Hostname: page-component-76fb5796d-vfjqv Total loading time: 0 Render date: 2024-04-26T18:55:44.925Z Has data issue: false hasContentIssue false
  • English
  • Français

Speed Adaptation in Learning from Demonstration through Latent Space Formulation

Published online by Cambridge University Press:  17 October 2019

Maria Koskinopoulou Open the ORCID record for Maria Koskinopoulou [Opens in a new window] ,
Michail Maniadakis  and
Panos Trahanias
Maria Koskinopoulou*
Affiliation:
Institute of Computer Science, Foundation for Research and Technology – Hellas (FORTH), Heraklion, Greece Department of Computer Science, University of Crete, Heraklion, Crete, Greece
Michail Maniadakis
Affiliation:
Institute of Computer Science, Foundation for Research and Technology – Hellas (FORTH), Heraklion, Greece
Panos Trahanias
Affiliation:
Institute of Computer Science, Foundation for Research and Technology – Hellas (FORTH), Heraklion, Greece Department of Computer Science, University of Crete, Heraklion, Crete, Greece
*
*Corresponding author. E-mail: mkosk@ics.forth.gr
Get access Rights & Permissions [Opens in a new window]

Summary

Performing actions in a timely manner is an indispensable aspect in everyday human activities. Accordingly, it has to be present in robotic systems if they are going to seamlessly interact with humans. The current work addresses the problem of learning both the spatial and temporal characteristics of human motions from observation. We formulate learning as a mapping between two worlds (the observed and the action ones). This mapping is realized via an abstract intermediate representation termed “Latent Space.” Learned actions can be subsequently invoked in the context of more complex human–robot interaction (HRI) scenarios. Unlike previous learning from demonstration (LfD) methods that cope only with the spatial features of an action, the formulated scheme effectively encompasses spatial and temporal aspects. Learned actions are reproduced under the high-level control of a time-informed task planner. During the implementation of the studied scenarios, temporal and physical constraints may impose speed adaptations in the reproduced actions. The employed latent space representation readily supports such variations, giving rise to novel actions in the temporal domain. Experimental results demonstrate the effectiveness of the proposed scheme in the implementation of HRI scenarios. Finally, a set of well-defined evaluation metrics are introduced to assess the validity of the proposed approach considering the temporal and spatial consistency of the reproduced behaviors.

Keywords

Human–robot interactionLearning from demonstrationArtificial temporal cognitionTemporal planningLatent representation
Type
Articles
Information
Robotica , Volume 38 , Special Issue 10: Human–Robot Interaction (HRI) , October 2020 , pp. 1867 - 1879
Copyright
© Cambridge University Press 2019

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Li, M., Bekiroglu, Y., Kragic, D. and Billard, A., “Learning of Grasp Adaptation Through Experience and Tactile Sensing,” Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems IROS, Chicago, IL (2014).CrossRefGoogle Scholar
Schaal, S., Peters, J., Nakanishi, J. and Ijspeert, A., “Control, Planning, Learning Andimitation with Dynamic Movement Primitives,” IROS 2003, Las Vegas, NV (2003) pp. 1–21.Google Scholar
Koskinopoulou, M. and Trahanias, P. E., “A Methodological Framework for Robotic Reproduction of Observed Human Actions: Formulation Using Latent Space Representation,” 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids), Cancun, Mexico (IEEE Press, 2016) pp. 565–572.CrossRefGoogle Scholar
Shon, A., Grimes, D. B., Baker, C. and Rao, R. P. N., “A Probabilistic Framework for Model-Based Imitation Learning,” Proceedings of the Twenty-Sixth Annual Conference of the Cognitive Science Society, Berkeley, CA (2004) pp. 1237–1242.Google Scholar
Calinon, S., Guenterlorent, F. and Billard, A., “On learning, representing and generalizing a task in a humanoid robot,” IEEE Trans. Syst. Man, Cybern. Part B Cybern. 37(2), 286–298 (2007).CrossRefGoogle Scholar
Vuga, R., Nemec, B. and Ude, A., “Speed adaptation for self-improvement of skills learned from user demonstrations,” Robotica 34(12), 2806–2822 (2016).CrossRefGoogle Scholar
Ficuciello, F., Falco, P. and Calinon, S., “A brief survey on the role of dimensionality reduction in manipulation learning and control,” IEEE Robot. Auto. Lett. 3(3), 2608–2615 (2018).CrossRefGoogle Scholar
Hofmann, T., “Probabilistic Latent Semantic Analysis,” Proceedings of the Fifteenth Conference on Uncertainty in Artificial Intelligence, Stockholm, Sweden (1999) pp. 289–296.Google Scholar
Maniadakis, M., Aksoy, E., Asfour, T. and Trahanias, P., “Collaboration of Heterogeneous Agents in Time Constrained Tasks,” Proceedings of the IEEE-RAS International Conference on Humanoid Robots (Humanoids), Cancun, Mexico (2016).CrossRefGoogle Scholar
Maniadakis, M. and Trahanias, P., “Time-Informed, Adaptive Multi-robot Synchronization,” Proceedings of the Simulation of Adaptive Behavior (SAB), Aberystwyth, UK (2016).CrossRefGoogle Scholar
de Rengervé, A., Hirel, J., Andry, P., Quoy, M. and Gaussier, P., “On-Line Learning and Planning in a Pick-and-Place Task Demonstrated Through Body Manipulation,” IEEE International Conference on Development and Learning (ICDL) and on Epigenetic Robotics (Epirob), Frankfurt am Main, Germany (2011) pp. 1–6.Google Scholar
Argall, B. D. and Billard, A. G., “A survey of tactile human-robot interactions,” Robot. Auton. Syst. 58(10), 1159–1176 (2010).CrossRefGoogle Scholar
Brys, T., Harutyunyan, A., Suay, H., Chernova, S., Taylor, M. and Now, A., “Learning from Demonstration and Reinforcement,” International Joint Conference on Artificial Intelligence (IJCAI), Buenos Aires, Argentina (2015) pp. 1–7.Google Scholar
Toris, R. and Chernova, S., “Goal-Based Learning from Demonstration for Mobile Pick-and-Place,” AAAI Fall Symposium on Artificial Intelligence and Human-Robot Interaction, Arlington, Virginia (2014).Google Scholar
Gupta, A., Eppner, C., Levine, S. and Abbeel, P., “Learning Dexterous Manipulation for a Soft Robotic Hand from Human Demonstrations,” 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2016, Daejeon, Korea (2016) pp. 3786–3793.Google Scholar
Kaelbling, L. P., Littman, M. L. and Cassandra, A. R., “Planning and acting in partially observable stochastic domains,” Artif. Intell. 101(1–2), 99–134 (1998).CrossRefGoogle Scholar
Evrard, P., Gribovskaya, E., Calinon, S., Billard, A. and Kheddar, A., “Teaching physical collaborative tasks: object-lifting case study with a humanoid,” Humanoids, Paris, France (2009) pp. 399–404.Google Scholar
Hou, S., Galata, A., Caillette, F., Thacker, N. and Bromiley, P., “Articulated Pose Estimation in a Learned Smooth Space of Feasible Solutions,” IEEE International Conference on Computer Vision (ICCV), Rio de Janeiro, Brazil (2007).Google Scholar
Quirion, S., Duchesne, C., Laurendeau, D. and Marchand, M., “Comparing gplvm approaches for dimensionality reduction in character animation,” J. WSCG 16(2–3), 41–48 (2008).Google Scholar
Hou, S., Galata, A., Caillette, F., Thacker, N. and Bromiley, P., “Real-Time Body Tracking Using a Gaussian Process Latent Variable Model,” ICCV, Rio de Janeiro, Brazil (2007) pp. 1–8.Google Scholar
Kramer, O., On Missing Data Hybridizations for Dimensionality Reduction (Springer, Berlin, Heidelberg, 2013) pp. 189–197.Google Scholar
Koskinopoulou, M., Piperakis, S. and Trahanias, P. E., “Learning from Demonstration Facilitates Human–Robot Collaborative Task Execution,” The Eleventh ACM/IEEE International Conference on Human Robot Interaction, Christchurch (2016) pp. 59–66.Google Scholar
Calinon, S., Pistillo, A. and Caldwell, D. G., “Encoding the Time and Space Constraints of a Task in Explicit Duration Hidden Markov Model,” Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), San Francisco, CA, USA (2011) pp. 3413–3418.Google Scholar
Ijspeert, A. J., Nakanishi, J., Hoffmann, H., Pastor, P. and Schaal, S., “Dynamical movement primitives: Learning attractor models for motor behaviors,” Neural Comput . 25(2), 328–373 (2013).CrossRefGoogle Scholar
Rozo, L., Jimenez, P. and Torras, C., “A robot learning from demonstration framework to perform force-based manipulation tasks,” Intell. Service Robot. 6(1), 33–51 (2013).CrossRefGoogle Scholar
Rozo, L., Silverio, J., Calinon, S. and Caldwell, D. G., “Learning controllers for reactive and proactive behaviors in human-robot collaboration,” Front. Robot. AI 3, 1–111 (2016).CrossRefGoogle Scholar
Ewerton, M., Maeda, G., Neumann, G., Kisner, V., Kollegger, G., Wiemeyer, J. and Peters, J., “Movement Primitives with Multiple Phase Parameters,” IEEE International Conference on Robotics and Automation (ICRA), Stockholm, Sweden (2016) pp. 201–206.Google Scholar
Dechter, R., Meiri, I. and Pearl, J., “Temporal constraint networks,” Artif. Intell. 49(1–3), 61–95 (1991).CrossRefGoogle Scholar
Morris, P., “Dynamic Controllability and Dispatchability Relationships,” In: Integration of AI and OR Techniques in Constraint Programming, Lecture Notes in Computer Science (Springer International Publishing, Cham, 2014) pp. 464–479.Google Scholar
Sigalas, M. P., Markos, M. and Trahanias, P., “Full-body pose tracking the top view reprojection approach,” IEEE Trans. Pattern Anal. Mach. Intell. 38(8), 1569–1582 (2015).CrossRefGoogle Scholar
Lawrence, N. and Hyvarinen, A., “Probabilistic non-linear principal component analysis with gaussian process latent variable models,” J. Mach. Learn. Res. 6, 1783–1816 (2005).Google Scholar
Levina, E. and Bickel, P. J., “Maximum likelihood estimation of intrinsic dimension,” NIPS (2004).Google Scholar
Rusinkiewicz, S. and Levoy, M., “Efficient Variants of the ICP Algorithm,” Third International Conference on 3D Digital Imaging and Modeling (3DIM), Quebec, Canada (2001).Google Scholar
Monnig, N. D., Fornberg, B. and Meyer, F. G., “Inverting non-linear dimensionality reduction with scale-free radial basis interpolation,” Technical Report (2013).CrossRefGoogle Scholar
Amorim, E., Brazil, E. V., Mena-Chalco, J. P., Velho, L., Nonato, L. G., Samavati, F. F. and Sousa, M. C., “Facing the high-dimensions: Inverse projection with radial basis functions,” Comput. Graph. 48, 35–47 (2015).CrossRefGoogle Scholar
Dubois, D. and Prade, H., Possibility Theory. An approach to Computerized Processing of Uncertainty (Plenum Press, New York, 1988).Google Scholar
Khadar, B., Rajesh, A., Madhusudhan, R., Ramanaiah, M. V. and Karthikeyan, K., “Statistical optimization for generalised fuzzy number,” Int. J. Modern Eng. Res. 3(2), 647–651 (2013).Google Scholar

Koskinopoulou et al. supplementary material

Koskinopoulou et al. supplementary material

Download Koskinopoulou et al. supplementary material(Video)
Video 7.7 MB