CASIA OpenIR  > 毕业生  > 博士学位论文
多机器人系统动态跟随与队形控制研究
其他题名Research on Dynamic Following and Formation Control of Multi-robot System
杨丽
学位类型工学博士
导师谭民
2009-05-25
学位授予单位中国科学院研究生院
学位授予地点中国科学院自动化研究所
学位专业控制理论与控制工程
关键词多机器人系统 协调与控制 动态跟随 队形控制 观察推理 Multi-robot System Coordination And Control Dynamic Following Formation Control Observation Reasoning
摘要随着机器人应用领域的不断拓展,多机器人系统以其柔性、并行性及鲁棒性等个体机器人所无法比拟的优越性受到普遍重视,多机器人协调的研究已经成为机器人学的重要研究方向之一。本文针对多机器人系统动态跟随与队形控制开展研究工作,主要内容如下: 首先,本文介绍了多机器人系统的特点,综述了多机器人系统的研究现状及主要研究内容,同时介绍了多机器人队形控制等方面的发展现状,并对研究背景和论文结构做了介绍。 其次,介绍了多机器人实体实验平台,包括主控上位机、驱动控制子系统和传感器子系统等,重点介绍了超声传感器和视觉传感器模型。给出一种基于无线局域网的多机器人通讯系统设计方案。 第三,给出一种基于模糊的多机器人协调跟随控制方法,机器人依靠自身的视觉信息跟随它的领航机器人,并与领航机器人保持期望的相对距离和相对观测方位角,不需要全局信息。 第四,研究了未知环境中的多机器人系统无碰协调跟随问题,提出一种跟随机器人的基于局部感知的多模式控制方法,分为到达、旋转角度调整、跟随、避障和随机搜索五种模式。 第五,提出两种基于Leader-Follower的队形控制方法:基于观察推理的队形控制方法和基于意图推理的队形控制方法。跟随机器人结合视觉与运动信息在自身的移动局部坐标系中多步递推预测领航机器人的运动,进而控制跟随机器人实现机器人之间的协调。 第六,提出一种不确定环境下的多机器人队形控制与变换方法。将队形问题转为跟随机器人离轴点对虚机器人离轴点的跟踪,通过这种局部跟随的方式,除领航者指引系统方向外,其它机器人均跟随自己的领航机器人,系统呈现出整体的队形形状。采用队形变换与避障相结合平滑躲避环境中的障碍物,设计故障容错策略应对机器人发生故障的情况以增强系统的鲁棒性。 最后,论文对所取得的研究成果进行了总结,并阐述下一步的工作。
其他摘要With the expansion of the robot application, multi-robot system has been paid much attention because of its advantages such as flexibility, parallelism and robustness that single robot cannot compare. The coordination among multiple robots has become one of the important aspects in robotics. This thesis focuses on dynamic following and formation control of multi-robot system. The contents are as follows: Firstly, the features of multi-robot system are described. The research development and main research directions of multiple robot systems are given. The survey of multi-robot formation control is presented. The background and structure of this thesis are also introduced. Secondly, real multiple mobile robot systems used to validate the coordination algorithms are introduced, which include PC, driving control subsystem and sensor subsystem and so on. The sensor models are introduced in more detail. The multi-robot communication system based on wireless LAN is designed. Thirdly, a fuzzy-based coordination control method is proposed to achieve multi-robot coordination following tasks. Robots endeavor to establish the relationship with the leader by keeping the desired relative distance and visual observation angle, and no global information is required. Fourthly, considering cooperative dynamic following problem of multi-robot system in unknown environments, a multi-mode control approach based on local sensing is proposed, which includes five modes: arrival, rotating-angle, following, avoiding-obstacle and random-search modes. Fifthly, two kinds of leader-following formation control approaches are proposed. One is the observation-based multi-robot cooperative formation control approach, the other is the multi-robot formation control approach based on intention reasoning. On the basis of the fusion of the encoders and the vision system, the followers may predict the leader by multi-step recursion in its own moving local coordinate systems and then achieve the cooperation. Sixthly, a formation control and switching approach for multiple robots in uncertain environments is proposed. The formation control problem is converted into the tracking problem of the-off-axis point of the follower to the-off-axis point on the virtual robot. Only one navigator robot guides the multi-robot system, and other robots follow the corresponding leader. By this local following mode, the whole desired shape may emerge. An obstacle avoidance strategy combining with formation sw...
馆藏号XWLW1349
其他标识符200618014628012
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/6158
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
杨丽. 多机器人系统动态跟随与队形控制研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2009.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
CASIA_20061801462801(1851KB) 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[杨丽]的文章
百度学术
百度学术中相似的文章
[杨丽]的文章
必应学术
必应学术中相似的文章
[杨丽]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。