“Working with Eric felt like working with a peer rather than supervising a student” – Fabian Gilson, nominator.

​

Working on a real-world engineering project to more effectively harvest pine pollen from New Zealand’s Pinus Radiata stocks, Eric Song impressed the ENVIs judges with his well-delivered pitch, iterative approach and validation testing, and creative use of a combination of existing technologies. 

​

Pine pollen has been used in traditional medicine for thousands of years. Recently there has been renewed interest in using pine pollen as a supplement to support the natural production of sex hormones. New Zealand-based companies aim to capitalise on this interest by harvesting and selling pollen from our abundance of Pinus Radiata. Currently, pine pollen is harvested by hand – restricting how much pollen can be harvested, as the parts of the tree that release the pollen (catkins) can be high up in the tree and are only ripe for a short period. We are working on an autonomous catkin-harvesting UAV to remedy this problem. 

​

Enabling a UAV to harvest catkins autonomously requires detecting and tracking the precise location of each catkin using an onboard camera - a task well suited to machine learning. We assigned this ambitious task to Eric as his final-year project. Eric exceeded our expectations significantly, completing the initial requirements for the project well ahead of schedule. He then worked as part of the wider team integrating his software with the UAV, enabling it to harvest catkins autonomously. Eric continued to work for us as an employee over the summer, further improving the capabilities of the UAV. His ability, hard work and initiative made him a valued member of our team.  

Ashley Gutteridge

University of Canterbury 
 

Taine Whare

University of Waikato