Original call for papers document.

The field of Visual SLAM (Simultaneous Localization And Mapping) has progressed
substantially since the last IEEE Transactions on Robotics special collection in 2008. Visual SLAM has attained a
consolidated level of maturity, finding applications in smartphones, AR/VR devices, and mobile
robots. Despite this progress, many questions remain open, such as: Are current methodologies
robust enough to handle outliers in data association effectively? Should the focus be on
learning-based end-to-end approaches, classical methodologies, or hybrid strategies? What is
the optimal map representation? 3D maps, NERF, Gaussian Splatting, or alternatives? Can
Visual SLAM leverage recent breakthroughs in Large Language Models and Foundation Models
to enhance its capabilities?

This special collection invites contributions from researchers to present their latest
advancements, propose solutions to open challenges and outline new avenues for research
within the realm of Visual SLAM.

We encourage submissions related, but not limited, to the following topics:

● Robust SLAM
● Representations in SLAM (NeRF, Gaussian Splatting, etc.)
● Semantic, Object-level, and Dynamic SLAM
● Large Scale SLAM
● Unconventional vision sensors for SLAM (event cameras, thermal cameras)
● SLAM for Robotics Navigation
● Multi-agents SLAM
● Visual(-Inertial) Odometry
● Learning-based SLAM
● Beyond Conventional SLAM (Large Language Models, Foundation Models, …)

We welcome top-quality original (unpublished) articles and review papers, following the IEEE
T-RO journal instructions for authors. Each submission will be peer-reviewed and the selection
of papers will be based on their originality, impact, and relevance to the scope of the special
collection. To ensure that all manuscripts are correctly identified for evaluation for the special
section, please select “Visual SLAM” in the submission process.