文档详情

An experimental comparison of imitation paradigms used in social robotics.pdf

发布:2017-04-07约字共6页下载文档
文本预览下载声明
An Experimental Comparison of Imitation Paradigms used in Social Robotics Joe Saunders, Chrystopher L. Nehaniv, Kerstin Dautenhahn Adaptive Systems Research Group, School of Computer Science, University of Hertfordshire College Lane, Hatfield Herts AL10 9AB, United Kingdom {J.2.Saunders, C.L.Nehaniv, K.Dautenhahn}@herts.ac.uk Abstract We study and contrast particular issues arising in two social learning paradigms that are widely used in robotics research: (i) following or matched-dependent behaviour and (ii) static observational learning. Ex- periments are carried out with physical Khepera robots whose controllers include motor schemas and new neu- ral network based methods for model agent-centred per- ception of angle and distance. The robots are trained to perceive the dynamic movement of a human or robot demonstrator carrying a light source. The robots learn the behaviour either through perception from a static location or while following. The differences and impli- cations of the results of both the following and obser- vation mechanisms are compared and contrasted. 1 Social Learning Paradigms Psychologists consider imitation1 to be one of the key elements in social learning, with social learning leading to the acceleration of the acquisition of intel- ligent behaviour [24, 8, 7]. If robots could imitate ei- ther each other or humans, both singly and in groups, then robot task acquisition would be easier and faster with the possibility of increased behavioural complex- ity and ultimately some form of cultural transmis- sion [1]. In fact we consider this social dimension to be the key to making robots behave more intelligently [6], an approach inspired from studies of social animals (e.g. apes) and the ‘social intelligence hypothesis’ [5], which proposes social origins for primate intelligence. In our research program we investigate social learn- ing and interaction between both human/robot and robot/robot pairs to understand the social dimension of imitative be
显示全部
相似文档