arXiv cs.AI· Lipeng Zhuang, Yingdong Ru, Shiyu Fan, Edmond S. L. Ho, Gerardo Aragon Camarasa, Paul Henderson·· 9 天前AI 评分22
MGMP：几何引导 Token 搜索的掩码生成式运动规划
Masked Generative Motion Planning with Geometry-Guided Token Search
AI 导读
研究者提出掩码生成式运动规划（MGMP），用掩码生成式 Transformer 并行生成离散轨迹候选，并由 Geometry-Guided Token Search（GGTS）依据场景几何决定编辑位置与候选方案，将测试时修复转为离散运动搜索，实现超越局部形变的路线级重构。
正文 · 原文
Abstract:Generative motion planners typically use learned trajectory priors for initial generation, while leaving test-time repair to local continuous refinement. We introduce Masked Generative Motion Planning (MGMP), which extends the learned prior from efficient parallel generation to structural repair. A masked generative transformer generates discrete trajectory candidates in parallel, and Geometry-Guided Token Search (GGTS) uses scene geometry to target where to edit and which prior-supported alternatives to evaluate. This turns refinement into an efficient search over discrete motion alternatives, enabling route-level restructuring beyond local trajectory deformation. MGMP achieves 96% success on Ring Maze and 82% repair success on Controlled Route Invalidation on Kuka, exceeding the strongest external baselines by 23 and 25 percentage points, respectively. It further generalizes to unseen layouts, additional obstacles, unseen geometries, single- and dual-arm planning, and real-world Baxter tasks.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.10646 [cs.RO] |
| (or arXiv:2610.10646v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10646 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lipeng Zhuang [view email]
[v1]
Wed, 7 Oct 2026 15:19:54 UTC (24,430 KB)
来源：arXiv cs.AI · arxiv.org