Releasing approximately 10,000 hours of mobile manipulator data and building a standard “data-to-decision” VLA pipeline with researchers and engineers worldwide

The AI Robotic Association (AIRoA) is pleased to announce that its workshop proposal “From Data to Decisions: VLA Pipelines for Real Robots” has been accepted for presentation at ICRA 2026. As one of the world’s largest international conferences on robotics, ICRA 2026 will be held in Vienna, Austria, from June 1–5, 2026.

The workshop will focus on end-to-end pipelines “from data to decisions” for applying Vision-Language-Action (VLA) models to real-world robots, and featuring two parallel tracks:

• A paper session on VLA pipelines (talks and posters)

• A mobile manipulation competition using approximately 10,000 hours of real robot data*

*The approximately 10,000 hours of robot data were obtained as the outcome of a project commissioned by NEDO (New Energy and Industrial Technology Development Organization).

Participation in both tracks will be invited through open calls.

◼︎Challenges in VLA Models and “Data Engines”

In recent years, Vision-Language-Action (VLA) models have rapidly emerged as a promising approach for robots capable of following natural-language instructions to perform a wide range of tasks.

At the same time, many open questions remain regarding the design of end-to-end pipelines, including:

• The cost and methodology of data collection

• Data curation and quality control

• The standardization of training and evaluation

• Safety and reliability in real-world deployment

Best practices and standardized metrics for these issues have yet to be established.

AIRoA aims to bridge this gap by building and releasing a “data engine” centered on real-world robot data.

◼︎Workshop Overview:

From Data to Decisions: VLA Pipelines for Real Robots

• Title: From Data to Decisions: VLA Pipelines for Real Robots

• Format: Full-day workshop (co-located with ICRA 2026)

• Website: https://icra2026vlapipeline.github.io/

• Planned program:

  • Invited talks on cutting-edge work in data collection, training, evaluation, and real-world deployment

  • Panel discussion: “How do we build trustworthy VLA pipelines?”

  • Paper session (lightning talks + poster session)

  • Mobile manipulation competition results and award presentations

Objectives of the Workshop

This workshop aims to:

• Clarify the overall structure of VLA pipelines, from data collection and curation to training, inference, and evaluation.

• Produce practical resources—datasets, evaluation protocols and baseline models—that the community can adopt.

• Build a shared foundation that enables researchers and engineers from both academia and industry to deploy VLA models on real robots.

◼︎Paper Session: Call for 2-Page Extended Abstracts on VLA Pipelines

We invite 2-page papers on any component of the VLA pipeline, including but not limited to:

Example topics

• Data collection via real robots, teleoperation, simulation, or web videos

• Best practices for data curation, preprocessing, and annotation

• Training strategies such as pretraining, fine-tuning, RL, and auxiliary objectives

• The design of evaluation metrics and benchmarks for generalization, safety, and robustness

• Inference and control for real-world deployment (e.g., latency, contact-rich manipulation)

Submission format

• 2-page paper (single-blind review)

Presentation format for accepted papers

• Lightning / spotlight talk (6–7 minutes)

• Poster session

• A selection of submissions will be selected as oral presentations (contributed talks)

We encourage authors to share links to code, configuration files, and “data recipes” to support reproducibility. AIRoA intends to actively archive these contributions. Accepted papers, slides, and video recordings will be made available online after the workshop.

Researchers interested in participating in the paper session are invited to submit via the following link:

https://openreview.net/group?id=IEEE.org/ICRA/2026/Workshop/From_Data_to_Decisions#tab-recent-activity

◼︎Competition: Mobile Manipulation with approximately 10,000 Hours of Real Robot Data

The second pillar of the workshop is an international competition for VLA models focused on mobile manipulation.

Competition overview

Target tasks (examples):

• Pick-and-place

• Insertion tasks

• Manipulation of articulated objects (doors, drawers, etc.)

• Tool-using tasks and other contact-rich behaviors

Dataset:

• Approximately 10,000 hours of real-world data collected with a mobile manipulator

• Multimodal data including RGB-D video, joint states, force/torque signals, language annotations, and action trajectories

• Data will be organized in the LeRobot format and released via storage services such as Wasabi.

Evaluation platform:

• A standardized environment based on Toyota’s Human Support Robot(HSR)

• On-site evaluation on physical robots, conducted every two weeks

Evaluation metrics:

• Task success rate

• Task completion time

• Robustness to environmental variations, etc.

Baselines:

• Official baselines based on existing VLA models such as SmolVLA and π0 will be provided, enabling participating teams to build and test improvements on top of them.

Prizes:

• 1st place: USD 2,000

• 2nd place: USD 1,000

• 3rd place: USD 600

Awarded teams will be invited to give talks at the workshop, sharing their methods, ablations, and failure cases.

Who should participate

• Academic labs and industrial research groups working on robot learning, reinforcement learning, or multimodal models

• Engineering teams conducting PoC projects to deploy autonomous robots in real-world settings

Important dates

• February 14, 2026: Dataset release

• From March 1, 2026: Bi-weekly evaluations begin

• April 15, 2026: Final model submission deadline

• May 1, 2026: Notification of results

Teams interested in joining the competition will be able to apply via a dedicated competition page.

https://icra2026vlapipeline.github.io/