【作者】 张继军；

【导师】 孟庆春；

【作者基本信息】 中国海洋大学 ， 信号与信息处理， 2004， 硕士

【摘要】 自主式移动机器人是目前机器人领域的研究重点之一，吸引着众多学者的注意，这是因为它有着巨大的应用潜力。近年来国内的科研院所都开始了对自主式移动机器人硬件系统的研制开发工作，我们在山东省自然科学基金的资助下，完成了新型智能导游机器人的研制工作，本论文完成的主要工作包括对机器人的底层硬件系统设计以及相应控制方法的实现。 论文首先讨论了机器人技术的研究发展现状以及在不同领域的应用情况。介绍了机器人领域发展的不同阶段和我国机器人研究的进展情况，分析了模糊控制在机器人系统中的应用特点，确定了机器人导航的模糊控制方法。 第1章详细分析了机器人的运动学特性。从独立双轮驱动的运动学模型出发，对运行速度和角速度之间的关系进行了探讨，指出机器人转弯半径与两轮线速度的差成复杂的非线性关系。推出了机器人纠偏过程的运动学方程，画出了机器人的运动学框图，指出其控制系统是一个复杂的非线性系统，采用传统的线性控制理论难以凑效，进一步从理论上确立了采用模糊控制方法的必要性。 第2章则对机器人硬件系统的设计展开了论述。针对导游机器人通常工作在结构化的环境中这一特点，自行设计了光电传感器来实现机器人的路标导航，采用中值滤波算法对传感器信息进行处理。根据渡越时间法原理设计实现了机器人的超声波测距功能模块，为了提高测量精度，采用单总线数字温度传感器进行温度补偿。然后对系统总体设计进行了探讨，提出了基于客户／服务器(C／S)思想的机器人开放式接口，接着介绍了系统软件的开发与调试工作，最后讨论了提高系统可靠性的抗干扰措施。 第3章论述了机器人小车转向的模糊控制方法。首先简述了模糊控制理论，根据小车运动特点提出了机器人的角速度—线速度函数模型，将两输出系统转化为单输出系统，设计了两输入单输出模糊控制器，简化了控制系统结构。将多传感器信息编码作为系统误差，对误差和误差的变化率模糊化后作为模糊控制器的输入，利用Mamdani推理方法将人工驾驶的知识经验蕴含到模糊关系中，最后解模糊得到控制量输出。为了减少实时控制过程中的运算量，对各种误差和误差的变化率分别利用Matlab进行离自主式移动机器人控制系统设计与模糊控制实现线计算，最后得到控制量查询表。我们利用单片机通过查询表方式实现了机器人小车转向的模糊控制，实际运行结果体现了控制意图，效果较为理想。最后，对全文工作进行了总结，指出了今后进一步的研究方向。更多还原

【Abstract】 The research on Autonomous Mobile Robot (AMR) is one of the emphases of mobile robot, and it is attractive to many specialists and scholars. The attraction of AMR should owe to its tremendous potential market. In recent years, many domestic research organizations and institutions have begun to develop the hardware of AMR. With the imbursement of science fund from ShanDong province, we have developed the New Intelligent Guide Robot (HiFully- I ). This paper focuses on the hardware system design and the fuzzy control implementation of our guide robot.First, we discussed the situation of the research on mobile robot and the applications of mobile robot in different fields. We introduced the developing decades of robot technology and domestic research situation on robot technology, analyzed the characteristic of fuzzy control and established the fuzzy control method for robot navigation.In chapter 1 we analyzed the kinematical characteristic of mobile robot. Depending on the kinematical model, we discussed the relationship between the line velocity and the angular velocity of mobile robot and indicated the linear relationship between swerve radius and wheel’s line velocity square difference. We calculated the kinematical equations of reclaim and planted the kinematical chart. Because the control system of mobile robot is a complicated non-linear system, it is difficult to implement with traditional method. The necessity of fuzzy control had been improved with theory analysis.We discussed the hardware design of mobile robot in chapter 2. Guide robot normally worked in fixed environment, so we chose the landmark to navigate our robot. We developed optical sensor to detect the landmark and deal with the sensor information with median filter. Developed ultrasonic distance measure system based on Time-of-Flight theory. To enhance the measure precision, 1-Wire digital temperature sensor was used to compensate the measure result. Then discussed the whole system design, and presented the open interface of mobile robot based on Client/Server idea. After that, introduced the development and debug of system software. At last discussed the methods for counter disturb in order to enhance the reliability of our robot system.In chapter 3, discussed the fuzzy control method on robot navigation. First introduced the fuzzy control theory and presented the "angular velocity-linear velocity" functional model. Thus transformed the two output system to one output system and simplified the structure of control system. Regarded the code of multi-sensor as the error input, we can get the error and the error change. After fuzzy input, artificial experience was involved in the fuzzy relationship through Mamdani reasoning. At last we got the control table by calculations with Matlab. Single chip unit implemented the fuzzy control and our experiments improved to be effective.At last part, summarized this paper and indicated the research directions for the future.更多还原