Learning Compliant Movement Primitives Through Demonstration and Statistical Generalization

Miha Deniša; Andrej Gams; Aleš Ude; Tadej Petrič

首页> 外文期刊>IEEE / ASME Transactions on Mechatronics >Learning Compliant Movement Primitives Through Demonstration and Statistical Generalization

【24h】

Learning Compliant Movement Primitives Through Demonstration and Statistical Generalization

机译：通过演示和统计概括学习合规动作原语

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

In this paper, we address the problem of simultaneously achieving low trajectory tracking errors and compliant control without using explicit mathematical models of task dynamics. To achieve this goal, we propose a new movement representation called compliant movement primitives (CMPs), which encodes position trajectory and associated torque profiles and can be learned from a single user demonstration. With the proposed control framework, the robot can remain compliant and consequently safe for humans sharing its workspace, even if high trajectory tracking accuracy is required. We developed a statistical learning approach that can use a database of existing CMPs and compute new ones, adapted for novel task variations. The proposed approach was evaluated on a Kuka LWR-4 robot performing 1) a discrete pick-and-place task with objects of varying weight and 2) a periodic handle turning operation. The evaluation of the discrete task showed a 15-fold decrease of the tracking error while exhibiting compliant behavior compared to the standard feedback control approach. It also indicated no significant rise in the tracking error while using generalized primitives computed by the statistical learning method. With respect to unforeseen collisions, the proposed approach resulted in a 75% drop of contact forces compared to standard feedback control. The periodic task demonstrated on-line use of the proposed approach to accomplish a task of handle turning.

机译：在本文中，我们解决了在不使用任务动力学的显式数学模型的情况下同时实现低轨迹跟踪误差和顺应性控制的问题。为了实现此目标，我们提出了一种新的运动表示形式，称为顺应性运动原语（CMP），该运动表示对位置轨迹和相关的扭矩曲线进行编码，并且可以从单个用户演示中学习。利用所提出的控制框架，即使需要高轨迹跟踪精度，机器人也可以保持合规性，因此对于人类共享其工作区而言是安全的。我们开发了一种统计学习方法，可以使用现有CMP的数据库并计算新的CMP，以适应新颖的任务变化。在Kuka LWR-4机器人上执行以下操作时评估了所提出的方法：1）进行离散的拾取和放置任务，对象的重量不同； 2）周期性的手柄转动操作。与标准反馈控制方法相比，对离散任务的评估显示出跟踪行为减少了15倍，同时表现出顺应性。它也表明在使用由统计学习方法计算的广义基元时，跟踪误差没有明显增加。对于不可预见的碰撞，与标准反馈控制相比，所提出的方法使接触力下降了75％。定期任务演示了在线使用所提出的方法来完成手柄转向任务。

著录项

来源
《IEEE / ASME Transactions on Mechatronics》 |2016年第5期|2581-2594|共14页
作者
Miha Deniša; Andrej Gams; Aleš Ude; Tadej Petrič;
展开▼
作者单位

Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jozef Stefan Institute, Ljubljana, Slovenia;

Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jozef Stefan Institute, Ljubljana, Slovenia;

Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jozef Stefan Institute, Ljubljana, Slovenia;

Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jozef Stefan Institute, Ljubljana, Slovenia;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词
Trajectory; Torque; Robot sensing systems; Biological system modeling; Dynamics; Mathematical model;

机译：轨迹;扭矩;机器人传感系统;生物系统建模;动力学;数学模型;

相似文献

外文文献
中文文献
专利

1. TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT . Statistical Learning with Sparsity: The Lasso and Generalizations . Boca Raton : CRC Press . TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT . Statistical Learning with Sparsity: The Lasso and Generalizations Statistical Learning with Sparsity: The Lasso and Generalizations . Boca Raton Boca Raton : CRC Press CRC Press . [J] . Kondofersky Ivan, Theis Fabian J. Biometrics: Journal of the Biometric Society : An International Society Devoted to the Mathematical and Statistical Aspects of Biology . 2018,第2期

机译：Trevor Hastie，Robert Tibshirani和Martin Wainwright。统计学习稀疏性：套索和概括。 Boca Raton：CRC压力机。 Trevor Hastie，Robert Tibshirani和Martin Wainwright Trevor Hastie，Robert Tibshirani和Martin Wainwright。统计学习与稀疏性：套索与稀疏性统计学习：套索和概括。 Boca Raton Boca Raton：CRC按CRC压力机。
2. TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT . Statistical Learning with Sparsity: The Lasso and Generalizations . Boca Raton : CRC Press . TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT . Statistical Learning with Sparsity: The Lasso and Generalizations Statistical Learning with Sparsity: The Lasso and Generalizations . Boca Raton Boca Raton : CRC Press CRC Press . [J] . Kondofersky Ivan, Theis Fabian J. Biometrics: Journal of the Biometric Society : An International Society Devoted to the Mathematical and Statistical Aspects of Biology . 2018,第2期

机译：Trevor Hastie，Robert Tibshirani和Martin Wainwright。统计学习与稀疏性：套索和概括。 Boca Raton：CRC压力机。 Trevor Hastie，Robert Tibshirani和Martin Wainwright Trevor Hastie，Robert Tibshirani和Martin Wainwright。统计学习与稀疏性：套索与稀疏性统计学习：套索和概括。 Boca Raton Boca Raton：CRC按CRC压力机。
3. Accelerated Sensorimotor Learning of Compliant Movement Primitives [J] . Tadej Petrič, Andrej Gams, Luca Colasanto, IEEE Transactions on Robotics . 2018,第6期

机译：顺应性运动原语的加速感觉运动学习
4. Learning and Generalization of Dynamic Movement Primitives by Hierarchical Deep Reinforcement Learning from Demonstration [C] . Wonchul Kim, Chungkeun Lee, H. Jin Kim IEEE/RSJ International Conference on Intelligent Robots and Systems . 2018

机译：通过演示的分层深度强化学习对动态动作基元进行学习和泛化
5. A framework for learning from demonstration, generalization and practice in human-robot domains. [D] . Nicolescu, Monica Nicolette. 2003

机译：一个从人机交互领域的演示，概括和实践中学习的框架。
6. Phase-Synchronized Learning of Periodic Compliant Movement Primitives (P-CMPs) [O] . Tadej Petrič 2020

机译：相位同步学习定期兼容运动原语（P-CMPS）
7. Balanced motor primitive can explain generalization of motor learning effects between unimanual and bimanual movements [O] . Ken Takiyama, Yutaka Sakai 2016

机译：平衡的电动机原语可以解释在不加凡和生理运动之间的电机学习效应的概括

Learning Compliant Movement Primitives Through Demonstration and Statistical Generalization

摘要

著录项

相似文献

相关主题

期刊订阅