西安科技大学 通信与信息工程学院,陕西 西安 710054

Linux; 模糊PID控制算法; 无线视频; 路径识别

Wireless video tracing car under Linux system
WANG Zheng-jin,WU Feng-bo,LI Guo-min,ZHANG Wen-hai

(College of Communication and Information Engineering,Xi'an University of Science and Technology,Xi'an 710054,China)

Linux system; fuzzy-PID control algorithm; wireless video; path identification

DOI: 10.13800/j.cnki.xakjdxxb.2015.0118


为研究如何缩小自动寻迹小车的在指定路线上的偏差率问题,采用基于Linux平台下的嵌入式操作系统和无线视频处理相结合的方法。主控单元用2块飞凌公司的32位ARM11系列的S3C6410处理器分别作为服务器和小车的客户端; 路径识别单元采用CMOS数字摄像头OV9650进行摄像,通过图像处理单元提取路径信息; 驱动控制单元采用SGS公司的L298N芯片。结果 表明,通过模糊PID控制算法对舵机转向和速度进行控制,能较好实现小车按预定路线平滑寻迹。为避免Wi-Fi传输信号不好或者说数据量过大,出现严重丢包现象,Wi-Fi传输协议上选择基于数据流式传输数据的TCP协议。经过反复测试,系统路径识别性能良好,运行比较稳定。

To study how to reduce the car's automatic tracking deviation rate on the specified routes,we use a method combined embedded operating systems with wireless video processing based on Linux.The main control unit uses 2 pieces of Forlinx's 32-bit ARM11 series S3C6410 processors as main control unit of server and client,respectively.The path identification unit adopts the CMOS digital camera OV9650 to obtain the path information through the image processing unit.The drive control adopts SGS's L298N chip.The results show that it's better to realize the smooth tracing car at predetermined route through the fuzzy-PID control algorithm to control steering gear and speed.We select the TCP protocol based on transmission data of data streaming at the Wi-Fi transfer protocol.After repeated testing,the system's performance of path recognition is good and operation is stable.