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: 动态目标多移动机器人主动合作观测方法研究
其他题名: Research on the Active Cooperative Observation for Dynamic Target
作者: 谷丰
导师: 王越超 ; 韩建达
分类号: TP242
关键词: 多移动机器人 ; 合作观测 ; 路径规划 ; 扩展集员滤波 ; 多旋翼飞行机器人实验平台
索取号: TP242/G64/2011
学位专业: 模式识别与智能系统
学位类别: 博士
答辩日期: 2011-11-30
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: 由于移动机器人系统作业范围的不固定性以及作业环境的高度动态性,在完成任务过程中其对环境的高效感知与正确理解是移动机器人系统所必须具备的能力之一,其中以对环境中感兴趣目标(尤其是动态目标)状态的在线、实时观测最具代表性。多机器人系统通过信息共享与融合可以实现对动态目标的合作观测,获得比单机器人具有更高精度的观测结果。但是现有的多移动机器人合作观测的研究只考虑了多机器人系统的传感器信息融合,而没有考虑多机器人观测状态对观测结果的影响。针对这一研究现状,本文提出了一种多机器人主动合作观测的概念:以获取较优的观测结果为目的,对携带同构/异构观测传感器的多机器人系统进行协调,以优化的观测状态完成数据融合,从而获得优化的合作观测结果。它充分利用多机器人系统的智能性,协调性以及测量信息分布性,完成观测信息的融合,获得高精度目标状态信息。围绕这一概念,本文对其中包含的动态目标观测,多机器人的观测信息融合,以及多机器人行为协调等问题展开了研究,提出了相应的解决方法,实现了三维环境下的多移动机器人主动合作观测。 另外,多机器人系统的实验研究一直是阻碍多机器人研究发展的一个瓶颈,尤其是针对多飞行机器人的实验研究,这主要是由于实验过程中存在的高风险性和实验条件的严格限制。针对这些问题,本文设计并实现了一款室内多旋翼飞行机器人实验平台,通过物理约束避免实验过程中的机器人碰撞,坠毁等事故的发生,同时很好的保留了旋翼飞行机器人的动力特性。而且这种室内实验平台可以避免天气环境等因素对实验的影响,提高实验效率。利用该实验平台,完成了本文所提出的主动合作观测方法的实验验证,验证了其在实际系统中的可行性与有效性。这也是基于集员滤波的合作观测方法首次在实际系统中得到验证。概括起来,本文的主要研究工作如下: 1) 现有的机器人动态目标观测方法中,大多基于概率的滤波方法,这类方法往往只能给出系统误差在白噪声假设条件下的目标状态的最大概率点,缺乏可靠性的描述。针对这类方法存在的问题,对一种基于集员理论的非线性滤波方法—扩展集员滤波方法(Extended Set-membership Filter, ESMF)展开了研究,并将其应用在三维环境中动态目标观测中,并验证了这种观测方法的可靠性和鲁棒性; 2) 针对现有基于ESMF的合作观测方法存在的问题,利用ESMF方法关于集合运算的特点,提出了一种基于测量信息的多机器人合作观测方法,将信息融合过程与估计方法本身存在的集合运算环节相结合,与单机器人的观测方法相比,在提高观测精度的同时,并没有显著增加计算量,具有较高的实时性,同时降低了通信负担; 3) 针对现有多机器人合作观测存在的问题,提出了一种主动合作观测的概念,并在该概念的框架下,提出了一种三维相对速度空间(Relative Velocity Coordinates, RVCs)下的多移动机器人路劲规划方法,该方法可以将多机器人路径规划这一非线性问题转化为线性规划问题,以实现具有较高实时性多机器人动态路径规划,通过将这种路径规划方法嵌入到前面提出的合作观测方法当中,实现了多机器人的主动合作观测。 4) 设计并实现了一款室内多旋翼飞行机器人实验平台,通过物理约束避免了多飞行机器人实验中存在的碰撞以及坠毁等风险。并在该实验平台上扩展了一套基于视觉的测量系统,利用这套实验平台完成了对所提出的主动合作观测方法的实验验证,在实际系统中验证了我们提出的主动合作观测方法的实时性和在线应用的。
英文摘要: Efficient modeling of environment is the basic abilities of the mobile robots that are required to work outdoor because of the dynamical environments and its huge uncertainties, while localization and tracking of dynamical targets is a typical problem in environmental model. However, using single robot to localize or observe (in the following contents, ‘observation’ instead of ‘localization’ is most often used) the moving target is not precise enough in most real applications. Thus, an effective substitution is to use multiple mobile robot to achieve cooperative observations which can improve the localization accuracy greatly.   Cooperative observation is not a new research direction, and most of existing work only focuse on the problem on how to fuse multiple sensor data from different mobile robots. However, it has been shown that observation state of each robot, for example, relative between robot and target and that among different robots, has obvious influence on the cooperative observation results. Thus, in this thesis, a new concept of active cooperative observation (ACO) is introduced and researched. That means, in order to improve the final cooperative observation results, we should not only focus on obsrevation data fusion, but also emphasize robot’s behaivor regulation.   In order to implement ACO, the following problems is researched in this thesis, estimation algorithm used to realize the continuous observation and filter of the moving target, data fusion method and multiple robots behavior coordination method to regulate the formation between robots and target on the purpose of improving observation imprecision further. In order to verify the feasibility and validity of the proposed algorithms, an indoor multiple rotor flying robots test-bed is developed and a great deal of experiments are conducted with respect to it in this thesis.   The detailed contents of this thesis are as following:   1) In order to avoid the problem of difficulties in measuring stochastic distribution of statistics based estimation algorithms which is often used currently, a new set theory based estimation method, Extended Set-Membership Filter (ESMF), is introduced and applied for the observation problem in the 3D surroundings in Chapter 2. Also, we test the feasibility and validity of it in moving target observation problem through plenty of simulations.   2) In Chapter 3, a new real-time cooperative observation method based on ESMF is proposed. The most absorbing advantage of this new proposed algorithm is that instead of using time-consuming complex computation, such as computing intersection and circumscribed ellipsoid, it makes full use of the computational ability of ESMF algorithm itself to decrease the computational burden greatly and thus can be used in many real time applications.   3) In Chapter 4, ACO algorithm is researched. The algorithm combines the ESMF based cooperative observation algorithm and behavior coordination algorithm through a concept of optimal observation angle, which means an optimal formation by which an optimal cooperative observation results can be obtained. Furthermore, in this Chapter, the LP based path planning method in 3D Relative Velocity Coordinates (RVCs) is proposed and used in behavior coordination of multiple robot systems.   Finally, in order to verify the feasibility and validity of the above proposed algorithms in real systems, an indoor multiple rotor flying robots test-bed is designed on which a great deal of experiment studies of the above theoretical algorithms are conducted.
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/173321/9375
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
动态目标多移动机器人主动合作观测方法研究.pdf(5998KB)----限制开放 联系获取全文

Recommended Citation:
谷丰.动态目标多移动机器人主动合作观测方法研究.[博士学位论文].中国科学院沈阳自动化研究所.2011
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[谷丰]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[谷丰]‘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