Designing Pin-pression Gripper and Learning its Dexterous Grasping with Online In-hand Adjustment

Hewen Xiao1,†, Xiuping Liu1,†, Hang Zhao2,†, Jian Liu3,*, Kai Xu2
1Dalian University of Technology     2National University of Defense Technology     3Shenyang University of Technology
†Equal Contribution     *Corresponding author
ACM Transactions on Graphics (SIGGRAPH 2025), 44(4).

The proposed pin-pression gripper offers object adaption and in-hand re-orientation through dynamically adjusting the extension and retraction of pins. The target object is successfully grasped and securely lifed with online pin adjustments on both fingers (shown on two sides) throughout the grasping process. As a comparison, we show the grasping of a parallel-jaw gripper (WSG50). The comparison of success rates tested on various datasets is also reported.

Abstract

We introduce a novel design of parallel-jaw grippers drawing inspiration from pin-pression toys. The proposed pin-pression gripper features a distinctive mechanism in which each finger integrates a 2D array of pins capable of independent extension and retraction. This unique design allows the gripper to instantaneously customize its finger’s shape to conform to the object being grasped by dynamically adjusting the extension/retraction of the pins. In addition, the gripper excels in in-hand re-orientation of objects for enhanced grasping stability again via dynamically adjusting the pins. To learn the dynamic grasping skills of pin-pression grippers, we devise a dedicated reinforcement learning algorithm with careful designs of state representation and reward shaping. To achieve a more efficient grasp-while-lift grasping mode, we propose a curriculum learning scheme. Extensive evaluations demonstrate that our design, together with the learned skills, leads to highly flexible and robust grasping with much stronger generality to unseen objects than alternatives. We also highlight encouraging physical results of sim-to-real transfer on a physically manufactured pin-pression gripper, demonstrating the practical significance of our novel gripper design and grasping skill.

Paper

Paper PDFPaper

Slides

Paper Slides

Supplementary/Data

Paper Supplementary

Method

Pipeline Image

Overview of our learning-based approach:

Our method obtains the current state information about the object, gripper, and their interaction to predict the appropriate action that moves the pins of the gripper finger in GtL or GwL grasping modes. Afer executing the online predicted action, the updated state is then passed through the same pipeline to predict the next action, so that a successful grasp is gradually formed.

Gripper Image

The specific constitution of the parallel pin-pression gripper:

The gripper finger is a 2D array of pins capable of independent extension and retraction. Each pin consists of a cylinder and a sphere. The gripper achieves full closure when every finger reaches its maximum forward movement.

Physical Results

Physical experiment of sim-to-real transfer. From (a) to (f) show several key frames of the dynamic grasping processes of our pin-pression gripper on five object models. The gripper approaches the target object from the top, forms a basic closure against the object with pin movements, and lifs the object while further performing in-hand re-orientation to improve the grasping stability.

Other Results

Video Presentation(Coming soon)

Poster

Thanks

This work was supported in parts by NSFC (62325211, 62495081, 62272082, 12494554, 62132021), the Joint Fund General Project of Liaoning Provincial Department of Science and Technology (2024-MSLH-352), and the Major Program of Xiangjiang Laboratory (No.23XJ01009)

BibTeX


      @article{hewen2025,
        title={Designing Pin-pression Gripper and Learning its Dexterous Grasping with Online In-hand Adjustment},
        author={Hewen Xiao and Xiuping Liu and Hang Zhao and Jian Liu and Kai Xu},
        journal={ACM Transactions on Graphics (SIGGRAPH 2025)},
        volume = {44},
        number = {4},
        year = {2025}
      }