Flat Earth: Are Engineering Tools About to Disappear?

0:00 This podcast is presented by NX AI, your partner for time series foundation models and physical AI. Hello everybody welcome to a new episode of our industrial AI podcast. my name is Robert Weber And it's pleasure to talk to Peter Sieber. Good morning Robert you're completely alive very much alive. what's happening?

0:32 Yeah It's AI on the volcano week right. so we record Monday. this episode is published on Thursday and normally Thursday early in the morning. And yeah, we have plenty of interesting groups again a lot of participants for STO KSB you name it...and VITRON! We are very happy to be invited by VITRO and I think that will be great event in Parkstein Yeah.

1:02 On the volcano, right? The very first listeners so may not be you, the Europeans but maybe you, let's see...the Japanese or the Singaporean who ever is first in this world because we make it go live I believe one minute after past early morning something like that. So at a time when we will have already met on the volcano and start going into our sessions.

1:30 But by then, what are we discussing today? What have you been doing. You had a very busy week as well last week right? I'd be very busy weaker it was let's say robotics focused weak right. Let's start maybe with the little bit of bias news but because Seb and his team published Kron LSTM And everybody is asking me Alice LSTM.

1:54 What is that? Yeah, it's a good question was grown. I asked them and then he said yeah technically clone lstm uses the chronicle structured information to update and control its recurrent memory state. okay what does It means up did you say? Did He Say chronic like Like The Time Cronos Crone LST M? so it's.

2:13 But he Said Like Did You Refer To Like Cronus Like The time? No no it's it's here refers to Kronecker. And And then now it's a quote that selectively updates individual key values associations instead of modifying. broader parts of the memory, right? And this is reducing significantly. The inference was previously stored information enabling more stable memory updates better long context retention and efficient recurrent processing.

2:59 so that's what we're going to see which I assume you're going to be doing another kind of announcement, which has all to do with his memory thing. And you say he's gonna base that maybe on this Kronos or the others? Ooh! Let me see... Krono. Yeah because i was gonna ask Because in German The word Korn, K-R-O-N as a meaning would mean crown and Maybe that is suggested That it should have.

3:33 No, it's Kronika. And in the second announcement as I think a lot of people knew that already we are working together with Nora building on memory tasks episodic memory for your friends, humanoid robots Peter. So humanoids and arm robots in all kind of robots and robot with wheels and what else? Not not only focusing on humanoid Robots but our kind of Robots.

4:04 Okay good so I guess the first announcement is four. dose of you listeners that like to go deep or deeper. i think that's going to be a certain share, maybe the major shares says okay thank you very much. I'm very interested in what it can do for me. how do i handle this? How can i get all of these things like show my results?

4:32 Can I use it? All the good things like that. and then many of you listeners, maybe myself will leave going into the details. Into the mathematical. let's call them the mathematical details to others. but thats not making one thing big or small. its just some. us go deep and find new things, which is maybe we talk about that again today as well.

5:00 Maybe I have this quote from... Well do it at the very end! But i got a lot of feedback through last episode because there we had also this episode not industrial AI focused episode, more on the research or more scientific side. But I got a lot of feedback from people and say it's great to have such kind of episodes all six eight weeks to get a little bit inspiration.

5:24 what is going when it comes to other domains right? And that we always say as long you are interested in things which bring us directly related maybe by listening to them, you know driving in your car sitting in the public metro tram whatever you're doing. You think oh no what if I would apply? What i just heard In a different area To my own job?

5:54 right yeah Right. so it was very interesting episode with Gunther and Gonzalo Very technically very Very deep when it comes to mathematics but I think It Was Every I don't Know Eight twelve weeks. We can do an episode like this because we got a lot of feedback on that, yeah? What do you have Peter? Oh I'm gonna make a bomb again.

6:15 message of the day is software is there? maybe we shall cut in a minute of silence here. Can't we do it out? Yeah sure This serious business really is not. so i am NOT saying this and i mean Software applications right indiana Because what are software? i'm Not Saying Is Because Brandon Thomas. He's a developer of Artcraft, suite-of-open source creative tools and he meant to replicate Adobe applications right for free.

6:47 And here is now announced apps who replace Microsoft Office Tools Excel PowerPoint Word. We've all been using them. some of us still do others Do not etc. I'm saying this because especially in our world of computer added X X for design, you know I happened to have studied computer at an architectural Design engineering the last two weeks.

7:13 It's unbelievable and Explosion i must say explosion of AI based Engineering announcements. And I was preparing this because it was traveling in just a day ago. yesterday You were sitting in the car driving into CV I was again looking at LinkedIn and there were another ten announcements. It's completely crazy, so i'm going to share some of them And then go a little bit more detail on that.

7:42 the main topic Of today sort of in the first section than is softwares that engineering. So they're as badder bua fear he says quote unquote. three months ago? I shared a demo of AI generating a standard CD model in Aviva. We call Aviva. we talked about last time as well from a single engineering drawing.

8:00 Today AI can build a fully detailed, sentry-fugal pump inside Aviva using eighteen engineering drawings with the level of detail comparable to Katia or Solidworks. Just first quote. I'm gonna give two or three quotes. then there's Aha Ose Zang. he says last month GPT-Six Astra Claude Opus VV reached zero point nine six on benchcat one dot o raising the question is a eye close to solving cat and three d modeling.

8:34 so we have already been building our challenge today. We're introducing BenchCat two or where? The best models course, it's only zero point five nine. So then we talked about you know that the benchmarks. If I becomes too good moving towards one point oh You know would they have to produce new ones? Spectralapse is another one.

8:55 To train AI models that automate CAT, we had to build our own CAT software suited to machine intelligence. now We have a fresh opportunity to enjoy It's rather interesting perspective. the art of making things and I will come back at the very end do this thought once again, by disencumbering engineers from all that is rage inducing about design software.

9:21 So their takeaway is cat built for agents steered by engineers. Wow! I think it's a very nice thought. not saying its true but i do feel the direction where things are moving. maybe. Yeah, I can't just tell all of them. You know here's a guy who is building his own engineering system. and then off course we have the guy.

9:45 you did an interview with our David Davey de Meyer from believe Belgium right? And he is doing... He's building plants overnight every morning when your looking to stream Here build a yogurt plant. not he it's Claude! He says Claude design a biofuel plant from scratch with six fermenters. That's all he says.

10:10 and then you wakes up in the morning, there is his biofuel plant. it's amazing! There a yogurt plant as a chocolate plant. that's it. and then Claude decides on the plot size designs the PFDs not PDF like Adobe. I had to look at up its process flow diagrams. We you and I mean, I thought it is a little bit very close but anybody can look at that.

10:41 And You see a plant not the engineer The persons who have been doing this for fifty years. They go into detail. But i mean as you know As do you recall? Davey's smart guy right he isn't just showing off. He so happy to share what today's possible. So I'm going to round this off by going a little bit deeper, Derek Alexander Molliter.

11:08 he says the same engineering is becoming a Venn Engineering AI. sorry it's becoming a vendor war and over the last months you know every major engineering tool vendors started to move faster everybody doing differently just to call them out that we have in our brain who we're talking about. They are Siemens, PTC, Dassault, MassWorks, Cadence, Synopsys and CIS Autodesk.

11:34 those represent probably ninety percent of the market together right? they all do it differently. now not going to go further here but he says that tool is no longer only a machine engineer. right becomes the assistant. In the end he's saying, you know in a few years engineering tools will look very different from what we use today.

11:55 Now and that is why I'm going to give my reply to say agree Dirk? And thats'the only thing for sure in a couple of years engineering tool looks very different than what were used today right. so in my humble opinion... by now know what I'm going to be saying. What do you think? I'm gonna be saying Robert, software is that!

12:19 I really believe the software's there. this engineering software i don't say is today but it will die very soon. The future value and so much convince is created by context data. I want to do an interview on very specifically context data and AI. For that reason, I believe the winners are going be one of three parties.

12:48 maybe there's a fourth or fifth party. so number one... Of course! The engineering tool vendors aren't saying they cannot change everything. They will And they ARE changing everything. They can buy AI capability from for example AI platform vendors. Number two the AI platform vendors, they can buy engineering context data.

13:09 You know? They can by industrial engineering companies. you know. let's think about what AWS is doing. forgot how to call it. What's that? yeah That exactly. number three Is The Customers The Industrial Organizations That So Far I've Bought These Tools From The Engineering Tool Vendors And now they buy AI platform vendors and develop their own solutions.

13:34 Number four, open stores. number five there's many different ones but I'm going to close this first section of with saying by the way the incumbents are forcing customers out. example of Autodesk started charging when software pulls design data out of its cloud. So as long you work with the Autodesk tools and access your work, it's your IP right then you're okay.

14:07 if you are taking Communities is getting completely crazy about right. You know, what do they think that? That day have the ride to the IP of the customer. So I'm gonna. i'm going to leave it there. It's so big What's happening in this? and this is only one section. as I said I started off with The guy who does the adobe axel powerpoint And we are here now concentrating on the engineering and it's really huge what is happening there.

14:43 But let me make a reference to our friend Bauer Scharinger because he already mentioned this in his book, right? He said that whole engineering process will be totally different for next two or three years. when it comes to AI in Engineering Let's say sweet spot when it comes to industrially I from Boris perspective was maybe not optimizing shop floors, more optimizing engineering processes.

15:15 And and here we are right? It is so easy too open a startup going into this domain Right? So you can see every week three five six seven eight nine ten new startups focusing on engineering topics The different. it's very difficult to understand what is the difference between these startups at the end.

15:40 At the end, It's about a context data. maybe What is your USP exactly? Do you have this context data running the best cat models in the market. I don't know, but this is very interesting because we see a lot of startups and this field will not see a tonne money going into these fields. let's say what was the outcome at the end.

16:00 i'm very looking forward to it And this is not bashing the engineering. No, no at all. any of them are doing well some of them we're things as I just said. i don't think they should do that because Why would they, and what I was saying here is quotes that I have picked up so i'm not bashing those but there are the incumbents.

16:26 There's a huge potential. if you say start-up. many of you listening already thought about doing it exactly this maybe one year ago or in the very middle. And then, of course question is what it's going to do knowing that you and not only the listeners at our podcast are together with a thousand other people thinking about something similar in same area.

16:52 That isn't saying don't do this but keep your brain that any engineers can today say design or buy if planned from scratch with six fermenters. Any of you can do that, so it must be your starting point at least that one. and maybe there's even a better one. I'm just using this quote sort to say from Davy And That is what is possible today.

17:20 So if you can't go beyond that i would almost say then think about because that Is What is Possible Today? Of course it's going to be the shit storm, you know. But then I'm gonna. I'm not gonna refer back today before that reason but i think he's Gonna say the same. You Know what? What is happening Today gives him I don't know if his sixty or ninety five percent and The missing four or five percent in relation To what we can do today It's gonna come over the next couple of weeks a month right?

17:51 There's a German saying. I have another news, right? So there is the German saying and Gutes Näschenhaben to have good... Yes! Have a good nose. it may be not the right translation. maybe you have some feeling for what's coming. You've got an eye of opportunities or something like this. yeah Three years ago i met a guy on Twitter now called Axe.

18:21 His name was Mr. Schmidt, right? So you and he was at Mila in this time with in Montreal And I did a recording with him on On climate change and AI and stuff like that. and then I met met Him two years later or one year later in Paris Oh yeah, and With his own company and this company was called ental pick And they did a little bit the same what Max Welling is doing with his company.

18:50 New materials, yeah? So research new material right and last week I saw news that NVIDIA's now investing in Antalpik. so if you listen to The Industrial AI Podcasts and want to invest such companies i think You listen first. here we have very good naysian Yeah, and yeah very interesting congrats. Were you in the house?

19:17 Jensen was on the house wasn't Berlin right with the chancellor? no I was not in their house. i was sick last week a little bit. yeah Jensen Was In The House. GTC taking place in berlin. yes as You Say many Many Announcements. Loads of them around robotics, right? Loads with him about making you know robotics.

19:37 I enjoy robots etc. a lot of robots and yeah It's the big show right GTC? maybe it's a small Hanover. messer You have your for quote on that, right Yeah from the ARC guy. sorry forget your name now but um you had that thought you were going there. he said Yeah, with things moving changing and I was thinking of it there.

19:59 I don't know when somewhere sometime last night Thinking here is one company Nvidia saying you know we're gonna We're going to come to town And You Know what? Gonna invite. you mean they come with a specific topic maybe but gonna invite. ten hundred companies have no idea And it's the complete ecosystem.

20:24 Their ecosystem joins, comes together for what? Two days or something... Three days of four days! You were there last time I believe. was that right? The Paris one? yeah And that's amazing, right? I mean it is also a little bit scary for the one company has. That um... Is The World? Yeah their ecosystem was very big ecosystem i would say.

20:52 But having said so.. I haven't met Jensen but he seems to be most modest guy in the world. You can leave this thing too Having you know huge huge, huge saying so to say right. Yeah and it's very interesting because in parallel to GTC there was an investment conference in Germany. Oh I saw that as well.

21:21 did you see the quote? Did you see what they were talking about? yeah exactly we have our Chancellor Mr Matz And at least the press release that i saw. The Quote Was That They Were Gonna Talk About AI In Industry Industrial AI. And I thought, isn't that amazing? You and i started this topic seven eight years ago.

21:42 Exactly it's amazing! The quote they used in German was the wording of a podcast we started with eight years back in the german language. It is amazing. I want to come back to Hannover Messer because the USP of Hannover was also always this political stuff on Monday morning or Sunday evening and stuff like that.

22:06 And now they are combining GTC with a huge investment conference in Germany, there's a lot of celebrities politicians Mr. Merz is doing talk with Jensen on stage etc. so be careful I would say be careful because this is very huge and NVIDIA makes no prisoners, right? Yeah. I didn't see the outcome uh... And of course we must say you know ...and We don't want to-we don't wanna Be political!

22:35 We never want To be. but where our Chancellor Is coming from? that mean That's what he has been doing for what have No idea. ten twenty years investment..that was his topic i believe. Right so ..I started listening to it But i didn't Have a good connection Where i Was at that time So i Didn't. But I think it was amazing.

22:55 It's amazing for us, it's amazing the podcast what we started that now. and then iIwas thinking because We had in the beginning And we were very proud of them still am Very proud That we're doing The autonomous driving as well right? And Then we have this Amazing interview and stuff. and then I'm Thinking back.

23:15 you know if We would've before than Decided which we never considered? No, we didn't. We never did but we took it because we thought It was a very interesting almost side topic. So to say if we would have said you know? We're gonna be doing the autonomous driving podcast where he wouldn't have been as successful As we are today in all modesty having sat that at the same time.

23:39 But I think And I must give that to you, Robert. You had an amazing nose started with this very topic at everybody in the global language even is calling industrial. thanks to Boris right. so it was also Boris idea most certainly. Yeah, you have two very quick ones on. I'm gonna give to other topics.

24:05 on my hypothesis that software is there. so There's the new layers through the automation pyramid and not sure he saw That one. So we had Knud Knut as a guy You did an interview with maybe already year ago right? And He came back from which one thing? From one of the cargo. No, he went to the to the IMES somewhere talking about the events, and he says that is a ninety five.

24:36 So That Is The Famous Automation Pyramid. Right It's Getting New Layers And He Describes A Different Lays Where We Were An I Recall We Had The Different Layers. You Need To Remind Me Again Because i think you Have Not Forgotten His Name. The Professor Knudler Salute Was The Guy Who Shared. Yeah Yes That'S Number One.

24:54 Then You Need to Remind me Of The Other Guy The Professor I Went To Japan With at that time. Mr. Industry, Professor Tullke very sorry I always keep on forgetting your name if you listen to us right now but you have Robert your agent? Very exactly recall. in the main Hall of Festo We were there with the OPC Foundation and Professor Zürke, he announced OPC at I believe.

25:26 He announced it that level five or six at a time And everybody had then understood what OPC was really about. now The point is Knutsi's eight layers So i'm not going to calling out all of them are from the bottom. There's physical machines in their software and then this somewhere data ops an ERP? Then come new layers Knowledge Graph, five six protocols MCP A to a MHS and then the agents number seven.

25:54 And on top of the application. so now together with Jan Wolkettle who I will be doing an interview as soon right? Do not agree that the pyramid is still The best way to represent it? i must say maybe i talk To Knut again or Maybe you Will Talk to Knut Again Sometime As Well Because It Is His Thought Right Now Jan Says because it says that the data flow is still one way, so from top to down through each layer and vice versa.

26:23 But the discussions today show that data directly from equipment needs to be available in the top end. I'm gonna close it off there And i go a bit for that So that as Jan maybe we hear more about this when i do the interview with him I go a bit further, say in my humble opinion the new world let's see.

26:42 The New Earth is flat. You remember when people thought you could fall off? Yeah exactly. and as you recall i was traveling...I did spend a lot of time on the sea..i did not smoke. but sometimes When you look across the Sea we know that it´s kind around shade. Actually, there were two or three thousand years ago that people thought so and not only five hundred years got a lever.

27:12 That was a different topic. I found out but people thought you could fall off now. But what i mean is when agents can hop across the layers As if they did not exist We don't need him anymore Really believe? So for me For Me The New World Is Flat And that's very, very much a very happy to have with Knud as well.

27:36 So one he say we're going to have new layers. I said no agents don't need the layers you know. so why would We need them? The title for this episode is Flat Earth. Yeah right software is there flat earth. I'm gonna just quickly bring in and then i'll be positive again. Karu Manchi says PLM is more than managing cat files.

28:04 It's how everything connects together, modern product life cycle management brings engineering manufacture. it's very similar. he has all these different...it's not like the vertical but its horizontal right. so one comes after another you go back and forth..its a horizontal thing. And of course I am the bad boy and say AI will integrate.

28:27 i make him superfluous. That's, again of course very bad and people don't always like that. And I do very much understand it... ...and thats why i will say to each one you make sure look at your own job Look what is happening in the job and change your own jobs. Don´t wait until its going be changed.

28:45 and having said so Not a person who has it right Robert is the guy with his nose What I was saying, and then realized that the first quote is about vertical integration. AI doesn't care that a sensor value is at the bottom, and let's say what we had called an app. An AppValue at the very top probably does not care... You are asking it to design your factory with six fermentators!

29:31 It will include whatever is there from the bottom sensor machines to top what today we call application. We don't really care, we called it application of software right? Do you remember I think one and a half year ago i told You that There's A new better Ground When It Comes To The New Domain Of Models.

29:52 Right And I Said Time Series Is A New Better Ground For The New. And We Can See This In These Days Because We Got A Lot Of New Models in The time series Market. So We Get Something from Mistral. We have something new for Google, something you from Toto DataDog, something new on LG from banking accounts and banking companies.

30:16 it's huge what is happening in the time series market at this moment. so I see that really a new battleground of companies or models not only LLMs but also Time Series Foundation Models. Okay, yeah. Kronos is there as well right? Krono's two mistral launched a new way of doing time series. then we have to get jam gamma T O better.

30:42 so French companies interesting very interesting. the New Battleground I think one new battleground. Yeah i was listening to plateau Socrates this morning and he did a reference too To Kronous which I think it's One Of The Gods. not sure He must've been the God of Time my guess or but I didn't go into details there.

30:59 Perfect, what do we have for the main part? Main part... What do i have should i not tell you know..I'm going to tell you other things later on. yeah very good no that's first. uh so i'm gonna be talking to. Torsten-i did already talk to Torston and so i am going to be talking about connecting agents AI agents with industrial robotics.

31:19 now at that time that we spoke a couple of weeks back specifically by means of MHS, SRCI. Now Torsten is the guy who in his spare time I believe like three months he just built a robot. okay explain what SRCI is? What MHS? we have heard about that one. so those are the topics. and then In The End He has view on how to bring AI agents into real world.

31:50 I must say, it's almost like... It shows how awfully fast our time is running. This was an interview from two weeks back and here we're talking about all these layers and all the acronyms where Dan just five minutes ago said are they still going to have them or not? That we're gonna continue. I'm going to continue that with Jan as well.

32:12 But for today, I hope you're gonna be enjoying my talk With Thorsten Brach here okay? He's with Bosch. yeah exactly Bosch manufacturing solution sir Okay and he is a firefighter. i see on his Oberfeuerwehrmann Very, very important here. Very important. thanks a lot. it's amazing to you that he. well I'm not sure i hope He hasn't had too many firefighting with the agents next To his job at Bosch and next to having built A robot completely by himself.

32:50 perfect Peter It was a pleasure Thanks a lot. Thanks, Robert. And take care and we will see each other at AI on the volcano in a couple of days here. looking forward to thank you very much. dear listeners hope to have you with us again soon. thanks bye-bye. my name is Peter Sieberg. I'm your host today.

33:11 i am going be talking to Torsten Brach. he's with Bosch Manufacturing Solutions And Torsten and I are going to be talking today about connecting AI agents with industrial robotics specifically by means of, not as two acronyms always need to be careful. MHS and SRCI. Hi Torsten! Hi Peter how you doing?

33:37 I'm doing good and you? Oh yeah well no oh yes i am um As i suggested before as I'm based in Munich and part of a marching band. We were playing yesterday, but I am perfectly happy to find this morning! Thank you very much. Let's start with you please introducing yourself to our listeners. Yeah, I'm Thorsten.

34:01 I am a software engineer with more than twenty-five years of experience in automation and robotics And yeah it's also kind of a hobby. so i spent quite a bit off my free time working on My own project but i think we'll get into that little later. That sounds great. having your hobby as job isn't the perfect thing everybody would like to go for right?

34:26 Yeah, that's great. But then it is also hard to find an end at a day you know? I understand! You never know which is the same thing when... because i tell people all of time It´s the same as now. How long do we have mobile computing and how long has been using notebooks And later on even worse mobile.

34:48 So as soon that started, I guess whatever twenty years ago and twenty-five years ago wherever we were where ever we are uh...we can always work. it's the same topic. nevertheless please tell us a little bit more about Bosch manufacturing solutions. I don't know, company or department sit within the big world of Bosch.

35:11 You may want to spend two three words on Bosch as well... ...I think most listeners will have heard about them. and what industries? What product areas do you use as a company sir? Yeah, Bosch Manufacturing Solution is a Bosch internal special proposed machinery division with different locations across Europe Asia and North America.

35:34 We are what you might call a full liner. They have lots of experience in areas such as battery, e-tribe sensor electronics and as well as hydrogen technologies. but we don't just work on an internal project. they also deliver projects for external customers And I personally worked at the department where we developed the next heat automation platform which provides foundation to most these projects.

35:57 Sounds great. Before we're going to move into today's topic, you... Oh that is an update! You just built a robot in I think around hundred days. where exactly at the topic? That we started? IN YOUR FREE TIME Tell us little bit about what was idea behind doing it. Yes, thanks for asking Peter. Yeah you're right I built this in around three months and the idea behind is yeah we all see the impact of physical AI And uh We See All The Humanites Coming and so on...and So On..And Uh Yeah i work In A Traditional Automation & Robotics Department of all these physical AI things.

36:41 But I think, uh...I will need the skills in the near future because the systems are coming and yeah so i decided to make a project. Uh..i built my own humanoid robot as an open source So it's available on github Yeah! And start building the hardware ..and already have running but with traditional PLC insights.

37:01 All the AI magic is still missing. ...but that exactly what I want to learn through all the topics which are needed to build such a system. And yeah, I think it's great playground for me to discover how things work. That is amazing! We're going. come back later and we'll be talking more about human noise as well... ...we want your perspective on what you do.

37:26 talk today basically. Maybe let us start with that one. Your experience in robotics. You say you have, I think that you said thirty years as a software engineer. now please share with us little bit more detail your past involvement? Your experience with robots and robotics because... ...I do assume the big share of our listeners will know and we'll have heard or would've been part of this world but there is most certainly probably many people that are relatively new, specifically coming from the humanoid part.

38:06 So tell us, share with us what have been the types of robots? The brands you've been working on and their capabilities or responsibilities in exactly this world. I think i already worked with all major robot friends but... not on the robot side itself. I mostly come from the PLC site, do integration of robots into our platform and write function blocks.

38:36 We have a technology package called Integrated Robotics where we move programming from the robot to the PLCs so that teach points are in the PLCS tools coordinate systems and then we enable the PLC programmer not only to program a PLC for COM, but also programmed the robot application. And that makes things easier because on one hand side you have kind of an abstraction layer so you can scale solutions across different robot vendors.

39:06 Because if your program in the robots system Then You Have To Develop Things Multiple Times For Each Robot Brand. If You Move The Programming To The PLC then it's easier to implement the applications and in case of service, for example. Then don't need to decide do I have to send a PLC programmer to the customer?

39:25 Do i send a robot engineer to the customers or does he has to send both? on this case The PLC program is enough. He can use those jobs And you Don't Have To Consider If The Person Has A KUKA Experience Or Yaskawa Experience For Example Exactly. And that is the problem because mostly your robot engineers only can support one or two, maybe three different robot brands and then you need another expert on this approach.

39:57 You don't need to deep vendor knowledge. We're in the middle off our topic and you trigger specific thoughts, specific questions I have but i'll ask them later. Remind me if I forget when we are going to be talking about just as one example of a communication architecture OPC UA because they have a very specific solution.

40:23 that is well... But we're not gonna start there! We do that later. AI agents with industrial robotics, which is maybe things that we're doing these days specifically through mhs srci. We are talking industrial robots and connecting them. So before we get to the topic of today Maybe you want to give us kind of an introduction into Kind of an overview of what is industrial communication?

40:59 There isn't different protocols. What are the types of protocols? what can they do, what has been your experience in using them. How does the PLC communicate with the robot? there's many different languages. you know that's profibus, profinat others. I don't know, maybe you have your own languages so to say that you are using?

41:21 Yeah mostly you have a field bus system between the PLC and the robot. yeah it depends on the brand of your plc system. i would say in the Siemens world always has the profi net. In the back of you have the ESA CAD, but there are a lot other feedback systems as well. And that's yeah little bit off the feedbacks more I would say still has.

41:44 when did they take place? That's mean world because i was part of provider of industrial communication solutions maybe just after...I think When I met people represented is exactly the solution. I did recall still the feeling that they didn't really want to talk together. That's maybe like twenty years ago, i guess or?

42:10 The Fieldbus Wars. Yeah it is still active but we will overcome this and have one common solution future for inter-operability between systems vendors devices. Very good, because we're moving into today's topic. We are talking other ways. maybe or maybe they sit on top when were talking MHSSRCI and I want to know from you first what their are.

42:38 i believe this is it on the higher level so say than the communication protocols right? So tell us first What does SRCI stand for? Yeah, SSCI stands for the Standard Robot Command Interface. And yeah that's a relatively new standard that aims to standardize communication and control of industrial robots across different robot vendors.

43:04 This has already been designed with future in mind. it is not limited to simple pick-and-place applications but also supports more advanced application such as Conveya tracking force controls or trigger functions. It was originally developed by Siemens together with several major robot vendors. Later it was moved under the umbrella of PI, ProfiBus and Profinet International which were a great move because its means that standard is now managed by neutral organization where they can maintain further development.

43:37 But does this mean for example... And number one? Was I correct in assuming that as standard like SRCI sits on a higher level than the communication protocol. In this case, maybe Profibus-Profinat? and does it mean because they are in the same organization that typically... ...the SRCI standard is being used by companies at lower levels using the past Profibust but today may be Profinat, a communication standard?

44:08 Yeah you're right! the field bus doesn't matter. The protocol is based on, I would say an anonymous byte array and it's... Doesn't matter which protocol you use to exchange this byte array there's a protocol inside And uh You have mechanisms to pass this telegram get information out of that. Okay i understand That.

44:29 so thats first one next one mhs. what does MHS stand for? What Does that do. Yeah, I would say it's better to start with MCP if this is okay for you because then I can connect MCP to MHS and i think its easier to understand. So MCP stands for the model context protocol. yeah It's an open standard that allows AI agents to communicate with external tools and data sources in a standardized way.

44:58 For example, an AI agent can use MCP to access file databases or other applications without require custom integration for every system. You can understand MCP as a universal connector between agents and software applications. I would say Very good. And it was introduced by Entropic a year ago, I guess or maybe longer?

45:21 Yes exactly. and we already see the huge potential of this protocol especially if it comes to making interaction with software more efficient and AI agents can automate tasks which would otherwise require a lot of manual work. So, and then we come to MHS. And there is where the things get really interesting.

45:43 I mean, MHS stands for the model hardware standard on that's the Pondor of MCP while MCP yeah it's more for connect the agents through the digital world. MHS is To Connect The Agent Through The Physical World. Okay very good That. so We have a little bit Of A Perspective Here at a certain level, the communication protocols.

46:09 And then at higher levels. now we're talking. maybe one question is but what will do that later? It's like... What may be interesting I don't know your perspective on this. as long as you've been talking about communications standards The companies typically would bring these standards shared to which are more European at least originated back off Siemens.

46:35 There's others there specific ones for the Japanese market, four Rockwell The American Market. they were typically automation companies and two Ones that you now talked about we aren't going into a detail on. And then SRCI, I think you were part of that one. Now we're going to move into that one. but the MCP and MHS now have been brought in to market by companies.

47:00 are not from an automation space right? Yeah That's Right. So We gonna see what it means maybe later on. You maintain open source implementation. four, and I'm going to just spell this one out. IEC six-one-one three-one dash three environments.". And of course we're gonna ask you what that means with the SRCI core profile already tested with robots from multiple manufacturers.

47:30 now those listeners are come from the automation space that can call out this IEC standard by heart. For those who are not, please tell us what that standard stands for? Yeah! This EEC-SixEleven-ThirtyOne is an international standard for programming PLCs. It's standardized programming language, program organization data types variables and execution concepts so that PLC programming follows a common structure across different manufacturers.

48:01 That's the idea behind IEC. Okay very good because we're gonna be finishing this. at least what i have in mind with it would a big question. Yeah, if we're still you know how far we humans. We the people that have been coding exactly along the lines of this way of coding programming PLC's If we humans are going to continue To do that but we keep that thought for the end.

48:32 now You have Maintained this open source implementation. Tell us a little bit more about it and maybe you want to move on or ready too how this standard you are now applying mhs, what is all about? I would start explaining why i implemented this interface as an open source. As already mentioned from my perspective SSEI is a game changer because it makes things easier if it comes to robot integration and because this protocol was introduced by Siemens, the CMATIC PLC system which had an integration for SSCI.

49:17 And I was completely excited about its interface. nothing happened. At the beginning I was checking their websites every month, then only every three months and at some point there is no implementation available. so why decided to implement SSEI interface by myself? What does it mean what? maybe share with us a big picture of you know providing these solution of you making that solution available.

50:00 What does it mean? what is now different from, what can somebody else do and maybe try to give us a use case an example? Yeah, I think the problem in industry is that every vendor has its own proprietary interface. If you have to integrate a robot and read all the manuals and do all of the integration work And yeah, if you have a standard interface like SSCI.

50:44 You just have to maintain one single interface which fits for all the robot prints available in the market. not all but we see that more and more robots when those are joining the SSCI working group. So maybe in the near future, we will have all major brands inside and that makes things a lot easier for their applications because they just to include one single interface.

51:08 And then I can work with different robot brands. now okay? That's already giving me. The answer is how does MHS make these hardware or implement this hardware standard, so to say. So what typically MHS... We have ten robots, industrial robots. we start with them standing in front of us. and what would MHS do?

51:41 MHS will go to brand number one. let's see it's a KUKA. and What Would MHS Do? but maybe read all exactly these manuals as an example And then decide, okay I've read everything. I know about how to communicate with the KUKA and now i provide you which is an agent or a human but it will do the same for brand two and number three Is that?

52:15 The following question is like How are you helping MHS by providing I forgot my acronym SRCI interface, right? Yeah. I think our industry is very slow. so if you would wait until every robot has an implementation for MHS then the standard does not play any role anymore. just take an existing solution like SSCI and make a, uh...like an adapter in bridge then mhs to srci bridge.

52:50 Then you don't need to implement image as directly on the robot system than your. just have this adaptor And you are able to connect all SSI capable robots with MHS. so that's basically the idea behind. Maybe this is the point where before I'm going to forget, i'm gonna be asking you and there's a disclaimer.

53:12 You know those of you that do not know Robert and I? we also produced he co-produced the OPC foundation podcast. it's just a question. It could be with any other ones but specifically an I've referred here couple times that OPC way maybe and they don't want to be called like a communication protocol.

53:38 They see themselves as an architecture with security embedded, etc., but the most important thing always has been this information model. at least I've understood it for last couple of years maybe also always had this potential for, you know doing for AI. For standards or artificial intelligence. meanwhile superintelligence is something with the machines and solutions we have now.

54:10 I know specifically that these information models they exist for specific subsets whatever kind of machine companies come together they produce this information model. And one of those exists, I know for robots. now where is that? Where's that overlap? are you aware of That One? and whereas the overlap or where Is your solution different from The one that it may be being provided by as an example OPC foundation.

54:39 Yeah i Know what You mean. It's a companion Specification. i think yeah i'm not so deep inside. Or What's going on in This direction. yeah i Think we Are more In. How can I say, in the practical way we communicate via field buses with systems to have high performance and real-time behavior. I think this OBCUA will not have that capability with Realtor maybe with OBCUAFX or TSN.

55:08 Maybe then it's possible for them also to perform with OPCUA And i'm very sure I think these standards will grow together. Okay, still can you give us an example of what you know typically? What you were doing a year ago or two or three four five years ago when When-when You wanted to start communicating with one of these robots that we talk about right.

55:35 What did you have to do? how long did it take? And then today Today you so want to say me or anybody else listening and interested What do they do instead today and how long does that take? So what is the difference? you are providing them with. Yeah, in the past to read all the manuals of each robot vendor, then you have to define an interface because we just had some inputs and outputs which I exchanged between systems.

56:10 but there is no standard on these bits in the field bus. So here they defined your handshake flexes between the system so that you can define values you want to exchange. Then you have a kind-of handshake for the robot program function block with a handshake inside where you're then able to trigger robot program on the robot side and we have to program everything in the robot.

56:38 I think that is what's done in the past. And depending whatever your job at hand, but it will take couple of weeks or months? Yeah this would take for sure. What can you do today? Today it's very easy. You install a technology package on the robot for SSCI, then download the SSCI open source project.

57:09 if your work with codices or Trinket-based PLC system Or if you have an CMATIC PLC, use the function block from CMATIQ and directly program your robot in the PLC without defining programs, without defining interfaces. Without defining which values for exchange everything is predefined in the standard and you can directly use it on directly access your robot.

57:35 now You're still programming I mean but you are programming at a higher level. maybe so. Is It more like what you had to do In The past after studying all the manuals? IEC six one no, that's wrong whatever. That standard is very specific standard. you exactly need to abide by the rules of how. that's a standard.

58:00 this is written. when You say yourself produce are you today? When you say programming do you say There was some kind of solution that you have? I assume you have Some kind of application. does something that needs a robot And that means it's going to give signals through the robot and jobs. To do move, to the left pick up a box put it down Pickup the red bin whatever.

58:27 is That still so? I Is there? It's The new way of programming at this higher level Of you know Do things an in the past was about You Know more cryptic Sticking to the rules off the language of the robot. What it is that you program these days? Yeah, I would say in the past. while all their robot language of the major brands there are very old programming languages because The base off of the robots software was developed.

59:02 But we'd say In the nineteen eighties around That's a time did i learn. So they're probably in Fortran or Pascal have no idea what they were. yeah and Because Windows downwards compatible, they were not able to modify the language a lot. That was the problem. or is it a problem in programming languages of robots?

59:25 If you move it to the PLC, as I mentioned then these things get a little bit easier. but yeah. The EC-SX-Eleven thirty one is also not in modern programming language like Python where we already have object orientation inside. But that's still still hard work through two program and the PLT. And uh i think In the future will be hard to get people who want to do.

59:50 Yeah, I understand. Now our topic is that we're talking about connecting agents with industrial robotics. so what and in the meantime i have discussed this a couple of times at least once would Robert it on use section and I didn't really know what it is that i wanted to discuss. but in the end, we need whoever wanted to build some kind of industrial automation solution or coding, you know whatever the language was.

1:00:37 You have to stick to these languages so that if a thought are you get a job and you think okay now I'm gonna start programming by using the languages that your learn. my hypothesis is what were doing. as one example is that we're skipping, so moving up again another layer. We talked about the communication standards at a certain level.

1:01:02 maybe there I think also global systems for discussing these layers let's say they sit in Layer three and then on top of it. now are talking about new layer didn't exist before. So how do you see working yourself? For people from your community are agents, but also humans. Are they gonna be programming?

1:01:29 quote unquote the robots in the future by talking their own language? So German or Dutch or French or whatever Chinese By saying you know go to the left goto right Whatever rather than Talking the cryptic languages of the standards that we have been using fifty years. What? what do think? Yeah, I think in future it will be like you described so that to just say for example picked a red cube and the robot would pick the Red Cube.

1:02:01 but i think inside of software we'll still have normal deterministic programming as before. And with who takes care? You know, and again I have been slaughtered before. When two years ago we talked about this kind of topic when it was about general coding so the biggest friend people who are capable.

1:02:30 but it's maybe a thought, an apostate that the agents will take care of. So software takes care exactly these deterministic layers. and today we're not discussing. I believe you cannot from today to tomorrow just you know, build your factory and have it completely done this way without any human being capable of understanding.

1:02:58 Any of the things happening at that layer we're talking. I'm suggesting these things may happen in the future. And what does that mean? That a higher level person who has less understanding can still be very useful And maybe he or she needs some person at the lower level who can go there if necessary.

1:03:27 I'm not sure of that. makes sense. Yeah, and i think it's a real problem in future. If all the software is AI generated and something breaks Nobody is able to understand how the systems works. Is that really so? The example I always give, and... ...I have this friend. And his friend is you know, eighty more years old.

1:03:57 So he learned programming before me. He was like the sixties or seventies i guess When he was programming on a chip level and had to move registers from left-to right. I've no clue what that means, but i can imagine it was something like you know making space somewhere on the chip free for Something else.

1:04:22 he had to do that himself. I assume That you have never done that? The point of Making is You Know this Is being taken care Of by I assume some software layer On top of the Chip. right so But a chip manufacturer whatever that chip is. So And I'm not saying that for the reason we can forget all, and then we've built so many layers on top of it.

1:04:48 But nevertheless... ...I am NOT going to try put any words in your mouth here! At least i would agree if were gonna be moving this direction.. ..i would agree that we need certain ways of confirming whatever agents are doing at a lower level makes sense. Still where we haven't talked about the agents kind of yet.

1:05:11 Do you have specific thoughts there? I mean, are you can share a project? Can you Share specific things that you are doing with Agents connecting to industrial robots? yeah experimenting visit. i can share an example and proof-of-concept. In our NECSEED Automation Platform, it's possible that you can release functions to third-party applications via an REST API.

1:05:41 I was experimenting with this and released some function of my machine. This was the robot. There is a predefined function for pick apart, check apart or place apart. And then i used FAST MCP because mhs was not announced. I used FastMTP and took this REST API and provided the capabilities of this West API via MCP to an AI agent.

1:06:09 And then I was able to tell the AI agents, pick a part on check the part or place apart... ...and that they were able to execute sequences. So is it like interpreter maybe? You're sitting there how do you write it? but could speak as well? so what's the receiver? in LLN you say, move to the right. Who is a receiver of this command?

1:06:36 Move To The Right. Yeah I talked with an agent but in the background they used MCP to address my REST API and then it was called Deterministic Function which has already been programmed on PLC. So execution as you said still happens deterministically is something we will see in future. if the AI agents just reacts as an orchestrator of predefined functions.

1:07:08 This had a advantage that you are able to have interlock mechanisms, so then you can take care of it and agent cannot break things because this sequence was predefined. but I think they cut their capabilities off the AI. So i think its very hard find something in between. What do we want AI to be capable of doing?

1:07:34 At the very moment, there's discussions at the highest political levels globally. What it is that we want an AI to have been able off-doing? right okay understand but maybe so when we talk age and can you a little bit more precise? Can You share with the type of agents, maybe a name. So are we talking?

1:07:59 you know foundation models? Are they also may be open source? could there also be closed-source? what is this thing that I have typically always been calling my personal assistant because That's what i believe in. even in couple years from now im just going to say telling things Assistant and receiving things from my assistant doesn't matter in my private life or In my work life as well.

1:08:27 So what is that? Can you share one two three names of tools our platforms are? Yeah, my examples was very easy. I just used my VS code And the VSCode i use to chat which was connected to an LLM and I Talked with them on say okay please robot pick the part and then uh, the agent was able to discover the mcmcp server.

1:08:52 And uh, to execute this command. very good let's come to a close by talking about what we started in the beginning. you built your own humanoid robot. I understand so now. So we've learned where you've been coming from. industrial robotics You're building, you say in your free time. You need to move into physical AI because that's where kind of everybody within our space is moving.

1:09:21 so with the experience that you have had but also looking or combining it with the solutions that you've been building with your colleagues for thirty years... ...you know the industrial space very well. and then what has used? a view? of humanoid, of human-oid robotics. Of the different modularities you know?

1:09:49 are these than the walking like a human type of robotics? Are they gonna be in wheels? what? how do see that space coming from the industrial robotics, where do you see solutions there? and that makes sense? Yes. I think first we will see a wheel based robots in an industrial environment because yeah it's easier to have the safety... to fulfill the safety with stable wheel-based system instead of balancing systems like a leg-based humanoid robot.

1:10:23 I think that's very clear to me. maybe later in future we also see systems with legs but i think the first humanoid introduction will be this wheels. and it absolutely makes sense because our factories are built for forklifts, for AGVs so on. The environment is there. they can directly interact into environments.

1:10:49 That's basically the first thing to do. So you mentioned the forklifts, the AGVs. yes I see exactly well at least let say in the newer larger factories with a typical shining concrete floor right? You know we sometimes say it so clean that you could eat from it and those factories not. maybe the small ones that have come out of age and are very old, then.

1:11:21 Maybe where that doesn't work. but you mentioned the AGVs, the forklifts... Of course the providers of those logistics solutions. they come to market by putting AI solutions, agent solutions onto their machines on to their vehicles is a better word as well. That's of course one solution, how do you see that in relation to the companies?

1:11:52 In a meantime we have how many I don't know not only hundred maybe five-hundred companies across the world that all come with their humanoid solutions. Do You See One With A Better Advantage Than The Other? One Who Do You Think Is Going To Be the first major players to be successful in that world. I think what we will see first, because the humanoids are currently do not have this safety... ...I think that we'll see first arm-air robots with a cobalt on it because that is something which already is safe and you can directly bring into the production.

1:12:32 I think this for now, that's a solution. but i'm very sure we will also see the humanoids in the production. We have seen them at test environments. it was important to start with trying things out two, three or five years. The humanoids will come and then we need all the experience how things work. what are good use cases?

1:12:57 What our bad use case is? How to solve problems, how to re-teach or relearn algorithms. We know that you need this knowledge To be competitive in future And maybe were soon going see your robot In an industrial environment. Oh I think my robot was not capable to move in the industrial environment, because it's built from sheep components.

1:13:25 Because for a hobby project... It looks like an industrial one! I have all of emergency stops and things which are familiar with automation world. That is what i recognize here right? But yeah you would never be able to interact in the industry but maybe at some time that can bring me beer The most important thing, bringing a beer to the man.

1:13:53 Nevertheless this is your version number one. I think it's... It's a wonderful thought! I said similar things week or two ago for any of you listeners You know? And then doing this even in your spare time with having knowledge of thirty years being software engineer and then having this drive, this feeling I want to do it in your personal time.

1:14:25 It's a wonderful thing! And you know... You're kind of driving your own future by doing that right? That is a wonderful idea. maybe closing off Torsten tell us about the typical day in the life of Thorsten Brach. That's a difficult question because there is no typical day in my job. That's also the nice thing in my job, so it is not everything.

1:14:58 The same isn't boring. for example at beginning of this week I was sitting on a car and was driving to customer because had problem with interface between PLC and robot Because there were new firmware available And that they are breaking changes. So i immediately have sit-in-the-car and moved There. Then the other day I sit in their office and programming some components for our framework.

1:15:25 And next, they are discussing solutions with customers. so it's a wide variety of tasks to do. It is fun! So i really enjoy my job. As you know ,I also take little bit of my job home during business hours. Perfect, thank you very much Torsten. Yes we can hear your drive to be doing all these wonderful things.

1:15:55 Thank You very much for having shared Your experience. I think it's Very important that in this industrial world In which We are that we think of the potential of doing new things with agents, with AI maybe with certain types of humanoids as we hear from you and you say yes they are. They will come not any of them but specific types first or may be other ones later.

1:16:25 thank you very much for having shared it with us. I'm looking forward to see robot version two three by Trosten Braach, making it an open source for everybody to use. Thank you very much! Yes thanks Peter and thanks for having me. Okay bye-bye.