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 Experiment Video

Updated: May 28, 2026

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

Hi-RAGrasp: A Human-in-the-Loop Experience-Augmented Method for Task-Oriented Grasping.

Yaxin Liu1, Yue Hu1, Yan Liu1

  • 1State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China.

Sensors (Basel, Switzerland)
|May 27, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Intention Reasoning for User Action Sequences via Fusion of Object Task and Object Action Affordances Based on Dempster-Shafer Theory.

Sensors (Basel, Switzerland)·2025
Same author

Object Affordance-Based Implicit Interaction for Wheelchair-Mounted Robotic Arm Using a Laser Pointer.

Sensors (Basel, Switzerland)·2023
Same author

Efficient Delivery of Antisense Oligonucleotides by an Amphipathic Cell-Penetrating Peptide in Acinetobacter baumannii.

Current drug delivery·2019
Same author

Liquid Nitrogen Activation of Zero-Valent Iron and Its Enhanced Cr(VI) Removal Performance.

Environmental science & technology·2019
Same author

A self-assembled perylene diimide nanobelt for efficient visible-light-driven photocatalytic H<sub>2</sub> evolution.

Chemical communications (Cambridge, England)·2019
Same author

Clinical value of MLH1-negative circulating tumor cells in lung cancer patients.

Medicine·2019

Hi-RAGrasp enables assistive robots to grasp diverse household objects by combining multi-stage reasoning with human feedback. This task-oriented grasping framework improves robot manipulation in complex, real-world scenarios.

Area of Science:

  • Robotics
  • Artificial Intelligence
  • Human-Robot Interaction

Background:

  • Assistive robots are crucial for aging societies, requiring robust task-oriented grasping in unstructured household environments.
  • Traditional grasping methods lack semantic understanding and human intervention integration, limiting performance in variable conditions.
  • Existing approaches struggle with unstable reasoning and effective incorporation of human feedback.

Purpose of the Study:

  • To develop a novel task-oriented grasping framework, Hi-RAGrasp, for improved robot manipulation in household settings.
  • To integrate progressive multi-stage reasoning, Human-in-the-Loop (HITL) interaction, and Retrieval-Augmented Generation (RAG) for enhanced grasping capabilities.
  • To address challenges in semantic constraint understanding, reasoning stability, and human intervention in robotic grasping.
Keywords:
affordance segmentationhuman-in-the-loopretrieval-augmented generationtask-oriented grasping

More Related Videos

Design and Use of an Apparatus for Presenting Graspable Objects in 3D Workspace
09:11

Design and Use of an Apparatus for Presenting Graspable Objects in 3D Workspace

Published on: August 8, 2019

Frame-by-Frame Video Analysis of Idiosyncratic Reach-to-Grasp Movements in Humans
10:51

Frame-by-Frame Video Analysis of Idiosyncratic Reach-to-Grasp Movements in Humans

Published on: January 15, 2018

Related Experiment Videos

Last Updated: May 28, 2026

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

Design and Use of an Apparatus for Presenting Graspable Objects in 3D Workspace
09:11

Design and Use of an Apparatus for Presenting Graspable Objects in 3D Workspace

Published on: August 8, 2019

Frame-by-Frame Video Analysis of Idiosyncratic Reach-to-Grasp Movements in Humans
10:51

Frame-by-Frame Video Analysis of Idiosyncratic Reach-to-Grasp Movements in Humans

Published on: January 15, 2018

Main Methods:

  • A coarse-to-fine pipeline refines predictions from object localization to part-level grounding for precise manipulation.
  • A HITL correction mechanism and a structured human experience database are combined with RAG for adaptive reasoning.
  • A Geometric Heuristic Segmentation (GHS) method is proposed for improved localization of textureless objects.

Main Results:

  • The Hi-RAGrasp framework achieved a 77.73% segmentation success rate on the evaluation dataset.
  • Real-world experiments demonstrated a 75% grasp success rate, significantly outperforming existing methods.
  • The system effectively maps human instructions to fine-grained, task-relevant regions, enabling robust grasping.

Conclusions:

  • Hi-RAGrasp offers a practical and effective solution for task-oriented grasping in open, unstructured environments.
  • The integration of HITL and RAG facilitates experience reuse and accumulation without retraining.
  • The proposed framework shows strong potential for advancing assistive robotics in aging societies.