DocumentCode :
186274
Title :
Generation of sensory reflex behavior versus intentional proactive behavior in robot learning of cooperative interactions with others
Author :
Murata, Shotaro ; Yamashita, Yukihiko ; Arie, Hiroaki ; Ogata, Takaaki ; Jun Tani ; Sugano, S.
Author_Institution :
Dept. of Modern Mech. Eng., Waseda Univ., Tokyo, Japan
fYear :
2014
fDate :
13-16 Oct. 2014
Firstpage :
242
Lastpage :
248
Abstract :
This paper investigates the essential difference between two types of behavior generation schemes, namely, sensory reflex behavior generation and intentional proactive behavior generation, by proposing a dynamic neural network model referred to as stochastic multiple-timescale recurrent neural network (S-MTRNN). The proposed model was employed in an experiment involving robots learning to cooperate with others under the condition of potential unpredictability of the others´ behaviors. The results of the learning experiment showed that sensory reflex behavior was generated by a self-organizing probabilistic prediction mechanism when the initial sensitivity characteristics in the network dynamics were not utilized in the learning process. In contrast, proactive behavior with a deterministic prediction mechanism was developed when the initial sensitivity was utilized. It was further shown that in situations where unexpected behaviors of others were observed, the behavioral context was re-situated by adaptation of the internal neural dynamics by means of simple sensory reflexes in the former case. In the latter case, the behavioral context was re-situated by error regression of the internal neural activity rather than by sensory reflex. The role of the top-down and bottom-up interactions in dealing with unexpected situations is discussed.
Keywords :
intelligent robots; learning systems; neurocontrollers; probability; recurrent neural nets; regression analysis; stochastic processes; S-MTRNN; bottom-up interaction; cooperative interactions; deterministic prediction mechanism; dynamic neural network model; error regression; initial sensitivity characteristics; intentional proactive behavior generation; internal neural activity; internal neural dynamics; network dynamics; robot learning; self-organizing probabilistic prediction mechanism; sensory reflex behavior generation; stochastic multiple-timescale recurrent neural network; top-down interaction; Context; Neurons; Predictive models; Robot sensing systems; Sensitivity; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Development and Learning and Epigenetic Robotics (ICDL-Epirob), 2014 Joint IEEE International Conferences on
Conference_Location :
Genoa
Type :
conf
DOI :
10.1109/DEVLRN.2014.6982988
Filename :
6982988
Link To Document :
بازگشت