Bridging Perspectives in Navigation: Lessons Learned and Future Directions – IROS 2026

Robot navigation in the wild is one of the most impactful and active problems in robotics today, and includes a wide variety of applications such as last-mile delivery, off-road driving, in-home assistance, and even digital agents navigating the web. This fundamental challenge brings together diverse research fields across machine learning, computer vision, and system design for planning, perception, and control. While these navigation domains share similar core technologies, the navigation research community remains fragmented, with distinct subfields, communities, and benchmarks.

This full-day workshop asks a simple question: What can different navigation domains learn from each other? We bring together leading researchers across navigation in its many forms, spanning aerial, ground, legged, humanoid, and even digital agents, from industry and academia. We envision this shared platform to foster dialogue across once isolated domains, with the goal of surfacing shared challenges and highlighting transferable solutions and dialogue. This will be accomplished via a series of talks from a diverse range of experts from both academia and industry, a panel discussion, and presentations of workshop papers.