AIoT Thesis Blog

A New Hope

2018-03-27T22:40:00+03:00

It is a period of research. A BBA student, striking from a hidden base, has scored their fourth research interview for their thesis.

During the interview, the student asked many questions on AIoT’s current and future state. The information gleaned from the interview will serve as a reality check and help to build a much more interesting technology overview than a mere literature survey.

Pursued by the sinister seminar deadlines, the student races home aboard Länsimetro, now the sole custodian of the interview recording that can save their thesis and restore freedom to the galaxy…..

After the final period of mandatory courses and an internship at Helsinki University I’m at last getting back to work with my thesis. Initially I thought that I could keep working at it at a low burn, something like a hobby, but I ended up seeing days going by and not having the energy to really put in the effort. Thus, all hope was lost. Not with a great crushing blow, but with a slow decay, gnawing away at the foundations until it becomes clear that the whole edifice is untenable, full of stale air with faint smell of rot and past dreams dreamed up too idly.

But not really. I’ve laid some groundwork with weekly and daily planning, using B. Franklin’s daily plannig method with Art Of Manliness’ interpretation as a starting point. As a side note, I’ve also added cold showers to my morning routine and I couldn’t recommend it enough. My writing toolbox has also been amended with an ortholinear keyboard from OLKB and a transcription tool from transcribe.wreally.com that I would recommend for anyone transcribing audio recordings or youtube videos. I think it would be worth your while to give it a try, especially in case you are working with recordings not in english language when, as in my knowledge, speech-to-text applications are not a realistic option (please prove me wrong). Also, there is the point of not willing to invest for a fully “professional product” with pedals and overpriced software. But those keyboards, however, are worth every damn penny and I have some serious difficulties of shutting up about them. So there’s that.

Details and technicalities aside, I am sincerily hoping (see title) that this process will help me to grow so after a few months I could review my work and see it –as it likely will be– flawed, unfinished and not very well thought out in the first place, but still an earnest first try at scientific writing.

There is much to be said and done on the actual work yet, but I will leave you with this quote from B. Franklin’s autobiography, hoping that those who might read this will be intrigued to research the subject themselves. The more astute readers will notice that B. Franklin is actually referring to his system of virtues and his journal he used to track his successes and failures in his attempt to live virtuously, but I believe the subject of daily planning and reviewing is close enough to warrant the use of this quote.

I was surprised to find myself so much fuller of faults than I had imagined; but I had the satisfaction of seeing them diminish. - B. Franklin

Big Tractors Never Tire

2017-10-17T13:12:00+03:00

KoneAgria 2017 exhibition at Tampere

”KoneAgria is an agricultural technology exhibition for agricultural professionals.”

While I’m certainly not an agricultural professional I still found a lot of goodies.

I set out early on Friday 13th, October 2017 with the morning train to Tampere. After a thankfully uneventful two-hour train ride and a minor weirdness event with Tampere’s public transport route planner I got to the exhibition site. It was pretty apparent right from the start that this was an all-out, big machines, real agriculture thing. They sure grow those combine harvesters big these days.

I had set my primary goal to meet with Pekka Isosävi from Valtra with whom I’ve exchanged emails on a prospect of a meeting and an interview. So far with limited success due to summer vacations and understandably more pressing business affairs on his part. He invited me to meet him at the exhibition and I hoped that a face-to-face meeting would nudge things forward a little bit —and he did not disappoint.

My other goal was set on meeting Katja Alhonoja or some of her colleagues from Yara, with similar intent as with Pekka since our correspondence had experienced a bit of a hiatus since last summer. I had not planned any other specific meetings or goals, other than that of the usual fact-finding and respondent-hunting: looking around, going about, seeing what’s what and who’s who.

As I was orienting myself to the layout of the exhibition, pretending to be rather lost and clueless plus superbly feigning an inability to read a simplified exhibition centre map, I came across a stand quite unlike the big machinery exhibits around it. There was an engine cowl of a Valtra tractor and behind it, where a cockpit of a tractor would be, a workstation stand with several screens, electronics enclosures connected by a lot of wires going about on the sides, few actuators at the bottom shelf and a sign ”Maatalouden Internet” (Internet of agriculture) hanging above. This must be the place.

I was greeted by Ari Ronkainen from Luke who kindly walked me through the system at display: a simulated tractor working on a field equipped with not only automatic steering but also with a seeding implement and an automated liquid fertilization system, all combined (please note that I may have mistranslated the terms here). The seeding was automated according to exact instructions and the fertilization was as well automated to precisely match the requirements of the field and the plants. This was all pre-programmed with a planning software that had written work instruction files for controller unit aboard the tractor, with the calculations for fertilization based on farmer’s own experiences and observations, soil samples, scanning and remote sensing like multi- or hyperspectral imaging with drones and/or satellites. All of this meant that the fertilization and the planting of seeds was done simultaneously in one go and the fertilization was precisely mixed from several different liquid components to match the need of the plants, saving fertilisers and avoiding over-fertilization. But this was not just another computer simulation: the system had actually been used just recently in planting of winter wheat (I think it was, but I could be wrong) and what was on display was the electronic systems and the computers used to control the tractor and the implements.

Since I seemed to have appeared to Ari as a reasonably sane person —despite the odd angles in my thesis plan— Ari introduced me to Mikko Laajalahti who was there as a secretary of Agricultural Automation of Finland (AAF). I have to admit that I didn’t recognise his name as I should have since he is —among many other things— the co-author of Alkutuotannon kyberuhat (with Jussi Nikander) that I believe I’ve mentioned in a previous post. Also, that paper is likely to be the only one of it’s kind in outlining the cyberthreats emerging in digitalization of agricultural production in Finland. You should read it.

We talked a while and Ari & Mikko gave me a few tips on what to look into, like Farmworks farm management and planning system, plugfests, EFDI (very much a work in progress), AgGateway (a non-profit consortion in U.S.) and Agricultural Industry Electronics Foundation (AEF). Ari also mentioned that how different systems appear to reflect the cultures and the requirements of their origin environments, as in the case of Farm Management software that has everything one could want from it —if one is in the business of growing corn in the United States. Not so much if the user is growing grass in Finland, as some of the required information and data structures required by farmers growing grass are simply not included in the software architecture in the first place.

— To me, this appears to highlight the need for a farm management and planning system that could respond either to a very wide range of requirements —or to a very specific requirements of a smaller market segment. I could imagine that so far farm management and planning systems have been built to respond to the requirements of, say, middle European farms or their counterparts in the U.S., leaving the farmers with different requirements to their own devices, literally. There have likely been attempts to remedy the situation by smaller operators with development of their own farm management systems. These would be a good fit to the requirements of their own environments, but this approach is likely to create a no small number of systems that can not be integrated or used in conjunction with other systems available. Combined with vendor locks, farmers could easily end up struggling with a host of different systems that can’t “talk” to each other and not seeing the big picture of how their farms are really doing as a whole.

As if meeting Ari and Mikko wouldn’t make me feel extraordinarily lucky, Liisa Pesonen, a senior scientist at Luke and the director of Efficient utilization of crop production data project whom I had previously shortlisted as one of the most relevant researchers on the subjects of precision agriculture, data management and data ownership, was also present at the Maatalouden Internet stand. Not to mention that her reaction to my short introduction of my thesis was so accommodating and positive that I’m still wondering whether my work will warrant such a warm welcome. Well, I better deliver the goods, then. :) We had a chat on data ownership in agriculture and she kindly gave me a tip that I should look into a farm called Toivon tila at Salo. That farm is according to her very active in many development projects, keen to take advantage of digitalization in their work and eager to openly share the knowledge of their experiences. Since their apparent positive attitude towards sharing information and new technologies they would make an excellent interview respondents in my work. I hope that I can get back on Toivon tila in a later post. Liisa also pointed out the day’s seminar presentation “Täsmäviljely ja digitalisaatio - uuden ajan työkalupakki viljelijälle” where she was presenting a commentary on Jussi Knaapi’s main presentation. See more on Jussi’s presentation below.

Since I had plenty of time before the presentations I moved on to Valtra’s exhibition stand where I met with Pekka who was at the moment demonstrating Valtra’s SmartTouch armrest. Although he was quite busy, he made the time to catch up with me and how he could help me along. He described to me how Valtra’s new tractors now come with telemetry capabilities (older ones with CAN buses can be retrofitted) and how they can be connected to Valtra Smart service to provide a remote view of a tractor’s systems. This is done by adding a listening device to tractor’s CAN bus and then sending the data to the cloud service. To give me a better understanding of the system, he introduced me to Olli Honkamäki who could demonstrate the system for me in actual use.

Before moving over with Olli who was busy at the moment, I asked Pekka about the market penetration and adoption of precision agriculture technologies in Finland and yeah, the numbers are still unfortunately low. It appears to me that agriculture is a tough sector to introduce new technologies, but considering what I’ve seen, I think that it would be within reason to anticipate quite a few changes within a few years. But let’s get back to business with Olli and Valtra Smart: with his tablet, he opened a map view of central Finland, where we could see all the tractors and service cars connected to the system. He picked a tractor and we could see where it was and what it was doing: he could see all of the telemetry data generated by the tractor, about 30 different signals in total and cross-reference them as he saw fit. For example, he picked engine revolutions per minute and graphed them together with engine oil pressure —with just a couple of taps on his tablet. He told me that the service was developed over the course of 10 months with Cinia Solutions and has had a great reception: almost without exception, tractor owners that he had demonstrated the system to would liked to have it installed and working right away. I could easily see why: service crew could see near real time what was happening with the tractor and could remotely diagnose and troubleshoot it, give instructions to the user and if necessary, pick up the correct parts from their warehouse even before leaving. This would mean substantial savings in down time. Likely the investment in subscribing to the system could be returned in just one service call when compared to old way of calling service to the field to perform diagnostics, then having them to drive to warehouse to pick up the parts needed and finally back to the tractor to install the parts. This kind of service has previously been a domain of the like of large truck fleets, but it is appears to me that this type of service is now being popularised in agricultural machinery service. In addition of being very useful to the service crews, the owners of the tractors could see in the application the situation on all of their machines right away with their smartphones, tablets or computers, near real time. Also, the data generated could be used in AgriSmart cloud service to further enable precision farming.

I have to admit, I was quite impressed. I asked if it would be possible to interview Olli on this and he was positive that it could be arranged —I hope that I can write a follow-up post on this in near future.

I made my way to the seminar where Jussi Knaapi, a freelancer journalist for Koneviesti gave a presentation “Millä tasolla täsmäviljelyn kärkiyrittäjät liikkuvat niin Euroopassa kuin Suomessakin?” (rough translation: On what level are the leading European and Finnish precision agriculture farms currently operating?). Jussi has written a lot on precision agriculture over the years and has an extensive hands-on experience on the subject, so I’ll see if I could have a chat with him on some future occasion. He mentioned that the presentation materials he used, at least the parts from Van Den Borne Aardappelen should be freely available for download and I found them —or at least something very much like it— here. As I can’t reproduce his presentation in it’s entirety in this post, I make do with some of his points that appeared to me as the most interesting ones —and I may have missed or omitted something important. Please let me know in the comments!

For this presentation he used the material from Van Den Borne Aardappelen as he had visited their farm and it appears to be a great example on how precision agriculture/smart farming can greatly enhance agricultural production. Contrary to usually held belief that in order for precision agriculture to have any effect the field sizes need to be enormous, it has now been proven that precision agriculture can produce great effects on very small and irregularly shaped fields. Jussi highlighted the importance of noting the location, time of the observation and the accomplished actions, which enables the continuous observation, leading to analysis and further actions in accordance to long term goals.

Jussi mentioned some technologies that enable the farmers to get deeper in to the field, like soil scanning with EM38-MK2 or Veris scanners that can produce a soil composition maps (EC potential maps) and portable gamma/X-ray spectrometry devices that are becoming available (read: cheaper) and can produce a great deal of information on soil contents like the amount of specific nutrients. While EC potential map of a field may sound like a bit academic at first it becomes a lot more interesting when compared to a harvest yield map, which usually strongly correlates. This can also be combined with a fuel consumption map that can further highlight areas of interest, especially if the ground is level. Other maps can be obtained with sensors like Yara’s N-sensor, drones and satellites. Composited maps can then be used as a base intelligence for precision agriculture actions like fertilization or soil treatments. Weather stations with internet connectivity can also give a much more detailed view to each field’s growing conditions. They have as well become quite affordable and according to Jussi, units can be purchased for about 100 € each. To me it feels like a small investment for the information they can provide, but I would like to interview a farmer who has implemented this kind of solution on their experiences.

One scenario of precision agriculture (or smart farming?) that Jussi described was that a farmer could take an EC map of his field as a base and making a drone image run in the morning. After getting an up-to-date intelligence on plant conditions on a map, using the combined maps to create a work instruction file for precision sprayer it can be used to spray the exact amount of fertilizers needed in the afternoon. While this sounds awesome, I’m not sure if it will be feasible for a couple of years —see a previous post on drone imagery where I wrote ”everything is hard and nothing is easy”. I hope I could get more info on this and perhaps I could have a chat with Jussi on this.

I’m sorry to write this, but writing this post has taken me longer than I expected and in order to publish this post in a reasonably timely manner, I have left Liisa’s commentary presentation “Mitä olemme oppineet? - Täsmäviljelyä 22 vuotta” and Jussi Pylväs’ presentation “AgriSmart-toiminnanhallinta koneyrittäjän tehotyökaluna sekä Säilörehun korjuun satokartoitus ja rehun reaaliaikainen seuranta”out of the scope of this post. Both presentations would merit a write-up and I hope I can get back on their work in both my thesis and this blog.

It was well past two in the afternoon at this point and after a break I thought I should see if I would be extraordinarily lucky and meet Katja Alhonoja at Yara’s stand. Longest of the long shots, but worth taking. I looked around but could not see her and after a while asked if anyone around would know her and whether she would be around today. Turned out that Katja had done a morning shift that day and had already left but Anne Kerminen knew her and agreed to take my note and send it to Katja. Many thanks, Anne. I’ll be writing to Katja and we’ll see what comes of this. Also, I think that I’ve used the word “hope” way too many times in this post already.

As I was winding down of the day’s operation and just strolling around, I noticed Viljelijän Berner’s stand that had Soil Scout on display. The same Soil Scout that I had read on and was first developed in Johannes Tiusanen’s dissertation, then called Peltotiedustelija. As I had somehow lost the track what happened to Peltotiedustelija I was very exited to find it on the market. I met with Teemu Helkala from Viljelijän Berner who gave me a short intro to Soil Scout’s application and hardware. I noted that the data generated by this system is downloadable, likely in a common spreadsheet format so users can really take ownership of their data. I immediately assigned extra points to Soil Scout for that feature. Teemu described that the use cases that could benefit most from Soil Scout’s capabilities were golf courses and speciality plant production where the precise control of irrigation systems could bring greatest benefits to users. He also encouraged to just call Tiusanen if I needed any more information. That will definitely make my to-do list.

Edit 18.10.2017 12:00: “tool” refactored to “implement”; minor styling; refactored Toivon tila’s description to match correction from L. Pesonen; added link to Toivon tila’s website.

Back to the Otaniemi

2017-08-03T23:42:00+03:00

Note slash disclaimer: this blog post has been written on the basis of author’s memories and notes of a conversation with J. Nikander at Otaniemi, 3rd August 2017, written on same day. All errors, misunderstandings and whatnot are author’s own. Please notify me on any errors in the comments or email me tatu-dot-polvinen(at)gmail-dot-com. Thanks!

As I browsed Luke’s Ajankohtaista news on 16th July I noticed an article published 9th June titled “Kyberturvallisuus on elintärkeä myös maataloudessa” by Mikko Laajalahti (luke.fi) and Jussi Nikander (luke.fi).

As I’ve written about IoT cybersecurity –or rather, the lack of it– in a couple of assignments before I read the article and the report Alkutuotannon kyberuhat linked to it, with interest. After going through the report and taking some initial notes I decided to cite it in my thesis and set it aside for further study. I meant to write this post on and about my process of reviewing my sources and notes so far, but things happened: I’m going to attend Worldcon 75 in Helsinki next week because, well, it’s basically next door to where I live. Also, WORLDCON –in Finland! In my lifetime!

But why am I writing about that? After attending Ropecon I thought I’d better do some planning ahead and go through the Worldcon programme, making notes on anything interesting I came across. Yeah, as usual, you guessed it, I marked about a third of everything in there BUT there’s a presentation titled “How Will We Grow Food in the Future?” by the same Jussi Nikander (worldcon.fi). That’s why.

After a bit of apprehension / hesitation I sent him an email last night, telling about my thesis and how I thought it would be useful to have his interview. I almost spilled my coffee when he replied this morning and told that we could meet right up today. I swear, it usually takes weeks, several emails and phone calls all over the place to get people to respond, let alone agree to meet. Not naming names here, but sheesh. But after all, I totally understand that for a busy CEO I’m basically just leeching valuable time and not adding much value to their firm. So there’s that –and no hard feelings. :-)

But this time, I got an opening and what the hell, straight to the deep end: take a bus to Otaniemi, go to talk with a Senior Scientist and a Doctor of Science. Me, a BIT student, writing about a field I haven’t got the foggiest –or had, before this process– about. And proposing a qualitative interview, no less.

Oh well. Here, hold my beer.

Considering the onset, our chat went very well. Heck, he came to the door wearing a Ropecon XX shirt and got me at ease right off the bat. While it was apparent that I could bring very little if nothing to the table that could be useful to him, he didn’t mind giving me a quick sitrep. We talked a bit about how on Gartner’s Hype Cycle for Emerging Technologies things like IoT and Big Data are already descending to the Trough of Disillusionment when people are beginning to realize that with IoT, like with everything else, everything is hard and nothing is easy, especially how hard it really is to produce a viable, useful and robust product with IoT technologies. Jussi also described how him being a specialist on digitalization among many other things (just see his page at research.luke.fi) brings in a lot of requests for assistance to projects that have pretty much nothing to do with digitalization itself. His take on the definition of the word is that in increasing amount things, tools and devices become capable of producing and consuming data, making them more –at least theoretically– useful as the data is analyzed and used to form knowledge. In turn, this knowledge can be automatically used to improve the user’s efficiency and/or experience. All the while reducing the need for user’s intervention / input and effort. And please don’t come telling me just in how many ways this could all go to hell in a handbasket. I already have an idea and I’m still somewhat optimistic. Somehow.

About that user input, Jussi told how fragmented the different systems currently are and how difficult it can be to pull together any kind of general idea how a farm is doing. By the data, that is. Since every system is different in almost every way, end users need to painstakingly analyze each one’s output and form a general view in their minds. Or in manually generated excel sheets, which makes me shudder. Jussi pointed out that farmers simply do not want a single system more. All of these different systems that a farm –or a greenhouse– uses to operate should be integrated (or interfaced?) to a general purpose one, effectively making the individual systems invisible to the end user. However, this kind of integration is really hard. I believe that there’s so many wildly different systems on the market that the task is borderline impossible in theory and totally impossible to pull off profitably. It really seems to me that all the major manufacturers are living in their own ecosystems, some intentionally blocking others, some by just the virtue of the way things are. At least, Agricultural Industry Electronics Foundation (AEF) was founded in 2008 to bring manufacturers together in attempt to create robust, real standards needed for integration. Go see aef-online.org. While integration of existing systems and machines is hard, integrating future ones may become easier with new standard AEF extended data interface to be finished in 2018. I will look in to this a bit more later, but for now, see AEF’s pages.

Jussi also noted that experience shows that general purpose farm management and measuring system should be privately produced instead of made by public institution. This is due to public institutions being very rigid and inflexible. This could way too easily lead to situations that when a key person is absent, support for the system could be compromised. Since the institutions and the personnel in them have no hard incentive to keep the users happy by providing guaranteed continuous service, especially if it would entail personal sacrifices from employees, service will likely fail at some point –leading to losses at farms. And those losses could well be substantial.

Jussi elaborated on the situation a bit: while the ISOBUS standard is a good thing to have and a lot better than nothing, it’s not a real de facto standard. Considering that the tasklist XML parsers in different equipment manufacturers’ machines work differently, user can not be 100 % sure that the machine will work the way user instructs it, if they mix machines from different brands. This is particularly bad if said machine is used to gather measurements from the field and there’s no way for user to know whether the data has been parsed into something different, ending up in costly mistakes in fertilization or irrigation later on when used in farm management / planning system. In addition to not being a real standard, ISOBUS was developed for far less data-intensive operations than the current farming equipment require. In example, Jussi noted that it’s no longer possible for tractor operator to see what goes under the wheels due to larger tractor frames, so a real time video feeds are needed and ISOBUS CAN doesn’t have the bandwidth for that. Not to mention that the basic server-client architecture does not allow both ends to send and receive data in order to automatically actuate automation in both ends. Now it’s just that a server sends commands to the client and that’s that. So 1996. I hope that I can learn to write that idea in a bit more clear manner sometime. :-D

Anyhow, consider that a tractor can have a N-sensor equipment on top of it with multispectral cameras, automatic steering system with GPS, different camera sensors on it’s front side, automatic weeding machine in tow (like in this commercial youtube video) and probably a drone flying above the field as well, all producing no small amount of data every second. To facilitate the data transfer the current can bus would need to be augmented with additional ethernet cabling and no one likes to do that kind of fragile modifications on farm equipment.

In closing, Jussi said that if there would be a single thing that he would like to raise on the subject of AIoT that is the system integration, as a challenge for the whole agricultural field and as the key enabler of functional AIoT. In my mind, pretty much everything else appears to be already available, but I may prove myself wrong in short notice.

But then again, I like to do that.

Finally, many big THANKS to Jussi for being awesome and giving me a free, private crash course on all things AIoT.

Now if you haven’t done it already, go buy your Worldcon membership and come see “How Will We Grow Food in the Future?”. Here’s the link again -> Jussi Nikander (worldcon.fi).

A Field Day

2017-07-28T14:50:00+03:00

A bit of a disclaimer may be in order: ever since I read about Robbes and Fujitsu’s project I’ve been a bit of a fan. Now, having met him along the team of Littlegarden, I’ve become a huge fan. Please let me know if gets too bad.

I had previously met with Kirmo and Jaakko from Little Garden Oy, first at circular economy event at Teurastamo and later at urban farming centre Kääntöpöytä where they have set up their office. I’m not sure why but I appear to have neglected to write about them before this post. Really, I haven’t? I should have, but I try to make amends today:

In the first place I heard about them through Helsieni and as both Helsieni and Littlegarden were at Teurastamo at the time I had a chance to chat with them. I was rather impressed with their project of building their own modular growth houses, very much like container-based plant factories but bigger. But when I heard that Robert Jordas was the third member of their founding team I knew I had to cover their case in my thesis. In case You’re not familiar with Robert’s work in greenhouse production business, well, you should. ´:D´ Go see his company’s website and this news article at hortidaily.com on their project with Fujitsu to build Finland’s first commercial plant factory. You can also google “robbes fujitsu” for good measure. The combination of Robbe’s long experience of greenhouse plant production –with continuous growth, investments and profits over many years– and technology from Fujitsu appears to me to be the most credible business case of plant factory production around.

Of course, please let me know if you know of other companies in the process of building plant factories. Preferably in southern Finland, but all info is much appreciated.

But back to Littlegarden: on it’s turn it presents an opportunity to prototype and build a growth space / plant factory system of it’s own without being vendor locked and aiming for modular, open sourced solution. I believe this kind of system would be very much sought after by pretty much anyone in greenhouse business that would like to keep up with the competition. I’ve tried to keep a lookout for a project like Luke’s Crop, Livestock and Forests Integrated System for Intelligent Automation, Processing and Control in Arable and Livestock Farms, and Forests but for horticultural production. But so far, no such luck. I’ve begun to toy with the idea to try and see if Luke would be interested in such a project since they already have the plant factory research project going on at Pyhäsalmi mine. Also, I would like to see if University of Helsinki would like to share the technology setup and experiences that they use in their growing rooms at Viikki campus.

Jaakko and Kirmo had set up a meeting with Robert to discuss the current situation of the project and to install new sensors on the pilot plant factory’s production line. They also invited me over to meet with Robert so we could see how could I help them with my research work. We set out from Kääntöpöytä in the morning with Kirmo’s biogas-powered Caravelle and headed to Lapinjärvi, stopping for a quick breakfast along the way.

At Lapinjärvi, we discussed a bit about my thesis, what I could do for them and then moved on to the project plans and blueprints on building the Pikkupuutarha’s pilot plant factory. Later we moved in to the current plant factory and I got to help Jaakko and Kirmo on installation of new sensors that will allow for more precise monitoring and recording of the production line’s conditions. I could also ask Robert more questions on the practicalities of the growing space and the limitations of their current setup. He also told me a few things on how he would like to build a new system, what would be the first improvements compared to the curent one and what it would ideally be capable of.

Going through one of *Robbe’s Little Garden Ltd*'s greenhouses to the pilot plant factory, installing new sensors with Kirmo and plant production line in action. Photos by author.

I also got to see the foundation and a bit of blueprints of the coming plant factory. Or “factories” as they are building one with cooperation with Fujitsu, Robbe’s Little Garden Ltd is building their own and the Littlegarden modular one is the third one under construction. Rather sizeable ones as well.

Jaakko checking the foundations and another picture taken a bit farther away. I couldn't get all of the foundatins in one picture. Photos by author.

When we got everything sorted and verified the new sensors were working we had a quick wrap-up in the office and headed home. I have a lot more to look into and hopefully can help both Robbe’s and Pikkupuutarha’s project to succeed and share their learning experiences in my work.

Finally, a big THANK YOU to Robert, Jaakko and Kirmo for having me. I really hope I can interview you in my thesis and that I can find a way to pay you all back somehow.

What Was Missing

2017-07-24T20:50:00+03:00

I’ve searched a couple of organizations’ sites with keywords that I feel that should return results relevant to my work, with mixed results. I have initially used each site’s own search engine since they have appeared to be well organized and established organizations, with history of scientific publication and large databases of research reports. I may consider searching the same sites with google and google scholar later.

I searched SYKE with keyword “data” and found a reference to Dataa tarjolla blog post from article Sateliittidatan jakelupalvelu FinHub avattiin that in turn pointed to Avointieto and Metatietopalvelu. I should probably return to Avointieto and Metatietopalvelu a bit later.

Next I tried with keyword “sensor” that returned Maatalouden kuormituksen vähentäminen that provided some insight how agriculture can cause stress to enviroment and local ecosystem, but not much relevant to my work. Keyword “internet” yielded a reference to KIVIKI project that focuses on how circular economy could boost competitiveness and growth to green economy. Again, not my research focus but since circular and green economies are growing trends and often connected to IoT technologies I thought it would be worth a mention. If you’re uncertain what the term “green economy” means and what practical implicatons it may have, this article helps: Vihreä talous - käsitteen jäsennys ja seuraukset eri toimijoille. The green economy’s implementation was researched more throughly in Syke’s project Biotalous ja vihreä tekniikka vihreän kiertotalouden perustoina (PEER GE) that was a part of PEER (Partnership for European Environmental Research)’s project Implementing the Green Economy. The report should provide an interesting read and it’s available as PDF here.

PEER was a new addition to my list of organizations to check out. Unfortunately for me, apparently their research focus is not in IoT technologies. At least none of their Flagsip projects appeared to relate to IoT tech and the site’s search engine didn’t return anything on search terms like “big data” or “internet of things”, and very little with “sensors”. At least I found a past event on precision agriculture, that may provide some leads: Precision agriculture: an opportunity for EU farmers. Well, one can hope.

I’ve wanted to re-check MTK’s site for a while now, just in case if there would be news or events that I haven’t noticed before. While nothing new in particular caught my eye, I browsed to the “agriculture” and “horticultural production” -category, where I came across a link to Puutarhaliitto’s pages. I decided to take stock of Puutarhaliitto since I had moment. Their publications were mostly books not relevant to my research –Can you see a pattern in here? Should I just give up and run home to google scholar?– but I picked the name of their greenhouse expert, engineer Jukka Tuominen and just googled ‘Jukka Tuominen puutarhaliitto’. I added that “puutarhaliitto” since his name appears to be so common that the search results with only his name were all over the place. I found a short article Energiatehokkuutta kasvihuonetuotantoon that described the use case for a container-size plant factory and some of it’s challenges. While nothing new to me, the article referred to two companies that I’m going to look into a bit closer:

Netled has developed water-cooled led lighting system in their vertical farming concept that can be combined to heat recovery system. I hope to get a closer look at one of those when I’ll get to visit Robbes Lilla Tädgård at Lapinjärvi.
Exsilio is developing a container based plant factory solution EkoFARMER in Salo. The project’s own pages are at ekofarmer.fi.

From Puutarhaliitto’s page I browsed to Kauppapuutarhaliitto’s pages where I found the basics of greenhouse plant production in Finland neatly wrapped up: Kasvihuonetuotanto Suomessa. Check it if you need an overview.

Next up on my list was LUKE but before I cranked up the site’s search engine, I checked –just a peek, no harm in looking, just a little bit… you know that I’m about to go down the rabbit hole and not coming back to my neatly planned and organized search. For a day or two, at least. But I checked what I could find in Research Focus: Innovative Food System. In the See also part of the page there was a link to Mushrooms and berries grown in forests of the future that while not in my reseach focus, was an interesting read. Perhaps some extraordinarily expensive mushrooms and berries could be cultivated in plant factories too?

“Such special mushrooms include chaga mushrooms and lingzhi mushrooms. Chaga mushroom tea, for example, may costs up to 960 euros per kilogramme.”

Well, that price point just made me sit up and take notice.

From Luke’s Current news section I also found a news article Drone project prepares ground for new business with Tekes funding that referred me to the DroneKnowledge project that has received Challenge Finland funding from Tekes, the Finnish Funding Agency for Innovation. Also, I learned that VTT Technical Research Centre of Finland is “developing a spectral camera which would be closer to the prices of consumer products” according to VTT’s researcher Heikki Saari. I think there’s a fair chance that I will use this article and look further in to DroneKnowledge project.

But I still got a couple of things from Luke I haven’t shared with you. Well, more than just a couple. Actually, quite a lot more. A whole bunch, really. Sheesh, they’ve been busy!

There’s this nice report about LED-lightning in greenhouse use: LED-valaisimet kasvihuoneviljelyssä. While not specifically and in itself an IoT tech, a nice report on LEDs themselves. I have also found a couple of interesting companies in LED growlight business like Netled I already mentioned and Valoya that makes wide spectrum LED grow lights. Browsing through Valoya’s pages I noticed a section “Valoya is a member of:” that looked interesting. There was International Society for Horticultural Science that we already know, but also some awesome new organizations –for me– like UK Controlled Environment Users’ Group and Association for Vertical Farming that shared a nice article from Hortidaily: Does vertical farming make sense? on their FB page. Here’s some thoughts from Martin Veenstra of Certhon.

“…Veenstra stressed that vertical farming creates a lot of possibilities. But what about the feasibility? With the right technology and knowledge, basically anything is possible. Growing bell peppers without daylight. Possible. Growing lettuce in a shipping container. Possible. The options are limitless. But does it make sense? Does it make any sense (and more important: money) when your competitor is growing them next door in a traditional greenhouse with daylight at 50% of the cost? “The purpose of the type of cultivation should be determined. When vertical farming allows you to grow a fresher, local product that you can market with a premium it makes sense, go for it. The goal is then to grow local and a vertical farm allows you to achieve this goal. Growing in a vertical farm simply because it is vertical may never be the goal.”

That appears to be yet another company for me to check out. No rest for the whatevs. :D But the article points that there’s another important point to be made: vertical and urban farming may be cool and trendy AF, but it’s business and not everyone makes it. Like in the case of FarmedHere.

Research Search

2017-07-22T02:00:00+03:00

A recent news article in Helsingin sanomat piqued my interest: in the deepest mine in Europe at Pyhäjärvi, Luonnonvarakeskus (Luke) is running a research project on how plants like potatoes and nettle could be grown deep underground. In this case, almost 700 meters underground. Here’s the article in Finnish: Euroopan syvin kaivos… and the project’s page on Luke’s website: Älykästä tutkimusta ja koetuotantoa. I’ve got in touch with Luke and sent initial emails to researchers working on this project, but as most of the people are on their summer vacations I’ve yet to hear back from them. I’ll get back on this a bit later in my future posts.

In the meanwhile, I’ve been looking for more research papers, researchers and articles to rebuild a better version of my literature survey. After looking into –just a glance, really– professor Salah Sukkarieh’s work on University of Sydney’s pages I got a hunch that it might be a good idea to further look into what I could find by searching for organizations first and then browsing through their publications. So, instead of the usual google + google scholar + keywords combination I’ve looked into different organizations like universities and agricultural unions, people working in those organizations, their publications and events.

But before going into that, I’d like to update you on how and what I’ve found recently: Via Sukkarieh’s articles I found Journal of Field Robotics and International Society for Horticultural Science, “The world’s leading independent organization of horticultural scientists”. ISHS has many publications, especially Acta Horticulturae where I’ve found several articles I would like to read, but unfortunately it’s not available in my school library and I’m not sure if paying for membership would be profitable enough. Same with the Journal of Field Robotics. So far I’ve marked the articles I’ve found most interesting and will return to them if needed. If so, I’ll see if I can get university library to get an access. It appears that the first issue of the current year of the Journal of Field Robotics is freely available here, but I didn’t find any articles there that would be relevant to my work.

While browsing some promising articles, I came across Field and Service Robotics, Results of the 9th International Conference which again appeared very promising, but you guessed it, I can’t get access to since neither my school Haaga-Helia, University of Helsinki nor Aalto university can be found in the organizations list of the Shibboleth / Open Athens login page, at least for this journal.

Also, Elsevier’s journal Computers and Electronics in Agriculture that I found via Acta Horticulturae is just as well behind a paywall. I hope I can find a way in, but I’m feeling a bit sads. I’m going home.

Which means, to Finnish universities like University of Helsinki and Aalto university. This was encouraged by finding Timo Oksanen’s Robottikylvön ja -kylvömuokkauksen mahdollisuudet ja haasteet Suomessa (PDF) via The Scientific Agricultural Society of Finland’s bi annual conference Maataloustieteen päivät, which was a find on it’s own. But since I forgot which way I found it, I’ve also started to keep track of which way I came across it in my notebook (pen and paper –shocking, I know), in addition of keeping track of what I find and where in my Evernote. Finally, sources making it to my work get added to my Zotero database. When I feel like I’ve got something to “write home about” I write about it in this blog to clear the image a little bit. Then, I’ll get around to actual writing. I think-feel this system could be improved upon a lot but right now it serves me well enough. Please comment or email me if you have any advice!

Messy first edition way of keeping track of my browsing. Photo by author.

Another case in point, I’ve got no clue how I came across Tyynelän tila, but their pages, both web and FB combined with Etelä-Saimaa’s article (PDF) and Kauppalehti’s article (PDF) convinced me that I’d better contact Juuso Joona, master of the farm and see if I could visit.

Anyhoo, now we’re getting to that part about browsing organizations’ pages: Inspired by my finding SMTS and Maataloustieteen päivät, I’ve branched out from universities to see what else I could find. Looking into other Finnish agricultural organizations I’ve come across Pro Agria, Luke and Finnish Environment Institute (Syke) with some success: while I haven’t found much from Pro Agria and Syke, going through Luke’s archive of news, publications and projects has yielded many notes, which equals stuff to read. Lots of stuff to read. The reason I haven’t found much from Syke’s site is just that there’s so much to go through that I’ll need to do some search engine voodoo, and I haven’t got around doing that just yet.

I actually found Luke’s OPAL-Life project twice: first I was googling MTT’s Ilmasopu project, leading to Luke’s news article Ilmastonmuutos tuo uusia viljelykasveja. As a side note, MTT has been a part of Luke since 2015, I found out. Since the link to MTT’s research programs was broken I hopefully went to Luke’s Research Focus page, chose Boreal Green Bioeconomy programme, where among Key Projects was a link to OPAL-Life.

The second time around I was taking a break from writing this post, scrolling my FB feed. A friend had shared YLE’s news article referring to two of Luke’s projects, ScenoProt and FutureCrops(in Finnish). Interesting! On that page there’s a section Projekti somessa that has tweets about the project and there was also a tweet about drones published just this morning, here. I clicked on the link that took me to Luke’s news article page (in Finnish) that mentions OPAL-Life project, which I had forgotten in just six days. I really need to review my Evernote and other notes before going much further…but there’s that cool project Tree seedling production under changing environment that gave me an idea to google ‘tree seedling plant factory’ and finding that it’s already a business, at least in greenhouses, like in Itasca Greenhouse Inc. in Minnesota. Hey I’d better read more about this…and I found more about it at Luke’s article Forest tree seedlings…. That adds one more researcher to my to-contact-list. :)

But back to Luke’s projects, theres also that Digitalization in Bioeconomy -roadmap which actually can be found at VTT’s site (PDF)… that I just happened to find through Luke’s article Digitaalinen ruokaketju…. Browsing the list of projects, I finally found an article about Cyberthreaths on Primary Production (in Finnish), full paper available here (PDF) (in Finnish as well). How awesome is that? Very awesome, I’d say.

Then there’s a couple of articles that I may cite in introduction that are about profitability of agricultural businesses and climate change. But more useful for my work may be the one about how Urban Greenhouses Could Cut Down on Product’s Carbon Footprint. That article also mentions Voimakas project that aims to improve the profitability of Finnish horticultural production.

Another take on plant production that I didn’t consider at all before I saw a presentation on it at North Food event is industrial microalgae production. Snooping around at Voimakas pages, I found an article about a report on Utilization of microalgae in industrial symbiosis, focus on Finland (full report PDF available here). While not edible for humans or anything with just a single stomach, this microalgae could be used –among other things– as a cattle feed, drastically reducing the need for arable feed farming (I hope I got that term right). While the article on Voimakas’s site describes microalgae production as challenging in larger scale, I think it’s still worth looking into. Also available in Finnish: Levien teollinen ja laajamittainen viljely Suomessa on haastavaa.

Also a post about L. Särkkä’s thoughts on her visit to Dutch agricultural university, which I guess means Wageningen University & Research. She writes about a research project that studies plant production in space that will involve a 9-month simulation period in Antarctica, where people are attempting to full Martian, so to speak. “I’m going to science the shit out of this.” Särkkä speculates that in future, ever more exotic plants could be grown in Finnish greenhouses. The business case for this could be that in northern climate it’s not needed to use as much pesticides and that in turn could be used as an advantage. Now this is getting interesting…she mentions plants like vanilla, papaya, mango, hops and moringa. I’ll see if I could find some of her research.

I’ve found so many interesting articles via Maataloustieteen päivät that I will probably write a separate post on my findings from there. This one is getting a bit long already. Also, I have to go through T. Oksanen’s and S. Sukkarieh’s works in a similar way. Perhaps I could actually contact Oksanen and ask for a chat? I’m a bit apprehensive but I think that I’ll dare myself.

As I’m likely have found out by now the main organizational players on this field and familiarized myself with their pages, I’m moving on to Doria and Theseus search engines / databases to find out relevant theses. I’ve got some promising notes already, but I guess the results are better served with separate blog post. Also, I’m writing this at three AM.

Good night,

PS: I noticed that I had mentioned Elsevier Open Access on a note about Computers and Electronics in Agriculture journal. Turns out several of the said journals articles are freely available here. I just needed to click on Open Access articles to bring them up. Writing notes along and about your surfing-browsing is useful. No more sads. :D

Drone Talk

2017-07-11T22:00:00+03:00

At the second day of North Food event I attended to Juho Kotala’s presentation of his ongoing research on aerial drone photography in crop monitoring. Although he’s very busy with his research he kindly agreed to meet me and show his work in a bit more detail than he could do in his presentation.

Juho's presentation & Koetilantie 5 sign. The drone in the middle is the one he uses in his research, another one is just a toy. Note the camera rig he constructed himself. Photos by author.

We met at Helsinki University’s Department of Agricultural Sciences at Viikki campus. While my knowledge of argotech is rather limited -which is kind of the point of doing this in the first place- he guided me through some key aspects of his work and I left with a much more detailed image of what can and can not be done with current technology.

I had a rather enthusiastic, starry-eyed even, vision of how multispectral drone photography could pretty much lead a revolution in crop production and everyone could enjoy massive amounts of cheap, accurate, up-to-date data of their crops. After Juho had run me through the real deal of the work, the data, the implications and the conclusion, well, damn. So much that I’m still debating should the title of this post be “Everything Is Hard And Nothing Is Easy”, but I’m uncertain of the capitalization.

While a drone can be programmed with a commercially available software (and there are MANY) to fly a specific flight path and take pre-programmed photos, one needs practice with the software, camera(s) and the practicalities of drone flight to name a few. Unfortunately neither can one simply walk into Mordor nor shoot cool pics with a drone to get actionable data. In order to have the drone and the camera to collect the images in a repeatable and reasonably reliable way all equipment needs to be carefully calibrated and the flights need to be conducted preferably in the same time of day, with as similar weather conditions as possible. If you’ve visited the great outdoors in any capacity in your life you probably see the problem here. Any deviation may yield very different measurements and that in turn need to be taken into account when reviewing the results. Those deviations and inconsistencies in the data bring in the need of an experienced reviewer and that particular kind of experience may be hard to learn even on one particular field, let alone generally. Especially in a place like Finland, where soil differences can be so big that virtually no two fields behave the same way. I’ll place my bets on AI and machine learning in near-ish future, but there’s more in the present. A lot more.

So there’s a lot of pictures gathered from each flight. If I noted Juho’s estimates correctly, there can be approximately 3000 high-res pictures for just six hectares of field. Flight level is also playing a factor here: the higher you go, the more you can cover with one picture -but with the cost of accuracy and resolution. In any case, the pictures need to be transferred from the drone / camera to a computer that will patch them to a single multilayered map. That computer needs quite a bit of processing power to do it in any reasonable time, like in a day. For example, Juho is using a reasonably powerful laptop to accomplish this and for one equally reasonably small field the stitching of images into one map can take about twelve hours of processing power, all out.

Of course, cloud servers are pretty nifty and come almost cheap these days but arises the question of data transfer, along the need for know-how of setting one up. Moving that amount of data is still not fast or convenient. Especially when most of fields to be analyzed are in areas where the cell network may be 1G or 2G at best and the “broadband” connections come with quotation marks so big you can use them as scarecrows. My own hunch is that the fastest, most reliable way to get the data for processing is to take the camera, drive it to the computer and get it done yourself.

I’m saying “yourself”, because growing crops is not such a profitable business that people would have the time or employees to drive to a big city for this kind of service. So there’s a need for local computing power. Although that can be arranged: a higher-end PC should do the trick. Another reason to do image processing and analysis locally is that the situation on the field (a pun so intended it hurts) changes fast -and this is where things get really hard. So much so that the previous steps to get your hands on some reasonably reliable data seems like a walk in the park.

Juho showed me two pictures of one field made from image data gathered on two separate flights just 5 days apart and the differences were remarkable. As a layman, I would have guessed the time interval to be around two months. Furthermore, one cannot just gather data whole summer and expect to use it next year as a reference, even as an approximate one. The field will be very different place next year, not to mention weather conditions. So for the data to be useful and even somewhat actionable, one needs to have it processed and interpreted within two days, give or take. He also noted on several occasions that while the image can show -among other things- the approximate plant biomass in near-infrared image and leaf color differences indicating plant’s stress levels, the image can’t point out the actual cause of plant’s condition. My notion is that it can only be guessed at, where the experience and knowledge of the said field’s history and current conditions come in. I don’t think that pretty much anyone besides the farmer working on that field has that kind of expertise.

All this leads to conclusion that to get actionable data with drones these days, a farmer needs to be rather computer / tech savvy, have several thousands of euros to invest, have a month or two of time to practice (full time, not just free evenings) during growth season and have time to conduct the flights on several sizeable fields on weekly basis. Plus evenings to analyze the data from each flight. A show of hands, please? Thought as much.

So could this be a business? It can only be done locally and needs to be done often and diligently for effect. Juho estimated that the price point of max one to two euros per hectare, which may be a problem. Anything more would not really sell because the information would just be an addition to what farmers already kind of know, not a game changer. Therefore, he thinks that drone photography could be a handy additional tool for field and crop analysts and his own estimate we could be seeing real commercial crop monitoring products in next, say, two to five years. But currently, he is to his knowledge the only person around nordics, possibly Europe, doing research on the subject.

Just please try not to flood his email and phone just now, he’s very busy with the research and he will publish the results as soon as possible. Besides, I still need / want to get a formal interview from him for my thesis.

Initial Pitch (in finnish)

2017-07-10T16:30:00+03:00

Moi,

Kirjoitan opinnäytetyönäni katsausta IoT-teknologioiden sovelluksista kasvintuotannossa. Internet of Things, Esineiden Internet on ollut esillä kuluttajamarkkinoille ja teollisuudelle tehtyjen sovellusten myötä, mutta maatalouden vastaava teknologiakehitys ei ole saanut julkisuutta eikä yleisesti osata hahmottaa kuinka suuri teknologinen muutos alalla on käynnissä.

Työni tarkoituksena on lisätä tieto- ja sensoritekniikan mahdollisuuksien tietämystä kasvintuotannon alalla, ensisijaisesti haastattelemieni toimijoiden kesken ja samalla mahdollistamalla verkostoitumista. Verkostoitumisen yhteydessä on mahdollista herätellä keskustelua, jakaa tietoa ja kokemuksia.

Työn toteutus on kaksiosainen: kirjallisuuskatsauksen osassa käydään läpi millaista kirjallisuutta, tutkimusta ja tuotteita on olemassa. Haastatteluosassa kartoitetaan millainen tilanne todellisuudessa kentällä on haastattelemalla erilaisia toimijoita, aina laboratoriosta työkonevalmistajien kautta pelloille ja kasvihuoneisiin asti. Tiedon ja kontaktien jako jo kirjoitusvaiheessa on tärkeä, vaikka epävirallinen, osa työtä ja samalla pyrin avaamaan keskusteluyhteyksiä joita ei ehkä muutoin olisi syntynyt.

Kirjoittelen työn edetessä tehdyistä havainnoista, työvaiheista ja tuloksista blogimuodossa, mikä on kiinnostuneille ihmisille helpompi seurattava: AIoT-Blogi

t. Tatu Polvinen

PS: Vielä linkki oppariksi työstettävään tiedostoon, jossa on ensimmäisen kirjallisuuskatsauksen tulokset ja alustavat tutkimusongelmat: Tämänhetkinen versio löytyy GitHubista: tpolvinen/Agri

Search for Research

2017-07-08T16:30:00+03:00

As I feel like the literature survey/theoretical in my research plan is rather minimal and not really where I want it to be, I’m getting back to work on searching more relevant and up-to-date research work and articles. Also, it’s about 20 pages too short. :P

I met some new people at the MycoHacklab meetup #4 and I was told (among other things) about Impact Iglu that “is an impact business accelerator that supports entrepreneurs to solve business and societal challenges together.” From their Facebook page I found a link to ii2030 event Precision Agriculture for Smallholder Farmers in Berlin, Germany at Oct 18 - 19, 2017. I’d love to be there for the visits to hydroponic farming project and satellite data provider. Not to mention all the people that I could meet there! I wish I could attend, but I would need some serious sponsorship to get there. $)

From Impact Iglu’s page I also found out about german Green Alley Award that was won 2016 by Green City Solution’s CityTree that’s an IoT-enabled and -controlled green wall much like Naava’s product, but instead of plants it’s growing moss that may be more effective solution.

However, I couldn’t find anything about plant production -as in agriculture- from Green Alley Award’s pages, nor at ArcticStartup so I’ll be concentrating on research papers and articles I’ve found from International Society for Horticultural Science publications on this weekend, leaving the world of start-ups for a more scholarly one. I’ll be back in a couple of days.

Also, professor Salah Sukkarieh’s extensive work appears more than promising. I found about it from The World Economic Forum’s article These are the top 10 emerging technologies of 2017. Go check it put, it’s pretty awesome.

Refactoring

2017-07-07T01:00:00+03:00

I have –finally– refactored my previous research plan and research report. After refactoring, I used them to compose a file for my actual thesis. Yay! Well, only to make clear to me that how much work still needs to be done. Not so much yay. Nay, actually.

The oppari.md file in my Github repository Agri is located along other mentioned files in source folder, here.

Too tired to write anything else right now. Good night.