Quantifying teaching behavior in robot learning from demonstration

International Journal of Robotics Research - Tập 39 Số 1 - Trang 54-72 - 2020
Aran Sena1, Matthew Howard1
1Department of Engineering, King’s College London, UK

Tóm tắt

Learning from demonstration allows for rapid deployment of robot manipulators to a great many tasks, by relying on a person showing the robot what to do rather than programming it. While this approach provides many opportunities, measuring, evaluating, and improving the person’s teaching ability has remained largely unexplored in robot manipulation research. To this end, a model for learning from demonstration is presented here that incorporates the teacher’s understanding of, and influence on, the learner. The proposed model is used to clarify the teacher’s objectives during learning from demonstration, providing new views on how teaching failures and efficiency can be defined. The benefit of this approach is shown in two experiments ([Formula: see text] and [Formula: see text], respectively), which highlight the difficulty teachers have in providing effective demonstrations, and show how [Formula: see text]–180% improvement in teaching efficiency can be achieved through evaluation and feedback shaped by the proposed framework, relative to unguided teaching.

Từ khóa


Tài liệu tham khảo

10.1145/1015330.1015430

10.3389/frobt.2018.00132

10.1109/ICRA.2015.7139728

10.1145/1228716.1228725

10.1016/j.robot.2008.10.024

10.1109/ICHR.2006.321361

10.1007/978-3-642-00982-2_1

10.1007/978-3-540-30301-5_60

10.2478/s13230-010-0001-5

Cakmak M, 2011, Proceedings of the ICML Workshop on New Developments in Imitation Learning

10.1016/j.artint.2014.08.005

10.1007/s11370-015-0187-9

10.1145/1228716.1228751

10.1075/is.8.3.08cal

10.1007/978-94-007-7194-9_68-1

10.1109/IROS.2010.5652040

10.2478/pjbr-2014-0005

10.2200/S00568ED1V01Y201402AIM028

10.1109/TSMCA.2012.2207108

10.3758/BRM.41.4.1149

10.1006/jcss.1995.1003

10.1561/1100000005

10.1007/s12369-012-0161-z

10.1109/ROBIO.2014.7090323

10.1515/pjbr-2018-0009

10.1177/0278364917690592

10.1007/s10846-015-0290-3

10.1109/ICRA.2018.8461079

Khan F, 2011, Neural Information Processing Systems, 24, 1449

10.3390/robotics2030122

10.1007/978-3-319-64816-3_8

Maeda G, 2017, Conference on Robot Learning (CoRL)

Ng AY, 2000, Proceedings of the Seventeenth International Conference on Machine Learning (ICML), 663

10.1145/860575.860614

Niekum S (2012) ar_track_alvar. http://wiki.ros.org/ar_track_alvar.

10.1145/2909824.3020252

10.1109/ROMAN.2016.7745110

10.1007/s11370-017-0235-8

Schaal S, 1997, Neural Information Processing Systems, 9, 1040

10.1007/4-431-31381-8_23

10.1027/1614-2241/a000016

10.1109/ICRA.2018.8461194

10.1145/2157689.2157784

10.1109/HRI.2016.7451753

10.1109/TRO.2015.2495003

10.1109/ROMAN.2009.5326274

10.1145/2909824.3020230

10.1109/DEVLRN.2010.5578841

Zhu X, 2015, Proceedings of the Twenty-Ninth Conference on Artificial Intelligence (AAAI)

Thông tin
Thông tin xuất bản

International Journal of Robotics Research

Tập 39 Số 1

54-72

Thông tin tác giả