基于内在动机的智能机器人自主发育算法

doi:10.11772/j.issn.1001-9081.2015.09.2602

计算机应用 ›› 2015, Vol. 35 ›› Issue (9): 2602-2605.DOI: 10.11772/j.issn.1001-9081.2015.09.2602

基于内在动机的智能机器人自主发育算法

任红格, 向迎帆, 李福进

华北理工大学电气工程学院, 河北唐山 063009

收稿日期:2015-04-20 修回日期:2015-05-25 出版日期:2015-09-10 发布日期:2015-09-17
通讯作者: 任红格(1979-),女,河北石家庄人,副教授,博士,主要研究方向:人工智能,414963045@qq.com
作者简介:向迎帆(1990-),男,河北唐山人,硕士研究生,主要研究方向:检测技术及智能装置;李福进(1956-),男,河北唐山人,教授,博士,主要研究方向:检测技术及智能装置。
基金资助:
国家自然科学基金资助项目(61203343);河北省自然科学基金资助项目(E2014209106)。

Autonomous developmental algorithm for intelligent robot based on intrinsic motivation

REN Hongge, XIANG Yingfan, LI Fujin

College of Electrical Engineering, North China University of Science and Technology, Tangshan Hebei 063009, China

Received:2015-04-20 Revised:2015-05-25 Online:2015-09-10 Published:2015-09-17

摘要/Abstract

摘要： 针对两轮自平衡机器人在学习过程中主动性差的问题,受心理学内在动机理论启发,提出一种基于内在动机的智能机器人自主发育算法。该算法在强化学习的理论框架中,引入模拟人类好奇心的内在动机理论作为内部驱动力,与外部奖赏信号一起作用于整个学习过程。采用双层内部回归神经网络存储知识的学习与积累,使机器人逐步学会自主平衡技能。最后针对测量噪声污染对机器人平衡控制中两轮角速度的影响,进一步采用卡尔曼滤波方法进行补偿,以提高算法收敛速度,降低系统误差。仿真实验表明,该算法能够使两轮机器人通过与环境的交互获得认知,成功地学会运动平衡控制技能。

关键词: 内在动机, 自主发育, 卡尔曼滤波, 平衡控制, 两轮机器人

Abstract: The initiative of two-wheeled self-balancing robot in the process of learning is poor. Inspired by intrinsic motivation theory of psychology, an autonomous development algorithm for intelligent robot based on intrinsic motivation was put forward. In the frame work of the reinforcement learning theory, the algorithm introduced human curiosity of intrinsic motivation theory as the internal driving force, and external reward signal into entire learning progress, and adopted double internal regression neural network for storage of knowledge learning and accumulation, which made robot gradually learn autonomous balance skill. Finally, aiming at the effects of measurement noise pollution on two-wheeled angular velocity of robot, further by adopting the method of Kalman filter to compensate, to speed up the algorithm convergence, and reduce the system error. Simulation experiments show that this algorithm can make the two-wheeled robot obtain cognition through interaction with the environment, therefore successfully learn balance control skill.

Key words: intrinsic motivation, independent development, Kalman filtering, balance control, two-wheeled robot

中图分类号:

TP183

任红格, 向迎帆, 李福进. 基于内在动机的智能机器人自主发育算法[J]. 计算机应用, 2015, 35(9): 2602-2605.

REN Hongge, XIANG Yingfan, LI Fujin. Autonomous developmental algorithm for intelligent robot based on intrinsic motivation[J]. Journal of Computer Applications, 2015, 35(9): 2602-2605.

参考文献

[1] YU H, ZHU C, LIU H. A survey on developmental robotics [J]. Transactions on Intelligent Systems, 2007,2(4):34-39.(于化龙,朱长明,刘海波.发育机器人研究综述[J].智能系统学报,2007,2(4):34-39.)
[2] JIANG G. The control of inverted pendulum based on Q learning algorithm and BP neural network [J]. Acta Automatica Sinica, 1998,24(5):662-667.(蒋国飞.基于Q学习算法和BP神经网络的倒立摆控制[J].自动化学报,1998,24(5):662-667.)
[3] PATHAK K, FRANCH J, AGRAWAL S K. Velocity and position control of a wheeled inverted pendulum by partial feedback linearization [J]. IEEE Transactions on Robotics, 2005,21(3):505-513.
[4] RUAN X, CAI J, CHEN J. Learning to control two-wheeled self-balancing robot using reinforcement learning rules [J]. Computer Measurement and Control, 2009,17(2):321-323.(阮晓钢,蔡建羡,陈静.基于强化学习规则的两轮机器人自平衡控制[J].计算机测量与控制,2009,17(2):321-323.)
[5] LI M, JIAO L, QIAO J. Hierarchical fuzzy control for self-balancing two-wheel robots [J]. Control Engineering of China, 2009,16(1):80-82.(李明爱,焦利芳,乔俊飞.自平衡两轮机器人的分层模糊控制[J].控制工程,2009,16(1):80-82.)
[6] OUDEYER P Y, KAPLAN F. What is intrinsic motivation? A typology of computational approaches [J]. Frontiers in Neurorobotics, 2007,1: Article 6.
[7] PFEIFER R, BONGARD J C. How the body shapes the way we think: a new view of intelligence [M]. Cambridge: MIT Press, 2006:75-77.
[8] CEDERBORG T, OUDEYER P Y. From language to motor Gavagai: unified imitation learning of multiple linguistic and nonlinguistic sensorimotor skills [J]. IEEE Transactions on Autonomous Mental Development, 2013,5(3):222-239.
[9] WHITE R W. Motivation reconsidered: the concept of competence [J]. Psychological Review, 1959,6(6):297-333.
[10] DENG Z. The information fusion filtering theory and application [M]. Harbin: Harbin Institute of Technology Press, 2007:120-124.(邓自立.信息融合滤波理论及其应用[M].哈尔滨:哈尔滨工业大学出版社,2007:120-124.)
[11] WANG R, SUN L, RUAN X. Reinforcement learning based on internal regression neural network [J]. Control Engineering of China, 2005,12(2):138-140.(王瑞霞,孙亮,阮晓钢.基于内部回归神经网络的强化学习[J].控制工程,2005,12(2):138-140.)
[12] QI Q, LI Z, TAN Z. Noise suppression of inverted pendulum system based on Kalman filter [J]. Control and Decision, 2010,25(8):1144-1148.(祁虔,李祖枢,谭智.基于卡尔曼滤波的倒立摆控制系统噪声抑制[J].控制与决策,2010,25(8):1144-1148.)
[13] SONG W, ZHANG Y. Kalman filter [M]. Beijing: Science Press, 1991:116-119.(宋文尧,张牙.卡尔曼滤波[M].北京:科学出版社,1991:116-119.)

[1]	张健铭, 施元昊, 徐正蓺, 魏建明. 基于误差预测的自适应UWB/PDR融合定位算法[J]. 计算机应用, 2020, 40(6): 1755-1762.
[2]	王守华, 李云柯, 孙希延, 纪元法. 基于低成本接收机的双天线测姿算法[J]. 计算机应用, 2019, 39(8): 2381-2385.
[3]	张榜, 朱金鑫, 徐正蓺, 刘盼, 魏建明. 基于贝叶斯网络的楼层定位算法[J]. 计算机应用, 2019, 39(8): 2468-2474.
[4]	汤启, 何腊梅. 带非线性等式约束无迹卡尔曼滤波方法[J]. 计算机应用, 2018, 38(5): 1481-1487.
[5]	商建东, 李盼乐, 刘润杰, 李润川. 基于加权时变泊松模型的出租车载客点推荐模型[J]. 计算机应用, 2018, 38(4): 923-927.
[6]	朱清超. 多跳吞吐量分析及邻节点实时估计算法设计[J]. 计算机应用, 2017, 37(9): 2484-2490.
[7]	游路瑶, 黄庆卿, 段斯静. 基于卡尔曼滤波估计的一致性时钟同步算法[J]. 计算机应用, 2017, 37(8): 2177-2183.
[8]	王龙, 章政, 王立. 改进扩展卡尔曼滤波的四旋翼姿态估计算法[J]. 计算机应用, 2017, 37(4): 1122-1128.
[9]	张恒德, 咸云浩, 谢永华, 杨乐, 张天航. 基于时间序列分析和卡尔曼滤波的霾预报技术[J]. 计算机应用, 2017, 37(11): 3311-3316.
[10]	罗元, 庞冬雪, 张毅, 苏琴. 基于自适应多提议分布粒子滤波的蒙特卡洛定位算法[J]. 计算机应用, 2016, 36(8): 2352-2356.
[11]	潘磊, 周欢, 王明辉. 适用于密集人群的异常事件实时检测方法[J]. 计算机应用, 2016, 36(6): 1719-1723.
[12]	李世宝, 陈瑞祥, 刘建航, 陈海华, 丁淑妍, 龚琛. 基于反向传播神经网络改进的增益修改卡尔曼滤波算法[J]. 计算机应用, 2016, 36(5): 1196-1200.
[13]	孟唐宇, 浦剑涛, 方建军, 梁岚珍. 姿态解算与外力加速度同步估计算法[J]. 计算机应用, 2016, 36(5): 1469-1474.
[14]	石刚, 赵伟, 刘珊珊. 基于无迹卡尔曼滤波估算电池SOC[J]. 计算机应用, 2016, 36(12): 3492-3498.
[15]	张朝霞, 王慧慧, 傅正, 杨玲珍, 王娟芬, 刘香莲. 基于卡尔曼最小均方误差准则的雷达旁瓣抑制研究及仿真[J]. 计算机应用, 2015, 35(5): 1488-1491.

基于内在动机的智能机器人自主发育算法

Autonomous developmental algorithm for intelligent robot based on intrinsic motivation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics