Robust Robot Learning from Demonstration and Skill Repair Using Conceptual Constraints
Robust Robot Learning from Demonstration and Skill Repair Using Conceptual Constraints
复制标题
使用概念约束从演示和技能修复中学习的鲁棒机器人
DOI：
10.1109/iros.2018.8594133
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Bradley Hayes
中科院分区：
文献类型：
--
作者：
Carl L. Mueller;Jeff Venicx;Bradley Hayes
从演示中学习（LfD）使机器人能够通过利用新手操作员提供的示例快速获得新的技能和能力。虽然这种训练机制是有效的，但由于操作人员无意的错误，这种训练机制可能会出现次优演示，从而对性能产生负面影响。在这项工作中，我们引入了基于演示的概念约束学习（CC-LfD），这是一种用于鲁棒技能学习和技能修复的新算法，它将现场演示过程中基于概念的约束（以规划谓词的形式）的注释合并到LfD过程中。通过我们的评估，我们表明CC-LfD可以用于快速修复技能，只需一个带注释的演示，而无需识别和删除低质量的演示。我们还为迁移学习的潜在应用提供了证据，据此，约束可以用来适应相关任务的演示，从而在很少需要新的演示的情况下达到熟练程度。
Learning from demonstration (LfD) has enabled robots to rapidly gain new skills and capabilities by leveraging examples provided by novice human operators. While effective, this training mechanism presents the potential for sub-optimal demonstrations to negatively impact performance due to unintentional operator error. In this work we introduce Concept Constrained Learning from Demonstration (CC-LfD), a novel algorithm for robust skill learning and skill repair that incorporates annotations of conceptually-grounded constraints (in the form of planning predicates) during live demonstrations into the LfD process. Through our evaluation, we show that CC-LfD can be used to quickly repair skills with as little as a single annotated demonstration without the need to identify and remove low-quality demonstrations. We also provide evidence for potential applications to transfer learning, whereby constraints can be used to adapt demonstrations from a related task to achieve proficiency with few new demonstrations required.