MOSU: Autonomous Long-range Robot Navigation with Multi-modal Scene Understanding
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908437415198720 |
|---|---|
| author | Liang, Jing Weerakoon, Kasun Song, Daeun Kirubaharan, Senthurbavan Xiao, Xuesu Manocha, Dinesh |
| author_facet | Liang, Jing Weerakoon, Kasun Song, Daeun Kirubaharan, Senthurbavan Xiao, Xuesu Manocha, Dinesh |
| contents | We present MOSU, a novel autonomous long-range navigation system that enhances global navigation for mobile robots through multimodal perception and on-road scene understanding. MOSU addresses the outdoor robot navigation challenge by integrating geometric, semantic, and contextual information to ensure comprehensive scene understanding. The system combines GPS and QGIS map-based routing for high-level global path planning and multi-modal trajectory generation for local navigation refinement. For trajectory generation, MOSU leverages multi-modalities: LiDAR-based geometric data for precise obstacle avoidance, image-based semantic segmentation for traversability assessment, and Vision-Language Models (VLMs) to capture social context and enable the robot to adhere to social norms in complex environments. This multi-modal integration improves scene understanding and enhances traversability, allowing the robot to adapt to diverse outdoor conditions. We evaluate our system in real-world on-road environments and benchmark it on the GND dataset, achieving a 10% improvement in traversability on navigable terrains while maintaining a comparable navigation distance to existing global navigation methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_04686 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MOSU: Autonomous Long-range Robot Navigation with Multi-modal Scene Understanding Liang, Jing Weerakoon, Kasun Song, Daeun Kirubaharan, Senthurbavan Xiao, Xuesu Manocha, Dinesh Robotics We present MOSU, a novel autonomous long-range navigation system that enhances global navigation for mobile robots through multimodal perception and on-road scene understanding. MOSU addresses the outdoor robot navigation challenge by integrating geometric, semantic, and contextual information to ensure comprehensive scene understanding. The system combines GPS and QGIS map-based routing for high-level global path planning and multi-modal trajectory generation for local navigation refinement. For trajectory generation, MOSU leverages multi-modalities: LiDAR-based geometric data for precise obstacle avoidance, image-based semantic segmentation for traversability assessment, and Vision-Language Models (VLMs) to capture social context and enable the robot to adhere to social norms in complex environments. This multi-modal integration improves scene understanding and enhances traversability, allowing the robot to adapt to diverse outdoor conditions. We evaluate our system in real-world on-road environments and benchmark it on the GND dataset, achieving a 10% improvement in traversability on navigable terrains while maintaining a comparable navigation distance to existing global navigation methods. |
| title | MOSU: Autonomous Long-range Robot Navigation with Multi-modal Scene Understanding |
| topic | Robotics |
| url | https://arxiv.org/abs/2507.04686 |