| 引用本文: | 黄静,阮晓钢,于乃功,张晓平,魏若岩,范青武.具有操作条件反射机能的人工感觉运动系统[J].控制理论与应用,2015,32(5):674~681.[点击复制] |
| HUANG Jing,RUAN Xiao-gang,YU Nai-gong,ZHANG Xiao-ping,WEI Ruo-yan,FAN Qing-wu.Artificial sensorimotor system with operant conditioning function[J].Control Theory and Technology,2015,32(5):674~681.[点击复制] |
|
| 具有操作条件反射机能的人工感觉运动系统 |
| Artificial sensorimotor system with operant conditioning function |
| 摘要点击 3094 全文点击 1487 投稿时间：2014-03-25 修订日期：2015-02-05 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2015.40235 |
| 2015,32(5):674-681 |
| 中文关键词 感觉运动系统 操作条件反射 斯金纳鸽子实验 维纳虫 认知机器人学 神经网络 |
| 英文关键词 sensorimotor system operant conditioning Skinner pigeon experiment Wiener worm experiment cognitive robotics neural networks |
| 基金项目 北京高等学校青年英才计划项目(YETP1610), 国家自然科学基金项目(61075110, 61375086), 北京市自然科学基金项目/北京市教育委员会科技 计划重点项目(KZ201210005001), 国家重点基础研究发展计划(973计划)项目(2012CB720000), 高等学校博士学科点专项科研基金课题 (20101103110007)资助. |
|
| 中文摘要 |
| 为探索认知机理、模拟认知机能进而提高机器人的认知及智能水平, 提出了一种具有操作条件反射机能的 人工感觉运动系统. 该系统以复现感觉运动系统的方式重现了生物的运动神经认知, 实现了从感知到运动的映射关 系, 同时借鉴了斯金纳的操作条件反射理论, 使得该系统具有操作条件反射机能, 遵从“刺激–反应–强化”的逻辑 形成了感知与运动之间的闭环系统. 为验证系统的正确有效性, 复现了行为心理学及《控制论》中的两个经典实验. 对比实验结果证明, 本系统成功的模拟了生物感觉运动系统, 使机器人具有类似生物的自学习能力, 能自主的认知 环境, 实现对环境的自适应. |
| 英文摘要 |
| In order to explore the cognitive mechanism and simulate the cognitive ability, we develop an artificial sensorimotor system with operant conditioning function (ASSOCF) to improve the cognitive and intelligent ability of robots. The system reproduces the biological motor cognition by simulating the sensorimotor system and realizes the mapping from perception to action. Meanwhile, the system is endowed with operant conditioning function based on Skinner’s operant conditioning theory. It forms the close-loop system between perception and action in compliance with the logic of “stimulus-response-reinforcement”. Two classic experiments in behavioral psychology and control theory, Skinner pigeon experiment and Wiener worm experiment, are performed to testify the validity of the system. Comparisons of the experiment results prove that the system successfully simulates the biological sensorimotor system. With such a system, the robots are endowed with the abilities of autonomous learning and self-adaptation to the environment. |
|
|
|
|
|