- 232 Downloads
Research in learning from demonstrations has focused primarily on the question of how to utilize demonstrations to learn new behavior assuming that the demonstrator (teacher) explicitly teaches the learner. In this paper, we focus our attention on learning from unplanned demonstrations. In such cases, the learner has to take the initiative and decide for itself what actions of the—assumed unaware—demonstrator are to be learned. Moreover, the actions of the demonstrator are not pre-segmented and the learner needs to discover the boundaries of interesting behaviors without any help from the demonstrator. We propose a fluid imitation engine that augments a traditional LfD system. The proposed engine casts the problem as a well-defined constrained motif discovery problem subject to constraints that are driven from object and behavior saliency, as well as behavior relevance to the learner’s goals and abilities. Relation between perceived behaviors of the demonstrator and changes in objects in the environment is quantified using a change-causality test that is shown to provide better results compared to traditional g-causality tests. The main advantage of the proposed system is that it can naturally combine information from all available sources including low-level saliency measures and high-level goal-driven relevance constraints. The paper also reports a series of experiments to evaluate the utility of the proposed engine in learning navigation tasks with increasing complexity using both simulated and real world robots.
KeywordsLearning from demonstrations Imitation learning Fluid imitation Action segmentation Motif discovery
This work was partially supported by Kyoto University’s GCOE Project.
- 3.Alippi C, Roveri M (2006) An adaptive CUSUM-based test for signal change detection. In: 2006 IEEE international symposium on circuits and systems, p 4 Google Scholar
- 10.EPFL: http://www.e-puck.org
- 11.Minnen D, Starner T, Essa I, Isbell C (2007) Improving activity discovery with automatic neighborhood estimation. In: Int. joint conf. on artificial intelligence Google Scholar
- 13.Mohammad Y, Nishida T (2009) Robust singular spectrum transform. In: The twenty second international conference on industrial, engineering and other applications of applied intelligent systems IEA-AIE 2009, pp 123–132 Google Scholar
- 15.Mohammad Y, Nishida T (2010) Mining causal relationships in multidimensional time series. In: Szczerbicki E, Nguyen NT (eds) Smart information and knowledge management: advances, challenges, and critical issues, studies in computational intelligence. Springer, Berlin, pp 309–338 Google Scholar
- 18.Mohammad Y, Nishida T, Okada S (2009) Unsupervised simultaneous learning of gestures, actions and their associations for human-robot interaction. In: Proceedings of the 2009 IEEE/RSJ international conference on intelligent robots and systems, IROS’09. IEEE Press, Piscataway, pp 2537–2544 CrossRefGoogle Scholar
- 20.Nagai Y, Rohlfing KJ (2007) Can motionese tell infants and robots. What to imitate? In: 4th international symposium on imitation in animals and artifacts, pp 299–306 Google Scholar
- 22.Scassellati BB (1999) Knowing what to imitate and knowing when you succeed. In: AISB99 symposium on imitation in animals and artifacts, pp 105–113 Google Scholar
- 23.V-REP: http://www.v-rep.eu/
- 24.Floyd MW, Esfandiari B, Lam K (2008) A case-based reasoning approach to imitating RoboCup players. In: AAAI Google Scholar