Kel Guerin is the Founder & CTO of READY Robotics, creators of the world’s first universal operating system for industrial automation, READY Robotics helps all manufacturers solve their labor challenges, boost output, improve quality, reduce costs, and augment their workforce through automation.
What initially attracted you to robotics?
I was interested in robots from an early age – I remember a particular book printed in the 80s I read as a kid that showed a bunch of fanciful images of what robots would be like. I also loved movies like Forbidden Planet and the original Lost in Space that showed robots as these helpful devices. I also understood from reading about robots like Dante 1 (a volcano explorer) and the robots we sent to Chernobyl and 3 Mile Island how important these devices were for tasks hazardous to people. However, I drifted towards Optical Engineering as a degree because I also had a love for space systems, specifically space telescopes.
From 2007 to 2009, you worked as an Engineer for Astrobotic Technology, Inc as the Optics Lead for a robotic lunar mission to explore and collect high definition imagery of the Apollo 11 landing site. Could you share some highlights of this experience?
This was an exciting time for me, because I was part of a small, extraordinary team, led by Dr. Red Whittaker, a professor at Carnegie Mellon University, and a pioneer and giant in robotics in space. It was an absolute pleasure to work with him, because he had such a clear vision for the mission. He already had success with several robotics challenges (Grand Challenge and later Urban Challenge) and set his – and our – eyes on a lunar mission. I had experience in designing camera systems from my undergraduate work and joined the team. It was an amazing experience because it was really our prerogative to design this lunar vehicle, which means we had to solve a lot of hard problems. One of the serious issues was how you drive this robot around, given that there was a 10-12 second time delay between the operator and vehicle on its lunar trek. That meant that any action or command you sent took a round trip of ~20 seconds to see what happened…take a left turn and you don’t realize you drove into a crater until it’s too late. This opened my eyes to the usability aspects of robots – sending a robot to the moon was a challenging problem, but driving it around was one crux.
You then went on to Johns Hopkins University, to focus on adaptive and modular industrial robotic systems for small manufacturers. What have you learned from this experience?
I had the opportunity at Johns Hopkins to work under Dr. Gregory Hager, which was a transformative experience because he is a serious expert in computer vision, who also has a lot of interest in how people can interact more effectively with computers and robots. I originally started under Greg working on robotic systems for surgery, and how you might make a surgical robot more usable for a surgeon. A large part of this was thinking about how the robot could assist the surgeon with a task, say typing a knot or suturing. That meant that the robot needed a model of what the task was so it could help, and the surgeon had to get what was in their head into the robot. We started looking at industrial manufacturing because we saw the same problem, robots could help in the manufacturing environment – and were more equipped to do so – but programming them, that knowledge transfer from the worker to robot, was still a huge barrier. The technology that came out of that research was an easy-to-use software application for programming robots, enabling anyone to program a robot, and learn how to do so in minutes. It was transformative. People compared it to playing a game on an iPhone because it was so approachable, and because of that we won the Kuka Innovation Award, the only US team ever to do so up until then.
In 2015 you launched READY Robotics. Could you share the genesis story behind this robotics company?
With the easy-to-use software from my PhD work, it was obvious we had a potential product. The typical robot cell would take weeks to program, and we could do the same in hours. It also took weeks of training courses to learn the average robot software, and with our software we could teach you in minutes. At Hopkins I had been working with Benjamin Gibbs, who had worked closely with me crafting an intellectual property strategy for the technology. When the time was right to spin the company out of Hopkins and licence that core technology, I asked Ben to join as CEO, with me in the role of CTO. At that point we raised an initial seed round of venture funding and started working on an MVP product. We initially targeted smaller manufacturers with a complete turnkey device called the Task Mate: a collaborative robot paired with our easy software. This really showed us the demand for this technology, and we saw a lot of interest from larger corporations, because they had the same issues deploying automation – the lack of people with robot programming skills. We also saw from this the need for this easy-to-use software in a wide range of robots and robot brands. People wanted easy-to-use software, but they wanted it on bigger, faster robots. That led us to create Forge/OS, a completely robot-agnostic operating system on which our easy-to-use programming software, which we named Task Canvas, could run.
Why is designing a standardized operating system for the industrial robotics industry so important?
If we look at the history of robotics, it closely parallels the computer industry. Both became commercially viable in the mid-60s, but computers took off and changed the world in a way robotics hasn’t quite managed. The reason behind this is two-fold. First usability; with companies like Apple pushing the usability of computers and allowing everyday people to leverage computers, there was massive growth and demand for these devices from consumers, because they could use and find value in them. The second is a common platform: Microsoft DOS and then Windows caused a massive consolidation in the fragmented PC industry of the early 80s and enabled developers who could create one solution that worked on any computer. This was huge, because it unlocked an entire ecosystem of computer OEMs, application developers, and users. The robotics space is still similarly fragmented. Every robot brand has their own operating system and software. There is almost no software interoperability. Usability has not been a focus until recently, with companies like READY now truly bringing usable programming software to market, not unlike Apple. But that isn’t enough unless, like Windows, you can learn one piece of software and program any robot. That is why we built Forge/OS. To enable users to learn one piece of software and use any robot, and to enable developers to build one application that works on any robot. This is a trend we saw first with PCs and then with smartphones – a common operating system and usable software driving massive adoption. There have been attempts at this, but Forge is actually a platform that shares these similarities in a real way. It unlocks application developers, making the robotics space accessible for the first time, because there is now a platform to develop for. It unlocks users, who will soon be able to buy and use those applications, from a real marketplace, like they can buy an app from the Android or iPhone app store.
Can you elaborate on the software solutions that READY Robotics offers and how it simplifies the set-up process for industrial robots?
A huge amount of time installing a robot is programming time, and that is assuming you know how to program the robot. The programming barrier makes robotics a tough lift for companies who don’t have in-house robotics expertise, or who can’t wait for an outside integrator. For some companies it’s just too difficult – learning a standard industrial robot programming language requires a lot of work. It’s hard and takes weeks just to learn the basics. This means that most companies who think about using automation either don’t, or spend a lot of time and effort to upskill their workforce and then do the programming. They also then have a hard time keeping that talent, who is now desirable because they can program these difficult systems. This also means that a company is now locked into that brand of robot they learned, because the cost of switching is learning it all over again. So if a cheaper or better robot comes along, they are stuck.
Our easy-to-program, no-code software, Task Canvas running on Forge/OS, solves that first problem. Manufacturers can now upskill their existing workers (the people who have the most process knowledge to boot) in a matter of days, because the software is so easy to learn. READY also offers curriculum at the READY.Academy as a resource for how to not only learn to program the robot with Forge/OS and Task Canvas, but how to do everything else you need to to – grab and hold parts, put them into your machine, to actually do the task with the robot. Most educational courses only teach programming and never teach you how to do something valuable with the robot. Forge/OS solves the second problem I described, that brand lock-in. Customers need to choose the right robot for the job, and with Forge/OS, they have that choice, because each robot has the same amazing usability, and can run the same applications. This common platform will be powerful when paired with READY.Academy because now students finish the course prepared to use any robot, not just a single one as with most courses.
How will robots and humans collaborate in the workplace of the future?
In manufacturing, you will see a transition from people doing the dull, dirty, and dangerous tasks – and more simply, the non-value creating tasks, less and less with robots performing those tasks more and more. Manufacturing workers are some of the most inspiring and creative people I have met, and these incredible people spend most of their time doing tasks like moving pieces of metal around, when they could be optimizing processes, tweaking machines for better production, or generally adding serious value. Robots will empower those people, not replacing them but unleashing them. Robots will take away the non-value adding tasks leaving people to do what they do best – solve hard problems. By any measure, automation is a job creator, and we specifically see a future of workers who are empowered with robots as another tool for them to be supremely productive, just as any good tool does.
What is your vision for robotics in everyday life?
We will see a lot more robots working side-by-side with people as safety technology advances, and with platforms like Forge/OS, robots will become much more ubiquitous, just as computers did. When everyday people can use and find value in a technology that was previously niche, and dramatically decreasing in cost (a robot that was $150K ten years ago is now $7-$25K), people will find creative ways to use it everywhere.
We also continue to solve hard robotics research problems every day that give robots the ability to navigate and interact with the world as we do. We could soon see capable, sub-$1000 robots that are in every coffee shop, every fast food restaurant, in our homes, all because they are usable by everyday people, like computers, and work on common software so developers can build amazing applications rather than reinventing the wheel. They might not all look like Robbie the Robot, but they will be something that makes people more valuable, by freeing us up to be the creative problem solvers we are.
Is there anything else that you would like to share about READY Robotics?
Our mission statement is “To Improve the World’s Quality of Life and Productivity through Automation”, and that really embodies our vision for the future of robotics. Just like computers, there is huge, transformative power locked away in the robotics industry, just waiting to be tapped. We need automation now more than ever – we have seen that with Covid-19 and the resulting supply chain issues – because in order to be prosperous as humans we need the things than manufacturing provides. But we also need to understand the value of people, who are incredibly talented and capable of things that robots may not be able to do for a long time, which is why we focus so heavily on education as well. As technologists, we need to prepare people for the future that our technology creates. Pioneers like Bill Gates and Steve Jobs did this in spades for computers, by enabling anyone to find value in a PC. Robots need to be an accessible tool for people as well, empowering more productivity, more value creation and less drudgery. That is READY’s goal.
Thank you for the great interview, I look forward to following your progress. Readers who wish to learn more should visit READY Robotics.
- Humayun Sheikh, CEO of Fetch.ai – Interview Series
- Research Team Aims to Create Explainable AI for Nuclear Nonproliferation and Nuclear Security
- Pieter VanIperen, Founder & Managing Partner of PWV Consultants – Cybersecurity Interviews
- Advance in Microchips Brings Us Closer to AI Edge Computing
- AI Researchers Create Video Game Playing Model That Can Remember Past Events