Achieving Synergy in Cognitive Behavior of Humanoids via Deep Learning of Dynamic Visuo-Motor-Attentional Coordination

TL;DR

This paper introduces a novel deep neural network model that enables humanoid robots to develop coordinated visuo-motor behaviors through iterative learning, enhancing their cognitive capabilities in object manipulation and gesture response.

Contribution

The study presents the Visuo-Motor Deep Dynamic Neural Network (VMDNN), a new model that integrates vision, motor, and higher-level networks for improved cognitive coordination in robots.

Findings

Synergetic visuo-motor coordination can be learned through iterative training.

Self-organization of spatio-temporal hierarchies enhances cognitive functions.

Higher-level networks effectively manipulate lower-level processes.

Abstract

The current study examines how adequate coordination among different cognitive processes including visual recognition, attention switching, action preparation and generation can be developed via learning of robots by introducing a novel model, the Visuo-Motor Deep Dynamic Neural Network (VMDNN). The proposed model is built on coupling of a dynamic vision network, a motor generation network, and a higher level network allocated on top of these two. The simulation experiments using the iCub simulator were conducted for cognitive tasks including visual object manipulation responding to human gestures. The results showed that synergetic coordination can be developed via iterative learning through the whole network when spatio-temporal hierarchy and temporal one can be self-organized in the visual pathway and in the motor pathway, respectively, such that the higher level can manipulate them with abstraction.