Dissertation
FUSED DEPOSITION MODELING OF A SOFT-ROBOTIC END-EFFECTOR FOR IMPROVED ROBOTIC APPLE HARVESTING
Doctor of Philosophy (PhD), Washington State University
01/2018
Handle:
https://hdl.handle.net/2376/108572
Abstract
Over the last century, the agriculture industry has seen tremendous growth as the world population continues to grow and mechanization in all areas of the farming cycle improve worker efficiency allowing a minimal labor force to cultivate more and more land to produce crops. This dissertation aims to utilize additive manufacturing technology, specifically Fused Deposition Modeling (FDM), to rapidly produce soft-robotic actuators and sensors that do not require casting or molding and operate as functional units to build a robotic end-effector for agricultural applications.
Several laboratory studies were conducted to characterize the mechanical and electrical properties of thermoplastic polyurethane (TPU) and thermoplastic polyurethane-multiwalled carbon nanotube composites (MWCNT) parts produced using FDM techniques, after which a functional actuator was designed, printed, and assembled into an end-effector to harvest apples. The soft robotic end-effector, fitted to a custom five degree of freedom robotic arm, offers several improvements over previous grasping end-effectors. The compliant actuators were unharmed in collisions with the tree canopy and trellis wires and offered a significant improvement in grasping speed over previous tendon-driven fingers, closing in a full order of magnitude (milliseconds versus seconds) faster. A secondary robot was utilized to catch and convey apples from the harvesting location to the storage bin.
Sensors were printed into the end-effector based on the conductive properties of TPU-MWCNT. Touch sensors were printed as the base of the actuator to detect contact with the apple. Flex sensors based on the piezoresistive properties of TPU-MWCNT were used to predict the inlet pressure, and therefore curvature of the actuators. Finally, a proximity sensor using a capacitive highly sensitive touch circuit was embedded into the end-effector palm to avoid collision and adapt to minor position errors.
Field testing was performed in a commercial apple orchard in Prosser, Washington over four days in the fall of 2017. The first two days of the experiment was conducted with Jazz apples, whereas experiments on days three and four were with Envy apples. An overall harvesting success rate of 67% was achieved with an average of 7.3 seconds per fruit from separation to fruit storage.
Metrics
13 File views/ downloads
22 Record Views
Details
- Title
- FUSED DEPOSITION MODELING OF A SOFT-ROBOTIC END-EFFECTOR FOR IMPROVED ROBOTIC APPLE HARVESTING
- Creators
- Cameron J Hohimer
- Contributors
- Changki Mo (Advisor)Aboutaleb Ameli (Committee Member)Joseph Iannelli (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Mechanical and Materials Engineering, School of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 125
- Identifiers
- 99900581621701842
- Language
- English
- Resource Type
- Dissertation