Whether the intention is to assist, replicate, or replace human labor, commercial robotics are emerging in response to labor needs. In commercial greenhouses, smart harvesting robots can help fill labour gaps, completing long hours of tedious and strenuous work that until recently, have mostly been people powered.
Arrell Scholar alumnus Evan Tollenaar helped create one such robot capable of automated harvesting in a greenhouse setting using artificial intelligence (AI). As a recent graduate student under the supervision of Dr. Medhat Moussa, Evan’s research focused on how robotics and AI can aid monitoring and harvesting of greenhouse-grown beefsteak tomatoes.
Training a robot to complete a certain task requires a specific AI algorithm; a robot designed to harvest tomatoes would not be able to harvest lettuce, for example. Harvesting tomatoes is also uniquely challenging due to the soft, easily damaged flesh of ripe tomatoes. Evan’s graduate work is the first to tackle these issues, as there are no commercial robotics yet available to address autonomous beefsteak tomato harvesting.
Evan’s work is part of a larger wave of intelligent robotics increasingly used to address food waste and labour shortages in the global food industry. Automated harvesting robots work independently and can handle produce more gently than humans while still being fast. Combined, these features increase harvest volume while reducing the amount of damaged produce that would otherwise go to waste.
With proper AI, the use of robotics can also offer cost savings. Robots can increase the efficiency of a food production system by avoiding mistakes from human error and lowering the cost of production, which could reflect in lower prices for consumers.
Since completing his research, Evan is working currently as the Head of Operations for FI Equipment, bringing his experience as an Arrell Scholar into his role in the ag-tech industry. Evan shared that communicating the details of his past research in plain language has easily translated into conveying complex technical information to industry partners, investors, and customers. Evan also says that connections with other Arrell Scholars and mentors have allowed him to approach both his research and current work with a broader systems lens.
As for the automated smart harvesting robot, next steps involve fine-tuning a more robust commercial model that can soon hold its own on any greenhouse floor.