Tối Ưu Hóa Đa Mục Tiêu Trong Điều Khiển Đèn Giao Thông

Trường đại học

De Montfort University

Người đăng

Ẩn danh

Thể loại

thesis

2019

187
0
0

Phí lưu trữ

30.000 VNĐ

Mục lục chi tiết

Abstract

Acknowledgements

Contents

1. CHƯƠNG 1: INTRODUCTION

1.1. Motivation

1.2. Traffic Signal Control Systems

1.2.1. Introduction to Traffic Signal Control Systems

1.2.2. Fundamental Definitions of Traffic Signal Control Systems

1.2.3. Overview of Traffic Signal Control Systems

1.2.4. Performance Measures of Traffic Signal Control Systems

1.3. Aims and objectives

1.4. Major Contributions of the Thesis

1.5. Simulation of Urban Mobility (SUMO)

1.6. Multi-objective evolutionary algorithms

1.6.1. Definition of Multi-objective Optimization Problems and Basic Concepts

1.6.2. General Framework of Multi-objective Evolutionary Algorithms

1.7. Surrogate-assisted evolutionary algorithms

2. CHƯƠNG 2: EVOLUTIONARY ALGORITHMS AND SURROGATE MODELS

2.1. Evolutionary algorithms vs. surrogates-assisted evolutionary algorithms

2.2. Strategies for managing surrogates

2.2.1. Model management: its roles and classification

2.2.2. Criteria for choosing individuals for re-evaluation

2.2.3. Techniques for constructing surrogates

2.2.4. Artificial Neural Networks

2.3. Multi-objective Traffic Signal Optimization

2.3.1. Traffic Signal Optimization using MOEAs

2.3.2. Multi-objective Traffic Signal Optimization using Local Search based MOEAs

2.4. Objectives in Traffic Signal Optimization

2.4.1. Optimization Objectives in Traffic Signal Control

2.4.2. Objective Calculation using Mathematical Programming Methods

2.4.3. Objective Calculation using Simulation-based Methods

2.5. Reducing Computational Cost using Surrogate Models

2.5.1. Computational Cost of Traffic Signal Optimization using MOEAs and Traffic Simulators

2.5.2. Techniques for constructing surrogates

2.5.3. Surrogate Assisted Optimization in Transportation

2.6. The local search strategy

2.6.1. Creating neighbours of a solution

2.6.2. Motivation of the local search method

2.6.3. The flow of the proposed local search

2.7. NS-LS algorithm

2.7.1. Overview of NS-LS

2.7.2. The flow of NS-LS

2.7.3. Design of the evolutionary search

2.7.3.1. Selection and Reproduction Operators

2.8. The surrogate model

2.8.1. Constructing a surrogate model

2.8.1.1. Choosing the model
2.8.1.2. The training algorithm
2.8.1.3. The error function

2.8.2. Updating a surrogate model

2.9. Fitness evaluation scheme

2.9.1. The motivation of the fitness evaluation scheme

2.9.2. The closeness of two solutions

2.9.3. The framework of the fitness evaluation scheme

2.10. SA-LS algorithm

2.10.1. Overview of SA-LS

2.10.2. The flow of SA-LS

3. CHƯƠNG 3: TRAFFIC SCENARIOS AND PERFORMANCE ASSESSMENT

3.1. Introduction to the traffic scenario of Andrea Costa

3.2. Introduction to the traffic scenario of Pasubio

3.3. Extracting optimization objective values from SUMO output

3.4. Indicators for Performance Assessment

3.5. Experimental design for evaluating the performance of the algorithms

3.5.1. Experiment 1 - Benchmark functions

3.5.2. Experiments using real-time traffic scenarios simulated by SUMO

3.5.2.1. Experiment 2 - Andrea Costa scenario
3.5.2.2. Experiment 3 - Pasubio scenario

7. CHƯƠNG 7: CONCLUSIONS, RECOMMENDATIONS, AND FUTURE WORK

7.1. Key findings of the research

7.2. Key contributions of the research

7.3. Limitations of the Research

7.4. Recommendations and Future Work

A Published Papers

B Mean hypervolume with standard deviation of the algorithms in Experiment 2

C Mean hypervolume with standard deviation of the algorithms in Experiment 3

Bibliography

List of Figures

List of Tables

Abbreviations

Symbols

Multi objective optimization in traffic signal control

Bạn đang xem trước tài liệu:

Multi objective optimization in traffic signal control