Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2504.20969 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908407604183040 |
|---|---|
| author | Zhang, Yiting Li, Shichen Shrestha, Elena |
| author_facet | Zhang, Yiting Li, Shichen Shrestha, Elena |
| contents | Mechanical search (MS) in cluttered environments remains a significant challenge for autonomous manipulators, requiring long-horizon planning and robust state estimation under occlusions and partial observability. In this work, we introduce XPG-RL, a reinforcement learning framework that enables agents to efficiently perform MS tasks through explainable, priority-guided decision-making based on raw sensory inputs. XPG-RL integrates a task-driven action prioritization mechanism with a learned context-aware switching strategy that dynamically selects from a discrete set of action primitives such as target grasping, occlusion removal, and viewpoint adjustment. Within this strategy, a policy is optimized to output adaptive threshold values that govern the discrete selection among action primitives. The perception module fuses RGB-D inputs with semantic and geometric features to produce a structured scene representation for downstream decision-making. Extensive experiments in both simulation and real-world settings demonstrate that XPG-RL consistently outperforms baseline methods in task success rates and motion efficiency, achieving up to 4.5$\times$ higher efficiency in long-horizon tasks. These results underscore the benefits of integrating domain knowledge with learnable decision-making policies for robust and efficient robotic manipulation. The project page for XPG-RL is https://yitingzhang1997.github.io/xpgrl/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_20969 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | XPG-RL: Reinforcement Learning with Explainable Priority Guidance for Efficiency-Boosted Mechanical Search Zhang, Yiting Li, Shichen Shrestha, Elena Robotics Machine Learning Mechanical search (MS) in cluttered environments remains a significant challenge for autonomous manipulators, requiring long-horizon planning and robust state estimation under occlusions and partial observability. In this work, we introduce XPG-RL, a reinforcement learning framework that enables agents to efficiently perform MS tasks through explainable, priority-guided decision-making based on raw sensory inputs. XPG-RL integrates a task-driven action prioritization mechanism with a learned context-aware switching strategy that dynamically selects from a discrete set of action primitives such as target grasping, occlusion removal, and viewpoint adjustment. Within this strategy, a policy is optimized to output adaptive threshold values that govern the discrete selection among action primitives. The perception module fuses RGB-D inputs with semantic and geometric features to produce a structured scene representation for downstream decision-making. Extensive experiments in both simulation and real-world settings demonstrate that XPG-RL consistently outperforms baseline methods in task success rates and motion efficiency, achieving up to 4.5$\times$ higher efficiency in long-horizon tasks. These results underscore the benefits of integrating domain knowledge with learnable decision-making policies for robust and efficient robotic manipulation. The project page for XPG-RL is https://yitingzhang1997.github.io/xpgrl/. |
| title | XPG-RL: Reinforcement Learning with Explainable Priority Guidance for Efficiency-Boosted Mechanical Search |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2504.20969 |