arXiv cs.AI· Jiahao Xu, Peiyuan Wang, Hanzhuo Zhang, Zihao Yu, Tianyu Fu, Hao Chen, Xuanhao Xiang, Zixuan Li, Jianbo Yu, Chenchen Fu, Wanyuan Wang·· 13 小时前AI 评分31
GTP-FA：带失败归因的“先抓取后规划”两阶段机器人操作框架
Grasp-Then-Plan with Failure Attribution: A Closed Two-Stage Framework for Precise and Generalizable Robotic Manipulation
AI 导读
研究者提出 GTP-FA，一个面向任务的“先抓取后规划”两阶段框架，通过失败归因模型对失败轨迹给出失败模式分布，并据此分别优化抓取候选评分与运动规划。在仿真和真机实验中，GTP-FA 在 RL、IL、扩散策略和 VLA 等基座设置上均提升了整体任务成功率。该工作已被 NeurIPS 2026 接收。
正文 · 原文
Authors:Jiahao Xu, Peiyuan Wang, Hanzhuo Zhang, Zihao Yu, Tianyu Fu, Hao Chen, Xuanhao Xiang, Zixuan Li, Jianbo Yu, Chenchen Fu, Wanyuan Wang
Abstract:In robotic manipulation, the tight coupling between grasping and motion planning often obscures the true source of failure, leading to inefficient trial-and-error. To enable efficient long-horizon manipulation, we propose GTP-FA (Grasp-Then-Plan with Failure Attribution), a task-oriented two-stage 'grasp-then-plan' framework that generates grasp candidates and performs downstream motion planning conditioned on the selected grasp. Given a failed manipulation trajectory, we learn a failure attribution model that generalizes to unseen grasps and produces a stable distribution over failure modes for diagnosis-guided optimization. Based on these attribution results, we then optimize both modules in a diagnosis-driven manner: on the grasping side, we inject task-level priors and risk penalties into grasp candidate scoring and optimization to suppress unstable or task-incompatible grasps; on the planning side, we target high-risk initial states through data collection and fine-tuning to address genuine planning bottlenecks. We evaluate the proposed framework in both simulation and real-robot experiments, and show that GTP-FA improves the corresponding base learners across RL, IL, diffusion-policy, and VLA-based settings, achieving substantially higher overall task success rates.
| Comments: | Accepted at NeurIPS 2026. 34 pages. Revised after peer review with additional experiments, updated results, and updated author information. Project page: this https URL |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.03385 [cs.RO] |
| (or arXiv:2606.03385v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2606.03385 arXiv-issued DOI via DataCite |
Submission history
From: Jiahao Xu [view email]
[v1]
Tue, 2 Jun 2026 09:29:03 UTC (41,070 KB)
[v2]
Tue, 27 Oct 2026 01:42:48 UTC (41,656 KB)
来源：arXiv cs.AI · arxiv.org