in partnership withAnalog Devices Inc.
On the outside, New England–grown tomatoes look much like tomatoes grown anywhere else. But in terms of flavor, they’re rarely anybody’s first choice. In fact, New England tomatoes are more likely to end up in soup or as ketchup than sliced for sandwiches or drizzled with olive oil and served with basil and mozzarella.
Determined to find out why the region’s tomatoes are comparatively tasteless, Analog Devices Inc. (ADI) started its Internet of Tomatoes project. This precision agriculture experiment uses technologies such as micro-electromechanical systems (MEMS) and sensors to figure out whether environmental monitoring could improve flavor.
The project stems from a 2014 MEMS & Sensors Industry Group conference in Scottsdale, Arizona. There, keynote speaker Francis Gouillart, president of the Experience Co-Creation Partnership, an education and consulting firm, challenged attendees to use technology to improve basic human needs: water, food, energy, health care, education, and freedom. ADI decided to tackle the food factor by examining two facets of tomatoes that could be remedied: temperature measurements and growing-degree days, a heat index used to predict when a crop will reach maturity.
The Hardware of Tomatoes
ADI launched its project in January 2015 with several local farm partners to test technology that could survive in the fields, connect to the cloud to send data gathered from sensors, and deliver information to farmers in real time. Among the initial challenges, the first farm on the project had no usable power or Internet access, which raised the issue of how to transfer data from environmental sensors in the field to the cloud. ADI examined different technologies, but it ultimately chose low-energy wireless Bluetooth sensors that connect to handheld devices hardy enough to survive New England’s unpredictable and often harsh weather. “Choosing the right devices is critical because they have to survive out there. We have to consider the temperature range and the New England weather,” says Rob O’Reilly, a senior technical advisor at ADI who’s been involved with the project since it began.
ADI is currently working on a complete package: temperature sensor, cloud provider, and gateways—the sum of which will be a turnkey solution that farmers could use to grow tastier tomatoes. Meanwhile, it’s still addressing battery life and condensation issues. “Every morning, everything is covered in dew,” O’Reilly notes. “Keeping the sensors dry in the morning dew can be a challenge.”
Leveraging “Growing-Degree Days”
Other analytics useful to farmers include growing-degree days. The average tomato’s growing-degree day cycle is 1,300 days, but, of course, that measurement doesn’t refer to actual days. Instead, it reflects a heat accumulation calculated using the average daily maximum and minimum temperatures in the farm’s location and compared with a base temperature. This formula determines when tomatoes should be harvested, as well as the effectiveness of the integrated pest-management system. Add in the gestation cycle of beetles and other insects that feast on tomato crops, and farmers know not only how long the plants have been growing and maturing, but also when the pests can be expected—all thanks to the data.
Another component of harvesting, knowing exactly when to harvest, has usually been managed by feel rather than science. But the Internet of Tomatoes adds data-gathering sensors and analytics that help farmers determine the optimal time to pluck their ripe tomatoes from the vines.
The Chemistry of Tomatoes
ADI, along with its farm partners, is working toward developing a chemistry profile that helps identify what elements make the tomatoes juicy and delicious. “We’re building a database around it to measure the quality of the tomato—not just individual items, but the taste profile,” O’Reilly says.
These measurements have revealed that the chemistry of New England tomatoes differs from their tastier counterparts from other parts of the world. They don’t have high fructose, glucose, or salt content—which is why when most people grab a New England–grown tomato, they also grab a salt shaker. “When you compare them to an Italian tomato, they’re not even close,” O’Reilly says, noting that from an aesthetic perspective, the New England tomatoes look like any other. But on the inside, the chemistry doesn’t match up, according to ADI’s analysis of thousands of tomatoes.
ADI is also seeking a non-invasive way to test tomatoes for their chemistry profiles without destroying the tomatoes. Currently, testers puree and pipette tomatoes into lenses to test for the different elements: fructose, sucrose, glucose, and salinity. ADI also measures the tomatoes’ Brix during testing—that is, their carbohydrate and mineral levels.
In addition to a tomato’s chemistry, its health is also critical. Lycopene is an antioxidant that affects not only the red color of the tomato, but also its quality. ADI’s goal is to look at acids, sugars, nutrients, and other measurable markers with the same technology as that used for tomato testing to assess the health and overall nutritional value of the tomato. “We have a non-destructive optical approach in the works so we don’t have to destroy the tomatoes we want to test,” O’Reilly says. That approach is a “tomato tricorder”—think of a Star Trek–inspired device to scan the tomatoes, using nearfield infrared technology like what is used in night-vision goggles.
Going Beyond Tastier Tomatoes
Although a precision design kit with some components will be released soon, ADI’s ultimate goal is to develop a kit for farmers to use, one that not only contains temperature and humidity sensors but also the components necessary to set up the system for data-gathering and tomato-scanning. Users will access the gateway and cloud via IOS or Android devices. Optical technology will measure the soil, leaf, and stem lengths and their stress levels.
Ultimately, the data collection will help create models to predict growth, bugs, and the all-important question of when to water. “Overwatering and humidity are the source of tomato crops being wiped out,” O’Reilly notes. “Farmers can lose a lot of tomatoes due to white blight mold and bug infestation.”
O’Reilly and his colleagues at ADI are also investigating other problems, including root stress and vapor pressure deficit (the difference between theoretical pressure and the actual pressure from humidity, which causes leaf stress). Mitigating those issues could lead to better-tasting tomatoes. Much of the data is being crowdsourced from the precision-design kit released last winter.
The potential economic impact to farmers is huge. “Getting the most from every crop is vitally important to them. A very large company can help without a huge amount of effort, and that’s why we’re doing this,” O’Reilly says. “Ultimately, farmers focus on where their tomatoes sit on the quality scale so that when they sell their harvest, it’s not going to the ketchup shelves”—but instead to higher-paying, more prestigious outlets such as supermarkets and restaurants.
In the long term, according to O’Reilly and his project colleagues, ADI’s Internet of Tomatoes project isn’t just about testing tomatoes. Instead, it’s part of a larger Internet of Things (IoT) effort intended to open the door for other optical applications for more crops and even for non-agricultural products.
“We’ve built our own IoT Farm lab at our Wilmington, Massachusetts, campus to further explore additional smart agriculture–sensing opportunities,” notes John Patrick O’Connor, an ADI senior systems application engineer. “Our ambition with smart agriculture is to provide ubiquitous sensing capabilities and easy-to-understand and factual data to growers worldwide,” adds Michael Murray, ADI’s general manager of industrial sensing. “This approach to crowdsourced, accurate, and economical data will help provide better outcomes for growers, higher-quality produce for consumers, and a socially sustainable business for ADI.”
Analog Devices’ technologies form the foundation of groundbreaking IoT solutions by sensing, measuring, and interpreting the world around us to bridge the physical and digital domains. For more information, please visit www.analog.com/IoT.
Capitalizing on machine learning with collaborative, structured enterprise tooling teams
Machine learning advances require an evolution of processes, tooling, and operations.
The Download: how to fight pandemics, and a top scientist turned-advisor
Plus: Humane's Ai Pin has been unveiled
The race to destroy PFAS, the forever chemicals
Scientists are showing these damaging compounds can be beat.
How scientists are being squeezed to take sides in the conflict between Israel and Palestine
Tensions over the war are flaring on social media—with real-life ramifications.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.