中国科学院沈阳自动化研究所机构知识库
Advanced  
SIA OpenIR  > 机器人学研究室  > 学位论文
题名: 拟人形机器人建模、控制与协调研究
其他题名: Study on Modeling, Control and Coordination of Humanoid Robots
作者: 刘英卓
导师: 王越超 ; 席宁
分类号: TP242
关键词: 拟人机器人 ; 建模 ; 交互作用力 ; 神经网络
索取号: TP242/L76/2003
学位专业: 机械电子工程
学位类别: 博士
答辩日期: 2003-10-12
授予单位: 中国科学院沈阳自动化研究所
学位授予地点: 中国科学院沈阳自动化研究所
作者部门: 机器人学研究室
中文摘要: 机器智能化是当代科学的主题之一,目前拟人形机器人被认为是工业和科研上研究智能的最佳载体,是传统的工业机器人走向人类社会和个人应用的标志,正在机器人领域掀起激动人心的热潮。本论文对中科院沈阳自动化所机器人学重点实验室承担的国家863计划“十五”项目——“宜人化双臂操作型服务机器人”的动力学建模、合理的控制器设计以及运动协调问题进行了比较系统深入的研究。 在广泛研究建立机器人动力学方程各种方法的基础上,结合当前已有的各种机器人模型,从智能控制的角度出发,本文研究了拟人机器人建模问题。由于拟人机器人机构复杂、具有多体结构和超多的自由度,使其运动学和动力学问题都成为挑战性的问题。 整体建模不能满足实时控制的要求,因而对拟人形机器人系统不再适用。针对这种情况,本文基于分体建模的思想来为拟人机器人建立动力学模型。分体建模使得各部分之间的交互作用力成为研究的中心,交互作用力的定义、计算和实现是解决问题的关键。通过深入研究拟人机器人各部分间的运动特性,本文将拟人机器人划分为三个子系统,较为合理地分配了整个系统的复杂度;采用径向基函数神经网络(RBFNN)建模交互作用力,以利于实时实现。概括地说,本文结合了机理建模方法的相对精确性和神经网络连接主义的优越性为拟人机器人建立动力学模型,并详细讨论了建模过程和其中的各种重要问题,如交互作用力的问题,对不同部分的交互作用力给出了仿真结果。 基于所建立的模型,本文研究了拟人机器人在关节空间内的控制问题。由于设计可靠、实用、高效的非线性控制器一直都是控制工程师的追求,而面对复杂的拟人机器人,控制器的设计更具有特殊的意义。因此,本文提出了一种基于NN的自适应鲁棒控制器,证明了其渐进稳定性,并通过了仿真验证。作为对这种控制方法的完善化、实用化,又提出一种改进的基于NN的自适应 位置跟踪控制器。鲁棒非线性 控制方法自然的与模型的直接自适应神经网络技术集成为一体。同样,证明了其鲁棒稳定性,进行了仿真研究。 由于通常给出的任务都是在任务空间描述的,因此,为拟人形机器人设计任务空间的控制器更为重要。本文采用类Lyapunov方法为拟该人形机器人设计了任务空间控制器,并进行了仿真。所提出的控制器具有类Lyapunov方法自身的优点:不需要雅可比求逆、有限确定时间内达到稳定,还对一定类型的外部干扰具有鲁棒性功能。通过研究,同样可以扩充到 意义之下。 最后,在分析宜人化双臂操作型服务机器人运动特征的基础上,本文提出一种结合模糊控制逻辑和基于事件行为方法的在线协调策略。
英文摘要: Machine intelligence is one of the main subjects of the present science. Humanoid robots have become exciting hotspot currently in robotics domain and indicate that traditional industrial robots are entering human society and personal are applications. This thesis studies systemically dynamics modeling, designing suitable controller and motion coordination problems for humanoid robots. This work is carried out at the Robotics Laboratory, Shenyang Institute of Automation, Chinese Academy of Sciences and supported by the Chinese National “863” Program project named “Friendly Two-armed Operable Service Robot (FTAOSR)”. After investigating different methods of robot dynamics modeling, this thesis combines the existing different robots’ model to research the dynamics modeling problem of FTAOSR with thinking about intelligent control. Because of humanoid robots’ complex mechanism with multi-body structure and multiple degrees of freedom, their kinematics and dynamics are fully challenging problems. Modeling FTAOSR as a whole body can not satisfy real time control, so we can not use the method of whole-body modeling. Aiming at this problem, we model FTAOSR based on decentralized-body modeling idea. However, decentralized-body modeling will bring the interactive force between different parts. The interactive force’s definition, computation and realization become the key points of solving the modeling problem. Through deep study on the motion characteristics of different parts of FTAOSR, we divide FTAOSR into three subsystems. So the degree of freedom of the whole system is reasonably assigned. Furthermore, we use radial base function neural network (RBFNN) to model the interactive force for real time realization. Conclusively, this thesis combines the relative accuracy of mechanism modeling method with the advantage of connectionism of the neural network for the modeling of the FTAOSR, the thesis discusses in detail modeling process and some relative problems such as interactive force. We also simulate the interactive force between different parts. Based on the proposed model, this thesis studies control problem of the humanoid robot in joint space. Because of the complicacy of humanoid robot, designing its controller has a special meaning. So, we propose a NN-based adaptive robust controller and prove its asymptotically stability. Simulation results show its validity. As a kind of perfect and practicality of the above-mentioned method, we again propose a modified NN-based adaptive position tracking controller. In this method, robust nonlinear control method naturally integrates with model’s direct adaptive NN technology. Similarly, we prove its robust stability and simulate its performance. Because the tasks executed by robots are almost described in task space, designing a task-space controller for FTAOSR will be more important. We design a task-space controller for our humanoid robot with Lyapunov-like method and simulate the performance of the controller. The resulting controller has the same advantage as the Lyapunov-like method, i.e., it needs not resolve the inverse of Jacobian matrix, reaches stability in a finite time and has a robust function to a kind of exterior disturbance. Furthermore, the above controller can be extended to possess performance. Finally, we propose a kind of on-line coordination strategy which combines fuzzy logical control and event-based action method.
语种: 中文
产权排序: 1
内容类型: 学位论文
URI标识: http://ir.sia.cn/handle/173321/9528
Appears in Collections:机器人学研究室_学位论文

Files in This Item:
File Name/ File Size Content Type Version Access License
拟人形机器人建模、控制与协调研究.pdf(1169KB)----限制开放 联系获取全文

Recommended Citation:
刘英卓.拟人形机器人建模、控制与协调研究.[博士学位论文].中国科学院沈阳自动化研究所.2003
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[刘英卓]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[刘英卓]‘s Articles
Related Copyright Policies
Null
Social Bookmarking
Add to CiteULike Add to Connotea Add to Del.icio.us Add to Digg Add to Reddit
所有评论 (0)
暂无评论
 
评注功能仅针对注册用户开放,请您登录
您对该条目有什么异议,请填写以下表单,管理员会尽快联系您。
内 容:
Email:  *
单位:
验证码:   刷新
您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。
标 题:
 *
内 容:
Email:  *
验证码:   刷新

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

 

 

Valid XHTML 1.0!
Copyright © 2007-2016  中国科学院沈阳自动化研究所 - Feedback
Powered by CSpace