SemGeoNav:A Safety-Guided Visual Navigation Approach with Semantic Reasoning and Geometric Planning

2026-06-15

AI Summary

This work addresses the limitations of existing end-to-end visual navigation methods, which lack explicit geometric constraints and exhibit unreliable obstacle avoidance in open environments, as well as the inability of traditional geometric planners to interpret high-dimensional visual goals. To bridge this gap, the authors propose a hierarchical visual navigation framework that tightly couples end-to-end semantic reasoning with geometry-based local planning for the first time, augmented by a temporal trajectory smoothing mechanism to jointly optimize high-level goal understanding and low-level safe navigation. Implemented on a Unitree Go2 quadruped robot, the system demonstrates significantly higher task success rates and reduced navigation times compared to state-of-the-art methods such as ViNT and NoMaD in real-world experiments.

Abstract

Learning-based visual navigation has enhanced semantic goal-reaching capabilities. However, due to their black-box nature, purely end-to-end models often lack explicit geometric constraints, leading to unpredictable and unreliable obstacle avoidance in open environments. Conversely, traditional geometric planners ensure safety but struggle with high-dimensional visual targets. To address these limitations, we propose SemGeoNav, a novel hierarchical visual navigation framework.It tightly integrates the high-level semantic reasoning of end-to-end models with the reliable local planning ability of geometry-based methods, achieving robust image-based navigation while significantly improving obstacle avoidance. Furthermore, we introduce a temporal trajectory smoothing mechanism to ensure continuous and stable robot motion. We evaluated SemGeoNav on a Unitree Go2 quadruped robot in real-world environments. The results demonstrate that SemGeoNav outperforms existing representative methods, including ViNT and NoMaD, achieving higher success rates and shorter navigation times.

Problem

• visual navigation
• semantic goal
• obstacle avoidance
• geometric planning
• safety

Innovation

• semantic reasoning
• geometric planning
• hierarchical navigation
• trajectory smoothing
• visual navigation