MIRROR:从机器人离线交互数据中反映隐藏对称性
MIRROR: Reflecting Hidden Symmetries from Offline Robot Interaction Data
AI总结:
MIRROR是从机器人离线交互数据中学习对称群在POMDP中对观测和动作的耦合变换的框架,在Robosuite任务上提升了对称性质量与策略成功率,其策略可零样本部署到真实Kinova GEN3机械臂
摘要:
感知对称性的机器人策略学习需要耦合的观测-动作变换。对于像素观测,即使是简单的物理对称性也会诱导出复杂的、依赖视角的图像变换,难以手动指定。现有的感知对称性方法通常依赖具有预定义变换的表示,需要针对感知设置和任务定制预处理。我们提出MIRROR,这一框架可从离线交互数据中学习已知对称群如何在部分可观测马尔可夫决策过程(POMDP)中共同作用于观测和动作。MIRROR通过对一步转移元组进行对抗性元组匹配来耦合这些变换,同时显式的群一致性正则化会惩罚违反群操作公理的行为。在四个Robosuite桌面操作任务上,MIRROR在四个对称性质量指标中的三个上实现了最低的聚合误差,包括相对于适配后的LaLiGAN基线,兼容性误差(衡量变换组合下的一致性)降低了61%。在帧平均行为克隆(BC)中使用学习到的变换,在所有四个任务上都比普通BC获得了更高的平均成功率,平均成功率从80.48%提升至85.12%,每个任务的提升幅度最高达9.66个百分点。我们还在使用高度图和分割掩码观测的仿真中重复进行对称性发现和策略训练,然后将得到的策略零样本部署到真实的Kinova GEN3机械臂上。项目页面:this https URL
英文摘要:
Symmetry-aware robot policy learning requires coupled observation-action transformations. For pixel observations, even simple physical symmetries can induce complex, viewpoint-dependent image transformations that are difficult to specify by hand. Existing symmetry-aware methods often rely on representations with predefined transformations, requiring preprocessing tailored to the sensing setup and task. We introduce MIRROR, a framework that learns how a known symmetry group operates jointly on observations and actions in partially observable Markov decision processes (POMDPs) from offline interaction data. MIRROR couples these transformations through adversarial tuple matching over one-step transition tuples, while explicit group-consistency regularization penalizes violations of the group-operation axioms. On four Robosuite tabletop manipulation tasks, MIRROR achieves the lowest aggregate error on three of four symmetry-quality metrics, including a 61% reduction in compatibility error (which measures consistency under transformation composition) relative to an adapted LaLiGAN baseline. Using the learned transformations in frame-averaged behavior cloning (BC) yields higher mean success rates than vanilla BC on all four tasks, increasing average success from 80.48% to 85.12%, with per-task gains up to 9.66 percentage points. We also repeat symmetry discovery and policy training in simulation with heightmap and segmentation-mask observations, then deploy the resulting policy zero-shot on a real Kinova GEN3 arm. Project page: https://pomdp-mirror.github.io.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。