How AI Secures The Product Lifecycle Management Of Agriculture

Texas A&M-Purdue works with Oracle Cloud to maximize the potential of machine learning and AI for the agricultural industry.

Securing a sustainable, secure food supply is one of the biggest challenges that humanity will face over the next century, especially as climate change continues to affect farmland. The world population will surpass eight billion within the next two years with nearly 700 million lives in constant hunger. Population growth has slowed in prosperous regions, but it continues to grow in parts of the world where famine is most common. To solve the problems of hunger, climate change, over-farming and drought, PLM technologies will have to play a larger role in agriculture than ever before.

Digitizing agriculture has the potential to be truly transformative for an industry that eight billion people will rely upon for survival.

Digitizing agriculture has the potential to be truly transformative for an industry that eight billion people will rely upon for survival.

Agriculture is big business, and the industry hardly resembles the mom-and-pop family farm that many still hold in their minds when thinking about food. Food is a commodity, and just like oil, its production and trade are obsessed over by business people and world leaders. The field is dominated by analytics, software, bioengineering, robotics, automation and new technology. Early estimates impact the global GDP at over $500 billion by 2030 if agriculture digitization continues on its current trajectory.

Every aspect of the food production industry must be optimized to stamp out hunger in a sustainable manner. Data science, artificial intelligence and machine learning help farmers understand when to plant, where to plant, how to avoid, and recover from, storm damage. By 2014, each farm was already generating close to two hundred thousand data points per day; that figure is expected to grow to 4.1 million by 2050. Being able to process that data efficiently is key to unlocking its value.

A joint effort by research teams at Texas A&M and Purdue University are leading efforts to bring even more digitization into agriculture. The effort is a collaboration between the Texas A&M AgriLife Research and Extension and the Lyles School of Civil Engineering at Purdue. The team behind the technology is comprised of crop experts, computer scientists and geomatics engineers. Their project first launched in 2015 as UASHub, a web-based system that relies on drones to monitor crop conditions. UASHub is now active at AgriLife’s 11 research centers, each focused on a specific region of Texas and its agricultural workings. The raw data from the drone imagery is processed using Structure from Motion (SfM) analysis, allowing 3D models to be rendered from overlapping 2D images.

Drones will play a large role in the future of agriculture, both from a data collection standpoint and daily farming tasks.

Drones will play a large role in the future of agriculture, both from a data collection standpoint and daily farming tasks.

UASHub offers various tools and datasets to aid in the digitization of agriculture, including data visualization, cotton simulation, 3D point cloud visualization, data processing and product lifecycle management tools. UASHub applies machine learning algorithms to massive sets of aerial imagery collected by drones. Drones have flown over millions of acres of fields collecting data that can be fed into a model. The power of artificial intelligence makes it possible to analyze fields down to a square foot and gives farmers insight on which crops need water and fertilizer, or are prone to disease.

To continue growing the program, and make it more accessible for researchers and agriculture teams in different parts of the country, the Texas-focused team is partnering with Oracle Cloud to process their massive trove of data–over 120 terabytes–in the Oracle Cloud Infrastructure.

The researchers from Texas A&M and Purdue sought out Oracle to help speed up their 3D rendering and lower the cost of analyzing terabytes of data on-premises. The 3D models the team generates are powerful and extremely useful to farmers, who can use them to evaluate and compute crop metrics that cannot be done with traditional remote sensing methods. 

Dr. Juan Landivar of Texas A&M, the director of UASHub, believes migrating to Oracle Cloud was a change that needed to be made to help his program realize its full potential and unlock more value in digital agriculture. Jinha Jung, an assistant professor at Texas A&M – Corpus Christi, who is also working on the project, echoed the sentiment that Oracle Cloud will help unleash the full potential of the data.

“Cloud gives us the flexibility to increase instances more efficiently without worrying about how to maintain security and scale-up computational resources if needed,” Jung said.

One of the first big tests for the team’s capabilities was the catastrophic cold spell that devastated Texas last winter. The freezing temperatures and ice storms caused over $600 million in crop damage, but the UASHub team believes they now have more capability to help provide relief to farmers thanks to their partnership with Oracle.

“This type of research will provide a quick and easy assessment of the damage by the recent storm and freeze,” Landivar said. “The system will speed up how we visualize UAS images, assess storm damage and plan the next step for agriculture fields … It’s opening the door to opportunities we didn’t have three or five years ago.”

Alison Derbenwick Miller, the vice president of Oracle for Research, felt that the project clearly fell within her team’s mission to help researchers improve their computing power and work with greater flexibility and nimbleness as they seek to make scientific breakthroughs that will have a great impact on humanity.

“The Texas A&M and Purdue scientists are doing work that will help solve one of the world’s most pressing and intractable problems. We’re proud to make it easier for them to not only help make farmers more financially secure but also deliver a more stable and reliable food supply to a world dealing with increasingly complex environmental challenges,” Derbenwick Miller said.

So, what comes next for the technology being developed at UASHub?

The team is hopeful that they can expand their offerings well beyond the state of Texas, commercialize their platform and turn it into something that can be used by agricultural companies across the globe. Ramping up their computing power will allow the team to increase the scope and reach of their work. The potential for this dataset to crossovers with other disciplines of agricultural research, like phenotyping and genetics, is huge. Machine learning applied correctly could help farmers determine which specific seeds will grow best in their unique soil conditions. Agriculture is a huge multi-variable equation waiting to be solved and powerful computing algorithms will enable farmers to combine their traditional methods and knowledge with cutting-edge technology that helps them optimize every aspect of their process.

“The digital revolution has enormous potential for transforming the agriculture and food system in ways not previously seen – by drastically lowering information asymmetries and transactions costs that plague the system, and, as a result, improving the lives of farmers, as well as the nearly 8 billion people who depend on them for food,” says the World Bank’s Kateryna Schroeder, Agriculture Economist and Task Team Lead.

Digitizing the agriculture industry and supply chain is how we get to a more sustainable, eco-friendly, equitable food system for everyone who calls this planet “home.” Waste can be reduced, resources distributed and produced more efficiently, the environment can be protected, with these practices being exported to developing nations to develop reliable farming methods to fight famine and hunger. Our food system and supply chain are fragile, as evidenced by the increasing severity of climate change, drought, or the empty shelves and wasted crops during the early days of quarantine. The challenges that come with feeding eight billion people will never be fully engineered away, but increasing investment and support for technology like that being developed by UASHub is a big step in the right direction.