Global workspace theory-based cognitive architecture for robots  

基于全局工作空间理论的机器人认知架构

作　　者：HUANG Wenjie CANGELOSI Angelo CHELLA Antonio HUANG Wenjie;CANGELOSI Angelo;CHELLA Antonio(曼彻斯特大学,英国曼彻斯特;曼彻斯特机器人与人工智能中心,英国曼彻斯特;巴勒莫大学,意大利巴勒莫;意大利国家研究委员会高性能计算与网络研究所,意大利巴勒莫)

机构地区：[1]The University of Manchester,Manchester,UK [2]Manchester Centre for Robotics and AI,Manchester,UK [3]University of Palermo,Italy [4]ICAR-CNR,Palermo,Italy

出　　处：《宁波大学学报(理工版)》2025年第2期49-58,共10页Journal of Ningbo University:Natural Science and Engineering Edition

基　　金：Supported by the European Union’s Horizon Europe research and innovation program(101120727-PRIMI).

摘　　要：This paper adopts the Global Workspace Theory as a neuro-scientifically plausible theory for developing conscious cognitive architecture.The Global Workspace Theory’s compatibility with the working mechanisms underneath human brains is enhanced by the implementation of different cognitive features based on this framework.Amongst the topics in the literature for intelligent systems,we start with attention,memory and learning mechanisms,and corresponding experiments are summarized here.We also discuss how other topics of cognitive robotics could be developed based on these three basic components,and their correlations.This provides a foundation for future long-term development of cognitive architectures of cognitive robots.The research in this paper follows the incremental research pathway for the architecture implementation,which is consistent with the Biologically Inspired Cognitive Architecture roadmap.采用适用于神经科学的全局工作空间理论开发有意识的认知架构.基于这一架构实现了不同的认知要素,使得全局工作空间理论与人脑工作机制的兼容性得以增强.从注意力、记忆和学习机制这些针对智能系统的研究主题开始进行探讨,并对相关实验进行总结.随后讨论了如何基于这三个基本要素及其相互关系研究认知机器人开发这一新主题.这些工作为认知机器人认知架构的未来长期发展提供了基础.此外,采用渐进式研究路线用以上述架构的实现,这与生物启发式认知架构的研究路线图相一致.

关 键 词：Global Workspace Theory cognitive architecture INCREMENTAL biologically inspired cognitive architecture 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]