Analog Devices, Inc. announced today a collaboration with The Cornucopia Project and ripe.io to explore the local food supply chain and use this work as a vehicle for educating students at ConVal Regional High School in Peterborough, N.H., and local farmers on 21^st century agriculture skills. The initiative instructs student farmers how to use Internet of Things and blockchain technologies to track the conditions and movement of produce from “Farm to Fork” to make decisions that improve quality, yields, and profitability. Together with the Cornucopia Project, the endeavor is funded by Analog Devices and ripe.io, with both companies also providing technical training.

For the project, Analog Devices is providing a prototype version of its crop monitoring solution, which will be capable of measuring environmental factors that help farmers make sound decisions about crops related to irrigation, fertilization, pest management, and harvesting. The sensor-to-cloud, Internet of Things solution enables farmers to make better decisions based on accumulated learning from the near-real-time monitoring. These 24/7 measurements are combined with a near infrared (NIR) miniaturized spectrometer that conducts non-destructive analysis of food quality not previously possible in a farm environment.

 

 

“This project expands on our ‘Internet of Tomatoes’ program which empowers farmers to make better decisions throughout the growing cycle, improving quality, economic, and environmental outcomes,” said Kevin Carlin, vice president, Automation, Energy and Sensors, Analog Devices. “Our crop monitoring solution will provide reliable and precise information to student farmers and local farmers so they can grow healthier, fresher, better tasting produce. It demonstrates how a crop monitoring solution extends the value and possibilities of the Internet of Things in truly transformative ways.”

The Cornucopia Project, a non-profit located in Peterborough, N.H., provides garden and agricultural programs to students from elementary through high school. Student farmers in its Farm to Fork program learn how to use advanced sensor instrumentation in their greenhouse, which provide valuable data to assess the attributes of tomatoes, and how these factors affect taste and quality. The program also educates students on how crops can be tracked throughout the agricultural supply chain to support food quality, sustainability, traceability, and nutrition.

“Analog Devices is helping us explore how advances in technology can support local food systems,” said Karen Hatcher, executive director, The Cornucopia Project. “We are training next-generation farmers in 21^st century agriculture to harvest tastier, more abundant and more sustainably grown tomatoes than ever before. This initiative will contribute to enhancing the economic health and vitality of local small- and medium-size farms and the communities that support them.”

ripe.io is contributing its blockchain technology to model the entire fresh produce supply chain, combining the crop growing data, transportation, and storage conditions. Blockchain – a distributed ledger, consensus data technology that is used to maintain a continuously growing list of records – will track crop lifecycle from seed to distributor to retailer to consumer, bringing transparency and accountability to the agricultural supply chain.

“This project is one of the first implementations of blockchain technology to build an open and transparent supply chain with farmers, suppliers, distributors, retailers, food service, and end consumers,” said Raja Ramachandran, CEO of ripe.io. “What is learned in the initiative not only will improve quality, economic, and environmental outcomes in the local farming community, but also can be extended to other farms and crop species around the country.”

Analog Devices (NASDAQ: ADI) is a global high-performance analog technology company dedicated to solving the toughest engineering challenges.

A future android brain like that of Star Trek’s Commander Data might contain neuristors, multi-circuit components that emulate the firings of human neurons.

Neuristors already exist today in labs, in small quantities, and to fuel the quest to boost neuristors’ power and numbers for practical use in brain-like computing, the U.S. Department of Defense has awarded a $7.1 million grant to a research team led by the Georgia Institute of Technology. The researchers will mainly expand work on new metal oxide materials that buzz electronically at the nanoscale to emulate the way human neural networks buzz with electric potential on a cellular level.

But let’s walk expectations back from the distant sci-fi future into the scientific present: The research team has developed neuristor materials to build, for now, an intelligent light sensor, and not some artificial version of the human brain, which would require hundreds of trillions of circuits.

“We’re not going to reach circuit complexities of that magnitude, not even a tenth,” said Alan Doolittle, a professor at Georgia Tech’s School of Electrical and Computer Engineering. “Also, currently science doesn’t really know yet very well how the human brain works, so we can’t duplicate it.”

Intelligent retina

But an artificial retina that can learn autonomously appears well within reach of the research team from Georgia Tech and Binghamton University. Despite the term “retina,” the development is not intended as a medical implant, but it could be used in advanced image recognition cameras for national defense and police work.

At the same time, it significantly advances brain-mimicking, or neuromorphic, computing. The research field that takes its cues from what science already does know about how the brain computes to develop exponentially more powerful computing.

The retina is comprised of an array of ultra-compact circuits called neuristors (a word combining “neuron” and “transistor”) that sense light, compute an image out of it and store the image. All three of the functions would occur simultaneously and nearly instantaneously.

“The same device senses, computes and stores the image,” Doolittle said. “The device is the sensor, and it’s the processor, and it’s the memory all at the same time.” A neuristor itself is comprised in part of devices called memristors inspired by the way human neurons work.

Brain vs. PC

That cuts out loads of processing and memory lag time that are inherent in traditional computing.

Take the device you’re reading this article on: Its microprocessor has to tap a separate memory component to get data, then do some processing, tap memory again for more data, process some more, etc. “That back-and-forth from memory to microprocessor has created a bottleneck,” Doolittle said.

A neuristor array breaks the bottleneck by emulating the extreme flexibility of biological nervous systems: When a brain computes, it uses a broad set of neural pathways that flash with enormous data. Then, later, to compute the same thing again, it will use quite different neural paths.

Traditional computer pathways, by contrast, are hardwired. For example, look at a present-day processor and you’ll see lines etched into it. Those are pathways that computational signals are limited to.

The new memristor materials at the heart of the neuristor are not etched, and signals flow through the surface very freely, more like they do through the brain, exponentially increasing the number of possible pathways computation can take. That helps the new intelligent retina compute powerfully and swiftly.

Terrorists, missing children

The retina’s memory could also store thousands of photos, allowing it to immediately match up what it sees with the saved images. The retina could pinpoint known terror suspects in a crowd, find missing children, or identify enemy aircraft virtually instantaneously, without having to trawl databases to correctly identify what is in the images.

Even if you take away the optics, the new neuristor arrays still advance artificial intelligence. Instead of light, a surface of neuristors could absorb massive data streams at once, compute them, store them, and compare them to patterns of other data, immediately. It could even autonomously learn to extrapolate further information, like calculating the third dimension out of data from two dimensions.

“It will work with anything that has a repetitive pattern like radar signatures, for example,” Doolittle said. “Right now, that’s too challenging to compute, because radar information is flying out at such a high data rate that no computer can even think about keeping up.”

Smart materials

The research project’s title acronym CEREBRAL may hint at distant dreams of an artificial brain, but what it stands for spells out the present goal in neuromorphic computing: Cross-disciplinary Electronic-ionic Research Enabling Biologically Realistic Autonomous Learning.

The intelligent retina’s neuristors are based on novel metal oxide nanotechnology materials, unique to Georgia Tech. They allow computing signals to flow flexibly across pathways that are electronic, which is customary in computing, and at the same time make use of ion motion, which is more commonly know from the way batteries and biological systems work.

The new materials have already been created, and they work, but the researchers don’t yet fully understand why.

Much of the project is dedicated to examining quantum states in the materials and how those states help create useful electronic-ionic properties. Researchers will view them by bombarding the metal oxides with extremely bright x-ray photons at the recently constructed National Synchrotron Light Source II.

Grant sub-awardee Binghamton University is located close by, and Binghamton physicists will run experiments and hone them via theoretical modeling.

‘Sea of lithium’

The neuristors are created mainly by the way the metal oxide materials are grown in the lab, which has some advantages over building neuristors in a more wired way.

This materials-growing approach to creating part of the computational structure is conducive to mass production. Also, though neuristors in general free signals to take multiple pathways, Georgia Tech’s neuristors do it much more flexibly thanks to chemical properties.

“We also have a sea of lithium, and it’s like an infinite reservoir of computational ionic fluid,” Doolittle said. The lithium niobite imitates the way ionic fluid bathes biological neurons and allows them to flash with electric potential while signaling. In a neuristor array, the lithium niobite helps computational signaling move in myriad directions.

“It’s not like the typical semiconductor material, where you etch a line, and only that line has the computational material,” Doolittle said.

Commander Data’s brain?

“Unlike any other previous neuristors, our neuristors will adapt themselves in their computational-electronic pulsing on the fly, which makes them more like a neurological system,” Doolittle said. “They mimic biology in that we have ion drift across the material to create the memristors (the memory part of neuristors).”

Brains are far superior to computers at most things, but not all. Brains recognize objects and do motor tasks much better. But computers are much better at arithmetic and data processing.

Neuristor arrays can meld both types of computing, making them biological and algorithmic at once, a bit like Commander Data’s brain.