SaferPath: Hierarchical Visual Navigation with Learned Guidance and Safety-Constrained Control
Lingjie Zhang, Zeyu Jiang, Changhao Chen · 2026
Read SaferPath: Hierarchical Visual Navigation with Learned Guidance and Safety-Constrained Control by Lingjie Zhang, Zeyu Jiang, Changhao Chen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-01898.txt.
Description
Lingjie Zhang, Zeyu Jiang, and Changhao Chen make visual navigation safer for mobile robots in tight indoor spaces. Their 2026 work starts from a flaw: end-to-end learned drivers generalize poorly and crash often in clutter. Dense rooms and narrow halls punish any model that steers straight from pixels to motion.
Their framework, SaferPath, keeps learned guidance but refines it through a safety-constrained control layer. Camera views first become a map of traversable area around the robot. An evolution-based optimizer then reshapes guide paths into safe trajectories within a few iterations. A predictive controller tracks those paths closely as the robot moves. Tests with unseen obstacles, messy dense spaces, and narrow corridors show higher success and fewer crashes. The system beats named navigation baselines and holds up in real-world settings.
Catalog record
| CREATOR | Lingjie Zhang, Zeyu Jiang, Changhao Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36011 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.01898