SIA OpenIR  > 机器人学研究室
基于多模态CPG模型的蛇形机器人仿生控制研究
Alternative TitleBiomimetic Control of Snake Robots Based on Multi-phase CPG Model
唐超权1,2
Department机器人学研究室
Thesis Advisor马书根 ; 王越超
ClassificationTP242
Keyword蛇形机器人 多模态中枢模式发生器 仿生控制策略
Call NumberTP242/T24/2012
Pages97页
Degree Discipline机械电子工程
Degree Name博士
2012-11-24
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract本文研究内容包括:1、提出了一种用于蛇形机器人控制的多模态CPG 模型。针对蛇形机器人CPG 控制中步态单一的问题,依据生物学原理,构建了多模态CPG 模型,可以实现蛇形机器人的多种步态,为蛇形机器人根据环境调整步态提供了基础。针对蛇形机器人CPG 控制模型稳定性证明不足的问题,提出了多模态CPG 模型任意节的稳定性证明,在理论上保证了多模态CPG 模型输出的平滑稳定。利用仿真以及实验平台验证了多模态CPG 模型的有效性。2、分析了多模态CPG 模型的参数及模型特性。为保证多模态CPG 模型能够满足蛇形机器人复杂的运动控制需要,对多模态CPG 模型中各项参数的意义以及对最终输出的影响和关系进行了分析并给出了表达式,并利用仿真对其进行了验证。从而为进一步的参数调整策略提供了基础。通过引入外部激励与模型参数之间的关系,从而使得多模态CPG 模型具有一定环境适应能力。3、提出了一种基于神经步进激励机制的仿生控制策略。目前用于仿生机器人环境适应的CPG 模型参数调整方法都是基于已有的优化方法,缺乏对于动物自身CPG 参数调整机理的研究,因而存在需要预先训练或在线搜索时间长等问题。通过已有解剖神经学研究,动物运动是通过不同的神经激励来获得运动幅度频率的调整以及步态的转换。因此,本文引入神经步进激励机制,结合蛇形机器人自身运动特点,提出了一种基于多模态CPG 模型的结构简单,调整快速的仿生控制策略。该策略很好的利用了蛇形机器人几种典型步态的特点,通过对于自身状态的评估,选择恰当有效的步态以提高运动效率。该仿生控制策略能够有助于提高蛇形机器人的环境适应能力。
Other AbstractThe main content of this dissertation includes: 1. Introduction of a multi-phase CPG control model for snake robots For filling the gap in gaits of snake robots with CPG control method, the multi-phase CPG model is proposed based on biomimetic theory. This model can produce several gaits of snake robots. This property of multi-phase CPG model is helpful to improve the capacity of adapting to environment. For the problem of lacking a proof of CPG model stability, the stability of arbitrary segments multi-phase CPG model is proven. This process grantees the CPG model output smooth and converged. The multi-phase CPG model is verified by simulation and experiment platform. 2. Analysis of characteristics of multi-phase CPG model. To ensure the multi-phase CPG model to meet the requirements of complex motion control, the performance of CPG model, referring to the parameters, is presented and verified with simulation platform. This analysis is helpful to the adjustment strategy of CPG’s parameters. By introducing the equations of the external inspiration and the model parameter, the multi-phase CPG model is capable to adapt to environments. 3. Proposal of biomimetic control based neural stepping mechanism The existing research about the parameter adjusting method of CPG model was based on optimization algorithm and lacked the study about the CPG parameters adjusting mechanism of animals. This would result in the consuming online search time or pre-training. Based on the neuroanatomy research, the motion of animals would modify the gaits or motion amplitude and frequency by varying the neural stimulation. This paper introduces the neural stepping stimulation mechanism. A simple but fast biomimetic control strategy based on multi-phase CPG model and motion analysis of snake robots are proposed. This strategy has fairly utilized the characteristics of several classical gaits of snake robots and has improved the motion efficiency by properly choosing gaits based on the sensory feedback of state of robots. The adapting capacity of snake robots is enhanced by this biomimetic control strategy.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/10651
Collection机器人学研究室
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院大学
Recommended Citation
GB/T 7714
唐超权. 基于多模态CPG模型的蛇形机器人仿生控制研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2012.
Files in This Item:
File Name/Size DocType Version Access License
基于多模态CPG模型的蛇形机器人仿生控制(3577KB) 开放获取LicenseApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[唐超权]'s Articles
Baidu academic
Similar articles in Baidu academic
[唐超权]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[唐超权]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.