Rummenie, Stacey Keith (2024) Control of Insect Pests Using Agricultural Robots - Development of an Insect Pest Eliminating End-Effector. [USQ Project]
![[img]](https://sear.unisq.edu.au/style/images/fileicons/text.png) |
Text (Project – redacted)
Rummenie_S_Thesis_Redacted.pdf Download (17MB) |
Abstract
The control of insect pests via pesticides is an issue of growing concern. While alternate methods exist, they are often slow, expensive or have issues with worker availability. Due to historical difficulties in the identification of insects using robotic visual systems, little work has been done in attempting to use robotics to help address these issues. With recent advances in robotics, this issue may soon be addressed, however, there are currently few end-effectors available for non-chemical control of insects. This is an area where further research is required. This project aimed to identify one or more non-chemical pest control approaches able to be integrated with robotics, through designing and testing a range of prototype end-effectors. Attempted approaches used for insect control by these prototypes included: application of a crushing force (the “Gripper” prototype); application of an electric field (the “Zapper” prototype); and application of a suction force (the “Sucker” prototype).
Prototype development occurred through use of an iterative, design-based methodology, with preliminary prototypes based loosely around devices of a similar nature. This was followed by viability testing, analysis and refinement of each of the three preliminary designs, so as to develop fully functional prototypes. The final prototypes were then subjected to three rounds of testing, including testing against Mealworms (Tenebrio molitor), Cabbage White Butterflies (Pieris rapae), and Ladybeetles (Family Coccinellidae).
The Gripper prototype, while showing poor effectiveness against killing larval Tenebrio molitor, proved highly effective in removing and killing pupal forms as well proving effective against larval forms of Pieris rapae and adult ladybeetles. From a build and host damage perspective this prototype showed good potential for integration with a robotic arm and limited potential for causing host plant damage. Results from the Gripper prototype were, furthermore, indicative of different control methodologies as being better suited to specific insect species and life cycle stages. While the Zapper prototype was the cheapest and smallest prototype, and ultimately proved to have some ability in killing both larval and pupal forms of Tenebrio molitor, it appeared far less effective against Pieris rapae and had very limited pest removal ability across all species tested. Though appearing well suited for integration with a robotic arm, this prototype was observed to cause damage to the host plant (Brassica oleancea) tissue and was prone to accidental discharge when in proximity to leaves. Testing against Tenebrio molitor (larvae & pupae) and Pieris rapae (larvae) proved the Sucker prototype to be extremely effective in removing and killing these specimens without damaging the host plant. This prototype was, however, noted to be the most expensive and largest of those examined, with significant work needed before it could be attached to a robotic arm.
In summary, the project demonstrated the Sucker prototype end-effector as having the most promise in controlling pest insects without chemicals, with the Gripper prototype also showing good potential. The Zapper prototype, while being the worst performer, showed sufficient potential to justify further investigation. Overall, the project effectively demonstrated use of non-chemical based control approaches, delivered via robot end-effectors, are a viable option for future insect pest control. With further development, such devices could significantly reduce the amount of pesticide currently used within agriculture, having the potential to provide great benefit to the health of agricultural workers, the community, and the environment. In consideration of recent advances in robotics with respect to machine vision and pest identification, the successful progression of end-effector technologies achieved by this project provides clear justification for the continuation of further work in this field.
|
Statistics for this ePrint Item |
| Item Type: | USQ Project |
|---|---|
| Item Status: | Live Archive |
| Faculty/School / Institute/Centre: | Current – Faculty of Health, Engineering and Sciences - School of Engineering (1 Jan 2022 -) |
| Supervisors: | Low, Tobias |
| Qualification: | Bachelor of Engineering (Honours) |
| Date Deposited: | 18 Mar 2026 03:14 |
| Last Modified: | 18 Mar 2026 03:14 |
| Uncontrolled Keywords: | insect pests; robotic visual systems; non-chemical control |
| URI: | https://sear.unisq.edu.au/id/eprint/53160 |
Actions (login required)
 |
Archive Repository Staff Only |
