Update:
Tommy Martin and Smartfield didn't like a particular statement in the original post because it suggested that our ideas were intentionally shared with them by USDA. After some phone and email conversations with him, I have agreed to edit this post to contain verified facts and clearly identifiable opinions only.
This is a story that I've been wanting to tell. It's really something that's been nagging me for years. Finally I have the motivation to go ahead and put it out here. Is it something to be concerned about? to take advantage of? to just be proud of? for which I should claim prior art? Or is it just sour grapes on my part? You tell me. One thing I know is that it's one of the more interesting things that has happened in my life so far.
It tells the story behind an exciting company in Lubbock, TX called Smartfield, whose Smartcrop product uses data from your crops to control your irrigation system/schedule, and hopefully conserve water.
Smartcrop uses data consisting of plant canopy temperatures (using infrared thermocouples) along with other environmental data (rain, humidity) to measure the stress of crops. If a time-temperature threshold is reached, research indicates that the plant is unable to adequately respire and should be watered. The science behind it was developed and patented (attached) by Dr. James Mahan, Dr. Donald Wanjura, Dr. John Burke, and Dr. Dan Upchurch at the USDA-ARS Crop Science Research Lab in Lubbock.
According to the article, Smartfield was founded when one day after church, Dr. Upchurch (no pun intended) proposed an idea about his research to Tommy Martin, an engineer at Accent Engineering who took Upchurch up on his idea.
"He asked what we were doing, and we'd just finished some work with wireless," said Martin, Smartfield's chief executive officer. "He perked up and said, 'We could use your help.' "
Within the week, Martin and McNeill were meeting with USDA researcher James Mahan, who was part of a team studying ways to measure heat stress in plants using infrared sensors.
"We saw a tangle of wires," Martin said. "It was a prototype, and they wanted our help to make it wireless."
It didn't take long before Martin and McNeill saw the potential and pointed out that for commercial use, such a sensing system wouldn't need expensive, research-grade hardware.
"Our background is consumer electronics," said Martin, who first got together with McNeill in 1983 at Texas Industries' Lubbock facility. "We suggested a clean sheet design, that resulted in a product with a more attractive price that could be commercialized."
I remember this time very well. It was the summer of 2001. I was a senior in college working for Dr. Mahan at USDA during the summer. Thomas Mullen, another engineering student who I knew from high school had been working for the previous several summers, and I had joined him as a technician on the BIOTIC research.
Dr. Mahan and Dr. Upchurch approached the guys at Accent to create a wireless thermocouple solution. It wasn't about commercialization at all - it was a solution to let them put thermocouples farther away from the BIOTIC base unit to give them more flexib
ility in their research. It's tru
e though that at that time, the BIOTIC unit was unwieldy (it probably weighed 75+ pounds) and very expensive. I attached a picture to this post of how it looked back then.
Thomas and I were responsible for most of the technological development behind the unit and its website, where daily crop data was automatically analyzed and posted. We had to troubleshoot it when the data didn't download right, we had to program the various components that comprised the device, and we had to deploy it to various test sites.
With our intimate knowledge of the technology and its issues, Thomas and I came up with the concept of the clean sheet design. We talked through all the details, and decided that this would be a great business opportunity. With that, we established a partnership that we called EDAS, or Environmental Data Acquisition Systems. We both invested our summer money into ordering microcontrollers, integrated circuits, buttons, screens, wire, etc, and started up what amounted to a "garage" business. Thomas worked through the electrical design, I programmed the microcontroller, and together we developed a prototype device. It only had 1 thermocouple input, but that was good enough for proof of concept - and the thing fit in your hand, a startling contrast to the monstrosity that we were lugging around during business hours. I've included some pictures of Thomas and I making our prototype with this post.
Read the article excerpt above a little closer. It didn't take long before Martin and McNeill saw the potential. Well, can you guess what happened between Accent's wireless work and when they "saw the potential?" The proposal, memo, and presentation details are included here. Thomas and I presented our prototype and opened the eyes of the researchers. We proposed a plan where USDA would fund product development (analogous to their agreement with Accent), we would meet certain device deliverables, and in return we would field test our device in the USDA trials. When the device design was finalized, we would commercially release the product.
The response from USDA was extremely enthusiastic, and Thomas and I started making plans about how to proceed. Pretty soon, though, things started to stall out, with sidesteps and ambiguous answers from USDA. Finally, they said no. The reasoning was never clear - there was some mention of a conflict of interest regarding the funding of BIOTIC's commercialization.
Update:
I'm not sure about all the events which occurred between our proposal and the agreement between USDA and Smartfield. What I do know is that a several years after Thomas and I made our proposal, Smartfield (after some prodding from the USDA researchers) secured the agreement with USDA that we had been pursuing, and they developed protoype devices for USDA that offered the same type of integrated design that we had proposed.
Although quoted in the article as saying that the "clean sheet design ... resulted in a product with a more attractive price that could be commercialized," Tommy Martin told me that the use of cheaper thermocouple sensors was the advancement that made it a viable commercial product. I disagree.
Martin and McNeill told me that the integrated design approach was an "obvious conclusion from the state of the project." It may be "obvious" to some engineers, but as I suggest above, the prototype device and the presentation of a more elegant, integrated design opened the eyes of the researchers (who are not engineers) as to how to productize their research. As an example, there are many obvious design changes I would make to Smartfield's implementation of the technology, but since SmartCrop was not created by automation and instrumentation engineers, the issues are not apparent to them.
Martin and company are personal friends with the researchers, but they are also established business partners. Maybe the Accent guys really are just that much better at the business side than us - in an email to me, Martin says that I should have had USDA sign a confidentiality agreement up front. I can see why USDA may have preferred to do business with Accent/Smartfield. I just wish they had taken a chance with some inspired young entrepreneurs instead.
It's really too bad, because I think Thomas and I could have done some great things with the idea.
So what's happened since then? Thomas and I are both now involved in the engineering of huge industrial automation projects. Martin and McNeill spun off Accent to form Smartfield (
website), which is apparently about to grow significantly.
They've received incentive money from the city of Lubbock to stay there, and are planning on hiring 12 employees (about $650k in salary). Did the researchers make the right decision? Certainly I question their ethics, but I don't know what would have been best. At least I've learned my lesson.
The BIOTIC Patent
(download)
An early prototype of our device; Dr. Mahan with the USDA's BIOTIC unit.
Thomas and I working on our device; Me with our finished prototype.
Our proposal presentation
(download)
Proposal summary
(download)
Proposal memo
(download)