Dr. Stefan Kreuzer has seen his fellow surgeons shrug at advancements involving robots in surgery.
If they can drive home without a GPS, why involve robots in a process that already leads to a good outcome?
For Kreuzer, though, getting an assist from automation is like utilizing GPS software to check traffic patterns or suggest the best route home, even if a driver already knows one way to get the job done.
“A good surgeon can have good outcome most of the time, but that’s just not good enough. We need to have a not-so-good surgeon have excellent outcome all the time, and I think that’s where robotics can play a huge role,” Kreuzer said. “Because, currently, there’s a frightening statistic – 70% of all knee replacements are done by surgeons who do 10 or fewer a year. If we can make that surgeon a better surgeon for those 10 operations, and [add] robotics, if we train them properly, just about anybody can incorporate that into their workflow. I think we’ve done a great deal of good for society.”
Seeking the greater good is what excites ARC Specialties President Dan Allford about linking up with Dr. Kreuzer. A patient of Kreuzer’s put the pair in contact after learning of Dr. Kreuzer’s interest in robotics. The Houston residents have been able to work together to consider how to improve current methods of robotic surgery, especially in joint replacement.
That’s resulted in Allford and ARC utilizing collaborative robots and an articulated arm rather than the SCARA arms currently in use.
“This is something I’ve been wanting to do for a long time. In my opinion, there’s no higher calling than medicine, and, as a roboticist, I don’t get to do a lot of medicine,” Allford said. “So, I was eager to learn something more about the industry.”
With influence from Dr. Kruezer, Allford hopes he can combine the best of human minds and robotic work to produce something that truly contributes to society.
Sitting in a classroom and learning about engineering is much different from actually being on the job.
That’s why Dr. Peter Loos, Professor in the Practice of Materials Science and NanoEngineering at Rice University, has such a critical role. He works with seniors on their capstone projects, taking their ideas and training from the theoretical to the practical and exposing some of the gaps that might exist.
There are many lessons to be learned, he said, not least of which the way materials are acquired in the real world vs. the plans students may draw upon paper.
“In college, most of the courses are just an endless array of possibilities, so typically students are really not taught much at all about industry practices, industry-standard procedures,” Loos said. “Every major company has vast sets of standards by which they purchase materials, so outside of academia whenever materials are bought and sold there’s a vast array of specifications. We really have to teach them quite a lot about that. It doesn’t happen in those other classes, so it tends to be part of those capstone projects.”
Loos is an old friend of Arc Specialties President Dan Allford, dating back to their days with Howard Hughes’ Hughes Tool Company, and they’ve enjoyed working together in their current capacity as well.
Allford helps shepherd along some of the projects and starts training the next generation of engineers. Some projects haven’t come to fruition, but one of the latest – a knee prosthesis project done in conjunction with the Rice MSNE department to test materials used in artificial knees, shows promise.
“This whole thing is a spin-off of robotic surgery. We saw a need for something better than the cobalt chrome material because even if you can tolerate the wear, all the wear particles that are created, you don’t want them floating in your body,” Allford said. “It’s more than just a replacement issue, we’re trying to avoid having those particles move about.”
With their experience with surface coatings, they now believe they have a solution using titanium as a bearing surface.
It’s all gotten students to take their knowledge outside of the classroom, where they can start working on solutions for the real world.
Allford is exploring another area that fits the three Ds by looking into automation in meat processing. There are challenges, but Allford is confident it won’t be long before both slaughterhouses and local butchers have tools that can improve the well-being of employees while taking them out of dangerous or repetitive environments.
With COVID outbreaks in many plants, Allford says modernizing the meatpacking industry is “long overdue.”
“You have to understand why. Every carcass is different,” he said. “That’s one really difficult thing on food processing. You need some kind of adaptive control or sensor system to deal with that, unless you just grind the whole thing up into sausage. That’s not going to work.”
Allford said cuts like going across the backbone are relatively easy to automate, but separating the meat from the bone is more difficult to teach a robot how to do. Even so, it’s a problem he’s eager to overcome.
“When it’s going to get challenging is when we’re trying to do things where we’re trying to remove the meat from the bone, yet not waste much,” he said. “That’s challenging and interesting to me. I’m looking forward to getting down to some of that work, because then we’ll have to use all sorts of sensor systems.”
That may include using technology ARC developed for human surgeries that detects where bones are before determining where to make a cut.
ARC sponsored a senior project at Rice University studying orthopedic implants with hard-layer coatings on top of them and saw that total perfection actually may not be ideal.
“What we’ve found by creating these coatings and then finishing them is sometimes better is not better. We’re finding that a perfect finish, a mirror finish, doesn’t last as long as one that has a slight amount of roughness,” he said. “I thought that was fascinating because, if you do get that new knee, you’re going to want it to last as much as possible.”
The finding is one of a number of fascinating new developments in the world of abrasive processing and in robotics as a whole.
“More and more customers are looking for more consistent, uniform and finer finishes in general, which is interesting when we look at the orthopedic knee project Dan references,” said Scott Barnett, Application Engineering Manager for Robotic Abrasive Processing at 3M.
Other exciting progress is being made in the area of abrasive wear detection, with more and more information available about when to change out devices and tools and how to teach robots to know when the tool is getting dull.
“As we get into stiffer discs like fiber discs, belts, abrasive wheels and things like that, it adds complexity,” Barnett said.
Those complex challenges are just the types of things ARC and 3M are working on solving today, and the learning process continues to deliver a better that really is best.
We all can use a little help from our friends – but what if that friend is a robot?
With collaborative robots, robots with safety features that allow more human interaction or for humans to work nearby, that’s possible.
It’s not only safety where collaborative robots shine, however, with programming a task that can be taken on even by those who haven’t programmed robots before.
“In the collaborative space, I think that’s been one of the big selling points. I can pick up a pendant (and) actually grab the robot and kind of move it around and guide it through the path so you don’t have to jog it, save a point, jog it again, save another point,” said Tyler Naatz, Advanced Applications Engineer, Robotics and Automation for 3M’s Abrasives Systems Division. “You can kind of do that through follow-my-path type programming using all those sensors that are in these robots. I think that’s a big reason people are looking at them.”
With that ability to minimize the complexity of programming also comes a few drawbacks. For one, the tool force will be limited, as will the speed at which a collaborative can run. That makes it perfect for projects like orbital sanding, but perhaps not the right fit for some other endeavors.
“To do that, we’re having to polish all the way around the circumference, around the corner, very complex shapes,” he said. “So, a robot was perfect for the application in that a robot could make all these moves. By using a collaborative robot with that sense of touch and the ability to maintain constant tool force, we’re able to, (even) as the part changed size or particularly as the abrasive changed size.”
“The collaborative robot, with that ability to translate and change trajectory on the fly based upon loads, made it a great application.”
Two Leading Automation Minds on Race Car Driving and Robotics
Both friends enjoy the thrill of mountain bike racing and the even bigger thrill of race car driving. The first race car Sooknanan competed with was a car Allford loaned to him. Despite the fact that Sooknanan blew that car up during his second race, they’ve continued to race together, with Allford helping instruct Sooknanan in the art of race car driving.
How Fast Cars and Automation Bring Friendly Competition
Another interest – and perhaps friendly competition – these two share is high-level robotics. Having so many interests in common allows Sooknanan and Allford to keep each other on their toes, both on the race track and in their work.
One such example is their ongoing debate regarding the differences between autonomous and teleoperated robots. Allford believes that a machine should run entirely on its own, without human instruction, whereas Sooknanan believes a robot can be autonomous with some level of human instruction, which he refers to as supervised autonomy.
Having a level of supervised autonomy is key in the space industry, according to Sooknanan, so that robots can do their jobs safely.
“You know, with some level of oversight, [we can] make sure we’re not going to take one of these massive manipulators on the outside of a space station and poke a hole in it or something crazy like that,” Sooknanan said.
Allford on the other hand is looking for autonomy where the robot will do the same thing over and over, but also wants it to have sensor systems that give it sight and touch sensors that encourage it to adapt. According to Allford, adding sensor systems is a key element of autonomy.
“It can not only repeat a repetitive task – it can adapt to its environment. It can adapt to its parts, and then do something that it wasn’t originally programmed to do,” Sooknanan said.
A Bond Built for the Long Haul
A little friendly competition is something Sooknanan and Allford enjoy, and they don’t mind disagreeing on robotics.
“I like to pick on Josh, largely to stimulate conversation, because we’re solving similar problems in very different ways. So, by provoking him, I can get him to talk a little bit more,” Allford said.
Robotics aside, the really big question is clear – who is the better race car driver?
“What I like to remind Josh of is that a good instructor is defined by the fact that their student exceeds their skill,” Allford said.
“I concur,” Sooknanan said with a laugh.
Is the answer I dread when I ask a technical question. Unfortunately when customers ask: “How accurate can you plasma cut parts with a robot?” I reply: “It depends”.
Recently we were asked to cut 1.575″ (40mm) holes in 3/16″ plate with a circularity of +/-.005″. This is a tight tolerance for plasma. Normally we would have recommended laser cutting. But I tell people that a watt of electrons is much cheaper than a watt of photons. Plasma has a number of advantages over laser: system cost, power efficiency, maintenance costs, and safety.
So we went to the lab to explore the limits of robotic plasma cutting. What made this project unique and doable was the fact that hole position tolerances were achievable. Our challenge was hole shape and size. We used a Hypertherm XPR300 High Definition Plasma and a KUKA North America robot. Jim Walker & David Luce programmed the cut in small segments to optimize torch path and cutting parameters and cut holes +/-.005″.
BUT your part accuracy depends on many variables. In the video, I stick out my neck and QUANTIFY best, worse and typical effects of 7 issues on part dimensions. This should start an argument!
ARC Specialties thrives on problems, send us yours!
Some things are challenging enough, but an extra element takes the degree of difficulty up several notches. That’s how it is for Josh Sooknanan, an aerospace engineer with NASA, who takes robots and puts them into space.
In addition to the already steep curve of working with robotics, Sooknanan also has to face things like delays in how long a robot takes to respond.
“Robots in general are hard. Anybody that is getting into robotics, the problem you’re getting into is always harder than you’ve scoped it to be,” he said. “That’s when the robot is in front of you on your lab bench. You put the robot 200 miles above your head orbiting the earth or, even more difficult, on the moon or on Mars or a probe doing deep-space research, and the delays are unbelievable.”
Dan Allford, president of Arc Specialties, is a friend of Sooknanan and noted there are different challenges he has to face in the industrial world.
“I think one of the biggest differences between your world and my world is gravity. I’ve got to work with this stuff called gravity and it’s 32 feet per second squared, and it’s pretty constant,” Allford said. “A lot of times when we have to design and build a robot, it’s because the loads exceed that of any available commercial robot, and that’s a gravity problem. You’re up in space and you don’t have to fool with gravity.”
Of course, while Allford has gravity, he also has more sources of power than simply the sun – and less delay between the time it takes to send a command to a robot and see it carried out.
The two friends, who race cars and mountain bikes when not talking shop about high-level robotics, come together on more things than they diverge on, with both eager mentors of high school students interested in robotics.
Both marveled at programs like the BEST Robotics team competition and U.S. First Robotics, which give kids like them who were obsessed both with science-fiction and tinkering with machines a year-round outlet to learn in a more structured environment.
During the last two months of the COVID-19 pandemic and the ensuing shutdown, ARC Specialties has continued to work to build the machines necessary for the critical infrastructure of America.
As I walked through the shop today, I was proud to see my team working on an interesting array of systems.
Most folks think of MIG when they think of robotic welding. At ARC, we consider all welding processes and match the solution to the problem rather than the other way around.
Currently, we are finishing up MIG, TIG, Subarc, resistance, drawn arc stud, laser and plasma welding robots, as well as diamond valve lapping and thermal spray systems. All of these innovative solutions are designed to provide highly tailored growth and success instead of pushing a tired, “one-size-fits-all” approach to unique operational needs.
I suspect that the tough economic environment we currently find ourselves in is the reason we are rebuilding three failed systems that were originally built by others.
We like the work, but we hate to see the customer have to pay twice.
This is yet another reason I am a huge advocate for the Robotic Industries Association Certified Integrator program, which ensures you can be confident an integrator has high-level capabilities regarding delivering safe, reliable and high-quality robotic solutions.
I am happy to report that ARC has not experienced any order cancellations. Our backlog is going up, and America is getting back to work. It’s an exciting time to be doing business, and we’re ready to keep pace.
If you would like to become more efficient and competitive, contact us today.
At ARC Specialties, we thrive on problems. Send us yours!
To view our video please see my recent post on LinkedIn.
Dan Allford, President, ARC Specialties
Dan Allford, President, Arc Specialties, and Brett Hubbard, OEM Sales Manager, Hypertherm joined the Roboticist chronicles to detail the long-standing partnership between Arc Specialties and Hypertherm.
Hypertherm is an industry leader in utilizing plasma for cutting alloys. Arc Specialties uses Hypertherm’s power supply and technology to cut parts with Arc’s robots.
“What we do is take plasma cutting and apply it to a three-dimensional world,” Allford said.
Hypertherm’s technology provides the cut, and Arc’s technology provides motion control.
In a collaborative partnership with Hypertherm, RobotMaster, and KUKA, Arc Specialties co-developed FlexFabTM, a flexible fabrication 3D Robotic Cutting Cell that converts CAD models into a 3D plasma cut steel part. This process saves time and labor hours due to its precision control and the ability to program offline while cutting work continues.
“The FlexFab can be just that – flexible to whatever your application may be in the 3D world,” Hubbard said.
What does the future hold in plasma cutting technology? Hubbard said Hypertherm is continuing to expand its plasma capabilities in the 3D space, as well as looking to its past in developing new non-ferrous metal cutting technologies.