Introduction
Humans and animals demonstrate remarkable robustness in manipulation, yet robots remain far behind. Despite progress in planning, control, perception, hardware, and learning, robotic systems often fail under uncertainty and variation. This workshop takes on one of the most fundamental questions in robotics: how can we achieve manipulation robustness comparable to that of humans and animals?
The goal of this half-day ICRA 2026 workshop is to identify and connect the underlying principles that enable manipulation robustness. We will bring together researchers across diverse areas to discuss definitions, mechanisms, and evaluation of robustness. By fostering cross-disciplinary dialogue, we aim to move beyond isolated approaches toward a more unified understanding of manipulation robustness.
The program features seven invited talks spanning robust grasp synthesis, in-hand manipulation, robust policy learning, robust control theory, embodied intelligence for robust manipulation, etc. These will be complemented by lightning talks, a poster session, and a moderated panel debate.
Invited Speakers
All speakers have confirmed. Listed alphabetically.
Workshop Schedule
| Time (Local) | Event |
|---|---|
| 09:00 – 09:10 | Opening and Goals |
| 09:10 – 10:30 | Invited Talks 1–4 (20 min each incl. Q&A) |
| 10:30 – 11:00 | Posters and Coffee |
| 11:00 – 12:00 | Invited Talks 5–7 (20 min each incl. Q&A) |
| 12:00 – 12:30 | Panel Debate: Principles and Pathways to Human-Level Robustness |
| 12:30 – 12:40 | Awards and Closing Messages |
Call for Papers
This workshop aims to bring together researchers from across robotics—including control, planning, learning, perception, and hardware—who are advancing the study of manipulation robustness.
Accepted short papers will be presented as posters, with a selection also invited to give brief spotlight talks. Important: both poster and spotlight presentations require in-person attendance.
We are especially eager to highlight real-world robotic systems. Submissions featuring videos of robots in action are strongly encouraged. Contributions are welcome in the following areas (non-exhaustive):
- Control for Robust Manipulation
- How can robust and adaptive control methods handle unstructured disturbances in manipulation?
- What is the role of compliance, variable stiffness, and hybrid control in achieving manipulation robustness?
- Planning under Uncertainty
- How can planning approaches account for uncertainty in real-world settings?
- What are effective strategies for robust task and motion planning in cluttered, contact-rich environments?
- How can planning and learning be integrated to balance robustness and efficiency?
- Perception and Sensing
- How can multimodal sensing (vision, tactile, proprioception) improve robustness?
- What role does active and interactive perception play in robust manipulation?
- How do we build perception systems that remain reliable under changes in lighting, background, or viewpoint?
- Learning for Robustness
- How can we improve robustness and generalization of learning-based approaches?
- What is the role of robust reinforcement learning, adversarial training, and domain randomization?
- Can we expect general-purpose “foundation policies” for manipulation that can predict, prevent or recover from failure?
- Hardware and Mechanical Intelligence
- How can robust manipulator and end-effector design contribute to system-level robustness?
- What benefits do compliance, soft robotics, or morphological computation bring for robustness?
- How do we co-design hardware and control policies for robust performance?
- How can micro- and nano-manipulation systems be designed for robust operation in safety-critic, compliant environments?
- Beyond Manipulation
- What common principles of robustness extend across manipulation, locomotion, and navigation?
- How do we benchmark and evaluate robustness in different domains of robotics?
- What lessons from biological robustness can guide robotic system design?
- Other related topics on robustness in robotics are welcome!
Submission Guidelines
- Submission Portal for Paper: OpenReview
- Paper Submission Guideline:
- Paper Length: Submissions could be short papers (extended abstracts) of 2-4 pages, excluding references, acknowledgements, and appendices.
- Format:
- The format requirement follows the ICRA 2026 main conference. Paper template: LaTeX and Word
- Please include the references and appendix in the same PDF as the main paper. Optional videos should be uploaded to Google Drive with access set to “Anyone with the link can view,” and the link should be included in the PDF.
- Dual Submission for Other Conference:
- Papers to be submitted or in preparation for submission to other major venues in the field are allowed.
- We also welcome published works as long as explicitly stated at the time of submission.
- Visibility: Submissions and reviews will not be public. Only accepted papers will be made public.
Timeline
- Submission Port Open: Feb 15, 2026
- Submission Deadline: Apr 15, 2026 (23:59 AoE)
- Notification: May 1, 2026
- Workshop Date: June 5, 2026
Award
- Two best workshop paper award: a robotic arm for each, sponsored by ARX Robotics
- We will organize an after-session dinner with workshop speakers, paper authors, organizers and other participants, sponsored by AlphaIgnis. Open for sign up soon.
