基于IOS的车辆行驶行为识别方法研究与实现

发布时间：2018-12-08 08:03

【摘要】：近年来,随着国民经济和社会的飞速发展,汽车保有量也急剧攀升,道路交通安全问题越来越受到社会的广泛关注。统计数据显示不良的驾驶行为是引起交通事故的主要原因,分析车辆在行驶过程中的行为对于规范驾驶行为、减少交通事故、提高交通安全具有指导意义。互联网技术的快速发展和智能移动终端设备丰富的功能给人们的生活带来便捷,智能设备已经成为人们生活中不可缺少的一部分。IOS智能设备中的加速计和陀螺仪能够感知设备的移动和状态,通过这些传感器可以实现低成本的数据采集功能。本文利用IOS设备传感器采集并识别车辆的行驶行为,主要包括:换道、加速、减速、刹车等行为,在此基础上提出了基于支持向量机的识别算法,具体工作如下:1.利用智能手机/pad等移动终端中嵌入的加速计和陀螺仪采集车辆行驶过程中的加速度和角速度数据,针对车辆行驶过程中因车辆颠簸产生的零点漂移及高频噪声的问题,对采集到的数据进行标零和低通滤波处理。2.针对车辆行驶过程中换道、加速、减速和刹车时采集到的加速度和角速度数据,建立能够表征其行为特征的特征向量,包括数据的最大值、最小值、最大最小值之间的差值、均值及方差,提出了基于支持向量机的车辆行为分类器。为了解决连续换道等行为的识别,提出了一种N-δ滑动窗口截取算法,实现了对包含多个行为的数据的快速划分。3.搭建实验测试平台,并在城市道路中对本文提出的车辆行驶行为识别方法进行了实验测试,测试结果表明,该方法能够有效地识别车辆行驶行为。
[Abstract]:In recent years, with the rapid development of the national economy and society, the number of vehicles has risen rapidly, and the problem of road traffic safety has been paid more and more attention by the society. Statistical data show that poor driving behavior is the main cause of traffic accidents. It is of guiding significance to analyze the behavior of vehicles in the course of driving to regulate driving behavior, reduce traffic accidents and improve traffic safety. The rapid development of Internet technology and the rich functions of intelligent mobile terminal devices bring convenience to people's life. Smart devices have become an indispensable part of people's lives. Accelerometers and gyroscopes in IOS smart devices can sense the movement and state of devices, and can achieve low-cost data acquisition through these sensors. This paper uses IOS device sensor to collect and recognize the driving behavior of the vehicle, including: changing track, accelerating, decelerating, braking and so on. On this basis, a recognition algorithm based on support vector machine is proposed. The specific work is as follows: 1. Accelerometers and gyroscopes embedded in mobile terminals such as smart phone / pad are used to collect acceleration and angular velocity data during vehicle driving, aiming at the problem of zero drift and high frequency noise caused by vehicle bumps during vehicle driving. Data collected by zero and low pass filter processing. 2. In view of the acceleration and angular velocity data collected during the course of the vehicle moving, accelerating, decelerating and braking, the characteristic vectors, including the difference between the maximum, the minimum, the maximum and the minimum values, are established to characterize the characteristics of the vehicle's behavior, including the maximum, the minimum and the maximum and minimum values of the data. A vehicle behavior classifier based on support vector machine (SVM) is proposed based on mean and variance. An N- 未 sliding window intercepting algorithm is proposed to realize the fast partition of data including multiple behaviors. The experimental test platform is built and the vehicle driving behavior identification method proposed in this paper is tested in the urban road. The test results show that the method can effectively identify the vehicle driving behavior.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495

【参考文献】