XPG-RL: Reinforcement Learning with Explainable Priority Guidance for Efficiency-Boosted Mechanical Search
This work addresses efficiency and robustness in robotic manipulation for autonomous systems, representing an incremental improvement through integration of domain knowledge with learning.
The paper tackles the challenge of mechanical search in cluttered environments by introducing XPG-RL, a reinforcement learning framework that uses explainable priority guidance, resulting in up to 4.5× higher efficiency in long-horizon tasks compared to baselines.
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/.