Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Automatic Processing and Automatic Social Behavior01:28

Automatic Processing and Automatic Social Behavior

Automatic processing refers to the cognitive operations that occur without conscious intent or awareness, playing a fundamental role in shaping social cognition and behavior. These processes enable individuals to navigate complex social environments efficiently by relying on mental shortcuts and pre-existing knowledge structures known as schemas. One of the most influential mechanisms underlying automatic processing is priming, which subtly activates mental representations through exposure to...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Acute Anterior Choroidal Artery Territory Infarction: A Retrospective Study.

Clinical neurology and neurosurgery·2020
Same author

Preference Ranking Procedure: Method Validation with Dogs.

Animals : an open access journal from MDPI·2020
Same author

Controlling defects in crystalline carbon nitride to optimize photocatalytic CO<sub>2</sub> reduction.

Chemical communications (Cambridge, England)·2020
Same author

Fabrication and Fireproofing Performance of the Coal Fly Ash-Metakaolin-Based Geopolymer Foams.

Materials (Basel, Switzerland)·2020
Same author

High-Resolution Chest X-Ray Bone Suppression Using Unpaired CT Structural Priors.

IEEE transactions on medical imaging·2020
Same author

Facile green synthesis of calcium carbonate/folate porous hollow spheres for the targeted pH-responsive release of anticancer drugs.

Journal of materials chemistry. B·2020

Related Experiment Video

Updated: Jul 2, 2026

Investigating Motor Skill Learning Processes with a Robotic Manipulandum
07:52

Investigating Motor Skill Learning Processes with a Robotic Manipulandum

Published on: February 12, 2017

Target-conditioned flow-matching policy for citrus harvesting robot pre-grasp approach behavior learning.

Luo Lei1, Han Li1, Zhijiang Zuo1

  • 1School of Intelligent Manufacturing, Jianghan University, Wuhan, China.

Frontiers in Plant Science
|July 1, 2026
PubMed
Summary

This study introduces a Target-Conditioned Flow-Matching Policy (TCFM Policy) for robotic citrus harvesting, improving fruit selection and approach accuracy in cluttered orchards with limited data. The TCFM Policy achieves a 76% success rate, demonstrating its effectiveness in real-world scenarios.

Keywords:
behavior learningcitrus harvesting robotflow matchingimitation learningmulti-step trajectory predictiontarget-conditioned policy

More Related Videos

Evaluating Skilled Prehension in Mice Using an Auto-Trainer
05:01

Evaluating Skilled Prehension in Mice Using an Auto-Trainer

Published on: September 12, 2019

Acquisition of a High-precision Skilled Forelimb Reaching Task in Rats
08:59

Acquisition of a High-precision Skilled Forelimb Reaching Task in Rats

Published on: June 22, 2015

Related Experiment Videos

Last Updated: Jul 2, 2026

Investigating Motor Skill Learning Processes with a Robotic Manipulandum
07:52

Investigating Motor Skill Learning Processes with a Robotic Manipulandum

Published on: February 12, 2017

Evaluating Skilled Prehension in Mice Using an Auto-Trainer
05:01

Evaluating Skilled Prehension in Mice Using an Auto-Trainer

Published on: September 12, 2019

Acquisition of a High-precision Skilled Forelimb Reaching Task in Rats
08:59

Acquisition of a High-precision Skilled Forelimb Reaching Task in Rats

Published on: June 22, 2015

Area of Science:

  • Robotics
  • Computer Vision
  • Machine Learning

Background:

  • Fruit harvesting in natural orchards is challenging due to cluttered environments and unstructured fruit distribution.
  • Key issues in multi-fruit harvesting include limited data, target ambiguity, and the need for precise approach motions.

Purpose of the Study:

  • To develop a novel policy for precise pre-grasp approach in robotic citrus harvesting.
  • To address challenges of limited data, target ambiguity, and motion stability.

Main Methods:

  • Proposed a Target-Conditioned Flow-Matching Policy (TCFM Policy) integrating image observations, robot state history, and target geometry.
  • Utilized a dual-branch visual representation for global context and local end-effector details.
  • Introduced a target-oriented visual augmentation strategy to mitigate overfitting.

Main Results:

  • Achieved a 76% success rate in 50 target-specified multi-fruit trials on a UR5 platform.
  • Demonstrated a 4% target-picking error rate and 20% picking-point offset rate.
  • Outperformed a diffusion-policy baseline in offline trajectory error and online approach performance.

Conclusions:

  • Explicit target conditioning, dual-branch visual encoding, and conditional flow matching enhance target selection and pre-grasp approach stability.
  • The TCFM Policy is effective in small-sample, multi-fruit citrus harvesting scenarios.
  • The method shows promise for improving autonomous harvesting systems.