CanonNav: Disentangling Navigation Behavior from Camera Geometry in Cross-Platform Visual Navigation

2026-08-31

AI Summary

CanonNav通过解耦导航行为与摄像机几何结构，并引入安全和局部进展监督，解决了跨平台视觉导航中模仿学习的一致性和隐式决策问题。

Abstract

While visual navigation has advanced through imitation learning from cross-platform demonstrations, fully leveraging such data remains challenging. First, directly learning from image-trajectory pairs entangles navigation behavior with platform-dependent camera geometry. This hinders consistent learning by forcing the policy to implicitly infer camera geometry from visual observations, an inherently ill-posed problem. Second, imitation learning from demonstrated trajectories captures the expert's chosen motion but leaves the intermediate decisions underlying that motion implicit. To address these issues, we propose CanonNav, a visual navigation framework that disentangles navigation behavior from camera geometry and incorporates complementary planning supervision into learning from cross-platform demonstrations. CanonNav introduces camera geometry canonicalization, which transforms visual observations and trajectories into a camera-consistent representation space. Building on this representation, we derive safety and local-progress supervision using pseudo-labels from an offline traversability estimator. Safety supervision penalizes unsafe trajectories, while local-progress supervision guides where the robot should advance. Experiments across diverse camera configurations and environments show that, despite using only RGB at inference, CanonNav consistently outperforms RGB-based baselines and even surpasses RGB-D-based methods in challenging scenarios.

Problem

• visual navigation
• cross-platform demonstrations
• camera geometry
• imitation learning
• navigation behavior

Innovation

• camera geometry canonicalization
• cross-platform demonstrations
• planning supervision
• safety and local-progress supervision