Learning Language-Conditioned Traversability Representations for Adaptive Visual Navigation
We present LaTraNav, a framework that learns language-conditioned traversability representations for adaptive visual navigation.
Key points
- Traversability is essential for visual navigation but varies with robot capabilities and user preferences.
- Moreover, viewpoint-dependent segmentation masks complicate asynchronous planning under perception latency.
- To train the system, we develop a simulation-based data generation pipeline with controllable trajectories, producing observations paired with language instructions, traversability maps, goal locations, and diverse trajectories.
- Evaluations on datasets from multiple sources demonstrate effective language-guided traversability segmentation and goal localization by the slow VLM, alongside adaptive pixel-space path planning by the fast planner.
Sources (1)
- [1]Learning Language-Conditioned Traversability Representations for Adaptive Visual NavigationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 10:03 AM
We present LaTraNav, a framework that learns language-conditioned traversability representations for adaptive visual navigation.
Traversability is essential for visual navigation but varies with robot capabilities and user preferences.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 7, 2026MeshCarve: Artisan Mesh Generation with Flow Matching in Compact Latent Spaces
- Oct 7, 2026EC-EarthFlow: Probabilistic emulation of daily transient global climate model simulations with flow matching
- Oct 7, 2026One Frame, Full Heartbeat: ECG-Free Cardiac Cine MRI Synthesis via Phase-Conditioned Flow Matching
- Oct 6, 2026One Frame, Full Heartbeat: ECG-Free 4D Cardiac Cine MRI Synthesis via Radial-Decomposed Flow Matching
- Oct 6, 2026Domain-informed Adaptive Sampling for Generalizable PINNs in Metal Additive Manufacturing via Conditional Flow Matching
- Oct 6, 20264D-HOF: Hand-Object Flow Matching for Feed-Forward 4D Interaction Reconstruction