Research map / Learning theory and algorithms / Combinatorial and search optimization
DAA*: Deep Angular A Star for Image-based Path Planning
Zhiwei Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 path angular freedom (DAA*-mix) to Neural A* | Neural A* · TMPD (maze) | SPR (%): 78.63 -> 88.59 | Improved |
| add path angular freedom (DAA*-mix) to Neural A* | Neural A* · SDD-inter | PSIM (%): 35.52 -> 38.78 | Improved |
| learn alpha (mix) instead of fixing alpha=1 (min) (ablation) | DAA* · TMPD (maze) | SPR (%): 81.91 -> 88.59 | Improved |
| learn alpha (mix) instead of fixing alpha=0 (max) (ablation) | DAA* · CSM (maze) | SPR (%): 74.08 -> 82.03 | Improved |
| use max path angle (alpha=0) versus min on drone-view SDD-inter (ablation) | DAA* · SDD-inter | PSIM (%): 35.71 -> 38.65 | Improved |
| use min path angle (alpha=1) on video-game maps (ablation) | DAA* · Warcraft | PSIM (%): 75.17 -> 76.71 | Improved |
| random-walk top-k node expansion instead of angular term (ablation) | Neural A* · Pokemon | SPR (%): 67.53 -> 52.98 | Hurt |
| train DAA* with path loss only on Aug-TMPD versus TransPath's PPM supervision | DAA*-path · Aug-TMPD | PSIM: 49.78 -> 56.37 | Improved |
| weight path loss 10x over PPM loss (DAA*-weight) (ablation) | DAA* · Aug-TMPD | SPR: 87.04 -> 96.87 | Improved |
| add DAA* search perturbation on maze datasets and SDD | Neural A* · Search efficiency (Hist) | Hist: increase of 0.5%--6.4% | Hurt |
Closest papers by meaning
- MAPF-GPT: Imitation Learning for Multi-Agent Pathfinding at Scale (AAAI 2025 · similarity 0.83)
- SPOC: Imitating Shortest Paths in Simulation Enables Effective Navigation and Manipulation in the Real World (CVPR 2024 · similarity 0.82)
- DiffTORI: Differentiable Trajectory Optimization for Deep Reinforcement and Imitation Learning (NeurIPS 2024 · similarity 0.82)
- Grounding Driving VLA via Inverse Kinematics (NeurIPS 2026 · similarity 0.82)
- Affordances-Oriented Planning Using Foundation Models for Continuous Vision-Language Navigation (AAAI 2025 · similarity 0.82)
- GravMAD: Grounded Spatial Value Maps Guided Action Diffusion for Generalized 3D Manipulation (ICLR 2025 · similarity 0.81)
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.
