×

Learning motor skills by imitation: A biologically inspired robotic model. (English) Zbl 1011.93078

A model for motor skills imitation that is biologically inspired is presented. Each module within the author’s model has a function that is based on a brain region responsible for the control of movement in primates. These modules include high level abstractions of the spinal cord, the primary and premotor complexes, the cerebellum and the temporal cortex. Each module is modeled at a connection level via neural networks. Learning of new combinations of movements is enabled in the premotor complex and cerebellum modules. A DRAMA (dynamic recurrent associative memory architecture) approach is used to allow learning of time series and spatio-temporal invariance in multimodal inputs.
The author’s work is motivated by a desire to develop a control mechanism for imitation in robots and to give a possible model of the neurological circuits that enable primates’ imitative skills.

MSC:

93C85 Automated systems (robots, etc.) in control theory
93A30 Mathematical modelling of systems (MSC2010)
92D50 Animal behavior
92B20 Neural networks for/in biological studies, artificial life and related topics
68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI Link