Benchmarking via Competitions in

Robotic Grasping and Manipulation

2nd Edition

______________

20th October 2025, 8:30 - 13:00 am (local time)

at 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems

Hangzhou, CHINA

Grand Ballroom A | Room 311A

Summary

Advances in robotics, perception, and artificial intelligence are driving the expectation of robots having high-level capabilities to autonomously, accurately, and robustly address various tasks. However, crucial robotic capabilities such as autonomous assembly and disassembly, cloth manipulation, in-hand manipulation, object picking in cluttered environments, or object hand-over are far from being solved. Many research activities have focused on these subject topics over the last decades, but solutions remained limited, and results are demonstrated on custom workflows and validation tests designed in individual laboratories considering distinct modalities and performance criteria. For this reason, benchmarking is important to enable reproducibility and comparison across different solutions and to advance the developments in challenging robotic manipulation problems. In addition, competitions have been an effective way to drive scientific progress in recent years, spurring the advances in robotics research. This workshop aims to stimulate discussions within the research community from both academia and industry on the current and future challenges of autonomous robotic grasping and manipulation for real-world applications where research innovation meets engineering solutions for developing efficient and elegant systems in relevant environments outside the laboratory with particular attention to benchmarking and competitions. To this end, we will engage with experts on the tasks and the winners of recent competitions. We will draw guidelines and directions for the reproducibility and comparison of future solutions and new challenges that arise from the discussion.

List of accepted Extended Abstracts (unpublished works)

See https://sites.google.com/view/iros2025-bench-competitions/contributions/extended-abstracts for details and PDFs.

Best extended abstract will be awarded $250
sponsored by the TC on Performance Evaluation & Benchmarking of Robotic and Automation Systems

No Title and Authors

1 Equivariant Triplane Representation for Volumetric Grasping

P. Song, Y. Hu, P. Li, R. Detry

2 RGB-SQ Grasp: Inferring Local Superquadric Primitives from Single RGB image for Graspability-Aware Bin Picking [Best extended abstract]

Y. Xu, F. Zhu, Y. Li, X. Huang, Y. Chen

3 Hybrid Gripper and Adaptive Strategy for Robust Grasping in Clutter: RGMC Champion Solution

S. Wang, C. Xin, Z. Zhang, G. Ma, X. Li

4 Towards Reliable Sequential Object Picking in Clutter: The Runner-up Solution to RGMC 2025

W. Yu, X. Zhang, Z. Zheng, W. Kong, H. Dong

5 DiffPort: Adapting Pre-trained Diffusion Models for Generalizable Robot Manipulation

C. Zhu, J. Seo, A. Ayouch, E. Senft, S. Kim, C. Oh

6 Object Pose and Shape Estimation for Grasping: Does it Work?

P. Karke, K. Shah, G. Singh, M. F. Karim, M. Krishna, R. Talak

7 Vision and Language-Guided Modular Task Reconfiguration with MoveIt Task Constructor: A Robothon Grand Challenge Solution

A. Al Abbas, D. Das, A. Ali, C. Murgia, E. Kerr, P. Greaney, S. Redmond, P. Long

List of accepted Abstracts (published works)

See https://sites.google.com/view/iros2025-bench-competitions/contributions/abstracts for details.

No Title and Authors

1 Flat'n'Fold: A Diverse Multi-Modal Dataset for Garment Perception and Manipulation

L. Zhuang, S. Fan, Y. Ru, F. Audonnet, P. Henderson, G. Aragon-Camarasa

2 GraspMamba: A Mamba-Based Language-Driven Grasp Detection Framework with Hierarchical Feature Learning

H. H. Nguyen, A. Vuong, A. Nguyen, I. Reid, M. N. Vu

3 Benchmarking of robot arm motion planning in cluttered environments

S. Yang; P. Liu; N. Pears

4 FLAME: A Federated Learning Benchmark for Robotic Manipulation

S. B. Betran, A. Longhini, M. Vasco, Y. Zhang, D. Kragic

5 RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning [Best abstract]

H. Geng, F. Wang, S. Wei, Y. Li, B. Wang, B. An, C. T. Cheng, H. Lou, P. Li, Y.-J. Wang, Y. Liang, D. Goetting, C. Xu, H. Chen, Y. Qian, Y. Geng, J. Mao, W. Wan,
M. Zhang, J. Lyu, S. Zhao, J. Zhang, J. Zhang, C. Zhao, H. Lu, Y. Ding, R. Gong, Y. Wang, Y. Kuang, R. Wu, B. Jia, C. Sferrazza, H. Dong, S. Huang, Y. Wang,
J. Malik, P. Abbeel

Invited Speakers

Xingdou Fu
Omron

Anh Nguyen

University of Liverpool

Minas Liarokapis

University of Auckland

and ACUMINO

Luis Figueredo

University of Nottingham

Yik Lung Pang
Queen Mary University of London

Winners of the 10th Robotic Grasping and Manipulation Competition (RGMC) tracks at ICRA 2025

Xiang Li

Winning team
Picking in Clutter

Xidan Zhang

Winning team
In hand manipulation

Program (local time)

8:30 - 8:40 Welcome and introduction

Salvatore D'Avella

[Session 1] Human-to-Robot Handover

8:40 - 8:50 Stereo Hand-Object Reconstruction for Human-to-Robot Handover

Yik Lung Pang, Queen Mary University of London

8:50 - 9:10 Beyond the Grip: Safety and Comfort in Physical Human-Robot Collaborative Grasping and Manipulation

Luis Figueredo, University of Nottingham

9:10- 9:20 Q&A

[Session 2] Grasping Heterogeneous Objects In Cluttered Environments

9:20 - 9:40 Fast Active Sensing Framework for Picking Under Challenging Conditions

Xingdou Fu, Omron

9:40 - 9:50 Towards Model-Free Universal Dexterous Grasping

Xiang Li, Tsinghua University

Representative of the winning team for the 2025 ICRA RGMC - Grasping in Clutter track

9:50 - 10:00 Q&A

[Session 3] Best abstract and extended abstract

10:00 - 10:10 RGB-SQ Grasp: Inferring Local Superquadric Primitives from Single RGB image for Graspability-Aware Bin Picking

Y. Xu, F. Zhu, Y. Li, X. Huang, Y. Chen

10:10 - 10:20 RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning

Haoran Geng, Feishi Wang, Songlin Wei, Yuyang Li, Bangjun Wang, Boshi An, ....

... Siyuan Huang, Yue Wang, Jitendra Malik, Pieter Abbeel

10:20 - 10:30 Q&A

10:30 - 11:00 Poster Session 1

[Session 4] In-hand Manipulation

11:00 - 11:20 Human to Robot Skill Transfer for Reliable, Dexterous In-Hand Manipulation: From Industrial Assembly to Home Robotics

Minas Liarokapis, University of Auckland

11:20 - 11:30 Manipulating objects in complex scenarios: Competition-Scale Solutions for In-Hand Reconfiguration and Cluttered Picking

Xidan Zhang , Zhejiang University

Representative of the winning team for the 2025 ICRA RGMC - In-hand manipulation track

11:30 - 11:40 Q&A

[Session 5] LLM in Robotics Grasping

11:40 - 12:00 Language-driven Grasping

Anh Nguyen, University of Liverpool

12:00 - 12:30 Panel Discussion

(Salvatore D'Avella)

12:30 - 13:00 Poster Session 2

13:00 End

Workshop organizers

Salvatore D'Avella

Assistant Professor

Alessio Xompero

Postdoctoral Researcher

Changjae Oh

Lecturer

Andrea Cavallaro

Full Professor

Kaiyu Hang

Assistant Professor

Berk Calli

Associate Professor

Yu Sun

Full Professor

Contact

For further questions please send an email to salvatore.davella@santannapisa.it

The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.