/Functionalgrasp

Functionalgrasp: Learning Functional Grasp for Robots via Semantic Hand-Object Representation (RA-L 2023)

Primary LanguagePython

Functionalgrasp: Learning Functional Grasp for Robots via Semantic Hand-Object Representation (RA-L 2023)

This is the official code of FunctionalGrasp_Learning_Functional_Grasp_for_Robots_via_Semantic_Hand-Object_Representation.

Introduction

Successful grasp is an important and long-standing issue for robots to interact with the real world. Most recent studies have devoted more attention to stable grasp rather than functional grasp, which cannot guarantee task-oriented postgrasp manipulation. To achieve human-like functional grasp, a semantic representation of functional hand-object interaction is introduced without labeling 3D hand poses, and a novel coarse-to-fine grasp generation network is designed to model this hand-object interaction. First, a coarse grasp is generated by combining the global hand pose and hand grasp type. Then, the fine pose will be optimized by guiding each finger to focus on the corresponding functional region of the object. Experimental results demonstrate the effectiveness of our method in achieving functional grasps for dexterous hands in the absence of high-DoF grasp poses annotation of the hand.

Get started

Common Packages: we use: (other version will be available)

    conda create -n functionalgrasp python=3.6.7
    conda activate functionalgrasp
    pip install torch==1.10.1 numpy==1.19.5

In this project, the grasp synthesis dataset processing is based on KPConv, and we use GraspIt! to view the results. Please configure the development environment according to the instructions in the links.

Installation process:

  • Download this code and unzip it. Note: 'cpp_wrappers', etc. are from [KPConv].

  • GraspIt!: It is recommended to install the ubuntu version.

  • Follow Toward-Human-Like-Grasp: to install its environment.

  • Pointnet/Pointnet++: It is recommended to install the ubuntu version.

  • If your research continues to be based on the above projects, please directly cite the original work.

Two stage

  • First, obtain the grasp type, which code is in Grasptype_bruch
  • Second, functional grasp synthesis, which code is in GraspNet

Citation

If you find our work useful in your research, please consider citing:

@article{zhang2023functionalgrasp,
  title={FunctionalGrasp: Learning Functional Grasp for Robots via Semantic Hand-Object Representation},
  author={Zhang, Yibiao and Hang, Jinglue and Zhu, Tianqiang and Lin, Xiangbo and Wu, Rina and Peng, Wanli and Tian, Dongying and Sun, Yi},
  journal={IEEE Robotics and Automation Letters},
  year={2023},
  publisher={IEEE}
}