Frank Tobe shares his experience from covering robotics in The Robot Report and creating the index Robo-Stox. Frank Tobe is the owner and publisher of The Robot Report. After selling his business and retiring from 25+ years as a provider of computer direct marketing and consulting to the Democratic National Committee, major presidential and other campaigns and initiatives, he has energetically pursued a new career in researching and investing in robotics. In 2013 he co-founded Robo-STOX™ LLC (renamed in 2015 to ROBO-Global) and developed a tracking index for the robotics industry: the ROBO-Global™ Robotics & Automation Index. Frank talks about how and why he shifted to robotics and how looking for an investment opportunity in robotics lead him to start ROBO Global, a robotics focused index company. Both companies have given Frank a unique perspective on the robotics scene as a whole, over a significant period of time, and we are pleased that he wanted to share some of his insights from that with us.
This podcast is part of the Wevolver network. Wevolver is a platform & community providing engineers informative content to help them innovate.
Learn more at Wevolver.com
If you are interested in sponsoring the podcast, you can contact us at richard@wevolver.com
Per: Welcome to the podcast version of Robots in Depth and this launch episode with Frank Tobe in cooperation with Wevolver. Robots in Depth is supported by Aptomica. Visit Aptomica.com to connect. You will find all past episodes and more on RobotsInDepth.com. Welcome to Robots in Depth, today I'm honored to have Frank Tobe here. Welcome.
Frank: Thank you for having me.
Per: We’re going to talk about robo stocks, we're going to talk about the robot report that you do but we'll start with how did you get into robotics. How did you discover the field and realize there's something here that's going to happen?
Frank: I used to be in the computer business. I did demographic work targeting direct mail, personalized direct mail for political candidates and it was a pretty successful business. During the 28 years I was in business it went from a mainframe computer business to a mini to a micro, to very powerful micros to PCs as we know them to being in the software business because the digital age had come. I burned out on politics. I had a small stroke and sold the business and closed the business. Then after a short while I tried to figure out well what's the next portion of the digital era and what's going to affect me. At that point in time which was around 2000 I was a passive investor. My idea was what's going to happen that I can invest in and sort of like a basket of stocks, a mutual fund literally that I can capture conservatively for sort of a retired person. What’s going to happen? The first thing that popped up in all my reading and research was the process of automation and regardless of the industries which seemed to progress at different rates the process was the same and robotics was at the core of it and software was a good part of that and I know software because that's what I did.
that's where I started and then I set up with a business plan thinking well I can compile a list because that's what I know how to do and then I can analyze those and see which stocks I want to buy myself. Perhaps I can find companies that I can invest in on an equity level so that's how I got started. During the process it dawned on me as I was sharing the information with my friends that maybe I should put up a website and share the information that way, all the while developing this database. The database had value, lots of different companies around the world kept writing in saying can we have a copy of your database and the answer was no after a short while when it dawned on me it had value. Ultimately a couple financial guys came along and said can we use your database, the publicly traded portion of your database and set up a fund. It sounded like free money to me plus it sounded like a solution to what I wanted.
I found three major bits of information about robotics and publicly traded stocks there. Amongst all the brokerages and analysts nobody covered robotics. There was no analyst for robotics. There were for automation but mainly that was on a machinery level company type company. The second thing was that robotics is not an American business it's a European and Asian business. America was users and initiators. A lot like they invented certain portions of TV screens or certain consumer products yet they didn't produce them in America. That was fascinating to me. Armed with that bit of information I started like I said the website. The website got very popular very fast and had good traffic statistics so there were actually advertisers who were willing to pay to advertise. It never really made a lot of money. I mean you can't expect that on a very limited audience type website but I learned how to write better than I had and I sort of enjoyed it. When the financial guys made a proposal that we could start an index rather than a fund. We could start an index and license the index to people that would like to bring out a fund we could actually make money on that.
Per: It's much easier them to start your own fund, many magnitudes easier.
Frank: Yes so that's what we did and we have two licensees of our index. One in London and one in America and that puts us on the American Stock Exchanges and all the European Stock Exchanges and the European banking system. Then later this year we'll probably do one in either Singapore or Hong Kong to cover the Asian thing. In the meantime the amount, the number of publicly traded stocks that fit our criteria was in the 80 range. We have I think 81 or 83 stocks in the index right now. The problem is that some of the really good, new, dynamic companies using a Universal is a perfect example. They’ve grown from zero to building robots at the rate of 5000 to 7000 a year, a run rate of that. That’s a dramatically good company. They’re not publicly traded right now. They may be sometime in the future. Another example of companies not traded is Schunk. Here is an end of arm tool company that's got grippers on every major robot around the world and every major robot user is aware of them, has some of their products and they're a family-owned business and pretty much intend to be that way.
Per: They are not even looking at that Universal might come to the market but the Schunk. How do you capture that?
Frank: You don't. What you do is you try and find companies that are representative of the marketplace. It’s true that there are a lot of family-run companies in the robotics industry particularly in Germany and the Netherlands. A good portion of those companies are family-run and have no interest in going public but there are a few older companies I mean KUKA, ABB. ABB is another problem. Two-thirds of ABB’s revenue comes from the power industry.
Per: How do you then sort out what comes from the robotic half?
Frank: How do you weight them so that it's representative and we couldn't figure that out. What we ended up doing is having two categories of member in our index. One we called the bellwether and one we called everybody else. The bellwethers were companies that were pretty much what they call pure plays. iRobot is a perfect example. Their whole business no matter what it is is robotics related. Intuitive Surgical is another one. Their whole business is building those devices and all the tools and maintenance material that goes with it. There aren't that many of pure play companies in fact I think we only have 20.
The other 60 companies range from John Deere. John Deere is a huge mechanical equipment manufacturer. A small portion is robotics related. It’s probably going to be a very big growing portion but it's not happening just yet but still they are an up and coming so they are one of the non-bellwether companies. There are vision companies that two-thirds of their revenue is involved in automation and robotics vision systems so we include them. It’s been fascinating. Amongst the universe of publicly traded stocks there are a lot of penny stocks and other uninvestable companies. There are a lot of companies whose market value isn't great enough or whose daily trading volume isn't significant enough for us to take them seriously as an investable product. The result is a pretty conservative portfolio which fits my personal objective because that's what I originally wanted. I wanted a sort of a retirement portfolio that was conservative and then I could just watch as the industry grew it would grow. That's how it came to be and also the robot report came to be.
Per: You capture the publicly traded companies in the index and you write about all robotics companies on the blog. You get this very deep and wide knowledge then because you see so much in robotics. We have major trends now growing that we see mobile manipulators becoming more accessible. Is there any key moment in this history that you'd like to share with the viewers that you think oh, this was really cool? Like for instance when the PR2 came out you just realized that this is going to lead to something awesome. I don't know what it is but it's going to be big. Are there any other movements like that that you'd like to share?
Frank: There are trends and I'm not sure who said it but they said whatever you see now in robotics in a decade that's what's really going to happen. I've been looking back what have I seen and truthfully what I've seen is in agriculture and in warehousing there are major trends occurring that are pushing both of those industries forward. In warehousing, material handling you've got bin picking which started 10 to 12 years ago and it's finally hit its stream. The concept of bin picking is simple but why it's important is because conveyor systems and space are costly and if you can eliminate that by a small amount of mobility bringing a bin of goods and picking it up you saved a lot of space, a lot of equipment and a lot of maintenance and a lot of engineers. That’s happening.
Another thing that's happening in material handling is that when the recession came the normal investment, capital investment in warehousing and warehousing systems stopped, stopped for five years during which the technology changed. That was the time where Kiva's concept which changed warehousing. Warehousing up until the Kiva System was man to goods. Man goes out, runs around, finds his stuff, puts it in a box, brings it back or gets it back. It was a very energetic method. Kiva System of goods to man is a very efficient system an economical in a variety of ways. It was economical in that it saved warehouse space. It sped up warehouse investment and development time. It enabled a smaller space to have double the amount of workable square footage all at lower costs.
Amazon and others quickly saw how that played out and they were one of the few that were investing. For them to buy Kiva and make them focus just on their own warehouses was a very logical good thing to happen but it's happening all over the industry. What's happening now is there's actual copycats of the Kiva methodology some of which will probably push patent infringement and some won’t but the concept of goods to man and then the software system to remove stationary shelving, that whole concept. The whole concept of high-rise shelving and mobile lifts to get skids of material way the hell up there and way down from there that's a thing of the past.
Per: That's also where we see robotics coming into an area that everybody feels is kind of set and then you are totally blowing all the old stuff away bringing in something totally new.
Frank: That's what's happening. You have it on all sorts of different planes. You have companies that are providing mobile navigation devices or kits in effect that you can put on a forklift and have it drive around by itself. You can have it tow goods out to where they need to be housed. This saves a human doing the same job for the cost of a navigation system. Same equipment but you save a human doing sort of a boring task towing stuff around. I just saw in the newspaper that Russians have an identical kit then are starting to have it installed in Finland and in Russia. That’s one portion. You have the Kiva copycats. There’s a big Indian firm, they have a product and they are selling it in India already and their volume of producing those products although not like Kiva's is growing, is in the thousands already. Kiva last year had 15,000 at work and they hope to do 15,000 this year which is just amazing. That’s 30,000 robots for a company that's producing at the rate of 15,000 robots. From a robo stock's point of view Kiva was taken off the market by being acquired by Amazon and their internal revenue is so small that we can't call Amazon a robotic company nor can we call Google a robotic company for the time being. They may with their autonomous driving system. Who knows where their revenue is going to come from but right this moment…
Per: It's such a small fraction for Google I guess it's nothing and for Amazon it's just a small fraction. Does Amazon also sell the Kiva systems to other companies?
Frank: They have left Kiva as a separate operating company so that it could but they have booked them to the extent that they cannot produce for anybody but Amazon for the next while.
Per: Amazon seem to need so many, their capacity is full.
Frank: They’re finding that so successful. Look at the difference, Amazon used to set up distribution centers that were a million and a half square feet. They were these high bay things. They had a huge investment in shelving and forklifts and other material handling devices and then conveyor systems and so scanning systems, big investment. Now you have these disposable plastic shelves that cost nothing on a single storied solid floor to the point where you can just build a second story and a third story in the same space as the high bay building so instead of a million and a half square foot building you have a 500,000 exterior sized building getting the same amount of square footage at lower cost and it's just fascinating.
Per: It shows you that a different way of doing things can really impact the results. Something as simple, it might sound simple as from switching from person to goods to switch for goods to person it doesn't look that big when you go into the detail it is huge.
Frank: It's a massive conceptual change but what's fascinating in addition to the concept of the Kiva robot is the full system. The Kiva robot is not just a robot. It’s where do you place the shelves. It’s a dynamic storage retrieval system and it's a dynamic traffic system so you've got 500 robots on the floor and they're all moving what's driving that?
Per: How do you optimize the whole setup so to speak? How do you remove shelves so that the most demand are the in the front and the closest?
Frank: So that's interesting that software. The Baxter robot has this theory that we provide the robot but what we really provide is the software. We can improve that robot through software for the life of the robot. That’s sort of what's happening with the Kiva system. They’re improving the software and maximizing the efficiency more and more as time goes on but that's happening in all of material-handling. Now these companies have put off the investment for five years have this new technology and are investing. I went to a show in Atlanta, a material-handling show and I had never seen, I've gone to a lot of shows and I stay in the hotel where most people stay and when you go to breakfast you hear people talking. I have never heard people talking like I did at that show where they were talking about buying. I'm going to the so and so booth because we need so and so and that was fascinating to.
Per: The fact that people stopped investing there for five years might actually have been a very important factor in bringing in robotics in to the warehouse because they simply had this pent up need for something new and also the opportunity because they haven't been investing for quite a while so the timing there was really right. The robotics technology was ready to serve the market and the market was ready to embrace the new robotic technology. Do you see any other areas, we talked about agriculture. Could you tell us a bit about what you think is going to happen in that area?
Frank: In agriculture you have a very psychological phenomena which is don't call my robot a robot. Agriculture is a very automated industry with a lot of robotic activity not called robotics. They call them machines. For instance almost every tractor delivered by John Deere or Caterpillar or any of the other manufacturers comes equipped with a self-driving kit but don't call it a robot. Those self-driving kits will drive using the extended GPS whatever it's called but it's down to an inch and a half of precision. That enables the tractor to drive properly. The other thing that it enables the tractor to do is to be a computer center for the implements or the tools that they're pulling or pushing. Those implements are more and more becoming digital and by that it means that let's say you've got an implement that has eight stations being pulled along each one is individual and it's computer controlled so that it needs to work now but not now, here but not there.
Per: Like a printer basically.
Frank: The concept is nothing new. It’s just a digital activity, a digitally controlled activity from information.
Per: I see that they do this they prune lettuce by applying fertilizers to the plants in succession so they plant a lot of them and then they give space to the viable plants by simply not applying fertilizer to the plants they want to go away. That’s so cool.
Frank: That's one method of trimming or pruning. Another method is to seek out the difference between like a lettuce and a weed and apply soap to the weed all as it's being moved along at speed.
Per: Instead of just like we did before indiscriminately distribute what is usually toxic materials to everything. We can apply it specifically to the things that we want to go away.
Frank: Exactly and they're getting that information from aerial surveys which they used to use planes and helicopters to get pictures of but now with the different scanners and different capabilities you can send up a pretty cheap drone gather information and with very cheap software you can compare that to previous versions of the same image and what you get is a digital distribution map which you can then feed to the implements that you're pulling on the tractors that are being driven automatically. That’s where things are going but if that sounds robotic to a farmer it is not robotic it’s precision agriculture but to anybody else that is a flying robot.
Per: The other one is a tractor robot.
Frank: John Deere's chief engineer and I once had a long discussion and he showed me pictures of a tractor without a cab, without any accommodation for a human driver and he said it costs 1/3. I said are you telling me that the creature comforts cost 2/3. He says yeah the creature comforts and control systems that are duplicated, the air conditioning, the TV screens, the cab, the space, yes that's 2/3 of the cost.
Per: If we could do an automated tractor we would gain is not only the actual automation of using it but we also cut the cost with two-thirds.
Frank: Then they're finding another thing as you build these bigger and bigger tractors to have bigger and more impressive cabs they weigh more so they're compacting the ground which means that at tilling time when you have to turn the ground over for a new crop you need a bigger and more powerful plough to cut through the compacted earth. There’s two ways to handle that. You can get a bigger tractor which is the present way or perhaps the future way is get 10 tractors that are really small and light and work as a swarm. The sum of their cost is about the same as this big monster tractor but these truly are autonomous. There’s no pretence to have a cab in them.
Per: It's interesting to see that that removing the cab has that enormous impact on the financial thing because if you're looking to buy a new tractor one that costs you a third as much is going to look very nice.
Frank: The farmers argument is there could be an animal out there, there could be a person in front of it, what about that big rock that I always avoid.
Per: I think the context of the field is reasonably easy to handle I mean if you compare to the Google self-driving car that's a totally other magnitude of problem but I also see this as a health and freedom aspect of the farmer. I would presume the farmer spends quite a lot of time in the tractors and it also so that I know for forest tractors which is another domain the constant motion of the platform is not good for you to sit all that time and then you move you back, humans are not built for doing that for that amount of hours so getting out of the tractor, allowing the tractor to do its business is going to cut your cost with a third, make you feel better and also make you productive doing other things that the tractor can't do.
Frank: The biggest buyer of tractors is in America and for whatever reason they haven't changed their mind just yet. The same John Deere people have told me regularly that they keep suggesting we'll give you a you know a cabless autonomously driven tractor and you can control them. We’ll set you up with a control room. No, I want to be out there. The interesting thing is that American dairy farmers are very similar. In Europe they have embraced the robotic milking devices which has changed dairy farming. You no longer have to be the person who is up at 5:00 in the morning instead the cows are milking themselves in effect.
Per: They enjoy it I know. They produce more milk and better milk. They were less sick and all these things.
Frank: That's exactly right. The figures all show that and yet here yet in America as old-style milking systems age and need replacement they're not really seriously considering the DeLaval or Lely or any of the other robotic milking systems. I saw the booths at an Agricultural Show and the salesmen all told me the same thing it's a very tough sell in America. Anywhere else it's an easy sell.
Per: How do you think we get from that state to the warehousing state where everybody's buying? Is there the need to be some form of a mind shift there? How do you how do how do you see that happening?
Frank: It could be happening by climate change. What happened in warehousing was this five year delay to a natural process of capital investment. That capital investment is a serious part of farming. As water shortages in California and serious winter storm damages in other portions of farming in America and in Europe are affecting what's planted, when it's planted, there could be a similar five year hiatus of investment followed by catch up and new technology. If you're looking at the technology that's 10 years ago that's going to show up and if you look at farming I would say five years and 10 years or five years forward you will have autonomous tractors.
Per: Do you see that the European market is more open to the autonomous tractors as we were more open to the autonomous milking?
Frank: According to Bosch which is making one of these little tractors that are swarm controlled, I'm going to visit them for just for that reason because it's so fascinating. They think there's a big market in it. There is European adoption faster than American for sure. All of the major farm implements that are digitally controlled start in the Netherlands, in Spain, they don't start in America.
Per: Very strange because we are used to California being such a pusher of new technology but in this field that seems not to be the case so that's interesting.
Frank: It's happening and it's often even initiated in California, Silicon Valley in particular but then it moves away.
Per: That's just the way things are. We've talked about these more or less industrial projects and things that is going to affect us but not directly. Do you see any trends in robotics that the general public will see and be able to enjoy beyond the vacuum cleaners that are coming out?
Frank: The next phase is probably not a humanoid robot.
Per: It’s too complicated and too expensive.
Frank: That's too far out into the future. Probably the next phase is this collaborative robotics. Every major robot manufacturer is attempting to expand into the area of small and medium sized businesses that could not afford a robotic installation for a variety of reasons but the primary reason is that big robots need to be stationary and prepared and integrated and that preparation and integration and stationary have costs. Generally that costs three times the cost of a robot so if you're thinking I’ll try a new robot for 80,000, 75,000 well that's 250,000 to 300,000 by the time you get it running. People buy phones and they buy apps. They want plug-and-play. They understand that. They even understand take a computer and a device with sensors and put on a variety of apps. I mean I can fly a drone from my phone. Why do I need that extra $250,000 worth of installation costs? Also why do I need an integrator when my guys who do the work know the kind of work that they need assistance with?
Per: That would be the Baxter then that you can put in easily.
Frank: Exactly the Baxter and Universal who are quickly programmed and programmable by the actual operator worker that is being augmented with robotic assistance. That’s going to play out over the next few years. It's going to play out because there'll be more participants, more and better companies including the big ones who are reducing the costs and weight and functionality of their products to get it into more hands. All the big robot manufacturers decided for whatever reason that they needed integrators and distributors. They didn't want to be specialists in agriculture or forestry or warehousing or fluid control or bottling plants. They wanted to sell single armed robots and let somebody else do all that so they built up this whole network of distributor integrators, all of which add value. I mean I know some integrators that know the candy business and they know the end of arm tooling that is unique to put on to a KUKA arm. KUKA gets what they want but this integrator is so specialized in the candy industry that they can work for Hershey, for Nestle, for all the other accompanies and because they add value.
Per: It also makes the whole thing much bigger and much slower. You can't do that for a small project. We also see this a lot where you have a big robotics production facility and then you have a human doing every 20 step there is a human doing something because it simply isn't possible to automate it today. I think Universal Robotics to a large has said, we are aiming at every 20 person thing and making it easy and quickly and I also think Baxter wants to be in this space to ease and quickly adapt to these tasks rather than do the full project.
Frank: Using those two companies as an example Rodney Brooks from Rethink is such a good salesman that he has described a need and the use perfectly even better than his product. So the beneficiary of his sales is twofold, everybody understands now that this is a brand new market that can be served with a product as Rodney Brooks has described. Baxter may not be that product, Universal may also not be that product but it's a better fit right now until more products come up. Last year at Automatica I counted eight competitors coming online this year and next.
Per: Now if now this definition that Rodney put up is really becoming a market and really taking off.
Frank: Exactly.
Per: This was very interesting to hear about global stocks and how you took this journey of wanting to
Invest in robotics to this amazing thing and I'm an avid reader of the blog. It’s an amazing source of information. Thank you very much for taking the time.
Frank: My pleasure, thank you.
Per: I hoped you like this episode of the podcast version of Robots in Depth. This episode is produced together with Wevolver. Wevolver is a platform and community providing engineers informative content to help them innovate. It helps engineers stay cutting edge. Aptomica is the founding sponsor for Robots in Depth. Aptomica runs anything in modular robotics. Dream, rent, build. Visit Aptomica.com to connect.
END OF TRANSCRIPT