary to accomplish the vision that Palme put forth for his “flexible” future factory populated by “flexible” collaborative robots. Especially one where he invokes a path from automation to the more rarified air of optimization. Measurements are critical to that hoped for success.

End-user perspective

Hour three consisted of a four-member panel moderated by Kara. Its charge was to take an end-user-community’s perspective of Palme’s flexible factory and Messina’s NIST measurements regarding collaborative-class robots, mainly, “the value proposition, human workforce integration, software and safety considerations as well as ongoing technology developments.”

By most any measure it was a large swath to cover in an hour, but all gamely jumped in, beginning with ROI expectations. An area where collaborative robots definitely outshine their caged brethren: under $50K as opposed to $300K and up.

For example, the Rodon Group, a small injection molding company about 30 miles northwest of Philadelphia, PA has a single Baxter. “‘Baxter’ as Industry Week described it, “has become the poster robot for a whole new category of industrial automation equipment–so called collaborative robots.”

The day USA Today visited with Rondon’s facility manager Tony Hofmann, Baxter had just finished a marathon work session of 2,160 straight hours or 270 8-hour shifts. At the going rate of $9.00 p/h, Baxter had just racked up the equivalent human paycheck of $19,440.

That’s a nice ROI.

A recent advert from Universal Robots touts “195 Days, Average Payback Period.” Again, collaborative robots seem to have unrivaled ROI capability.

Although both Rethink and Universal Robots sell their machines at full price, Yaskawa offers a decided end-user advantage with both leasing arrangements and rentals for their robot workers.

One of the best of the collaborative robot high points is the ease of programming: Universal’s robots can be operated by non-specialist line workers, while Baxter needs no programming–it learns each job function after being taught a single time.

The story that’s just beginning

Collaborative robot by collaborative robot, measurement science and smart system building all in quest of flexible manufacturing seems like a story that’s not ending any time soon. Rather, as Carl Palme said at the workshop’s opening: “The new revolution in robotics is the collaborative robot.”

It’s coming on as a massive part of advanced manufacturing and bears watching from here on out.