【作者】 左建昌；

【导师】 王国庆；

【作者基本信息】 长安大学 ， 机械设计及理论， 2008， 硕士

【摘要】 移动机器人作为机器人学科中一个比较活跃的研究分支,一直以来得到了学界的关注。移动机器人的避障,导航和定位是移动机器人研究领域内一个重要的研究方向,同时也是移动机器人研究的关键技术。随着信息技术,自动化技术,计算机技术和导航技术的不断发展,各个学科间不断地交叉融合。这也极大地推动了移动机器人在避障、导航和定位方面研究的发展。本文以工作在部分未知或者完全未知环境下的移动机器人为研究对象,从工程应用和理论两个方面对移动机器人的超声波避障与惯性导航进行了研究,并且着重研究了系统的硬件设计。通过综合国内外现有的技术文献,提出了一种融合超声波避障与惯性导航的综合导航平台系统方案。导航平台具有超声波避障探测、模拟信号采集、MEMS惯性导航模块和与GPS模块进行通信等功能。为了能够在满足系统性能要求的基础上尽可能地降低系统成本,该导航平台是以MSP430F135为核心。在完成其外围电路的基础上,采用T/R40作为超声波传感器,并对超声波的发送和接收电路进行了设计。该导航平台对模拟信号的采集可分为对高频模拟信号和低频模拟信号的采集。其中高频模拟信号采集是由AT89S52控制8位高速A/D-TLC5540INSR来实现的,而低频模拟信号采集是由MSP430F135控制其集成的12位ADC来实现。为了能够建立起主CPU和同样由MSP430F135作为主控CPU的MEMS惯性导航模块间的通信,文中采用同步SPI作为主从CPU通信方式。另外,为了能够扩展系统功能,文中采用RS232作为通信接口,实现主MSP430F135与GPS模块之间的通信。然后,在完成以上硬件设计的基础上进行了系统软件的仿真与调试。在实现各个模块的基本功能之后,初步进行了系统的综合调试。为了提高惯性导航的导航精度,本文从理论上对惯性导航系统中的MEMS陀螺仪的漂移问题和卡尔曼滤波中的野值问题进行了理论分析,并研究了野值剔除方法和修正野值的卡尔曼滤波算法。论文最后对全文进行了总结并对移动机器人在避障和惯性导航方向的研究进行了展望,阐明了移动机器人未来可能的研究方向。更多还原

【Abstract】 As a correspondingly active research branch of robotics, the mobile robot has being drawn more attention in this field all the time. The obstacle avoidance and navigation of mobile robot was an important research field, as a key technique as well. As the development of information technique, automation technique, computer technique and navigation technique, each field has been continuously intercrossing and syncretizing with others. It impelled the researching development of mobile robot in obstacle avoidance and navigationThis thesis took the robot which worked under partly unknown or absolutely unknown environment as research object, and briefly focused on the design of hardware.Comparing the existing literature both domestic and overseas, we designed a kind of mobile robot navigation system platform which integrated the ultrasonic obstacle avoidance and inertial navigation system. This navigation platform had the function such as ultrasonic obstacle avoidance detecting, analog signals acquisition, MEMS inertial navigation module and communicating with GPS module. In order to satisfy the system requirement and low the cost, we selected the MSP430F135 as the Signal Processor. Basing on completing the peripheral circuit designing, we selected the T/R40 as the ultrasonic sensor and designed the ultrasonic eradiating circuit and receiving circuit. The analog signals acquisition module was composed of high frequency and low frequency analog signals acquisition. Thereinto, the high frequency acquisition system was implemented by that AT89S52 controlled the high-speed 8-bit analog-to-digital converter, while the low frequency acquisition system was implemented by MSP430F135 which integrated a 12-bit ADC. In order to build the communication between the master and MEMS inertial navigation module, the synchronous SPI was used as the communication mode between the master and slavery. In addition, in order to extent the system function, the RS232 communication interface was used to implemented the communication between the master CPU and GPS module. And then, basing on the completed hardware design, we did some system software simulation and debugging to implement every module function. Finally, after finishing the every module function, we made some simple system synthetically debugging. At the end, in order to enhance the navigation precision, we analyzed the random drift of the MEMS gyroscope in inertial system and the drift of Kalman filter,and did some research on how to eliminate the drift and a Kalman filter arithmetic which could properly compensate the drift.更多还原