摘 要

随着经济的发展，交通堵塞，交通状况越来越差，人们对于良好交通的需求正在逐渐上升，进而引发了智能交通的研究与发展。现如今，智能交通发展的越来越迅速，规模越来越大，自动驾驶，智能规划路线，辅助驾驶等正在改变着人们的交通生活，并且智能手机已经成为人们生活中不可或缺的必需品，本文在智能驾驶以及智能手机发展迅猛的背景下，对智能交通系统中的车道变道行为检测方法进行研究。

本文主要研究一种智能手机的车辆变道行为检测方法。从车辆的变道行为特征出发，介绍了单次变道，连续同向变道，连续反向变道等各种情况下的车辆变道行为特征，并根据这些特征，设计了基于车辆变道过程中横向加速度变化的模型，在理论分析的基础上，对算法模型进行了工程实践，并通过真实驾车行驶的数据进行测试与分析，其中包含了独立单次车道变换，同向连续车道变换，反向连续车道变换，连续两次同向连续车道变换，连续两次反向连续车道变换，先后进行同向连续车道变换与反向连续车道等场景，将算法检测的结果与实际变道行为进行对比，得出结论，测试算法的准确性和鲁棒性。

通过测试与分析证明，本文提出的基于智能手机的车辆变道行为检测方法，能够成功检测到各种类型，各种情况下的车辆变道行为，并在实际驾车行驶的过程中，检测算法有良好的准确性和鲁棒性。

关键词： 智能手机，车辆变道行为，变道检测

Abstract

With the development of economy, traffic jam and worse traffic condition, people's demand for good traffic is gradually rising, which leads to the research and development of intelligent traffic.Nowadays, the development of intelligent transportation is more and more rapidly, the scale is more and more big, the autopilot, intelligent route planning, such as auxiliary driving are life changing people's traffic, and smart phone has become indispensable in people's life necessities, based on the intelligent driving and intelligent mobile phone under the background of developing rapidly, the lane lane change behavior of the intelligent transport system detection method were studied.

This paper mainly studies a smart phone vehicle lane change behavior detection method.From the lane changing behavior characteristics of vehicle, this paper introduces the single lane change, continuous direction change lanes, continuous reverse change lanes and other vehicle lane change behavior characteristics of various circumstances, and according to these characteristics, design based on the vehicle lateral acceleration in the process of changing a lane change model, on the basis of theoretical analysis, the algorithm model for the engineering practice, and through the data of the real driving test and analysis, which includes the independent single lane change, synthetic continuous driveway transform, reverse continuous driveway transform, two consecutive synthetic continuous driveway transform, two consecutive reverse continuous driveway transform,In the same direction continuous lane change and reverse continuous lane change scenarios, the algorithm detection results were compared with the actual lane change behavior, and the conclusion was drawn that the accuracy and robustness of the algorithm were tested.

Through testing and analysis, it is proved that the vehicle lane change behavior detection method based on smart phone proposed in this paper can successfully detect various types and situations of vehicle lane change behavior, and the detection algorithm has good accuracy and robustness in the actual driving process.

Keywords: smartphone, lan change behavior, lane change detection

目 录

摘 要 I

Abstract II

目 录 3

第1章 绪论 1

1.1 课题研究背景 1

1.2 课题研究的目的和意义 1

1.3 国内外研究现状 2

1.4 课题研究内容及预期目标 3

第2章 车辆变道行为研究 4

2.1车辆变道行为基础 4

2.1.1 基础理论 4

2.1.2 数据采集 5

2.2车辆变道行为分析 6

2.2.1 单一车道变换 6

2.2.2连续车道变换 7

2.3 本章小结 10

第3章 车辆变道行为建模 11

3.1 独立单次车道变换模型 11

3.2 同向连续变换车道模型 11

3.3 反向连续车道变换模型 12

3.4 本章小结 13

第4章 测试与分析 14

4.1车辆变道行为检测方法实现 14

4.2 车辆变道行为检测方法测试 17

4.3 本章小结 20

第5章 总结与展望 21

参考文献 22

附录A 24

致 谢 27

第1章 绪论

1.1 课题研究背景

随着经济的快速发展，车辆的数目正在快速增加，与之伴随的是交通状况的越来越差，人与道路，与汽车之间的矛盾越来越深，为了利用新科技来缓解这一矛盾，学术界提出了建立一个智能交通系统来处理这些矛盾^[1]，让新兴的技术服务人类，提高人们的生活质量。智能交通系统指的是人，车，路之间互相协调的一种系统，也是解决人，车与路之间种种矛盾的途径之一^[2]。

近年来，智能手机的快速发展也给人们的生活带来了极大的便利，智能手机的处理速度越来越快，越来越智能化^[3]。这为智能手机能够作为一台便携的数据采集器与数据处理器提供了可能。传统上，智能手机一般利用GPS得到一个位置信息，用以导航以及定位等功能^[4]，但面对着要求越来越准确，智能化的智能交通系统而言，GPS的误差范围达不到要求。