Lane Change Maneuvering in AVs for Collision Avoidance and T
发布时间：2021-03-19 14:19
自动驾驶（AD）是提升舒适性,道路安全性和效率的最新兴领域。自动驾驶基于对人类驾驶员行为的研究。驾驶行为是驾驶员在不同环境下,加速和减速,速度,刹车踏板,经度和横向运动以及车辆之间的最小安全距离方面的概念性研究。本文大纲基于以人为中心的自动驾驶汽车。我们特别关注换道时的交通流量的安全性与效率。现代社会要求安全无障碍道路。根据车联网（VANET）的各种用途所需的特定需求,在大多数情况下进行换道,设计稳定的避撞（CA）和安全车道变换已经转变为在密集环境中控制车辆的基础。为了减少现代交通系统中的碰撞并提高效率,高效的嵌入式系统和汽车之间进行有效的通信是非常重要。本文将VANET和车载系统安装在一起,以实现我们的目标。本文根据,速度和各类车辆之间的距离,提出了高效的换道选择和无碰撞的车辆运输系统。在本文中,我们首先描述了基于耐心的策略,并利用博弈论方法重新确立了交通流的安全性和效率。然后,我们利用从车辆通信中得到的汽车的距离和速度信息,改进了用于车辆换道和碰撞控制的算法。本文算法在分析了两种不同的轨道变换和无碰撞驾驶算法之后;考虑各种交通密度,形成基于随机耐心的算法和汽车跟随算法。因此,在实验...
【文章来源】：大连理工大学辽宁省 211工程院校 985工程院校 教育部直属院校
【文章页数】：65 页
【学位级别】：硕士
【文章目录】：
摘要
Abstract
1 Introduction
1.1 Research Background
1.2 Research Objective and Significance
1.2.1 Research Objective
1.2.2 Research Significance
1.3 Domestic and Overseas Progress
1.3.1 Significance of Artificial Intelligence in Lane Change Maneuvering
1.3.2 The Recent Development of VANET in Autonomous Vehicles
1.3.3 Existing Methods to Overcome Safety and Efficiency Issues
1.4 Story Outline and Methodology of Dissertation
1.4.1 Story Outline
1.4.2 Research Methods
2 Challenges and Existing Models
2.1 Characterization of Driving Automation
2.2 Challenges in Driving Automation
2.2.1 Collision Avoidance and Trajectory Generation
2.2.2 Drivers Behavior Prediction
2.2.3 Intend Recognition and Skill Learning
2.2.4 Detection of Potential Threats
2.2.5 Minimum Safety Spacing
2.3 Existing Methods to Overcome Safety and Efficiency Issues
2.3.1 Stochastic Switched Autoregressive Exogenous
2.3.2 Cerebellar Model Articulation Controller（CMAC）
2.3.3 Hidden Markov Model
2.3.4 Neural Network
2.3.5 Gauss Mixture Model
2.3.6 Fuzzy System
2.3.7 Bayesian Network
2.3.8 Support Vector Machine
2.3.9 Car-Following Model
3 Set up a Trajectory Planning Model via Artificial Neural Network and VANET
3.1 Lane Change Analysis and Evaluation
3.1.1 Lane Change Maneuvering
3.1.2 General Methods of Lane Change Maneuvering
3.2 Setup Safety Model via VANET
3.2.1 Data Collection and Individual Situation Analysis
3.2.2 Cooperative Situation Analysis
3.2.3 Disseminated Knowledge Management
3.2.4 Individual Consequences and Driver Interface
3.2.5 Consistent Assessment of the Existing Driving Situation
3.2.6 Communication and Information Broadcasting
3.3 Learning Approach in Bayesian Networks
3.3.1 Parameter learning
3.3.2 Structure Learning
4 Bayesian Structure Employment
4.1 Data Preprocessing
4.2 Traffic Scenario
4.3 Application of Bayesian Network in Lane Change Maneuvering
4.3.1 Pre Mature Model
4.3.2 Mathematical Model Setup
4.3.3 Algorithm Design
4.4 Summary
5 Experimental Setup and Results
5.1 Simulation Environment
5.2 Simulation result
5.2.1 Input data
5.2.2 Output/results
5.2.3 For Safety Application
5.2.4 Improvement in Traffic Efficiency in the traffic
Conclusions
References
Research Projects and Publications in Master Study
Acknowledgments
【参考文献】：
期刊论文
[1]应用于换道预警的驾驶风格分类方法[J]. 王畅,付锐,彭金栓,毛锦. 交通运输系统工程与信息. 2014(03)
本文编号：3089718
【文章来源】：大连理工大学辽宁省 211工程院校 985工程院校 教育部直属院校
【文章页数】：65 页
【学位级别】：硕士
【文章目录】：
摘要
Abstract
1 Introduction
1.1 Research Background
1.2 Research Objective and Significance
1.2.1 Research Objective
1.2.2 Research Significance
1.3 Domestic and Overseas Progress
1.3.1 Significance of Artificial Intelligence in Lane Change Maneuvering
1.3.2 The Recent Development of VANET in Autonomous Vehicles
1.3.3 Existing Methods to Overcome Safety and Efficiency Issues
1.4 Story Outline and Methodology of Dissertation
1.4.1 Story Outline
1.4.2 Research Methods
2 Challenges and Existing Models
2.1 Characterization of Driving Automation
2.2 Challenges in Driving Automation
2.2.1 Collision Avoidance and Trajectory Generation
2.2.2 Drivers Behavior Prediction
2.2.3 Intend Recognition and Skill Learning
2.2.4 Detection of Potential Threats
2.2.5 Minimum Safety Spacing
2.3 Existing Methods to Overcome Safety and Efficiency Issues
2.3.1 Stochastic Switched Autoregressive Exogenous
2.3.2 Cerebellar Model Articulation Controller（CMAC）
2.3.3 Hidden Markov Model
2.3.4 Neural Network
2.3.5 Gauss Mixture Model
2.3.6 Fuzzy System
2.3.7 Bayesian Network
2.3.8 Support Vector Machine
2.3.9 Car-Following Model
3 Set up a Trajectory Planning Model via Artificial Neural Network and VANET
3.1 Lane Change Analysis and Evaluation
3.1.1 Lane Change Maneuvering
3.1.2 General Methods of Lane Change Maneuvering
3.2 Setup Safety Model via VANET
3.2.1 Data Collection and Individual Situation Analysis
3.2.2 Cooperative Situation Analysis
3.2.3 Disseminated Knowledge Management
3.2.4 Individual Consequences and Driver Interface
3.2.5 Consistent Assessment of the Existing Driving Situation
3.2.6 Communication and Information Broadcasting
3.3 Learning Approach in Bayesian Networks
3.3.1 Parameter learning
3.3.2 Structure Learning
4 Bayesian Structure Employment
4.1 Data Preprocessing
4.2 Traffic Scenario
4.3 Application of Bayesian Network in Lane Change Maneuvering
4.3.1 Pre Mature Model
4.3.2 Mathematical Model Setup
4.3.3 Algorithm Design
4.4 Summary
5 Experimental Setup and Results
5.1 Simulation Environment
5.2 Simulation result
5.2.1 Input data
5.2.2 Output/results
5.2.3 For Safety Application
5.2.4 Improvement in Traffic Efficiency in the traffic
Conclusions
References
Research Projects and Publications in Master Study
Acknowledgments
【参考文献】：
期刊论文
[1]应用于换道预警的驾驶风格分类方法[J]. 王畅,付锐,彭金栓,毛锦. 交通运输系统工程与信息. 2014(03)
本文编号：3089718
本文链接：https://www.wllwen.com/kejilunwen/qiche/3089718.html
