ECMR Workshop 2025
About
Despite decades of SLAM research, real-world deployment remains challenging, especially in perceptually degraded or structurally complex environments. This workshop aims to unite researchers across Europe to address SLAM robustness in such conditions. Topics include radar-visual-inertial fusion, low-cost sensor configurations, resilience to sensor dropouts, and architectural designs for long-term autonomy. A key goal is fostering discussion on robust SLAM backends, adaptive fusion strategies, and benchmarking under degenerative conditions. The workshop will feature invited talks, poster sessions, a World Café-style discussion, and a SLAM competition to evaluate new methods on real datasets collected with quadruped and wheeled robots using diverse sensors.
Call for papers
During this workshop, we will also hold a poster session to present recent developments. We invite you to submit your novel contribution aligned with the following indicative (and non-exhaustive) list of topics.
Important dates
- 01. 08. 2025 — Deadline for submission
- 15. 08. 20025 — Notification of acceptance
- 05. 09. 2025 — Deadline for final paper submission
- Two weeks before ECMR 2025 — Competition deadline
- 02. 09. 2025 — Workshop
Program
Full-day schedule with keynotes, technical sessions, and competitions: 02. 09. 2025
- 09:00–09:10 — Welcome & Introduction
- 09:10–10:10 — Invited Speakers Session 1
- 10:10–11:00 — Poster & Competition Session 1 + Coffee
- 11:00–13:00 — Presentations of Accepted Papers
- 13:00–14:00 — Lunch Break
- 14:00–15:30 — Invited Speakers Session 2
- 15:30–16:15 — Poster & Competition Session 2 + Coffee
- 16:16–17:15 — World Café Discussion
- 17:15–17:30 — Closing Remarks
Invited speakers
Matteo Matteucci
Politecnico di MilanoTopic: Radar-based SLAM for Robust Localization
Giorgio Grisetti
Sapienza University of Rome (online)Topic: TBD
Achim Lilienthal
Technische Universität MünchenTopic: TBD
Konstantinos Alexis
NTNUTopic: TBD
Martin Saska
CTU in PragueTopic: TBD
Topic: TBD
Competition
The workshop hosts a SLAM benchmarking competition with datasets that include radar, multispectral imagery (NDVI, thermal), and low-cost sensors on quadruped and wheeled robots.
Evalution
- Absolute Trajectory Error (ATE)
- Relative Pose Error (RPE)
- Dockerized ROS pipelines evaluated via CodaBench.
Datasets
- Outdoor forest, tunnel, and alpine trail datasets with radar and GNSS
- Indoor dataset using a quadruped with RGBD and motion capture ground truth
Participants can access training datasets ~3 months before the workshop. Testing data is hidden and evaluated post-submission.
Organizers
Main organizer
Vsevolod Hulchuk
CTU in Prague
Co-organizers
Martin Magnusson
Örebro University, Sweden
Vladimír Kubelka
Örebro University, Sweden
Konstantinos Alexis
NTNU
Matteo Luperto
University of Milano, Italy
Tomasz Kucner
Aalto University, Finland
Piotr Kicki
Poznan University of Technology, Poland
Rudolf Szadkowski
CTU in Prague