机器人类脑智能研究综述  

A Survey of Research on Robotic Brain-inspired Intelligence

作　　者：王瑞东 王睿[1] 张天栋 王硕[1,2] WANG Rui-Dong;WANG Rui;ZHANG Tian-Dong;WANG Shuo(State Key Laboratory of Multimodal Artificial Intelligence Systems,Institute of Automation,Chinese Academy of Sciences,Beijing 100190;School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing 100049)

机构地区：[1]中国科学院自动化研究所多模态人工智能系统全国重点实验室,北京100190 [2]中国科学院大学人工智能学院,北京100049

出　　处：《自动化学报》2024年第8期1485-1501,共17页Acta Automatica Sinica

基　　金：科技创新2030-“脑科学与类脑研究”重大项目(2022ZD0209600);国家自然科学基金(62276253);北京市科技新星计划项目(Z211100002121152,20230484457)资助。

摘　　要：传统机器人经过长时间的研究和发展,已经在生产和生活的多个领域得到了广泛的应用,但在复杂多变的环境中依然缺乏与真实生物类似的灵活性、稳定性和适应能力.类脑智能作为一种新型的机器智能,使用计算建模的方法模拟生物神经系统的各类特性,进而实现对各类信息的推理和决策,近年来受到了学术界的广泛关注.鉴于此,综述了国内外面向机器人系统的类脑智能研究现状,并对类脑智能方法在机器人感知、决策和控制三个研究方向的成果进行了整理、归纳和分析,最后从软硬件层面分别指出了机器人类脑智能目前存在的主要问题和未来的发展方向.After a long time of research and development,traditional robots have been widely used in many fields of production and life,but they still lack the flexibility,stability and adaptability similar to real organisms in complex and changing environments.As a new type of machine intelligence,brain-inspired intelligence uses computational modeling methods to simulate various characteristics of biological nervous systems and realize the reasoning and decision-making based on all kinds of information.In recent years,it has received extensive attention from the academic community.The main applications of brain-inspired intelligence methods in robot perception,decisionmaking and control problems are introduced.The related research results are analyzed and summarized.Finally,the main problems of software and hardware are pointed out and future development directions of robotic brain-inspired intelligence are proposed.

关 键 词：机器人 类脑机器人 类脑智能 脉冲神经网络 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]