Learning object relationships which determine the outcome of actions

Abstract

Infants extend their repertoire of behaviours from initially simple behaviours with single objects to complex behaviours dealing with spatial relationships among objects. We are interested in the mechanisms underlying this development in order to achieve similar development in artificial systems. One mechanism is sensorimotor differentiation, which allows one behaviour to become altered in order to achieve a different result; the old behaviour is not forgotten, so differentiation increases the number of available behaviours. Differentiation requires the learning of both sensory abstractions and motor programs for the new behaviour; here we focus only on the sensory aspect: learning to recognise situations in which the new behaviour succeeds. We experimented with learning these situations in a realistic physical simulation of a robotic manipulator interacting with various objects, where the sensor space includes the robot arm position data and a Kinect-based vision system. The mechanism for learning sensory abstractions for a new behaviour is a component in the larger enterprise of building systems which emulate the mechanisms of infant development.

This is a preview of subscription content, log in to check access.

References

  1. [1]

    Frank Guerin, Dirk Kraft, and Norbert Krüger. A survey of the ontogeny of tool use: from sensorimotor experience to planning. IEEE Transactions on Autonomous Mental Development, 5(1):18–45, 2013.

    Article  Google Scholar 

  2. [2]

    J. Piaget. The Origins of Intelligence in Children. London: Routledge & Kegan Paul, 1936. (French version 1936, translation 1952).

    Google Scholar 

  3. [3]

    Jeffrey J. Lockman. A perception-action perspective on tool use development. Child Development, 71(1):137–144, 2000.

    Article  Google Scholar 

  4. [4]

    L. B. Smith. Dynamic systems, sensori-motor processes and the origins of stability and flexibility. In J. Spencer, M. Thomas, and J. McClelland, editors, Toward a unified theory of development: Connectionism and dynamic systems theories reconsidered. Ox. Uni. Press, 2009.

    Google Scholar 

  5. [5]

    Kurt W. Fischer. A theory of cognitive development: The control and construction of hierarchies of skills. Psychological Review, 87(6):477–531, 1980.

    Article  Google Scholar 

  6. [6]

    P. Willatts. Development of problem-solving strategies in infancy. In D.F. Bjorklund, editor, Children’s Strategies: Contemporary Views of Cognitive Development, pages 23–66. Lawrence Erlbaum, 1990.

    Google Scholar 

  7. [7]

    J. Piaget. The Construction of Reality in the Child. London: Routledge & Kegan Paul, 1937. (French version 1937, translation 1955).

    Google Scholar 

  8. [8]

    Benjamin Rosman and Subramanian Ramamoorthy. Learning spatial relationships between objects. The International Journal of Robotics Research, 30(11):1328–1342, 2011.

    Article  Google Scholar 

  9. [9]

    Richard S. Sutton. Verification, the key to ai, 2006. Unpublished document, available on author’s webpage http://www.cs.ualberta.ca/~sutton /IncIdeas/KeytoAI.html.

  10. [10]

    A. Stoytchev. Some basic principles of developmental robotics. IEEE Trans. Auton. Mental Development, 1(2):122–130, 2009.

    Article  Google Scholar 

  11. [11]

    Esther Thelen. Rhythmical stereotypies in normal human infants. Animal Behaviour, 27(3):699–715, Dec 1979.

    Article  Google Scholar 

  12. [12]

    T. G. R. Bower. Development in Infancy. San Francisco: W.H. Freeman, 1982.

    Google Scholar 

  13. [13]

    Kristine S. Bourgeois, Alexa W. Khawar, S. Ashley Neal, and Jeffrey J. Lockman. Infant manual exploration of objects, surfaces, and their interrelations. Infancy, 8(3):233–252, 2005.

    Article  Google Scholar 

  14. [14]

    Jean Piaget. The Child’s Conception of the World. Harcourt, Brace and Co., New York, 1929.

    Google Scholar 

  15. [15]

    Harold Chaput. The Constructivist Learning Architecture: A Model of Cognitive Development for Robust Autonomous Robots. PhD Thesis, (The University of Texas at Austin, Artificial Intelligence Laboratory), 2004.

    Google Scholar 

  16. [16]

    Severin Fichtl. Making Artificial Intelligence which Copies the Way Babies Learn. Master Thesis, (Aberdeen University, Department of Computing Science), 2011.

    Google Scholar 

  17. [17]

    M Rolf, J J Steil, and M Gienger. Online Goal Babbling for rapid bootstrapping of inverse models in high dimensions. In Development and Learning (ICDL), 2011 IEEE International Conference on, volume 2, pages 1–8, 2011.

    Article  Google Scholar 

  18. [18]

    Gert Kootstra, Mila Popovi., Jimmy Alison Jostate spacergensen, Kamil Kuklinski, Konstantsin Miatliuk, Danica Kragic, and Norbert Krüger. Enabling grasping of unknown objects through a synergistic use of edge and surface information. Int. J. Rob. Res., 31(10):1190–1213, September 2012.

    Article  Google Scholar 

  19. [19]

    J. A. Jorgensen, L.-P. Ellekilde, and H. G. Petersen. Rob-WorkSim — an Open Simulator for Sensor based Grasping. In ISR/ROBOTIK 2010 (41st International Symposium). VDE-Verlag, June 2010.

    Google Scholar 

  20. [20]

    SørenMaagaard Olesen, Simon Lyder, Dirk Kraft, Norbert Krüger, and JeppeBarsøe Jessen. Real-time extraction of surface patches with associated uncertainties by means of Kinect cameras. Journal of Real-Time Image Processing, pages 1–14, 2012.

    Google Scholar 

  21. [21]

    Kourosh Khoshelham and Sander Oude Elberink. Accuracy and Resolution of Kinect Depth Data for Indoor Mapping Applications. Sensors, 12(2):1437–1454, 2012.

    Article  Google Scholar 

  22. [22]

    R B Rusu and S Cousins. 3D is here: Point Cloud Library (PCL). In Robotics and Automation (ICRA), 2011 IEEE International Conference on, pages 1–4, May 2011.

    Google Scholar 

  23. [23]

    N. Pugeault, F. Worgotter, and N. Krüger. Visual primitives: Local, condensed, and semantically rich visual descriptors and their applications in robotics. International Journal of Humanoid Robotics (Special Issue on Cognitive Humanoid Vision), 7(3):379–405, 2010.

    Article  Google Scholar 

  24. [24]

    Martin Riedmiller and Heinrich Braun. RPROP — A Fast Adaptive Learning Algorithm. In Proceedings of the International Symposium on Computer and Information Science VII, Institut fur Logik, Komplexitat und Deduktionssysteme, Universitat Karlsruhe, 1992.

    Google Scholar 

  25. [25]

    M Riedmiller and H Braun. A direct adaptive method for faster backpropagation learning: the RPROP algorithm. IEEE International Conference on Neural Networks, 1993.

    Google Scholar 

  26. [26]

    Miroslav Kubat, Robert Holte, and Stan Matwin. Learning when negative examples abound. In Proceedings of the 9th European Conference on Machine Learning, ECML’ 97, pages 146–153, London, UK, UK, 1997. Springer-Verlag.

    Google Scholar 

  27. [27]

    Severin Fichtl, John Alexander, Dirk Kraft, Jimmy Alison Jorgensen, Norbert Krüger, and Frank Guerin. Rapidly learning reconditions for means-ends behaviour using active learning. In 2nd IEEE International Conference on Development and Learning and Epigenetic Robotics, 2012.

    Google Scholar 

  28. [28]

    Matti Kaariainen. Active Learning in the Non-realizable Case. In Jose L. Balcazar, Philip M. Long, and Frank Stephan, editors, Algorithmic Learning Theory, volume 4264 of Lecture Notes in Computer Science, pages 63–77. Springer Berlin Heidelberg, 2006.

    Google Scholar 

  29. [29]

    Donald E Knuth. Big Omicron and big Omega and big Theta. SIGACT News, 8(2):18–24, April 1976.

    Article  Google Scholar 

  30. [30]

    Sara Stolbach. Active Learning Models and Noise. Course Paper, Columbia University, (Advanced Topics in Computational Learning Theory, COMS 6253), 2007.

    Google Scholar 

  31. [31]

    Ina C. Uzgiris and J. McV. Hunt. Assessment in infancy: ordinal scales of psychological development. University of Illinois Press, 1975.

    Google Scholar 

  32. [32]

    E. Bates, V. Carlson-Luden, and I. Bretherton. Perceptual aspects of tool-using in infancy. Infant Behavior and Development, 3(2):181–190, 1980.

    Google Scholar 

  33. [33]

    Matthew Schlesinger and Jonas Langer. Infants’ developing expectations of possible and impossible tool-use events between ages 8 and 12 months. Developmental Science, 2(2):195–205, 1999.

    Article  Google Scholar 

  34. [34]

    P.J. Kellman and M.E. Arterberry. The Cradle of Knowledge. MIT-Press, 1998.

    Google Scholar 

  35. [35]

    J. M. Mandler. How to build a baby: II. conceptual primitives. Psychological Review, 99(4):587–604, 1992.

    Article  Google Scholar 

  36. [36]

    Denis Mareschal. Computational perspectives on cognitive development. Wiley Interdisciplinary Reviews: Cognitive Science, 1(5):696–708, 2010.

    Article  Google Scholar 

  37. [37]

    Paolo Tommasino, Daniele Caligiore, Marco Mirolli, and Gianluca Baldassarre. Reinforcement learning algorithms that assimilate and accommodate behaviours: an experimental comparison. In 2nd IEEE International Conference on Development and Learning and Epigenetic Robotics, 2012.

    Google Scholar 

  38. [38]

    A.G.E. Collins and M.J. Frank. Cognitive control over learning: Creating, clustering and generalizing task-set structure. Psychological Review, 120:190–229, 2013.

    Article  Google Scholar 

  39. [39]

    Terry Zimmerman and Subbarao Kambhampati. Learning-assisted automated planning: Looking back, taking stock, going forward. AI MAGAZINE, 24(2):73–96, 2003.

    Google Scholar 

  40. [40]

    J. Mugan and B. Kuipers. Autonomous learning of high-level states and actions in continuous environments. IEEE Trans. Autonomous Mental Development, 4(1):70–86, 2012.

    Article  Google Scholar 

  41. [41]

    Kira Mourao, Ronald Petrick, and Mark Steedman. Using kernel perceptrons to learn action effects for planning. International Conference on Cognitive Systems (CogSys 2008), pages 45–50, 2008.

    Google Scholar 

  42. [42]

    E. Ugur, E. Oztop, and E. Sahin. Goal emulation and planning in perceptual space using learned affordances, 2011.

    Google Scholar 

  43. [43]

    A. Ferry, S.J. Hespos, and S. Waxman. Language facilitates category formation in 3-month-old infants. Child Development, 81(2):472–479, 2010.

    Article  Google Scholar 

  44. [44]

    S. J. Hespos and T. Piccin. To generalize or not to generalize: Spatial categories are influenced by physical attributes and language. Developmental Science, 12(1):88–95, 2009.

    Article  Google Scholar 

  45. [45]

    Marianella Casasola. Can language do the driving? the effect of linguistic input on infants’ categorization of support spatial relations. Developmental Psychology, 41(1):183–192, 2005.

    Article  Google Scholar 

  46. [46]

    Katharina J Rohlfing. Facilitating the acquisition of under by means of in and on.a training study in polish. J Child Lang, 33(1):51–69, 2006.

    Article  Google Scholar 

  47. [47]

    Jean Matter Mandler. The foundations of conceptual thought in infancy. Cognitive Development, 7(3):273–285, 1992.

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Norbert Krüger.

About this article

Cite this article

Fichtl, S., Alexander, J., Kraft, D. et al. Learning object relationships which determine the outcome of actions. Paladyn 3, 188–199 (2012). https://doi.org/10.2478/s13230-013-0104-x

Download citation

Keywords

  • Developmental Artificial Intelligence
  • Vision
  • Infant Development
  • Means-end Behaviour
  • Learning Preconditions