 Well, good morning. It's 9 30 so we should probably get started. I'm Bernie angle head of ag and biological engineering Great to see all of you here this morning So before I do introductions assuming we have some time at the end for questions. Let me remind you now You have speakers at your desk So if you'll press the button continue to hold that and so when the green light is on you're able to speak and we'll Pick that up in the recording. So so hopefully we'll remember to remind you yet again at the end for that So let me introduce Dean Meng Cheng from from engineering. So he will introduce our speaker So mung is the John A. Edwards and Dean of the College of Engineering His research received the 2013 Allen T. Waterman Award very prestigious award his online courses and textbooks have reached over 250,000 people And he's co-founded several startup companies and a nonprofit consortium. So Dean Cheng Thank you, Bernie. Good morning everyone Well, it is a delight to be here and to introduce the outstanding lecture We have today as the third installment of the new series the Purdue engineering distinguished lecture series that started two months ago and on a roughly monthly basis we have One distinguished lecture across all engineering fuses and today We're particularly delighted with a B. E. As we all know one of the very best in the world And joined between the College of Agriculture of a dimplot over there just joining us and With the College of Engineering Now I know that this is being streamed and recorded and I will be tweeting it momentarily But first of all, what a fantastic topic and what an outstanding lecture today Professor met Dar is the Professor in Agriculture and Bound System Engineering and The King's Manufacturing Fellow at Iowa State University received his PhD from Ohio State and His research and teaching focus on digital agriculture agriculture broadly defined Especially in the use of electronic technology and data analytics to solve applied engineering challenges in the agriculture industry And he leads large team of graduate students and staff working on precision agriculture telematics data analytics on many aerial systems next-generation machinery automation and all of these are music to our years as here at Purdue We are Looking at exactly the set of exciting directions that will transform how We look at agriculture engineering and how we feed ourselves in the centuries to come and Professor Dar has received numerous awards including the New Holland Young Research Award from the American Society of agriculture and biological engineers the W. Ferrell Young Educator Award from the American Society of Agriculture Biological Engineers the supplier innovation award for yield sensor technology from Deer & Company and Precision Act Excellence Award in Education Research from Precision Act Institute and today's topic is commercializing academic innovations in digital Agriculture and I have to say I am personally a big fan of commercializing academic Innovations is a win-win it fulfills the land-grant mission in unique and powerful way and it translates Fundamental research into societal impact. I am eager to learn a lot today from Professor Darce Distinguished lecture. Thank you for joining us. Thank you. Thank you Can everything back here be fine? If not, there's plenty of seats up front, so you feel free to move down at any time It really is a pleasure to be here today I've you know the ag and bio system engineering community is relatively small or disciplines relatively relatively small and a lot of us already have great networks and to be here for the next really two days and just Continue to grow the network between Iowa State and Purdue and in collaborative opportunities is really what what has me excited to be here I will tell a quick story. You know Purdue is the is a highly highly ranked ag engineering program. Is that is that correct? What's the ranking so I was given it I was given a presentation I gave a presentation a Few months ago to a group of executives from John Deere and I opened up with Iowa State's the number one ranked engineering program I had an asterisk says, you know But we're actually tied with Purdue and their comment was you know I've heard that a lot from Purdue, but they never put the asterisk on there And I had to tell them that you know well I is before P in the alphabet So when we we list them alphabetically, I guess we'll take that But it really is a pleasure to be here and talk about two things that that personally I have a lot of passion both digital agriculture as well as Commercialization I think you among mentioned the importance of this within the academic sector I think there's a lot of differences depending on what disciplines are in and in our discipline for us to really have an impact Right to go to go home at the end of the day and feel like we have moved the needle in agriculture Commercialization getting technologies to the market is something that we really feel is very important to the to the process And and that's different depending on the discipline you're in but that's that's the field we come from My intensity. I'm going to spend a few minutes going introducing my program I'm going to talk about You know the the breadth and scope of what digital agriculture is probably a little bit of background And then I'll dive a little bit more into some examples of ways we've commercialized digital agriculture Hopefully to use that as a way to maybe Inspire open up some thoughts and then close with some general comments around commercialization The folks at Purdue told me that this is over when I'm done talking so so just just have a good seat here We'll work through this so The the program I wanted I was to have been I would say for ten years I run the the digital ag innovation lab is what we call ourselves. We are a Multidisciplinary team meaning within my own research team. We have PhD level agronomists and soil scientists And we have PhD level computer engineering staff and graduate students So not only are we interdisciplinary across campus, but we're also interdisciplinary within our own organization And that is that is key in digital agriculture digital agriculture is There is no such thing as a digital Agriculturalist, right? It's not a it's not a thing. It's a it's a blend and emerging of Technologies we are very focused on tech transfer. That is what what drives us what gets us going every day and at this point almost exclusively working with industry partners, so we have 15 industry partners we do about four million dollars a year and an annual research with those partners through through our team The rest of my group we have we have four research faculty that that worked for me in these in this in this area and then a host of professional staff that Help lead projects again most with masters and PhD level backgrounds to to lead the scientific aspect of what we're doing here And the last bullet so the global footprint has been something that's really important in the commercialization space if we're going to Commercialize you know I always remind our students and our most of our students and our undergraduate programs are from the Midwest, right? For most of the you know in a relatively central area, but Agriculture is global and if we're going to think about making an impact We have to make an impact outside of the states we serve but also to the global breadth of what exists and in agriculture today Why we do what we do So I have a lot of staff I do I spent a lot of time on employee development And I think that's a key aspect of what makes our team successful And I asked this question to my staff every single year when they do their self evaluations of performance Why we do what we do what is what is it that really gets us up in the morning gets us out of bed Gets us focused on being innovative and drive technology and for us that mission is all about advancing Agriculture through the innovative use of technology Agriculture is a is a very complex industry The decisions that have to have to be made to execute agriculture the risks that are involved in agriculture are very dynamic And the use of technology has an opportunity to De-risk and enhance productivity profitability sustainability of agricultural way that we think is really powerful and that's that's why we get up every day That's why we that's why we come to work The the model that we've set up I think is at Iowa State It's relatively unique, but I think Purdue is also a real leader in the area of engaging with industry and thinking about how industry partnerships can really be fruitful and and lead to strong partnerships that branch beyond just traditional aspects of funding so for us industry partnerships are really at the heart of everything we do and Through strong partnerships and the right kinds of partnerships We can create relationships that branch from basic research to maybe applied research and all the way through the commercialization phase We believe the benefits of being able to be involved in all aspects of that Allows us to better prepare our students to enter into industry and gives us a better microphone or a forum to To have an impact, right? I relate this quite a bit to some of the colleagues I have in in agronomy at Iowa State and I see at least one agronomist in the room. Is there anybody else from agronomy? Risha holding up the flag so, you know in a lot of ways I Envy the voice that our extension agronists have to direct change and inflect change and help to progress agriculture in the right way and For engineering, you know if we have an innovation Iowa State's not going to manufacture it Purdue's not going to set up a Manufacturing Center and start selling products. So that partnership with with industry is our microphone It's how we make sure that we get the success and the innovation and the ideas out To a place where they can actually impact producers on the on the farm Our research team members do become integrated parts of product teams And so I think when you hit a level of engagement with a company There's there's a certain inflection point where rather than being a Outside entity that you're doing some some research on pretty soon. You're part of the core team Okay, from a graduate student development perspective. That is a huge opportunity Right. Does anybody have graduate students that maybe didn't always take the advice you gave them or you gave them tips on a presentation and that I see some people smiling at least right Well when they're engaged with industry all the way through their process That's that's that that gives us the ability to grow their skills professionally from a technical side as well as from a project management and overall refinement of their professional development and That's what it's about, right? I mean, that's that's why we're that's why we're here to make the you know To add value to those individuals along the way The last piece I'm really proud that you know Purdue is very innovative about this as well but in today's day and age at a tier one university in my opinion you have to have flexibility in how you manage IP if you want to be competitive with other tier one institutions and You know, there's several of us now There's a there's a sort of a core group of 12 that have kind of led this effort Purdue and Iowa State are both in that group that has Created flexible models for industry. So when they come to work with the universities, they understand There's a pathway for them to To eventually get to a commercialization phase and we've really leveraged that pretty pretty intently Okay, so what is digital agriculture? This is the this is going to be the audience interactive part of today How do you define digital agriculture of always faculty we always complain when students don't talk and engage, right? So using data to try to optimize the system, right? Absolutely. What else robotics and automation, right? That's a key element, right is agriculture is not getting more not getting simpler over time, right? Automation plays a key role. What else? Sensing sensing things we couldn't sense before potentially using a high-resolution arrow imagery to detect things We've never seen before in agriculture. These are all parts of it, right? Is this a new field? Is digital agriculture something that's Brand new Absolutely, not. Yeah, I like that answer right in my mind This is where I start and this is plant-based digital agriculture So animal base is a whole different thing But I think you hate to draw lines and things but to me we can start to draw lines around There is a major emphasis for digital agriculture in genetics development phenomics genetics plant breeding There's there's a ton of opportunities there to apply a digital agriculture There is also a ton of opportunities and decision support How do we make better decisions through the course of a growing season and and lead to better overall productivity? And there's a whole bunch of stuff happening in in autonomy now the term digital agriculture has been around since 2014, right? We're bleep. We're redoing plant breeding prior to 2014 I think so, right? So are these new are these new areas? No, absolutely not, right? So what's happened is in In digital agriculture, we have really taken data science and applied Data science innovations and machine learning and and computing power Into these disciplines that have been around for a long time, right? And so when I hear somebody talk about like I want an industry in digital agriculture I say well again, I don't think there actually is a digital agriculturalist I don't think that's necessarily a job, right? I think the career opportunities and the real innovation here is around how do we take the fundamentals of data science the fundamentals of Again computing technology we have today and leverage that into these existing fields to get more out of them, right? Taking data adding value improving decision-making and and driving forward So I do have several examples though in terms of what it means today, right? So digital agriculture is information technology can connecting all aspects of production systems So in an ag one of the great examples is applying Factory automation or lean manufacturing to a lot of our systems It takes multiple steps in a process to produce a crop of corn, right? And there are in lean manufacturing. What do we do right? We measure we just we define waste and we drive it out of The system it's all about efficiencies, right? And that's empowered through data. So if you walk through a factory today There are computer screens next to every robot showing throughput and downtime and uptime and and risks and prognostics and The live data streams that we can now tap into in agriculture Allow us to start to apply this in our areas So, you know simply stated these data streams are making us smarter and more data-driven, okay? Is agriculture always data-driven? You know if you if you think about our growth and progression of agriculture Certainly it is right, but there's there's a there's a underlying production agriculture always underlying sort of a lag function there that Generally speaking takes takes a while for new technologies to be adopted and digital agriculture plays a plays a role in that What else is it? Integrating biological engineering systems through data science I think this is is right in the wheelhouse of university expertise is right Real-time things like real-time crop growth modeling and linking those with hydrology and Nitrogen cycles to be able to provide information on risk for crop production management, right? This is an inherently interdisciplinary field. Where do you find experts that like to work together? Interdisciplinary universities. It's a great place to us to to merge merge strengths Digital ag is creating instantaneous and actionable information. This is these are really key terms and this is In my opinion, this is really driven by the sort of user experience and the use of mobile technologies to get information into people's hands Most of the data much of the data that we use in digital ag is not necessarily new, okay? Bruce, when did you write the textbook on precision ag? 2018 years ago, right? This is a lot of these things about here, right? But but the ability for us to move information get it into decision-makers hands apply analytics to that to give some Directional indication that's new, right? That's technology that hasn't been here forever, right? In fact, when you look at surveys today of growers and say where do you get your information? How do you access information? Mobile technologies is number one the list, right? So if we're not thinking about delivering insight in that way, then we miss that opportunity, okay? Why is the actionable part of this so important? Why is Actionability of information in digital agriculture? Why should that be a major thrust for us? Drives adoption, right? How many those of you that do research, right? Think about think about if your research lab Only got to plan and execute a single experiment a year That's all you had you got one shot a year to run an experiment, okay? Think about what it could do for your lab if all of a sudden maybe you could correct something that was going wrong and get a Second year of information out of that season right of that year. That's what that's what Actionability is and for agriculture We don't have that opportunity, right? A typical, you know, if you look at a 40 year career of a corn, soybean producer in Indiana That means that producer gets 20 chances to grow a corn crop in a field 20 experiments to get it perfect Okay, think about that think about that in terms of how we think about replications experimentation and and driving technology So the Actionability information the ability to effectively sort of create a mulligan, right? Get it get a redo in that season take a step forward allows us to Effectively get maybe 50 years of farming knowledge packed into a 40-year career it allows us to progress more quickly into Into into new technologies management practices or or in some cases simply ward off potential potential real challenges Digital agriculture is accelerating this this process by infusing more data in customers hands. So this is this graph here Which may scare some of you in the room is is From a company called farm business farmers business network, right guess who owns them or who's a major investor? Google okay, so Google is very involved in agriculture at this point for for many reasons. So this is a This is a response of seeding rate for one variety across 320,000 acres within within this within this network. Okay now Purdue Iowa State Illinois we all do these kind of trials in our research We deliver this through extension is Purdue planting 300,000 acres of single varieties to do yield response trials No, so the power of having this much information Right offers a lot of opportunities for where we can grow with this and where this technology can go Are there risks associated with that? So so I'm showing you a lot of data that that that doesn't belong to me But some producer, you know released into this network into the system for for the purposes of this of this growing and sharing And data analytics, so this does bring along a lot of questions about security and privacy and things that are certainly major research questions across to across the university for across the nation for us to to grapple with Reducing the environmental footprint of agriculture. This is one that I think gets overlooked often in terms of the the viability of digital Ag and what it can mean for water quality and environmental stewardship. Okay, is water quality important in Indiana? It's important everywhere, right? This is an issue. This is a major issue that we have to grapple with in agriculture, right? If we applying data to these scenarios applying decision structures, whether it through like Solar erosion modeling or terrain analysis or nutrient pathways The digitalization of agriculture not only to provide the information the knowledge on how to make the decisions with the delivery mechanism to execute Is is key. This is a this this this picture here is a variable rate tillage field So all these highly erodible areas we didn't till, right? So it's it's an automated process now We can we can be able to control and change some of those mechanisms that maybe maybe years ago We're we're more challenging to to cover and Then the last thing here digital ag is automating these complex agricultural systems And there's a number of examples of this UAVs tend to be one of the main examples This is a quick little animation of a of a scouting activity within a production cornfield So to actually go out and manually scout that field would be a couple hour process Okay, with a UAV we can fly it in about 30 minutes at a with a with a Person that's a much lower cost that is probably going to go out and do the initial scouting and Use that for some virtual analysis get information and specialist hands at the frankly at the same level of resolution and detail and Often greater detail than we would get from a from a from a traditional scouting operation In addition to some of these activities, you know machine learning data science, you know holistically is getting embedded into The majority of our of our ag machines purely to make them smarter. Okay in agriculture We really one of the biggest challenges. We have an automation is around dealing with the environment So weather changes crop conditions shifts soil moisture is all that kind of stuff and machine learning is a real key area That's started to open up those opportunities for us and making us ask ourselves for undergraduate students You know, where's the line of sort of the the minimum amount of data science that they really need to understand to be To be a professional engineer, right? It's no longer. I think a field that is unique to itself as much as a complementary aspect a complementary part of all engineering and all all sciences Okay, so to put this in a in a succinct way and again There's there's no wrong answer here because nobody can define digital agriculture, right? so to me what did what digital agriculture means to me is data science in action Right to improve the economic environmental societal and sustainability of agriculture We did a survey at Iowa State about a year ago when we started to get big into digital agriculture to figure out Who's working in digital agriculture in our College of Ag and Life Sciences? Guess what we figured out if you think about the term agriculture and digital that means data and agriculture Pretty soon it figured out that it was much easier to put a list together of who's not using digital anything in their research Right, this is a very very wide spectrum. Okay, so Holistically though, right the integration piece of data science really indicates that we're leveraging The latest advances in technology data mining real-time data access information flows to turn that into the Actionable information that leads to the improvement right the action information that we can take in and in progress for so I want to I want to share a few examples of of Some things my team has done this area and really talk about why we believe now is the right time to to grow these programs And and I think Purdue would agree with this given where you're going with digital ag programs So digital ag is naturally inter-disciplinary, right? That in of itself makes it a phenomenal opportunity for universities to work and we we Universities have a unique advantage to be more inter-disciplinary and tap into more expertise areas than Any of the companies that we might work with this is I think this is in our wheelhouse these complex issues are bringing together particularly mixing Biological environmental physical system modeling and bringing those into decision structures that is I think inherently a University-grounded university-led effort It's going to be empowered and certainly will will leverage cloud computing data science and those innovations The third bullet's really important for us in the room, right in today's market the federal funding opportunities for digital agriculture are pretty broad Okay In my discipline of ag engineering and sort of the machinery automation side Ten years ago if you try to get a federal grant to automate combine technologies You're out of luck those didn't exist Okay, we have an environment today We're both USDA and us and NSF are putting funds into digital agriculture, right? It's been a long time since we've had this much sort of opportunity from these these federal programs to really have an impact Right, so for our young faculty, they're getting started for for even for our established programs This is an excellent opportunity to grow and again the time is right for this A fourth point the employees are expecting it, right? Are you are your employers asking for more data competency in? Your students ours are right so Visual ag is a great way to start to bleed those into into those programs and then in the in the ag market today any any Advance is good. So when you see the the tight margins that exist in agriculture today that has continued to push Companies to deliver solutions and universities deliver solutions knowledge-based solutions that help increase efficiency, right? So on-farm profitability is driven by revenue and cost So in times where prices of quantities are low You know your revenues going to come down you've got to maintain your profitability your sustainability through Through working better with those with those tight margins As far as the direct value back there are I could spend all day just talking about This is this one slide, but I Think we all recognize the the graph on the right. That's not for me Many of us have probably seen that in the room before but Federal funding on basic research is for the last ten years has been flat and in many areas has been declining Okay, has Purdue is Purdue have more faculty than it had ten years ago about the same okay, so This this pool is not getting any larger So we're all competing for that same pool right and so with our growth opportunities This is just makes it tighter and tighter look at the curve for corporate spending on this is on basic R&D Okay, what have they done in the last ten years? More than doubled Okay, now there is a you know pharmaceutical is a large part of this This is an all just agriculture of course, but it provides an indication that if we're really looking at Opportunities to drive programs and we're just looking at at federal funding opportunities to To create those innovations. We're probably missing a big opportunity and You know finding your niche with industry is is really the key there to to establish those those relationships The other thing I'll note is that strong industry partnerships really can be very sustainable very consistent I think this is a a myth about industry funding that it's not reliable if you get a federal grant You're gonna get three years of funding support, right? But you know the next federal grants gonna come Or maybe there'll be a shutdown during that time and there'll be gaps I mean they have you know, I assume Iowa State's on the one that has funding gaps between federal projects So with industry relationships, I think I think those that have been successful in this field have really found that these can be Extremely consistent in supporting long-term research, right? And it's the it's really the relationship there and the quality of the relationship that that drives that the excellent educational programs is also a huge here the I Personally feel the way we work with industry helps to ensure that we do a better job in the classroom We get I think more direct feedback into our curriculum We have a little bit of a chance to see Insights on what's coming down the way in a way that we can adapt and develop curriculum to to prepare students for for where They need to go So in addition to funding what's what are their valuable resources do we need to do research? If you're gonna accelerate a recent fast faculty, what what are the what are the keys need for research program money is always the first one Right, so let's take that off table after that. What do we need? What do we need to be successful people? places where you can test things data data helpful This was as a young faculty this was something that shocked me as I started my program in Iowa State and We started with some federal federal projects to get us started and I found myself on those projects spending 75% of the time and money Developing test stands instrumentation collecting data right spent all this time collecting data And then we have this limited amount of time we actually focus on analytics and innovation and applications that data and one of the things that's really Turned us on to industry partnerships is the ability to really leverage those partnerships for data Right because most industry partners usually already have that information are looking for higher-grade analytical skills And it has shifted our mindset of how we run our operation to be more more data-focused and data-centric and maybe eliminate some of these other Other activities that we used to do So maybe a couple examples that we've been involved with the first one is around cyber physical systems Right cyber physical systems is a new buzzword. We used to call it telematics that came out of mechatronics So the term is not necessarily new the idea, but it's it's a this is an NSF term So it's it's kind of got an appeal to it now cyber-physical systems simply mean that we're gonna connect things together right this is the the Internet of things type approach so in Iowa in 2010 we started to become a focal point of cellulose biofuels the cellulose biofuels means taking these bales of corn stover and Turning those into liquid fuel. Okay, so as a side question The 10 bales on the back of that wagon right there. How far do you think we can drive on those? If we turn those an ethanol and put them in your car, how far can you go? 50 miles There's a price for 10,000 miles From just just the just the bales This the stack that would be would be hundreds thousand miles just just the bales right there 10,000 miles worth of ethanol now that's in a fuel efficient vehicle. That's not in a one ton pickup, right? Let's get that straight, but So we had a huge opportunity I was becoming the focal point of this and this was going this had to go from not being an industry at all To being a fully efficient cost-effective supply chain that competes on cost with the petroleum industry in a matter of about eight years So how do you do that without having a heavy focus on data? Information modeling to drive out some of those scenarios and so ourself and the two Dupont and poet are both the companies in Iowa who have been developing these technologies We we went out this from a perspective of applying cyber physical systems to achieve Optimization and in agriculture. We're lucky. We have a lot of data This is a kind of a standard graphic of what a a typical tractor would look like and tractors have can buses on them Which are effectively the the ethernet of the ag vehicle and everything you want to know about that machine is on the can bus Right, so we started a multi-year investment first and just the the basic acquisition of that data That's the data origination piece get the data into a format you can use and then start to leverage it We we were able to leverage that with Virtual engineering the screwed element type modeling and system modeling to really bring in the weather piece of it Because in eight years when you have Large weather swings in agriculture. We can't just test and wait for a wet year wait for a dry year So we do a lot of virtual engineering to to do that We we also learned a term called infotainment infotainment means I Didn't either until so one of the things that shocked us is from this data No matter how much we trained the supply chain operators who had zero most of them had zero experience here No matter how much we trained them. We weren't seeing change If you're a controls engineer, you should be able to affect change and control the system But when you still have a human in the seat We had to think differently about how we were delivering digital information and technologies to truly drive change One of the simplest things was is relating to how many bales per day of cornstover we produced So a baler should make 400 bales per day. It's pretty simple math There's so many per hour it makes and you got so many hours a day and that should work right and no matter How much you pound in that we were basically usually we were on the order of 30 to 40 percent under what the supply change should be able to achieve Okay, and that really was driven by the fact that folks would go out in the morning They would check the fields. It's not quite ready to go I'm gonna wait another hour and then you know and so what we did is we turned it into a game We tapped into the we used data to tap into the innate sort of Competitiveness of operators. So these dots up here are all the balers running in the supply chain This is a real-time information. So we threw a telematics module on there We ran it through a server right you do some quick analytics and you put stuff on a website This is it's not not rocket science. It's the application of the science and technology to do this, right? So the dots the red versus green if it's moving or not and the number in the middle is How many bales they've made that day, okay? All of a sudden right this is like post-in name and grade on the wall, right? And in the matter of a single I mean literally the week after we released this productivity dropped 35% Okay Purely by getting the right simple information, right? This is where we have to sometimes You know there are the analytics report behind this like a hundred pages, right? Well, that's not nobody's gonna read that so you got to figure out how to how to tap into the human aspect of this through data delivery To affect change, right? Remember the comment I made earlier about having it's not just about knowledge development It's about affecting change the ability that we have to to correct and and reinforce the system The other thing we're able to do is implement prognostics. What's prognostics? Prognostics is predicting failures, right? So we were able to get to the point where we were we could send a service team to a machine before the machine broke Okay, and through that we reduced cost of Maintenance, but also increased productivity substantially from that just through data streams, right? This is the the classic leverage day. It's available set up infrastructure and make decisions from that the Economic impact of this we took 40% directly 40% of the cost out of the supply chain. The best story I'll tell here is There was in this picture. There's about ten different crews or or businesses here small businesses These are farm room businesses and about seven of them were local Iowa businesses three of them had come from across the country that were really technical experts in the area and the one that had the Longest experience in this wanted nothing to do with this. They didn't need it. They had experience We could take a fresh Iowa group of mostly Iowa State college students and a few farmers and within a week they could outpace a Company with 30 years of experience just by using data, right? And that's that's the kind of change that we're talking about in the information flows Other areas sensor fusion we use sensor fusion a lot of projects to a machine learning to help make ag machine smarter so Green yield monitoring grain yield monitoring has been around a long time This is a yield map. So red is red is high yield. Sorry red is a low yield green is high yield This technology has been around for 25 years. So I see an ag leader hat in the back Ag leader was the company that invented precision agriculture and launched the launch the whole industry, right? Yield Mars over design mostly for large grains in the Midwest and they have some real challenges They get out globally and if anybody's worked with yield data, you probably know that it doesn't always look pretty, right? Sometimes it's it can be challenging the yield monitors themselves utilize a sensing technology that that basically resolves the impact force from grain and turns that into a Through a strain gauge turns into something we can measure and then we relate that to mass flow The problem is a crop moisture changes or the coefficient of Restitution of the grain changes or density changes. It doesn't work anymore. We lose calibration pretty fast through telematics data We were able to assert that The typical producer in the Midwest at best calibrated their combine once a year. Okay, so just wasn't happening but yet yield data is The fundamental piece of information we use for every optimization project So if yield data isn't right, this whole thing doesn't work and that's the Midwest you get to Brazil There's no infrastructure to even ground truth against same thing with much of Europe or Ukraine big issues globally So we took we had some innovation around how to solve this through data fusion so this this involves we put a couple a couple sensors in a grain tank and Used other data streams from within the machine as well as use some physical or some kind of a discrete element model analysis of how grain piles and flows into a combine and as part of that We were able to create an innovation of data fusion that every single time we fill the combine with grain We produce on machine automatically a calibration load for the combine So effectively as the as the producer drives that the combine gets calibrated, right? It just keeps self-caliberating and if you think about a regression curve in Statistics we're always worried on the end of the regression curve You're trying to fit a curve through a couple data points the accuracy of a single point, right? Data science central limit theorem says if we just throw a ton of calibration points at it We don't have to be perfect with every calibration point We just have to be perfect on the average of all the calibration points, right? And that's that's the philosophy behind this so this this got released We licensed this to John Deere and they released as part of an active yield Program and this is what it looks like a default calibration from the factory can be huge swings in accuracy The novice this is what a typical producer can achieve at once a year Our active yield system is as good as an expert Calibrating every single field that they that they go into we can't beat that we can't beat an actual ground truth if you do it in every single field, but We also know nobody does that 2% of people do that And even more important here is the fact that this this technology has opened up yield data accuracy in parts of the world that simply don't have the infrastructure to do this and that's the thing that I Think really gets us excited about you know folks in Brazil and in the Ukraine and in parts of Europe that now have have better data It has it has also created some unique opportunities to get feedback so We all get interesting comments sometimes with student evaluations or maybe peer-reviewed journals And one of things about licensing technology is pretty soon people start tweeting about the stuff that you was a PhD project, right? and Of course, I cherry-picked the ones that are really positive But most of most of them are but you know this is for us. This is really fun, right? This is John Deere liking active yield data science on the go, right? That's impact That's a PhD project that now is creating better data for agriculture, right? That's that's the value of commercialization. I think when we really scale this up and and Create huge opportunities and then you know, we've we've been able to you know, of course Christian doesn't know that we were involved and that's okay, right? We have other ways to recognize our innovations through other reward metrics and and things to make sure that we We can evaluate those at the university The last one is the use of supervised machine learning and pattern recognition. So to me Pattern recognition and supervised machine learning is again one of these areas that there's still a lot of research going on But for an ag engineer, this is a tool in the toolbox, right? There's just one more thing we can we can use In agriculture, we have we have we have used ultrasonic sensors and load cells and pressure sensors to every potential opportunity And so vision systems are the next frontier of what of what we're doing in agriculture we've been partnering with Both deer as well as a Carnegie Mellon the national box engineering center Carnegie Mellon in this space for for automation of agriculture In the automotive sector automation means taking the driver out of the loop for us to do that It's easy to steer. We are a student vehicle, right for us is about automating the functions And so we started with with automating combine combine threshing systems and used machine learning and trained images To be able to predict differences in broken content differences in foreign matter And then use that to close the loop on on combine control and operation, right? So automating key aspects of the of the supply chain The other area that's that we've been able to truly leverage this is in the the sugarcane industry Sugarcane is the is the highest producing crop in the world Okay So there's more tonnage of sugarcane produced than any other crop in the world Most of it's in Brazil some of it's in the southern u.s. As well sugarcane has a long history of having Not having a sustainable yield monitor product right not having a really high quality yield monitor solution This is a picture of embersenish cane harvester. It's it's kind of a science experiment on steroids and They go through and they cut the cane it gets chopped up in little billets Then then goes out the the back of the machine We had a chance through some again some some machine learning and stereo vision type technologies to use stereo vision to measure total flow rates of material through the through the cane elevator and And turn this into a licensed commercial product So dear currently sells this as the it's the only commercially available sugarcane yield monitor on the in the world In my career as far as highlights of things that we've done the about two years ago when we were You know, can you know sharing some of this information with folks and particularly in Brazil where the sugarcane industry is Has a lot of work to do logistically But it's actually a fairly data-driven process and the opportunity to really Show them the potential this has for their industries and get the feedback that they're saying was just just huge In addition to yield You know the neat thing about cameras is you can do a lot with them So sometimes you see there's more trash and loose stuff in here than other times so we actually measure trash content as well and then can can close the loop and and And control the you know the cleaning system to a full system automation, right? That's that's that's that's digital act So a couple comments here on on I guess my my view on industry partnerships and and things that we need to do to Establish good industry partnerships, then I'm going to open this up for for questions First of all from a university level I actually break this apart and say there's things the university level we should do things that the Some of the faculty level at the university level We need to be intentional about what we want if we want faculty to work with industry create commercialized partnerships We need to ask for it that line at the bottom is one that is starting to pop up more and more in job descriptions Successful Canada's expected to develop high-impact research program with specific emphasis on technology transfer industry engagement If we if we want it if it's important to us we need to verbalize that we need to we need to communicate That's a goal that we have The P&T process is also extremely extremely important. This is a This is this is my publication history, right? So I started in 2008 at Iowa State I had a very traditional pathway of federal grants and and peer-reviewed journal publications I got tenure in 2012 over here. So most of these are journals We made an intentional shift in that year to say we are going to really start focusing on tech transfer That's that's what we want to do and so at Iowa State We've been very proactive about saying when we look at peer-reviewed publications. It's all of these together There isn't we're not going to do you know journals on one and then and then go over and say here's a little table of tech transfer We do right What's what's the definition of peer-review? Independent assessment of quality of scholarship, right? Many of us have reviewed papers Sometimes we spend a lot of time on it sometimes Maybe not so much Think about the the process and we license technology right think about the the the it's the same thing Right, you've got industry experts who are viewing your technology in detail and then writing a check for it Does that have the same level of scrutiny for scholarship? We think so and so we've been really pragmatic about making sure we say this is a goal for us And we're gonna and we're going to to use that the last bullet's also really important Providing shared services for basic data needs. I have found the particularly in some of the agronomic sciences areas that the collaborators I work with they have Excellent ideas don't know anything about it at all. I mean they really don't have a the functional background, right? So within our team, we've actually started to provide basic data services to their sciences So, you know, we help them set up databases. We help them set up information transfer We help them set up cloud computing right because a little bit of that basic support is able to amplify what they do Right take take innovations and just blow it up from there And I think that's something that you know universities need to be well if you want to grow in this area It's one of those basic sort of shared services. That's necessary to really develop critical partnerships And then at the faculty level there's a there's a similar level of sort of needs first and foremost You've got to build the right team and I think deep industry connections Usually many times will require a little different mix of who you hire, right? It's difficult to maintain Long-term highly engaged industry partnerships that they cover multiple years with just graduate students because we love graduate students They do excellent work And then what happens? They leave they graduate so continuity through programs is is forced us and in fact That's why we have several Research faculty our staff who really lead programs right there. They're really excellent excellent individuals Probably that culture of entrepreneurship right making sure that the folks that work for us understand the same vision and mission of the impact of Collaboration is is key Acknowledging and overcoming ceilings that exist within the university system. So maybe Purdue doesn't have any of these but there are Oftentimes as we as faculty take on more and more opportunities There isn't necessarily the Sometimes it doesn't make work easier right sometimes that that just makes makes more work to get done maybe the same amount of effort and so We think that if you're really going to create these relationships that have such a touch point back to educational programs and and other benefits University making sure that you don't That you protect yourself there and make sure you've got that top-end capability to To be productive and engaged is really key Communication issues I at this point We will not do a project with an industry sponsor unless they commit to a one-hour conference call every other week If it's not important enough to them to put somebody on the project to talk to us and stay in connection Then we don't we won't do it right communication is key in these and that in my opinion Is what translates from a sort of one-off industry project into a really strong? Strategic relationship. That's a that's a key difference So we've got about 10 minutes here. I'm gonna I'm gonna go ahead and wrap up. I'll leave you with this This is a statement I tell our students quite a bit of so the value of the innovation is measured by the Impact of the answer not the complexity of the solution. I think that totally encapsulates digital agriculture there are a lot of low-hanging fruit frankly in Digital agriculture where just a little bit of the right information the right hands can have a huge impact, right? And so sometimes we tend to maybe we maybe the academics in us We we want to take things to the nth degree of analysis or or or technology or bring in a bear And we've always got to come back to what's the real innovation? What's the impact? How are we using this to redirect and And really move the needle within within production agriculture So with that willing to take any questions that you have I'm gonna go ahead and use the microphone in case everybody can hear we're recording. I don't even know so I'm curious about this notion of Having the university and or your college make a commitment to having tech transfer and Commercial and industry relationships be critical. Did you you know when you started talking about, you know, how patents became important in the P&T process? Did you were you engaged in the discussions to have that be like a Like a switch in the and the way you worked on things Or were you just lucky enough to be in a position when the university or the department was thinking about making that kind of a Transfer because I think it's a very important distinction It's a great. That's a very very great question So our faculty handbook has had tech transfer scholarship right in the arts of science is certainly as well It's not a lot of that's not scholarship. It's not done through traditional peer reviewed articles So it's always been there is language our provost actually is pretty progressive in this space our provost John the wicker was co-authored a document from kind of a u-group on how to how to Enhance the visibility of tech transfer So I had that working for us that we had a very senior person then call the university administration saying this is important and I think every university sort of has a culture right? I mean there I mean we all have things that we know were really really exceed at and I would say we believe This is a cultural element of what we do and so I will say not every Department has that same vision But certainly within the College of Agriculture and Life Sciences and through the majority of the College of Engineering It's it's widely accepted now that peer reviewed tech transfer is the same level as journal articles Now I'm careful there say that that doesn't mean a patent disclosure is a peer reviewed article Right a patent disclosure is not peer reviewed But when it gets to a point where it is truly licensed truly patented has gone through a standardized process then it's it's an equal Lots of communication though there. Yes I just want to hear a little bit more about the team that you have How you built it and how many disciplines are involved and yeah So in full disclosure my wife is the vice president of human resources for Iowa State University, so I know more about personnel development and the HR system than any normal faculty should right So how do we build the team first and foremost First of all we I go through a strategic planning every every year We revise our internal strategic plan and make sure that we're heading in the right direction as we go through that We pick in the next three-year cycle. What are the things? We're really going to target and do we have those right people on our team to do that? That has driven us to go outside of our own current bounds right five years ago I had half the size team, but it was all people that were kind of like me right and as we've looked at these strategic areas We've said no we really need somebody in soils right because soil machine interaction is a big deal for us that's something we need and we need a phd Computer engineering person leading our data science machine learning efforts right because that's really important, but you know I can't do that myself so And then you then you take risk right because we have we have a phenomenal amount of Soft-funded staff on one-year contracts So every faculty member has to decide where they want to be in that in that risk portfolio to make sure that you you manage that in our case we have So there's there's like 15 kids in daycare that their parents work for me So there's a responsibility there to make sure those jobs are stable and continual And then you know I think I have to have the philosophy as a leader of that team that I may be steering the ship But you know there are there are leaders underneath that are really driving programs And of those individuals you have to be competitive with the package you provide them And you have to give them opportunities to for upward mobility and the upper mobility piece is one that You know you you can't look at him as being an assistant scientist the university for ten years right those are That may not be the type of person that you want and so you've got to have a plan for how to let those individuals grow and it really feel Feel like we're checking all the boxes for them professionally Yeah, John So you showed a lot of kind of slides with trends and Projections so if I asked you to make a projection what the farm of the future is going to look like and say ten years or Fifteen years whatever you want to pick how would you how would you describe what the impact of digitalized might be in that The farm of the future right so I Believe so I believe a lot in technology. I will tell you I I am not as bullish on full automation and and small-scale Robotics in production ag is maybe some some colleagues are As far as the trend and direction If it if I was starting over today and start a new program Machine learning is huge right and and there's a whole field of time series based machine learning that we could leverage Tremendously an ag that has a lot of scientific questions Still to still to answer in that space. We generally all of our all of our machine intelligence today is still mostly driven by Linear regression models with a very few sets of variables And yet we're generating gigs and gigs of data that we're just thrown away on the machines That's that's a huge part the second thing that I think is shifting agriculture and we'll continue to shift on the on the on the farm technology side is We're quickly going to get to the place. I believe where when that machine leaves the factory that software that's on it is no longer The software that just stays on it right today If you have a machine that in one part of your field doesn't act right every time you drive that field It doesn't act right there every time right because it's the same control system It's same software happening over over with cloud connectivity. We have a chance to start learning between machines We have a chance start adapting machine technologies regionally. Okay, if you sell tractors into 10 countries and Multiple continents you're going to shoot for the middle in terms of performance specs in order to make sure all your customers have a realistic, you know set of Expectations the ability with cloud computing to really automate that and begin to self-tune to environments Which is a totally different way to do software development on machines. I think is going to unlock another tremendous set of tools for us an optimization So ten years, they're still going to be people in in tractors and combines They are but but the level of augmentation through Control systems is going to I think we're on an escalating curve there That's a great question. I think the industries we work with they really they do struggle with the nebulous that Exists within universities and understanding how to get to the right person is a real key I'm sure not where the only ones that I mean sometimes a contact will come in through an industry relationship somewhere and You guys probably don't forward emails this university, do you? But sometimes you can email it's forwarded life forward from a from an industry relations research park out to Colleges out to department chairs out to faculty right and and there's no fall I don't I don't speak to how Purdue does things. There's not always a good fall through there That is I think we have a tremendous not more we could provide if we could connect the dots The other piece is you have to be committed there has to be a mutual benefit you have to be committed to understanding what the company expects right and I think particularly younger I met a lot of younger faculty and I think in Younger faculty today the PhDs that come out. They're not all necessarily set. I'm going to academia. There's a lot more job opportunities for PhDs in industry there were maybe 30 years ago at least in the fields I work in and so they tend I think to have a more entrepreneurial spirit right to them and And that's a culture that I think universities should try to really maximize in the next couple decades Yes, sir. Do you see a role for open source in digital ag? That's a that's a great. I mean, that's a whole that's a whole multiple. That's a I don't know how to answer that in a couple minutes It's tricky right and when you talk about Open source from a like from a pure software perspective I mean talking about like like control and machine software and service ability and those sort of things There there are there are safety implications that come in that are that are real right that you have to you have to manage through On the data side, you know You can sit on either side of the fence and here right so you can you know There's there's reasons that we need to make sure data privacy. We have good standards in for data privacy At the same time we also need data mobility We have to go move data around between places and and I think to me whether it's through a True open source solutions or whether it's through sort of industry guided solutions As long as the farmer is in the seat as long as they're the ones that get to make the decision on Where data goes and who accesses it and it's it's under their control. I'm good with however, that's that's implemented Related to that how much of your intellectual property is licensed exclusively versus non-exclusive? That's an easy question. All of it is exclusively licensed So we have Iowa State has a flexible solutions very similar in parallel to the flexible solutions that Purdue has So I think your flex number three. I think is the is the upfront fee that buys royalty-free exclusive licensing and We use that we're the half of that we use base of half of those counters Iowa State to use that we're 50% of all those contracts So we were heavily invested in that our industry partners have loved that it has taken out any back-end negotiation It allows them to budget the cost or understand the cost structure up front So when they get ready to commercialize they're not trying to figure out how to set the price because they don't know Iowa State's going to charge for intellectual property. So that I think has been I think universities that don't offer that and Purdue does Are going to be behind in the next next 10 years because it's it's an it's too broad an option now at institutions Yes, sir. Are you working at all with the blockchain or? Any plans for incorporating that yeah We certainly certainly aware of it and that's that's about we're not doing we're not actively engaged right now Yep, yeah Hey Matt, so you mentioned about soils being important and in doing You know research managed fields seems as though we still don't have a good solid understanding of what's going on there From a soil lab perspective. We have some machines coming etc Is there anything that you're learning or working on that's going to improve our understanding of the soil? Yeah, that's tough right I get a kick out of a You know a lot of UAV high-resolution stuff we can do with plants like oh, there's a problem with that variety It's like no, it's a soil thing right the it's not a variety thing We we have some we do some R&D in that space Physics you can't fight you can't you can't cheat physics right and physics just says measuring things with with what our understanding of Physics is today measuring things below the soil surface with any level of accuracy even even in the near surface area is Just tough and I wish I was more optimistic about breakthroughs there I think we're going to be I think we make it it may be a The more productive solution is going to be through modeling through better understanding of of weather events and tire traffic and soil dynamics to give us some indications because direct measurements of We've been trying to measure compaction for decades right and we're not we're not getting any closer Just a follow-on to that is one of the things that's happening currently is you know RML defense fund nature Conservancy's soil health initiative, etc They're wanting to create a soil health benchmark starting in the Corn Belt, right? And they want farmers to get compensator monetize for maintaining it But you know tools to do that they want to identify cover crops, so that's machine learning with imagery Yeah, right. They want to identify tillage practice. So again, but Fundamentally our soil tests you send to four different labs you get six different answers Yeah, so agreed. So nothing there though in the soil lab side now. No, it's that's it's soils are tough Okay, thank you so much I know you can answer questions all day, but maybe one final one What do you see as the role of startups in the whole innovation entrepreneurship process? That's a great That's a great question. That's a very good question. So in some ways we compete against startups Right and we could we could spin businesses off and do this we choose to do with inside the university environment I think there's a real role for both, right? I think So in the way industry look how industries investing data research right industry isn't all that industry Still does their own internal research, but they're also leveraging venture capital to sort out the winners and losers, right? So it's kind of a unique environment where you might overpay to buy a company But you for that one company, but you've saved yourself a whole lot of money in rnd by You know trying 100 different things to get to that one company that that survived In the ag sector, I think that's only going to continue to grow I think you're going to continue to see a lot of opportunities for Smart startups in this did particularly in this digital ag space that are going to be highly desired by some of the more majors To add into their their portfolio