Artificial Life and Robotics

, Volume 15, Issue 4, pp 515–521 | Cite as

Rapid behavior adaptation for human-centered robots in a dynamic environment based on the integration of primitive confidences on multi-sensor elements

Original Article

Abstract

This article presents a method for tele-operated mobile robots to rapidly adapt to behavior policies. Since real-time adaptation requires frequent observations of sensors and the behavior of users, rapid policy adaptation cannot be achieved when significant data are not differentiated from insignificant data in every process cycle. Our method solves this problem by evaluating the significance of data for learning based on changes in the degree of confidence. A small change in the degree of confidence can be regarded as reflecting insignificant data for learning (that data can be discarded). Accordingly, the system can avoid having to store experience data too frequently, and the robot can adapt more rapidly to changes in the user’s policy. In this article, we confirm that by taking advantage of a significance evaluation not only of proposition of behavior, but also of each proposition of each piece of sensor-level data, a robot can rapidly adapt to a user’s policy. We discuss the results of two experiments in static and dynamic environments, in both of which the user switched policies between “avoid” and “approach.”

Key words

Bayesian network Rapid adaptation Degree of confidence Dirichlet distribution 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Datar M, Gionis A, Indyk P, et al (2002) Maintaining stream statistics over sliding windows. Siam J Comput 31(6):1794–1813CrossRefMATHMathSciNetGoogle Scholar
  2. 2.
    Kifer D, Ben-David S, Gehrke J (2004) Detecting change in data streams. Proceedings of the 30th VLDB Conference, TorontoGoogle Scholar
  3. 3.
    Widmer G, Kubat M (1996) Learning in the presence of concept drift and hidden contexts. Mach Learn 23:69–101Google Scholar
  4. 4.
    Klinkenberg R (2004) Learning drifting concepts: example selection vs. example weighting. Intel Data Anal, Special Issue on Incremental Learning Systems Capable of Dealing with Concept Drift 8(3):281–300Google Scholar
  5. 5.
    Billsus D, Pazzani MJ (2000) User modeling for adaptive news access. User Modeling and User-Adapted Interaction 10:147–180CrossRefGoogle Scholar
  6. 6.
    Chiu P, Webb G (1998) Using decision tree for agent modeling: improving prediction performance. User Modeling and User-Adapted Interaction 8:131–152CrossRefGoogle Scholar
  7. 7.
    Jordan MI, Jacobs RA (1993) Hierarchical mixture of experts and the EM algorithm. Proceedings of the International Joint Conference on Neural Networks, pp 1339–1344Google Scholar
  8. 8.
    Haruno M, Wolpert DM, Kawata M (2001) MOSAIC model for sensorimotor learning and control. Neural Comput 13:2201–2220CrossRefMATHGoogle Scholar
  9. 9.
    Kaelbling LP, Littman ML, Moore AW (1996) Reinforcement learning:a survey. J Artif Intel Res 4:237–285Google Scholar
  10. 10.
    Inamura T, Inaba M, Inoue H (2004) PEXIS: probabilistic experience representation based adaptive interaction system for personal robots. Syst Comput Jpn 35(6):98–109CrossRefGoogle Scholar
  11. 11.
    Inamura T, Inaba M, Inoue H (2000) User adaptation of human-robot interaction model based on Bayesian network and introspection of interaction experience. Proceedings of the International Conference on Intelligent Robots and Systems (IROS 2000), pp 2139–2144Google Scholar
  12. 12.
    Inamura T, Inaba M, Inoue H (1999) Acquisition of probabilistic decision model based on interactive teaching method. Proceedings of the 9th International Conference on Advanced Robotics, ICAR, pp 523–528Google Scholar
  13. 13.
    Nicolescu M, Mataric M (2001) Learning and interacting in human-robot domains. IEEE Trans Syst Man Cybern Part A: Syst Hum 31(5):419–430CrossRefGoogle Scholar
  14. 14.
    Tareeq SM, Inamura T (2008) A sample discarding strategy for rapid adaptation to new situation for Bayesian behavior learning. Proceedings of the IEEE International Conference on Robotics and Biomimetics pp 1950–1955Google Scholar
  15. 15.
    Pearl J (1988) Probabilistic reasoning in intelligent system; network of plausible inference. 2nd edn. Morgan Kaufman, Los AltosGoogle Scholar
  16. 16.
    Michel O (2004) Webots: professional mobile robot simulation. J Adv Robotics Syst 1(1):39–42Google Scholar

Copyright information

© International Symposium on Artificial Life and Robotics (ISAROB). 2010

Authors and Affiliations

  1. 1.Graduate University for Advanced Studies, National Institute of InformaticsTokyoJapan

Personalised recommendations