Sandesh Poudel
As a Ph.D. student in Agricultural & Biological Engineering at the University of Illinois at Urbana – Champaign, Sandesh Poudel is combining robotics, artificial intelligence, and precision agriculture to address small-scale specialty crop farming in Illinois. Under the mentorship of his advisor, Dr. Sunoj Shajahan, and with the help of the rest of the team at the Digital Precision Agriculture lab, he is working to develop an autonomous robotic system for in-field sustainable farming solutions. In this article, Poudel and Shajahan discuss the gaps in technological innovation for specialty crops and the approach they are taking to address them.
Key Takeaways
- Addressing a Technological Gap: While major commodity crops like corn have seen massive innovation, specialty crops like horseradish have been largely overlooked by big agricultural companies and still rely on manual labor.
- A Multi-Tech Approach: Poudel is integrating robotics, AI, and drones to manage weeds. Drones identify “hotspots” from above, while ground robots physically remove weeds in real time.
- Sustainable Weed Control: Because horseradish grows underground, chemical herbicides pose a contamination risk. The robotic system provides a chemical-free, sustainable alternative for farmers.
- Commitment to Open-Source Innovation: The project aims to make technology accessible to small-scale growers. The team plans to provide design files and source code so farmers can build or customize their own robots using affordable hardware store components.
Illinois may be famous for being the land of Lincoln and home of “Da Bears,” but few are familiar with one of its lesser known claims to fame, which lies underground.
The commercial cultivation of horseradish in Southern Illinois took root in the 1850s, when German immigrant farmers discovered that the region’s sandy, nutrient-rich soil created an ideal environment for the unassuming crop to thrive.
Today, Illinois is recognized as one of the world’s leading suppliers of horseradish root. Despite this, the crop remains one of the few in the state that has seen minimal innovation in targeted agricultural technologies. That is where Sandesh Poudel comes in.
In the winter of 2025, Poudel joined Dr. Sunoj Shajahan’s Digital Precision Agriculture Lab at the University of Illinois at Urbana-Champaign as a Ph.D. student. With a Master’s degree in Agricultural Engineering from the University of Georgia, Poudel’s research focuses on the integration of robotics, AI, and precision agriculture to develop sustainable solutions for farming systems. Upon his start at Illinois, he and Dr. Shajahan identified a unique opportunity to modernize farming for the state’s most pungent export.
“Robotics and automation have been a big driver in major commodity crops like corn,” says Dr. Shajahan. “But Illinois also leads in production for certain specialty crops, like horseradish and pumpkin. These specialty crops are often overlooked by big ag companies, and mechanization is close to zero.”
Recognizing this gap, Dr. Shajahan’s team sat down with local horseradish growers to assess their immediate needs. “Currently, horseradish farmers rely very heavily on manual labor for weeding, which is at a shortage in the United States,” Poudel explains.
Chemical herbicides, as it turns out, are also not an option. “Horseradish is a below-ground crop. With herbicide applications, there is a concern of contamination, so there are very limited chemical solutions to address weeds in horseradish.”
With limited labor and no chemical solutions, horseradish growers in Illinois turned to the Center for Digital Agriculture with a need for an autonomous solution to manage weeds in a crop that had seen very little technological progress. Their request aligned directly with the goals of the USDA-funded AIFARMS (Artificial Intelligence for Future Agricultural Resilience, Management, and Sustainability), whose technology adoption thrust focuses on turning real production challenges into new innovations.
With support from AIFARMS, the project progressed quickly. Today, Poudel is leading research efforts to develop an autonomous robotic system for removing weeds in horseradish fields.
Because the project is targeted towards a smaller group of Illinois growers, the team has built strong relationships within that niche community. “We’re developing something exclusively for about 20 to 25 farmers in Southern Illinois,” Shajahan says. “They are all very supportive. They let us run experiments in their fields, store our robot, even help move it around. These partnerships are what help us create a field-ready solution.”
While some may have hesitated to take on a project targeting a specialty crop like horseradish, Shajahan and Poudel saw it as an opportunity. “It was actually a very good space for us,” Shajahan says. “It demonstrated a gap we could fill as a research institution. Working on a small-scale, modular platform lets us build something meaningful that will have a larger impact within the community.”
The team hopes that their research will make advanced agricultural technology more attainable for smaller growers, envisioning a do-it-yourself robotic weeding system. “All the components can be purchased from the hardware store,” Shajahan says. “We’ll provide the source code and design files so farmers can scale it to fit their own operations.”
When it comes to data collection, Poudel explains that the team takes a two-step approach, combining ground and aerial perspectives to weed the field. “We collect data using both a robot and a drone,” he says. “The robot is equipped with two cameras: one for detecting weeds, and another for identifying crop rows so it can navigate through the field. We also fly a drone overhead to capture image data from above.”
The on-ground images are used to help the robot distinguish between crops and weeds, while the aerial data helps record the exact locations of weed-heavy hotspots in the area. “We use that geolocation data from aerial imagery to guide the robot to those spots where it can detect and remove the weeds.” Poudel explains.
Currently, the team is working on testing different computer vision models to assess which works best for distinguishing between weeds and crops. According to Poudel, the work can get complex. “The robot integrates several interdependent subsystems,” he says. “Weed detection, path planning, and the weeding mechanism. Ensuring that all these components operate smoothly in real time can be challenging, considering the variable nature of field work.”
Poudel is also no stranger to the logistical challenges of operating a robot outdoors. He laughs as he recalls their run-in with a bout of bad weather.
“We were out in the field flying the drone when it suddenly started raining hard,” he says. “The wind picked up, we had to rush to cover everything. Each one of us was holding a canopy pole so it wouldn’t blow away, while trying to protect the robot and the drone from the rain at the same time.”
The team’s robot was manufactured by farm-ng, a robotics company which also served as Poudel’s sponsor for the FFAR (Foundation for Food and Agricultural Research) fellowship, which he recently received in summer of 2025.
The 3-year program provides mentorship, professional development, and career guidance to food and agricultural scientists across the United States.
“You get the opportunity to build a large network of researchers and students from different universities and agricultural backgrounds,” Poudel says. “It’s been a really great experience.”
Poudel believes that establishing professional goals early on is a key for success as a graduate student. “Agriculture is a very broad and multidimensional field,” he says. “And on top of that, artificial intelligence is growing. The opportunities for exploration are expanding, so identifying your research focus and long-term career direction becomes essential.”
Poudel is already charting the next steps in his research journey. Just recently, he was selected as a recipient for the 2025 Digital Green Talent award by The German Federal Ministry of Research, Technology and Space. This award, created to promote international collaboration on digitization and sustainable development projects, will allow him to spend three months conducting research in Germany next summer.
“It’s a big recognition,” says Shajahan. “The emphasis in his project on small-scale farming is part of what helped him achieve this.” he pauses, then laughs. “He’s taking the horseradish back to Germany.”
More than 150 years after horseradish was first brought to Illinois, Poudel will make the journey back to its homeland, equipped with modern agricultural expertise and a vision for the future of specialty crop farming.
“We’ll be really happy if we can provide a 100% working solution for horseradish farmers to remove weeds,” Poudel says. “If it’s successful in horseradish, we can adapt it for other specialty crops as well. There’s really no boundary to how far this can go.”
Learn More and Get Involved
Visit the Digital Precision Agriculture Lab to learn more about their research.
Contact the Office of Data Science Research if you’re aware of other people or resources we could feature here. ODSR is a campuswide convening organization that facilitates collaborations, resource sharing, and public engagement focused on data science research activities at the University of Illinois.