基于图像信息采集的自动寻迹平衡小车设计  

Design of Automatic Tracing Balance Car Based on Image Information Acquisition

作　　者：崔渊 姬丰欣 陈祝洋 高倩 钱铮 CUI Yuan;JI Fengxin;CHEN Zhuyang;GAO Qian;QIAN Zheng(School of Electricity and Information Engineering,Jiangsu University of Technology,Changzhou 213001,China)

机构地区：[1]江苏理工学院电气信息工程学院,江苏常州213001

出　　处：《江苏理工学院学报》2021年第6期31-41,共11页Journal of Jiangsu University of Technology

基　　金：国家自然科学基金青年基金项目“典型含核粒子对无衍射波束的散射特性研究”(61601208);2020年大学生创新创业训练计划项目“基于自动循迹的图像信息采集平衡小车设计”(202011463029Y);江苏理工学院科研项目“精密电源/测量单元测量仪器研发与产业化”(KYH17036)。

摘　　要：针对循迹自平衡车的两个关键技术难点——直立控制和视觉导航,采用STM32F407VET6单片机为控制核心。一方面,采集MPU6050六轴传感器的加速度与角速度数据,经卡尔曼滤波后得到精准的直立倾角信息,结合速度-姿态串级PID控制器,加快电机对误差的响应速度,实现对平衡车直立的精确控制;另一方面,利用DMA机制采集OV7670摄像头的图像信息,并对其进行灰度化、高斯图像滤波、图像二值化、路径提取和拟合等处理,得到精确的路径信息,实现平衡车的视觉导航。为了增强人机交互体验,平衡车的关键数据还将实时传输到Android端App,并可通过App对平衡车进行控制。Aiming at the two key technical difficulties of tracking self-balancing vehicle,namely upright control and visual navigation,STM32F407VET6 single-chip microcomputer is used as the control core in this paper.On the one hand,the accurate vertical angle information is obtained by Kalman filter using the collected acceleration and angular velocity data of MPU6050 six-axis sensor.Combined with the speed attitude cascade PID controller,the response speed of the motor to the error is accelerated,and the accurate control of the upright of the balance vehicle is realized.On the other hand,the image information of ov7670 camera is collected by DMA mechanism,and processed by graying,Gaussian image filtering,image binarization,path extraction and fitting,so as to obtain accurate path information and realize the visual navigation of balance vehicle.In order to enhance the human computer interaction experience,the key data of the balancing car will also be transmitted to the Android App in real time,and the balancing car can be controlled through the App.

关 键 词：STM32 自平衡车 卡尔曼滤波 循迹 串级PID Android 

分 类 号：TM935.3[电气工程—电力电子与电力传动]