Abstract
This paper presents a new method to evaluate multiple end effectors for a given task, which was applied to four custom-designed robotic end effectors for apple fruitlet thinning in a controlled laboratory environment. These end effectors’ specific mechanisms are stem cutting (cutting), grasp and roll (paraboloid), suction and twist (suction), and piercing and retract (piercing). An artificial fruitlet structure that replicates real apple fruitlets critical characteristics and suitable for different mechanisms of tested end effectors was used to ensure a consistent test platform across all end effectors. Each end effector was tested on single fruitlet at approach angles varying from 90 to 30 ° and five different positions of targeted fruitlet within the cluster at approach angles varying from 90 to 60 ° (cluster). A vision system and path planning system were integrated to facilitate the evaluation. The results showed that all effectors except piercing are capable to remove fruitlets from 45 to 90 ° for single fruitlet, and suction performed best in cluster evaluation. Cutting, paraboloid and suction end effectors were deemed feasible for further development and real field evaluation.
| Original language | English |
|---|---|
| Title of host publication | Australasian Conference on Robotics and Automation, ACRA |
| Publication status | Published - 2024 |
| Externally published | Yes |
| Event | 2024 Australasian Conference on Robotics and Automation, ACRA 2024 - Auckland, New Zealand Duration: 27 Nov 2024 → 29 Nov 2024 |
Conference
| Conference | 2024 Australasian Conference on Robotics and Automation, ACRA 2024 |
|---|---|
| Country/Territory | New Zealand |
| City | Auckland |
| Period | 2024-11-27 → 2024-11-29 |
Bibliographical note
Publisher Copyright:© 2024 Australasian Robotics and Automation Association. All rights reserved.
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver