本論文提出一供機器人自我學習之互動式認知教學平台系統。教導者可以透過此互動式教學平台系統讓機器人進行動作行為與動作名稱的命名和物件名稱的學習，並透過結合已學習之知識來組合成更高階之認知。透過人類互動式教導，機器人可經由語音、視覺與馬達等動作接收環境中資訊來擴增自我認知範疇。經由人類教導基礎認知後，機器人內建之互動式認知系統便能將語音、視覺、馬達動作等所學習到的知識以累積的方式互相連結，自行擴大所學之知識。此外，機器人亦能夠自主地以基礎認知為基準透過認知組合的方式轉換成為新的高階複合認知，最終能夠依照不同種之物件作出相對應之高階動作，自主性地完成任務。

This thesis presents an interactive learning method to teach a simulated humanoid robot to name actions, objects and to combine this knowledge to acquire higher -order concepts. Through the teaching of human interaction, the robot is enabled to integrate input information of linguistic, visual and sensorimotor actions received from the environment to expand its own knowledge of the world. Following a human tutor’s linguistic instructions, the robot’s cognitive system learns to link linguistic, visual and sensorimotor knowledge altogether by using cumulative method to expand its knowledge. Moreover, the robot is able to autonomously transfer the grounding from basic knowledge to new higher level composite concepts, and execute the corresponding higher level actions on the specific objects.

[1] Honda_Asimo_Robot, http://world.honda.com/HDTV/ASIMO/.
[2] Toyota_Partner_Robot, http://www.toyota-global.com/innovation/partner_robot/
[3] Metta G., Natale L., Nori F., Sandini G., Vernon D., Fadiga L., von Hofsten C., Rosander K., Santos-Victor J., Bernardino A., and Montesano L. (2010). The iCub Humanoid Robot: An Open-Systems Platform for Research in Cognitive Development. Neural Networks, special issue on Social Cognition: From Babies to Robots. 23(8-9)
[4] Schaal S. (1999). Is imitation learning the route to humanoid robots?. Trends in Cognitive Sciences, 3(6) : 233-242
[5] Aude B., Sylvain C., Rudiger D., and Stefan S. (2008). Robot Programming by Demonstration. Handbook of Robotics: MIT Press
[6] Cangelosi A., Metta G., Sagerer G., Nolfi S., Nehaniv C.L., Fischer K., Tani J., Belpaeme B., Sandini G., Fadiga L., Wrede B., Rohlfing K., Tuci E., Dautenhahn K., Saunders J., and Zeschel A. (2010). Integration of action and language knowledge: A roadmap for developmental robotics. IEEE Transactions on Autonomous Mental Development, 2(3) : 167-195
[7] Harnad S. (1990). The symbol grounding problem. Physica D, 42: 335-346
[8] Harnad S. (1993). Grounding symbols in the analog world with neural nets. Think, 2: 12-78
[9] Cangelosi A. (2010). Grounding language in action and perception: From cognitive agents to humanoid robots. Physics of Life Reviews, 7(2) : 139-151
[10] Steels L. (2002). Grounding symbols through evolutionary language games. In Cangelosi A, Parisi D (Eds) Simulating the Evolution of Language, London: Springer, (p.211-226)
[11] Harnad S., Hanson SJ., and Lubin J. (1995). Learned categorical perception in neural nets: Implications for symbol grounding. In Honavar V, Uhr L (Eds) Symbol Processors and Connectionist Network Models in Artificial Intelligence and Cognitive Modeling: Steps toward principled integration. Academic Press (p. 191-206)
[12] Plunkett K., Sinha C., Moller M., and Strandsry O. (1992). Symbol grounding or the emergence of symbols? Vocabulary grout in children and a connectionist net. Connection Science, 4(3-4): 293-312
[13] Cangelosi A., Greco A., and Harnad S. (2000). From robotic toil to symbolic theft: Grounding transfer from entry-level to higher-level Categories. Connection Science, 12: 143-162
[14] Cangelosi A., and Riga T. (2006). An embodied model for sensorimotor grounding and grounding transfer: Experiments with epigenetic robots, Cognitive Science, 30(4): 673-689
[15] Cangelosi A., and Harnad S. (2000). The adaptive advantage of symbolic theft over sensorimotor toil: Grounding language in perceptual categories. Evolution of Communication, 4: 117-142
[16] Greco A., Riga T., and Cangelosi A. (2003). The acquisition of new categories through grounded symbols: An extended connectionist model. In O. Kaynak, E. Alpaydin, E. Oja & L. Xu (Eds.), Artificial Neural Networks and Neural Information Processing - ICANN/ICONIP 2003. Berlin: Springer, pp. 773-770
[17] Cangelosi A., Hourdakis E., and Tikhanoff V. (2006). Language acquisition and symbol grounding transfer with neural networks and cognitive robots. In Proceedings of 2006 IEEE World Congress on Computational Intelligence (IJCNN 2006), IEEE Press, pp. 2885-2891.
[18] Elman J., Bates E., Johnson M., Karmiloff-Smith A., Parisi D., and Plunkett K. (1996). Rethinking Innateness: A Connectionist Perspective on Development. Cambridge: MIT
[19] Stramandinoli F., Cangelosi A., and Marocco D. (2011). Towards the grounding of abstract words: A neural network model for cognitive robots. Proceedings of IJCNN-2011 International Joint Conference on Neural Networks, San Jose
[20] Cangelosi A., and Riga T. (2006). An embodied model for sensorimotor grounding and grounding transfer: Experiments with epigenetic robots. Cognitive Science, 30(4): 673-689
[21] ODE (Open Dynamics Engine), www.ode.org
[22] 賴科樺.(2013)."互動遊戲人形化機器手之研發",國立台灣科技大學機械工程系，碩士論文。
[23] Abbas Q., Ahmad J., and Bangyal W. (2010). Momentum term heals the performance of Back Propagation Algorithm for digit recognition, Emerging Technologies (ICET), 2010 6th International Conference on , vol., no., pp.16,20, 18-19 Oct doi: 10.1109/ICET.2010.5638387
[24] Lin C.Y., Chuang L.W., Huang C.C., Lin, K.J., and Fahn C.S. (2013). Development of hand posture recognition system for finger gaming robot, Advanced Robotics and Intelligent Systems (ARIS), 2013 International Conference on , vol., no., pp.86,91, May 31 2013-June 2
[25] Chuang L.W., Lin C.Y., and Cangelosi A. (2012). Learning of Composite Actions and Visual Categories via Grounded Linguistic Instructions: Humanoid Robot Simulations, Proceedings of the WCCI 2012 IEEE World Congress on Computational Intelligence, Brisbane, Australia, June, 10-15