Abstract
Parsing of action sequences is the process of segmenting observed behavior into individual actions. In robotics, this process is critical for imitation learning from observation and for representing an observed behavior in a form that may be communicated to a human. In this paper, we develop a model for action parsing, based on our understanding of principles of grounded cognitive processes, such as perceptual decision making, behavioral organization, and memory formation.We present a neural-dynamic architecture, in which action sequences are parsed using a mathematical and conceptual framework for embodied cognition—the Dynamic Field Theory. In this framework, we introduce a novel mechanism, which allows us to detect and memorize actions that are extended in time and are parametrized by the target object of an action. The core properties of the architecture are demonstrated in a set of simple, proof-of-concept experiments.
References
[1] Gutemberg Guerra-Filho and Yiannis Aloimonos. The syntax of human actions and interactions. Journal of Neurolinguistics, 25(5):500–514, 2012. 10.1016/j.jneuroling.2009.12.006Search in Google Scholar
[2] Yiannis Aloimonos. Sensory grammars for sensor networks. Journal of Ambient Intelligence and Smart Environments, 1(1):15–21, January 2009. 10.3233/AIS-2009-0002Search in Google Scholar PubMed PubMed Central
[3] Florentin Wörgötter, Eren Erdal Aksoy, Norbert Krüger, Justus Piater, Ales Ude, and Minija Tamosiunaite. A simple ontology of manipulation actions based on hand-object relations. IEEE Transactions on Autonomous Mental Development, 5(2):117– 134, june 2013. 10.1109/TAMD.2012.2232291Search in Google Scholar
[4] Georg Layher, Martin A Giese, and Heiko Neumann. Learning representations of animated motion sequences—A neural model. Topics in Cognitive Science, 6(1):170–182, January 2014. 10.1111/tops.12075Search in Google Scholar
[5] Gregor Schöner. Dynamical systems approaches to cognition. In Ron Sun, editor, Cambridge Handbook of Computational Cognitive Modeling, pages 101–126, Cambridge, UK, 2008. Cambridge University Press. 10.1017/CBO9780511816772.007Search in Google Scholar
[6] Yulia Sandamirskaya. Dynamic neural fields as a step towards cognitive neuromorphic architectures. Frontiers in Neuroscience, 7:276, 2013. 10.3389/fnins.2013.00276Search in Google Scholar PubMed PubMed Central
[7] Yulia Sandamirskaya and Gregor Schöner. An embodied account of serial order: How instabilities drive sequence generation. Neural Networks, 23(10):1164–1179, December 2010. 10.1016/j.neunet.2010.07.012Search in Google Scholar PubMed
[8] Yulia Sandamirskaya and Gregor Schöner. Serial order in an acting system: a multidimensional dynamic neural fields implementation. In 9th IEEE International Conference on Development and Learning (ICDL 2010), 2010. 10.1109/DEVLRN.2010.5578834Search in Google Scholar
[9] Mathis Richter, Yulia Sandamirskaya, and Gregor Schöner. A robotic architecture for action selection and behavioral organization inspired by human cognition. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2012), pages 2457–2464, 2012. 10.1109/IROS.2012.6386153Search in Google Scholar
[10] Yulia Sandamirskaya, Stephan K.U. Zibner, Sebastian Schneegans, and Gregor Schöner. Using dynamic field theory to extend the embodiment stance toward higher cognition. New Ideas in Psychology, 31(3):322 – 339, 2013. 10.1016/j.newideapsych.2013.01.002Search in Google Scholar
[11] Shun-ichi Amari. Dynamics of pattern formation in lateralinhibition type neural fields. Biological Cybernetics, 27:77–87, 1977. 10.1007/BF00337259Search in Google Scholar PubMed
[12] Hugh R Wilson and Jack D Cowan. A mathematical theory of the functional dynamics of cortical and thalamic nervous tissue. Kybernetik, 13(2):55–80, 1973. 10.1007/BF00288786Search in Google Scholar PubMed
[13] Stephen Grossberg. Nonlinear neural networks: Principles, mechanisms, and architectures. Neural Networks, 1(1):17–61, 1988. Search in Google Scholar
[14] Stephan K U Zibner, Christian Faubel, Ioannis Iossifidis, and Gregor Schöner. Dynamic Neural Fields as Building Blocks for a Cortex-Inspired Architecture of Robotic Scene Representation. IEEE Transactions on AutonomousMental Development, 3(1):74– 91, 2011. 10.1109/TAMD.2011.2109714Search in Google Scholar
[15] Y Sandamirskaya and T Storck. Neural-dynamic architecture for looking: Shift from visual to motor target representation for memory saccade. In IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL EPIROB 2014), 2014. 10.1109/DEVLRN.2014.6982951Search in Google Scholar
[16] Radu Bogdan Rusu and Steve Cousins. 3D is here: Point Cloud Library (PCL). IEEE International Conference on Robotics and Automation (ICRA 2011), pages 1–4, May 2011. Search in Google Scholar
[17] Martin A. Fischler and Robert C. Bolles. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM, 24(6):381–395, June 1981. 10.1145/358669.358692Search in Google Scholar
[18] Michael Berger, Christian Faubel, Joseph Norman, Howard Hock, and Gregor Schöner. The Counter-Change Model of Motion Perception : An Account Based on Dynamic Field Theory The Dynamic Field Model of Counter-Change Motion. Lecture Notes in Computer Science, 7552:579–586, 2012. Search in Google Scholar
[19] Rudolph Emil Kalman. A new approach to linear filtering and prediction problems. Transactions of the American Society of Mechanical Engineers - Journal of Basic Engineering, 82(1):35– 45, 1960. 10.1115/1.3662552Search in Google Scholar
[20] HaroldWKuhn. The hungarian method for the assignment problem. Naval Research Logistics Quarterly, 2(1-2):83–97, 1955. 10.1002/nav.3800020109Search in Google Scholar
[21] James Munkres. Algorithms for the assignment and transportation problems. Journal of the Society for Industrial & Applied Mathematics, 5(1):32–38, 1957. 10.1137/0105003Search in Google Scholar
[22] Christian Faubel and Gregor Schöner. Learning to recognize objects on the fly: a neurally based dynamic field approach. Neural Networks, 21(4):562–576, 2008. Search in Google Scholar
[23] Yulia Sandamirskaya, Mathis Richter, and Gregor Schöner. A neural-dynamic architecture for behavioral organization of an embodied agent. In IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL EPIROB 2011), pages 1–7, 2011. 10.1109/DEVLRN.2011.6037353Search in Google Scholar
[24] Boris Duran and Yulia Sandamirskaya. Neural dynamics of hierarchically organized sequences: A robotic implementation. In 12th IEEE-RAS International Conference on Humanoid Robots (Humanoids 2012), pages 357–362, 2012. 10.1109/HUMANOIDS.2012.6651544Search in Google Scholar
[25] Kyuhwa Lee and Yiannis Demiris. Towards incremental learning of task-dependent action sequences using probabilistic parsing. IEEE International Conference of Development and Learning (ICDL), 2011. Search in Google Scholar
[26] Falk Fleischer, Vittorio Caggiano, Peter Thier, and Martin a Giese. Physiologically inspired model for the visual recognition of transitive hand actions. The Journal of neuroscience : the official journal of the Society for Neuroscience, 33(15):6563–80, April 2013. 10.1523/JNEUROSCI.4129-12.2013Search in Google Scholar
[27] Sohrob Kazerounian, Matthew Luciw, Mathis Richter, and Yulia Sandamirskaya. Autonomous reinforcement of behavioral sequences in neural dynamics. In International Joint Conference on Neural Networks (IJCNN), pages 1–8, 2013. 10.1109/IJCNN.2013.6706877Search in Google Scholar
[28] Brenna D. Argall, Sonia Chernova, Manuela Veloso, and Brett Browning. A survey of robot learning from demonstration. Robotics and Autonomous Systems, 57(5):469–483, May 2009. 10.1016/j.robot.2008.10.024Search in Google Scholar
[29] Stefan Schaal. Is imitation learning the route to humanoid robots? Trends in Cognitive Sciences, 3(6):233–242, 1999. 10.1016/S1364-6613(99)01327-3Search in Google Scholar
© 2015 David Lobato et al.
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.