Based out of Toulouse France, Naïo Technologies is a clear leader in agricultural robotics, active in Europe, and now poised to expand globally.
Let’s take a moment to look at technology designed to gather data on the farm. The TerraSentia robot developed by EarthSense is just that. A robot who’s primary purpose or mission is to gather crop data, not just for the farmer, but for scientists, breeders, product developers, and agronomists.
Precision agriculture is enabled by building machine learning models using data gathered from crops (and livestock). However this assumes a steady supply of data, especially localized data, as each farm is unique, and growing conditions vary widely.
While companies use test farms to help gather data and build models, those tests are only applicable in conditions or environments that are similar. Not to mention that our environments are dynamic and constantly changing, especially given increasing climate volatility.
As we dig deeper into our coverage of robots and AI in agriculture, let’s take a moment to look at technology designed to gather data on the farm. The TerraSentia robot developed by EarthSense is just that. A robot who’s primary purpose or mission is to gather crop data, not just for the farmer, but for scientists, breeders, product developers, and agronomists. From their website:
Our first robot—TerraSentia—improves the quantity, accuracy, cost and speed of in-field plant trait data collection, especially for under-canopy traits that cannot be obtained by other technologies.
Our machine vision and machine learning based analytics seamlessly convert field data to specific, actionable information about plant-traits.
The benefit of a robot like TerraSentia is that it makes it easy for a farmer or researcher to collect data where they are, rather than rely upon someone else’s crop data or machine learning models. Especially when it comes to creating or selecting seeds that will work in a given location and under specific conditions.
While farmers may be curious or interested in using automated tools for precision farming, they may not be in a position to understand how this technology works. In particular the more data they have about their crops and fields, the more accurate they can be, and the more likely the technology will bend to them rather than expecting them to bend to the technology. Sometimes this might be as simple as knowing which seeds to use.
TerraSentia has currently been deployed in corn, soybean, wheat, sorghum, vegetable crops, orchards, and vineyards. The more data the company collects, the more specific measurements they’re able to make and monitor.
Part of what sets TerraSentia apart from other agricultural robots is its size, as this relatively tiny robot is only a bit more than a foot high and a foot wide. Here’s a good overview video of how it works and moves:
While drones and other machines tend to gather data from above the crops, TerraSentia does so from below. Another quote from their website:
TerraSentia uses a combination of sensors—including visual cameras, LIDAR, GPS, and other on-board sensors—to autonomously collect data on traits for plant health, physiology, and stress response.
TerraSentia’s unique dataset delivers high-value under-canopy plant traits including stand-count, stem width, plant height, LAI, etc.
This under-canopy approach allows them to gather even more precise data, which is a big reason they’ve received attention and recognition for their work.
While farmers will increasingly desire and benefit from the kind of data and modelling that TerraSeentia can produce, the immediate and profound impact of this technology will be felt by researchers and breeders.
The data collected by the TerraSentia is changing breeding from a reactionary process into a more predictive one. Using the robot’s advanced machine-learning skills, scientists can collate the influence of hundreds, even thousands, of factors on a plant’s future traits, much like doctors utilize genetic tests to understand the likelihood of a patient developing breast cancer or Type 2 diabetes.
“Using phenotyping robots, we can identify the best-yielding plants before they even shed pollen,” said Mike Gore, a plant biologist at Cornell University. He added that doing so can potentially cut in half the time needed to breed a new cultivar — a plant variety produced by selective breeding — from roughly eight years to just four.
The potential success of this robot has also attracted interest from big AgSciencee companies like Corteva, who recognize that TerraSentia could be a powerful tool in their research and development process.
EarthSense has produced roughly 80 of these robots so far, and are in the process of making 100 more. This means that the robots are mostly being used in trials that build both the machine learning models for its use as well as interest in it as a concept. More from the NY Times article above:
Dr. Chowdhary and his colleagues hope that partnerships with big agribusinesses and academic institutions will help subsidize the robots for smallholder farmers. “Our goal is to eventually get the cost of the robots under $1,000,” he said.
Farmers don’t need special expertise to operate the TerraSentia, either, Dr. Chowdhary said. The robot is almost fully autonomous. Growers with thousands of acres of land can have several units survey their crops, but a farmer in a developing country with only five acres of land could use one just as easily. The TerraSentia has already been tested in a wide variety of fields, including corn, soybean, sorghum, cotton, wheat, tomatoes, strawberries, citrus crops, apple orchards, almond farms and vineyards.
While it is essential that these robots remain easy to use, interest from farmers, at least on a large scale, may be along way off. The company still needs to develop plans for how these devices will be maintained or charged, with work currently being done on a kind of maintenance barn that emulates a combined drive shed and dog house.
In the meantime the real impact of this technology will be on research and the speed by which it allows researchers to collect field data.
TerraSentia is the first robot from EarthSense, but not the last. It’s been in development for several years, but is now reaching a point to be useful and valuable to the agricultural industry.
One of the features of TerraSentia is that it is designed to be teachable. This allows the use of the robot to be expanded over time, in particular identify and collect new kinds of data and crop attributes. From their website:
We've developed a cloud-based platform that will let you easily teach TerraSentia to automatically measure a variety of key traits.
In 2019, we have worked with leading private- and public-sector organizations to accurately detect and quantify high-value traits in corn, soybean, wheat, sorghum, etc.
We are now teaching TerraSentia to measure early vigor, corn ear height, soybean pods, plant biomass, and to detect and identify diseases and abiotic stresses.
In order to see TerraSentia become a successful product, and create additional products, that leverage the technology and software developed through the robot, EarthSense is forging partnerships and connecting with the rest of the AgTech ecosystem, in this case as an active partner of the new AIFARMS institutue.
On Wednesday, the National Artificial Intelligence Research Institutes announced a $20 million award to the Center for Digital Agriculture at the University of Illinois at Urbana-Champaign to develop a new Artificial Intelligence for Future Agricultural Resilience, Management, and Sustainability (AIFARMS) institute. The White House-backed program, a joint effort between the National Science Foundation and the USDA’s National Institute of Food and Agriculture, supports AI research designed to impact and improve society.
The AIFARMS institute is led by Vikram Adve, principal investigator and Donald B. Gillies Professor of Computer Science at Illinois’ Grainger College of Engineering. “I’m excited and humbled to be leading the AIFARMS Institute. Illinois and our partner institutions are world leaders in the areas of computer science, artificial intelligence, and agriculture research, and these strengths are reflected in the breadth and depth of the AIFARMS team,” Adve says. “By fostering close collaborations between these researchers, and by growing and diversifying a workforce skilled in digital agriculture, we have an exciting opportunity to help address some of the most daunting challenges faced by world agriculture today.”
A multi-department and multi-institutional collaboration, AIFARMS brings together 40 researchers from Illinois, University of Chicago, the Donald Danforth Plant Sciences Center, Michigan State University, Tuskegee University, and USDA Agricultural Research Service to accelerate AI and promote foundational advances in agriculture. At AIFARMS, world-class scientists; industry partners EarthSense, Microsoft Research, and IBM Research; doctoral students; and postdoctoral researchers come together to address major agricultural challenges such as labor constraints, animal health and welfare, environmental crop resilience, and soil health. The institute leverages expertise and resources to revolutionize the agriculture industry through technological advances, training programs that support workforce diversification, and sustainability management.
It’s interesting to see this level of collaboration and it offers an important insight towards what is necessary to find success when developing new technology like the TerraSentia robot.
Similarly EarthSense demonstrates that not all technology in the agricultural sector is designed for farmers, even if this robot is designed with farmers in mind. The dynamism and diversity of the industry shows that technology that helps make better seeds faster can be just as important and essential as helping farmers with their day to day operations.
We’ll certainly keep an eye on the AIFARMS institute and the efforts of EarthSense. We also plan to continue profiling robotic and AI startups in the agricultural sector.
For more detail on the founder and their long term vision, check out this presentation Girish Chowdhary made just under a year ago: