We are pleased to announce that the full-day workshop “CoRoboLearn: Advancing Learning for Human-Centered Collaborative Robots” will take place as part of the Conference on Robot Learning (CoRL 2024) on November 9th in Munich, Germany.
This workshop aims to bring together leading experts and newcomers from the robot learning and human-robot interaction communities to discuss the specific challenges and opportunities in advancing learning-based human-centered cobots. Invited and spotlight talks, a poster session, and a panel discussion will provide participants with an up-to-date view on open challenges and recent advancement in the field and opportunities for interactive discussions.
We solicit the submission of papers covering recent or ongoing work and early results on the workshop's themes, to be presented during an interactive poster session and selected spotlight talks.
Important Dates
Submissions open: September 13th
Submission deadline: October 4th 8th, 23:59 Anytime on Earth (AoE)
Notification of publication decision: October 25th
Final version due: November 1st, 23:59 Anytime on Earth (AoE)
Workshop: November 9th (full day)
Abstract
Collaborative robots (cobots) are designed to work alongside humans in shared environments, distinguishing them from traditional robots that operate in isolation. The defining trait of cobots is their ability to interact safely, efficiently, and intuitively with human partners, adapting to dynamic and uncertain environments. Recent advancements in learning-based systems propelled progress in enhancing such interactions, yet the seamless integration of cobots into everyday environments remains a significant challenge. This workshop aims to bring together leading experts from the robot learning and human-robot interaction communities to discuss the specific challenges and opportunities in advancing learning-based human-centered cobots.
Workshop Topics
The guiding topics of the workshop, together with potential research questions, are the following:
Key Challenges in Human-Centered Collaboration:
What are the primary obstacles when deploying cobots in human-centric environments?
What are the unique benefits and challenges of using learning-based approaches for human-centered collaboration?
Learning to Adapt and Improve in Real-World Situations:
How can cobots leverage information from humans and the environment to learn and adapt their actions in real-time?
Which approaches can ensure continual learning and adaptation in collaborative tasks?
Innovative Learning Methods and Intuitive Interfaces:
How can recent progress in generative AI, such as foundation models and LLMs, be leveraged to enhance the capabilities of cobots in human-centric environments?
How can we utilize new communication interfaces (e.g., wearables and gestures) to ease and enhance human-robot interaction?
Simulation and Real-World:
What are the pros and cons of learning from real-world data compared to simulated data for collaborative tasks? How to effectively model human behavior in simulation?
What are the potential approaches to bridge the sim-real gap for collaborative robotic tasks?
Submission Instructions
Papers covering ongoing work, under peer review, or published recently (including in the CoRL 2024 main conference track), are welcome.
Paper submission format:
Format: CoRL conference paper template
Papers can be submitted through OpenReview
Submissions must be anonymous, since papers will be peer-reviewed in double-blind mode. This includes potential links to supplementary videos and repositories. Submissions and reviews will not be released publicly.
Paper length: 4 to 8 pages, excluding references, acknowledgments, and appendices
One optional supplementary video is allowed:
Maximum 50MB
Maximum 3-minutes duration
Only MP4 video format is allowed
The video must be compressed and uploaded within a ZIP file ("supplementary_video.zip")
Please note: supplementary files can be uploaded from the OpenReview Author Console while uploading the main manuscript
Accepted papers will be presented in-person during an interactive poster session, and selected papers will also be presented as spotlight talks.
At least one of the authors needs to be present at the venue for each paper. Accepted papers will (optionally) be made available on the workshop website. There will be no proceedings for this workshop.
We are committed to diversity and inclusion. We actively seek participation from individuals across gender, ethnicity, nationality, and career stages.
For any inquiries or technical assistance, please contact the organizers at: corobolearn@gmail.com