基于BP神经网络的被动步行稳定不动点的估算

摘要
图/表
参考文献
相关文章 (10)

全文: PDF (2887 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
倪修华, 陈维山, 刘军考, 石胜君. 基于BP神经网络的被动步行稳定不动点的估算[J]. 机器人, 2010, 32(4): 478-483.. NI Xiuhua, CHEN Weishan, LIU Junkao, SHI Shengjun. Estimation of the Stable Fixed Point of Passive Dynamic Walking with BP Neural Network. ROBOT, 2010, 32(4): 478-483..

摘要建立了被动步行机器人的动力学模型．使用BP神经网络对被动步行稳定不动点进行估算，并将估算值作为Newton-Raphson迭代的初值来求解稳定不动点．该方法解决了以往利用非最简模型求解不动点时，由于初值选取不当所造成的搜索成功率低以及搜索时间长的问题．在获取训练样本时，参数变化较小时不动点的变化很小；利用这一特点，在计算相邻两个样本的不动点时，只使一个参数发生较小的变化，并将本次使用Newton-Raphson迭代搜索得到的稳定不动点作为搜索下一样本不动点时的初值．使用1000组随机产生的参数对所得神经网络进行测试，结果表明该方法可以大幅提高稳定不动点的搜索成功率，大幅缩短搜索时间．

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	倪修华
	陈维山
	刘军考
	石胜君

关键词 ：被动步行, 不动点, BP神经网络

Abstract：The dynamic model of passive dynamic walking robot is given.BP(backpropagation) neural network(NN) is used to estimate the stable fixed point,and the estimated fixed point is taken as initial value for Newton-Raphson iteration to search the stable fixed point.This method solves the problems of low successful search rate and long search time caused by the improper initial value for the walking model which is not the simplest in previous study.Small variation of parameters leads to small variation of fixed points when acquiring training samples.When calculating fixed points of two successive samples,only one parameter has a small variation.The fixed point obtained with Newton-Raphson iteration is used as initial value for the next sample.1000 groups of parameters are generated randomly to test the NN performance.The results show that this method increases the successful search rate of stable fixed points significantly and shortens the search time greatly.

Key words： passive dynamic walking fixed point BP(backpropagation) neural network

收稿日期: 2009-09-17 录用日期: 2010-04-07

TP24

通讯作者: 倪修华 E-mail: nixiuhua@gmail.com

作者简介: 倪修华（1980- ），男，博士生．研究领域:被动步行．
陈维山（1965- ），男，博士，教授，博士生导师．研究领域:仿生机器人，仿生机器鱼，超声波电机．
刘军考（1973- ），男，博士，副教授．研究领域:仿生机器人，仿生机器鱼．

链接本文:

https://robot.sia.cn/CN/ 或 https://robot.sia.cn/CN/Y2010/V32/I4/478

[1] McGeer T.Passive dynamic walking[J].International Journal of Robotics Research,1990,9(2):62-82.

[2] Alexander R M.Walking made simple[J].Science,2005,308(5718):58-59.
[3] Collins S,Ruina A,Tedrake R,et al.Efficient bipedal robots based on passive-dynamic walkers[J].Science,2005,307(5712):1082-1085.

[4] Garcia M,Chatterjee A,Ruina A.Speed,efficiency,and stability of small-slope 2-D passive dynamic bipedal walking[C]//IEEE International Conference on Robotics and Automation.Piscataway,NJ.USA:IEEE,1998:2351-2356.
[5] Schwab A L,Wisse M.Basin of attraction of the simplest walking model[C]//ASME Design Enoneering Technical Conference.New York,USA:ASME,2001:531-539.
[6] 柳宁,李俊峰,王天舒.用胞胞映射计算被动行走模型不动点的吸引盆[J].工程力学,2008,25(10):218-223.Liu Ning,Li Junfeng,Wang Tianshu.Study of the basin of attraction of passive models by the aid of cell-to-cell mapping method[J].Engineering Mechanics,2008,25(10):218-223.
[7] 苏学敏,赵明国,张楫,等.被动行走周期性步态不动点搜索的新算法[J].清华大学学报:自然科学版,2009,49(8):1109-1112.Su Xuemin,Zhao Minggno,Zhang Ji,et al.Identifying fixed points in periodic gaits during passive walking[J].Journal of Tsinghua University:Science and Technology,2009,49(8):1109-1112.
[8] Garcia M,Chatterjee A,Ruina A,et al.The simplest walking model:Stability,complexity,and scaling[J].Journal of Biomechanical Engineering,Transactions of the ASME,1998,120(2):281-288.

[9] Hass J,Herrmann J M,Geisel T.Optimal mass distributionfor passivity-based bipedal robots[J].International Journal of Robotics Research,2006,25(11):1087-1098.

[10] 李洁.BP网络的算法及在Matlab上的程序仿真[J].西安航空技术高等专科学校学报,2009,27(1):41-43.Li Jie.Computation of BP network and Matlab's programming emulation[J].Joumal of Xi'an Aerotechnical College,2009,27(1):41-43.