For generations, farmers have had two ways to tackle weeds: spending hours pulling them out by hand or relying on herbicides that can affect both the environment and the food we eat. Researchers at the Aga Khan University's Arusha Climate and Environmental Research Centre (AKU-ACER) are exploring another approach: using artificial intelligence (AI) to identify weeds and eliminate them with precision lasers rather than chemical herbicides.
Working with Simon Fraser University (SFU) in Canada, the team has been testing LiteWeed, an AI-powered robotic platform that precisely destroys weeds while leaving crops untouched.
Professor Emmanuel Sulle, Director of AKU-ACER, still remembers the first time he watched it move across a field in Arusha. "The first thing I thought was, 'Wow'. It could identify the weeds, destroy them and leave the spinach untouched."
His first reaction was amazement. His second was a question: could a machine developed in Canada work for farmers in East Africa?
That question has influenced the partnership ever since.
Since October 2022, AKU-ACER has hosted 10 field schools in Arusha, bringing together more than 300 students from different specialisations, including mechatronic engineering students from SFU, to test agricultural technologies alongside local farmers. Rather than arriving with ready-made solutions, students spend time understanding how farmers work, the challenges they face and the realities of everyday farming.
"It is not enough to build technology," says Professor Sulle. "You have to understand the culture, the priorities and the needs of the people you want to serve."
Farmers, researchers and Maasai community leaders all contribute to the work, including respected elder Mzee Alais Morindat, who now serves as an adjunct faculty member, ensuring local knowledge guides the research.
Field trials also challenged some assumptions. Arusha's hilly terrain made it difficult for drones to operate consistently, while hard soil sometimes had to be irrigated before equipment could collect samples. Conditions that appeared straightforward in the laboratory proved far more demanding in the field.[ES1]
Some questions remain unanswered. Although the robot destroys weeds without harming nearby crops, researchers want to know whether repeated laser exposure could affect the nutritional quality of vegetables. They are also examining how drones can reduce farmers' exposure to pesticides without allowing chemicals to drift into neighbouring fields.
The robot may be an engineering achievement, but Professor Sulle sees a much bigger challenge. Agriculture remains one of East Africa's largest employers, and introducing autonomous machines too quickly could come at the expense of the people who depend on farming for a living. Rather than focusing solely on automation, the team is exploring technologies such as drones that monitor soil health and nutrient levels, helping farmers make better decisions while keeping people at the centre of the work.
AKU is leading a proposal with Aga Khan Foundation Canada and Global Affairs Canada to expand the partnership, bringing together universities in Canada and Tanzania to train farmers, support local technology companies and strengthen policies that encourage the responsible use of emerging technologies.
Years from now, Sulle hopes people will remember more than the robot itself. He hopes they will remember that the best innovations are built with the communities they are meant to benefit.
"Technology is for the people; it is the social scientists who help contextualise that technology and make it meaningful to the people who will use it."
For AKU-ACER, every new idea begins with the same question: not simply whether it works, but whether it works for the people who need it most.
