Harvesting End-effector Design and Picking Control
Jonah Mapes, Andong Dai, Yunjun Xu, Shinsuke Agehara
- 发表年份
- 2021
- 引用次数
- 7
摘要
As an enabling technology in robotic harvesting, an end-effector requires precise positioning and picking without bruising strawberries or sabotaging strawberry beds. Here, an end-effector design with a spiral-curve tooth is presented, which can contain strawberries in a spherical shell before pulling them away from their stems. Visual feedback with a camera-in-hand configuration is used to identify strawberries of sufficient size and ripeness. A two-step estimation and control approach is proposed including aligning the end-effector with the target strawberries in the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$X$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$Y$</tex> directions and then reaching them along the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$Z$</tex> direction. Specifically, the control logic is designed to correct the depth error based on the changing rate of a strawberry diameter in the pixel frame. The unique feature of this approach is that it can precisely move the end-effector to the desired position regardless of the initial depth and size of the target. The end-effector, XYZ table, and control algorithm were extensively tested in an artificial strawberry bed and also partially experimented in a commercial farm.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991