Research map / Reinforcement learning / Visual and representation RL
Coupling Vision and Proprioception for Navigation of Legged Robots
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add proprioceptive collision detection for unseen obstacles (ablation) | VP-Nav · 2x Inv-Obstacle | success rate: 68.45 -> 74.15 | Helped |
| Add proprioceptive collision detection for unseen obstacles (ablation) | VP-Nav · 4x Inv-Obstacle | success rate: 45.85 -> 59.20 | Helped |
| Add proprioceptive collision detection for unseen obstacles (ablation) | VP-Nav · 8x Inv-Obstacle | success rate: 24.35 -> 39.25 | Helped |
| Add proprioceptive fall prediction and speed adjustment (ablation) | VP-Nav · Randomized | success rate: 80.25 -> 87.40 | Helped |
| Use proprioceptive fall prediction and speed adjustment (ablation) | VP-Nav · Randomized | time (s): 105.68 -> 117.65 | Hurt |
| Use varying gaits instead of fixed trotting (ablation) | VP-Nav · Flat | energy consumption (K): 252.56 -> 233.05 | Helped |
| Use varying gaits instead of fixed trotting (ablation) | VP-Nav · Flat | success rate: 93.80 -> 95.20 | Helped |
| Use legged locomotion rather than wheeled navigation | VP-Nav · RoughTerrain | success rate: 69.10 -> 95.05 | Helped |
| Use legged locomotion rather than wheeled navigation | VP-Nav · RoughTerrain | time (s): 146.85 -> 80.87 | Helped |
| Use continuous velocity commands instead of four discrete actions (ablation) | VP-Nav · Flat | time (s): 110.27 -> 80.87 | Helped |
| Use continuous velocity commands instead of four discrete actions (ablation) | VP-Nav · Flat | success rate: 95.35 -> 95.20 | No clear effect |
| Use proprioceptive collision feedback to map and replan around visually missed glass | VP-Nav · Real-world glass wall | success rate | Helped |
Closest papers by meaning
- Hybrid Internal Model: Learning Agile Legged Locomotion with Simulated Robot Response (ICLR 2024 · similarity 0.86)
- Reasoning in Visual Navigation of End-to-end Trained Agents: A Dynamical Systems Approach (CVPR 2025 · similarity 0.84)
- SPIN: Simultaneous Perception Interaction and Navigation (CVPR 2024 · similarity 0.83)
- Learning to Navigate Efficiently and Precisely in Real Environments (CVPR 2024 · similarity 0.83)
- Let Humanoids Hike! Integrative Skill Development on Complex Trails (CVPR 2025 · similarity 0.83)
- Versatile Navigation Under Partial Observability via Value-guided Diffusion Policy (CVPR 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.