AI for Agriculture and Environment

AI and Machine Learning: Revolutionizing Agriculture and Environmental Management

Artificial intelligence (AI) and machine learning (ML) are transforming the way researchers address agricultural and environmental challenges. At Purdue Agriculture, scientists are harnessing these technologies to enhance efficiency, sustainability, and precision in both fields. AI enables computers to mimic human intelligence, while ML, a subset of AI, allows systems to learn from data and identify patterns that might otherwise go unnoticed.

One of the key researchers in this area is Brady Hardiman, an associate professor of forestry and natural resources. Hardiman co-edited a special issue of Frontiers in Ecology and the Environment, highlighting the importance of AI and ML in processing large 3D datasets. His work focuses on urban ecosystems, where 80% of the U.S. population resides. By analyzing remote sensing data and high-resolution imagery, Hardiman’s team uses AI and ML to detect patterns in urban landscapes, such as identifying tree species to improve urban forest management.

Tracking Invasive Species with AI

Invasive species pose a significant threat to both urban and rural ecosystems. In Chicago, Hardiman’s team employs aerial light detection and ranging (LiDAR) data combined with machine learning to map the spread of invasive buckthorn shrubs in city forest preserves. This technology helps forest managers target conservation efforts more effectively.

In rural areas, ML is used to identify tree species for timber production and other applications, demonstrating the versatility of AI in managing natural resources.

AI-Enhanced Medical Robot for Cattle Health

Upinder Kaur, an assistant professor of agricultural and biological engineering, is developing an AI-enhanced medical robot to monitor cattle health. This innovative device collects data on nutrition, sustainability practices, and methane output, providing farmers with detailed insights into their herd’s condition. The robot, designed to operate inside a cow’s stomach, measures biomarkers like temperature and pH, offering a comprehensive view of the animal’s health.

Kaur’s team is also working on miniaturizing the robot to make it small enough for cows to swallow. Currently, the robot is inserted through a cannula. The goal is to improve animal welfare and reduce the burden on farmers by providing real-time health data. Additionally, Kaur is developing a robot "dog" equipped with sensors to locate ticks and identify tick activity hotspots, aiding in the control of tickborne diseases.

Simulating Crop Yields Under Climate Scenarios

Climate change is a growing concern for agriculture, and AI is playing a crucial role in predicting its impact on crop yields. Sajad Jamshidi, a postdoctoral scientist in agronomy, co-authored a study that used an ensemble of ten machine learning models to assess the effects of climate change on rice yields. By combining multiple models, the team achieved more accurate predictions, helping breeders develop resilient rice varieties.

The study analyzed data from the Southern U.S. rice growing region from 1970 to 2015 and found that public breeding programs had unintentionally developed rice varieties resistant to warming climates while focusing on higher yields. This framework allows breeders to simulate rice variety responses to climate, temperature, and rainfall, enabling them to adapt to changing conditions.

AI-Powered Decision Support Tools for Crops

Ankita Raturi, an assistant professor of agricultural and biological engineering, is developing technologies to help farmers make informed decisions. Her web-based recommender system, dubbed the "Netflix for crops," allows users to filter data by location, soil, weather, and goals to identify the best crop for specific conditions. This human-centered approach helps farmers optimize their yields and adapt to environmental challenges.

Raturi’s projects also include using agent-based models to represent foodsheds for data-driven policymaking. These models help policymakers and food coordinators balance food security with sustainability goals, promoting regenerative and small-scale agriculture.

Detecting Crop Diseases with Machine Learning

Crop diseases can devastate agricultural productivity, but AI is providing new tools to combat them. Somali Chaterji, an associate professor of agricultural and biological engineering, uses ML to build algorithms that help farmers scout for diseases more efficiently. Her work on semi-supervised semantic segmentation is particularly effective for detecting rare disease outbreaks.

Chaterji’s Innovatory for Cells and Neural Machines (ICAN) lab developed Agile3D, a LiDAR-based perception algorithm for resource-constrained platforms. This technology enables devices to run sophisticated models without draining batteries or requiring constant connectivity, making it ideal for agricultural surveillance. By processing data close to where it is generated, ICAN’s work supports real-time decision-making and reduces energy use.

The Future of AI in Agriculture and Environment

The integration of AI and ML into agriculture and environmental management is still in its early stages, but the potential is vast. As these technologies continue to advance, they promise to revolutionize how we grow food, manage natural resources, and adapt to climate change. Researchers at Purdue Agriculture are at the forefront of this revolution, using AI to solve some of the most pressing challenges facing our planet.