Underwater Marine Robotics Background Cover Image

IROS 2026 Workshop

Perception and Navigation for
Marine Robotics

Innovations in Marine Robotics: Bridging Robotics, Industry, and Ocean Science

Date: October 1st (Morning Session) Location: Pittsburgh, PA, USA

Workshop Overview

Marine robotics operates in environments that break the assumptions behind standard perception and navigation pipelines, from sudden turbidity and GPS-denied conditions to constant current perturbations. This workshop brings together researchers to discuss resilient sensing, state estimation, and mapping methods, grounded in a real-world challenge through an invited talk from a marine biologist on coral reef conservation.

IROS 2026 Conference Logo

GPS-Denied Operations

Marine robots must operate without GPS, under constant current and wave perturbations, with intermittent communication that limits remote supervision and correction. Localization and navigation pipelines must contend with persistent drift and the difficulty of obtaining reliable ground truth over long-duration missions.

CUREE autonomous underwater vehicle surveying a coral reef

Ocean Autonomy & SLAM

Robust ocean autonomy requires multimodal sensing and fusion across vision, sonar, inertial, and acoustic positioning; mapping and SLAM tailored to low-visibility and dynamic environments; and planning methods that explicitly account for uncertainty, drift, and sensing dropouts.

Aqua2 autonomous underwater robot exploring a ship

Bridging Science & Conservation

A unique aspect of this workshop is its grounding in a concrete, real-world mission: an invited talk from a marine biologist will share the on-the-ground challenges of coral reef conservation, giving marine robotics researchers a common reference point to align their technical work with genuine conservation needs.

Confirmed Speakers

John Leonard

John Leonard

Massachusetts Institute of Technology

Website →

Professor of Mechanical and Ocean Engineering at MIT and CSAIL member who leads the Marine Robotics Group, specializing in navigation and mapping for autonomous underwater and terrestrial robots.

Gregory Dudek

Gregory Dudek

McGill University

Website →

Distinguished James McGill Professor of Computer Science who leads the Mobile Robotics Lab, known for marine field robotics including the amphibious Aqua robot for coral reef monitoring.

Katherine Skinner

Katherine Skinner

Stony Brook University

Website →

Associate Professor of Mechanical Engineering whose research spans perception, computer vision, and machine learning for field robotics across air, land, sea, and space.

Lauren Olinger

Lauren Olinger

University of the Virgin Islands

Website →

Coral reef ecologist who directs UVI's Virgin Islands Center for Autonomous Research, using underwater robotics, 3D imaging, and AI to monitor reef health and marine ecosystems.

Seth Mccammon

Seth McCammon

Woods Hole Oceanographic Institution

Website →

Marine roboticist at WHOI developing autonomy and active sensing methods, using AUVs and passive acoustics to map complex coral reef ecosystems.

Call for Papers

Now Accepting Submissions

We are currently accepting one-page abstract submissions (not including references), which should follow the IEEE Conference Template. Submissions are reviewed and accepted on a rolling basis until the final deadline. Abstracts will undergo a light peer-review process, and accepted authors will be invited to present a poster and a lightning talk highlighting their key results during the workshop.

Final Deadline: September 1, 2026
Submit to OpenReview →

GPS-Denied Localization

Robust state estimation, acoustic positioning, and inertial navigation under persistent drift from currents and waves.

👁️

Low-Visibility SLAM

Simultaneous localization and mapping in turbid, dark, or otherwise unstructured underwater environments.

📡

Multimodal Sensing & Fusion

Cooperative integration of vision, sonar, inertial, acoustic positioning, and GNSS (when available) for resilient perception.

🪸

Mapping & 3D Reconstruction

Large-scale, multi-session mapping and structure-from-motion for dynamic, low-visibility environments such as coral reefs.

🧠

Marine Robotics Autonomy

Learning-based autonomy and decision-making, including deep learning models for sonar denoising, underwater image restoration, and semantic segmentation.

🗺️

Planning Under Uncertainty

Navigation and planning methods that explicitly account for uncertainty, drift, and sensing dropouts.

Organizers

Edwin Meriaux Profile Photo

Edwin Meriaux

McGill University

Steve Wen Profile Photo

Steve Wen

McGill University

Kalvik Jakkala Profile Photo

Kalvik Jakkala

Texas A&M University

Monika Roznere Profile Photo

Monika Roznere

Binghamton University

Marios Xanthidis Profile Photo

Marios Xanthidis

SINTEF Ocean

Jason O'Kane Profile Photo

Jason O'Kane

Texas A&M University

Alberto Quattrini Li Profile Photo

Alberto Quattrini Li

Dartmouth College

Organizing Institutes