Master of Engineering Science
Electrical and Computer Engineering
Dr. Mehrdad Kermani
The agricultural industry is developing autonomous systems to sustain the demands of global population growth. There are many challenges associated with the development of autonomous systems for the agricultural industry because of their dynamic and constrained environments. An example is the mushroom harvesting industry as it requires an indoor, dark, and highly humid environment for the rapid growth of mushrooms on narrowly stacked compost beds. Manual labour is currently the only acceptable method of harvesting mushrooms, but overtime, the harsh conditions cause high worker turn-over rates, driving the cost of manual labour up while overall reducing the potential of the industry.
A mushroom scanner was developed to scan a mushroom bed in real-time and to provide the data to a recently developed mobile harvesting unit, designed to pick mushrooms using a robotic claw, which up until now, lacked the ability to find them. The scanner can precisely determine the 3D position and orientation of all visible mushrooms using a custom real-time image processing algorithm. The algorithm then filters the data for target sized mushrooms and sends commands to the harvesting unit for picking with high precision. The mushroom scanner system was designed, developed, and tested in both laboratory and industrial settings, and has proven its potential in the mushroom harvesting industry.
Glibetic, Stefan, "Development of a Real-Time 3D Mushroom Vision System for Autonomous Mushroom Harvesting" (2017). Electronic Thesis and Dissertation Repository. 5161.
Available for download on Saturday, August 31, 2019