基于嵌入式的数字控制器设计与实时性分析
|
摘要
自行车机器人的自平衡控制是机器人领域的一个难点问题,国内外目前的研究大多还停留在初步的理论探讨上。本课题在前人研究的基础上,对自行车机器人的控制作了更深刻的探讨,设计出了完整的数字控制系统,并且做了大量的实验对其进行验证,分析了数字控制器的实时性。
本文首先总结了前人在自行车机器人动力学模型上的研究成果,在依靠调节车把保证自行车机器人平衡的非线性动力学模型的基础上,为自行车机器人设计了基于二次型最优控制理论的数字控制器,并分析了该系统的稳定性、可控性与可观测性。接下来,详细介绍了用于验证该控制系统有效性的基于嵌入式平台的硬件设计方案和软件模块。设计了一套数据采集系统,该系统通过使用USB存储设备,可以在自行车机器人行驶过程中随时记录其倾斜角度和同一时刻的车把旋转角度,为每次实验后的总结工作提供了第一手的材料。
在自行车机器人的运行过程中,控制系统需要及时根据自行车机器人的当前状态来向执行机构发出指令以确保车体的平衡,这就对系统的实时性提出了很高的要求。本文定量分析了程序运行的每一次任务调度和中断响应所耗时间,与传统的控制系统策略下的实时性表现作了详细对比,证明了本控制系统对系统相应的实时性作了很大的改进,很大程度上满足了自行车机器人对实时性的要求。
Bicycle Robot Self-balance control is a difficult issue in the field of robotics, and the current studies are mainly about the initial theory researches of the system. After some profound studies were made on the basis of previous theories, an integrated digital control system was designed. A mass of experiments had been conducted to verify the designing and to analyze the real-time performance of the system.
At the beginning of this paper, the previous research results about bicycle robot dynamic modeling were summarized. A kind of steering control dynamic model for bicycle robot based on Lagrange method was indicated. Based on the model and the theory of quadratic optimal control theory, digital controller of bicycle robot control system was established. The stability, observability and controllability of the bicycle control system were analyzed. In order to validate the effectiveness of the controller based on an embedded platform, the designing details of the control system's hardware and software modules were expatiated. Then, a simple data acquisition system with USB storage devices had been built up. The data acquisition system was used to get the tilting angle of the bike's body and the steering angle of the bike's front wheel at a time. The data of every experiment can be used as first-hand materials for farther analyzing work.
In the operating processes of the bicycle robot, in order to ensure the balance of the bike, the controller should send instruction to the executing mechanism according to the current status of the bike as soon as possible. So the system's real-time performance should be high. The time data spent by task scheduling and interrupt response were saved. By quantitative compared to the traditional control strategies, the real-time performance of the controller was improved greatly. And the performance can satisfy the real requirement to a great extent.
本文首先总结了前人在自行车机器人动力学模型上的研究成果,在依靠调节车把保证自行车机器人平衡的非线性动力学模型的基础上,为自行车机器人设计了基于二次型最优控制理论的数字控制器,并分析了该系统的稳定性、可控性与可观测性。接下来,详细介绍了用于验证该控制系统有效性的基于嵌入式平台的硬件设计方案和软件模块。设计了一套数据采集系统,该系统通过使用USB存储设备,可以在自行车机器人行驶过程中随时记录其倾斜角度和同一时刻的车把旋转角度,为每次实验后的总结工作提供了第一手的材料。
在自行车机器人的运行过程中,控制系统需要及时根据自行车机器人的当前状态来向执行机构发出指令以确保车体的平衡,这就对系统的实时性提出了很高的要求。本文定量分析了程序运行的每一次任务调度和中断响应所耗时间,与传统的控制系统策略下的实时性表现作了详细对比,证明了本控制系统对系统相应的实时性作了很大的改进,很大程度上满足了自行车机器人对实时性的要求。
Bicycle Robot Self-balance control is a difficult issue in the field of robotics, and the current studies are mainly about the initial theory researches of the system. After some profound studies were made on the basis of previous theories, an integrated digital control system was designed. A mass of experiments had been conducted to verify the designing and to analyze the real-time performance of the system.
At the beginning of this paper, the previous research results about bicycle robot dynamic modeling were summarized. A kind of steering control dynamic model for bicycle robot based on Lagrange method was indicated. Based on the model and the theory of quadratic optimal control theory, digital controller of bicycle robot control system was established. The stability, observability and controllability of the bicycle control system were analyzed. In order to validate the effectiveness of the controller based on an embedded platform, the designing details of the control system's hardware and software modules were expatiated. Then, a simple data acquisition system with USB storage devices had been built up. The data acquisition system was used to get the tilting angle of the bike's body and the steering angle of the bike's front wheel at a time. The data of every experiment can be used as first-hand materials for farther analyzing work.
In the operating processes of the bicycle robot, in order to ensure the balance of the bike, the controller should send instruction to the executing mechanism according to the current status of the bike as soon as possible. So the system's real-time performance should be high. The time data spent by task scheduling and interrupt response were saved. By quantitative compared to the traditional control strategies, the real-time performance of the controller was improved greatly. And the performance can satisfy the real requirement to a great extent.
引文
[1]科依费特,奇罗兹 机器人技术导论 国防科技大学出版社1991
[2]R.格兰曼著 回转仪的理论和应用北京 科学出版社 1959
[3]刘延柱 自行车的受控运动力 力学与实践 1995,17(4):39-42
[4]Neil H.G.Internal equilibrium control of a bicycle.Proceedings of the 34th Conference on Decision & Control.New Orieans:IEEE 1995:4285-4287
[5]Y Yavin The Derivation of a Kinematic Model from the Dynamic Model of the Motion of a Riderless Bicycle Computers and Mathematics with Applications,2006,51:865-878
[6]侯爽,宋颖慧 一种实时系统中的多任务可预测调度算法 计算机工程 2004,8:67-69
[7]金宏,王宏安等 一种任务优先级的综合设计方法 软件学报 2003,14:376-382
[8]郭磊 自行车机器人非线性系统中若干问题的研究[学位论文]中国博士学位沦文全文数据库 北京邮电大学 2007
[9]Katsuhiko Ogata(尾形克彦)著离散时间控制系统(原书第2版)北京:机械工业出版社 2006
[10]王耀南 著 机器人智能控制工程 北京:科学出版社 2004
[11]Jean J.Labrosse嵌入式实时操作系统μC/OS-Ⅱ(第2版)北京航空航天大学出版社 2003
[12]周立功 ARM嵌入式系统基础教程 北京航空航天大学出版社 2004
[13]任哲 嵌入式实时操作系统μC/OS-Ⅱ原理及应用 北京航空航天大学出版社2005
[14]郭磊 自行车机器人动力学建模与MIMO反馈线性化 北京邮电大学学报2007年第1期:80-84,95
[15]刘淼,王田苗,魏洪兴等基于μC/OS-Ⅱ的嵌入式数控系统实时性分析 计算机工程,2006,32(22):222-224
[16]季志均,马文丽,陈虎等 四种嵌入式实时操作系统关键技术分析 计算机应 用研究,2005,9:4-8
[17]施华,薛广涛 机器人控制系统实时性的研究 计算机工程 2003,29(1):91-92,125
[18]R J A Buhr,D L Bailey An Introduction to Real-time Systems:From Design to Networking with C/C++.Englewood Cliffs,NJ:Pretentice Hall 1998
[19]陈燕俐,洪龙 自由调整中断优先级的研究和实现 南京邮电学院学报 2005,12:35-39
[20]王田苗 嵌入式系统设计与实例开发(第2版)北京:清华大学出版社 2003,10
[21]杜春雷 ARM体系结构与编程 北京:清华大学出版社,2003,2
[22]刘豹 编著 现代控制理论 机械工业出版社 2000
[23]周杰,陈伟海,于守谦 基于ARM的嵌入式系统在机器人控制系统中应用 微计算机信息 2007,2:271-274
[24]孙鹤旭,林涛主编 嵌入式控制系统 北京:清华大学出版社 2007
[25]鲁比尼LINUX设备驱动程序(第二版)北京:电力出版社 2002
[26]邹思轶 嵌入式Linux设计与应用 北京:清华大学出版社 2002
[2]R.格兰曼著 回转仪的理论和应用北京 科学出版社 1959
[3]刘延柱 自行车的受控运动力 力学与实践 1995,17(4):39-42
[4]Neil H.G.Internal equilibrium control of a bicycle.Proceedings of the 34th Conference on Decision & Control.New Orieans:IEEE 1995:4285-4287
[5]Y Yavin The Derivation of a Kinematic Model from the Dynamic Model of the Motion of a Riderless Bicycle Computers and Mathematics with Applications,2006,51:865-878
[6]侯爽,宋颖慧 一种实时系统中的多任务可预测调度算法 计算机工程 2004,8:67-69
[7]金宏,王宏安等 一种任务优先级的综合设计方法 软件学报 2003,14:376-382
[8]郭磊 自行车机器人非线性系统中若干问题的研究[学位论文]中国博士学位沦文全文数据库 北京邮电大学 2007
[9]Katsuhiko Ogata(尾形克彦)著离散时间控制系统(原书第2版)北京:机械工业出版社 2006
[10]王耀南 著 机器人智能控制工程 北京:科学出版社 2004
[11]Jean J.Labrosse嵌入式实时操作系统μC/OS-Ⅱ(第2版)北京航空航天大学出版社 2003
[12]周立功 ARM嵌入式系统基础教程 北京航空航天大学出版社 2004
[13]任哲 嵌入式实时操作系统μC/OS-Ⅱ原理及应用 北京航空航天大学出版社2005
[14]郭磊 自行车机器人动力学建模与MIMO反馈线性化 北京邮电大学学报2007年第1期:80-84,95
[15]刘淼,王田苗,魏洪兴等基于μC/OS-Ⅱ的嵌入式数控系统实时性分析 计算机工程,2006,32(22):222-224
[16]季志均,马文丽,陈虎等 四种嵌入式实时操作系统关键技术分析 计算机应 用研究,2005,9:4-8
[17]施华,薛广涛 机器人控制系统实时性的研究 计算机工程 2003,29(1):91-92,125
[18]R J A Buhr,D L Bailey An Introduction to Real-time Systems:From Design to Networking with C/C++.Englewood Cliffs,NJ:Pretentice Hall 1998
[19]陈燕俐,洪龙 自由调整中断优先级的研究和实现 南京邮电学院学报 2005,12:35-39
[20]王田苗 嵌入式系统设计与实例开发(第2版)北京:清华大学出版社 2003,10
[21]杜春雷 ARM体系结构与编程 北京:清华大学出版社,2003,2
[22]刘豹 编著 现代控制理论 机械工业出版社 2000
[23]周杰,陈伟海,于守谦 基于ARM的嵌入式系统在机器人控制系统中应用 微计算机信息 2007,2:271-274
[24]孙鹤旭,林涛主编 嵌入式控制系统 北京:清华大学出版社 2007
[25]鲁比尼LINUX设备驱动程序(第二版)北京:电力出版社 2002
[26]邹思轶 嵌入式Linux设计与应用 北京:清华大学出版社 2002