SemGeoNav:A Safety-Guided Visual Navigation Approach with Semantic Reasoning and Geometric Planning
2026-06-15 • Robotics
Robotics
AI summaryⓘ
The authors developed SemGeoNav, a new robot navigation system that combines smart understanding of scenes (semantic reasoning) with reliable geometric path planning. This mix helps robots avoid obstacles better than using only one method. They also added a feature to make the robot move smoothly. Tests on a real robot showed that SemGeoNav was more successful and faster than some popular existing methods.
visual navigationsemantic reasoninggeometric planningend-to-end modelsobstacle avoidancetrajectory smoothingquadruped robotrobot motionhierarchical navigationreal-world evaluation
Authors
Yu Liu, Zongyang Chen, Yan Guo, Chao Liu, Xianfei Pan
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.