Call for Papers
Submission opens
25 May, 2026
Submission deadline
29 July, 2026
Notification
27 August, 2026
Workshop day
27 September or 1 October, 2026
Topics of Interest
Robots navigating complex, dynamic environments must go beyond geometry, they need to understand the underlying semantics of the world. This workshop addresses the critical gap in current navigation systems by exploring how semantic representations, learned scene understanding, and cognitive mapping enable truly adaptive autonomy.
We focus on the intersection of semantic 3D explicit maps and Multi-Modal Large Language Models (MM-LLMs) for complex spatial reasoning, moving beyond structured indoor environments to address the challenges of scaling semantic navigation to unstructured outdoor settings and human-populated spaces.
We invite submissions on topics including but not limited to:
- Semantic SLAM
- 3D scene understanding
- 3D scene graphs and hierarchical maps
- Neural-symbolic spatial reasoning
- Topological and metric-semantic maps
- MM-LLMs and VLAs for robot navigation
- Vision-language grounding
- Foundation models for mobile robots
- Open-vocabulary scene understanding
- Semantic mapping in outdoor environments
- Embodied AI and vision-language-action models
- Sim-to-real for semantic navigation
- Semantic exploration and active mapping
- Human-robot interaction and semantics
- Multi-modal perception
- Long-term autonomy in dynamic environments
- Mapping for interaction and planning
Submission Guidelines
Paper Format
Submissions must follow the format guidelines below:
- Document Type
Submissions must be in the form of an Extended Abstract.
- Format
Papers must be submitted in PDF (up to 10MB) format and must strictly adhere to the official IEEE RAS double-column format. You can find the LaTeX and MS Word templates on the official IEEE RAS template page or via the IROS 2026 call for papers guidelines .
- Page Budget
Submissions must be between 2 and 4 pages in length. Note: References do not count toward this page limit.
- Visuals
Authors are highly encouraged to include figures, diagrams, and graphs to clearly illustrate their contributions and experimental results.
Submissions that do not adhere to these guidelines may be rejected without review. Please ensure that your submission is complete, well-formatted, and clearly communicates the significance of your work.
Review Process
The review process will adhere to the following policies:
- Reviewers number
Each submission will be reviewed by at least two independent reviewers.
- Double-Blind Review
The review process is strictly double-blind to ensure an objective evaluation and eliminate any bias. Authors must remove all identifying information from the manuscript. This includes omitting author names, affiliations, and explicit acknowledgments. Self-citations should be written in the third person (e.g., "As demonstrated by Author et al." instead of "As we demonstrated in our previous work"). Please refer to the Rules for the Double-Anonymous Review Process from IEEE Robotics and Automation Society (RAS).
- Non-Archival Policy
Workshop proceedings are non-archival. We welcome submissions that summarize recently published work, or research currently under review at other conferences or journals. This workshop aims to foster discussion and exchange ongoing ideas.
Ready to submit?
Submit your paper via the workshop submission system. For questions, reach out to the organizing committee.
To submit your paper, press the button below and follow the instructions on theOpenReview page. We look forward to receiving your contributions and fostering a vibrant discussion at the workshop!
Visit the workshop OpenReview page
Poster Sessions
Accepted papers will be presented as posters, organized into two thematic clusters. Each cluster will be facilitated by a senior researcher to encourage in-depth discussions and mentorship. At least one author of each accepted paper must attend the workshop to present their work.
Cluster 1
Semantic Mapping and World Representations
Cluster 2
Semantic-Aware Navigation and Exploration