Jump to content
Object-Level Fusion for Surround Environment Perception in Automated Driving Applications / Titelei/Inhaltsverzeichnis
Object-Level Fusion for Surround Environment Perception in Automated Driving Applications / Titelei/Inhaltsverzeichnis
Contents
Chapter
Expand
|
Collapse
Page
Titelei/Inhaltsverzeichnis
Details
1–26
1 Introduction
1–26
Details
1.1 Motivation
Details
1.2 Automation in Driver Assistance and Safety Systems
Details
1.2.1 Level 0 – No Automation
Details
1.2.2 Level 1 – Driver Assistance
Details
1.2.3 Level 2 – Partial Automation
Details
1.2.4 Level 3 – Conditional Automation
Details
1.2.5 Levels 4 and 5 – Towards Fully Automated Driving
Details
1.3 Problem of Object Detection
Details
1.4 Contribution and Outline of the Thesis
Details
27–45
2 Sensor Data Fusion Architectures
27–45
Details
2.1 Overview
Details
2.1.1 Low-Level
Details
2.1.2 High-Level
Details
2.1.3 Hybrid
Details
2.1.4 Comparison
Details
2.2 ProposedModular Sensor Data Fusion Architecture
Details
2.2.1 Object Model
Details
2.2.2 Sensor-Level
Details
2.2.3 Fusion-Level
Details
2.2.4 Application-Level
Details
46–63
3 Fusion Strategy and Object Association
46–63
Details
3.1 Data Alignment
Details
3.1.1 Spatial
Details
3.1.2 Temporal
Details
3.2 Fusion Strategy
Details
3.2.1 Sensor-to-Sensor
Details
3.2.2 Sensor-to-Global
Details
3.3 Association
Details
3.3.1 Architecture
Details
3.3.2 Feature Selection
Details
3.3.3 State Vector
Details
3.3.4 Geometrical
Details
3.3.5 Association Validation
Details
3.3.6 Multi-Object Association
Details
64–92
4 State and Covariance
64–92
Details
4.1 Sensor-Level Processing with Tracking Algorithms
Details
4.1.1 Feature Extraction
Details
4.1.2 Data Association
Details
4.1.3 Filtering
Details
4.1.4 TrackManagement
Details
4.1.5 KinematicModels
Details
4.2 Correlation and Sequence of Sensor Data
Details
4.2.1 Process Noise
Details
4.2.2 Common Information History
Details
4.2.3 Out-of-Sequence Data
Details
4.3 Track-to-Track Fusion with the Common State
Details
4.3.1 Adapted Kalman Filter
Details
4.3.2 Covariance Intersection
Details
4.3.3 InformationMatrix Fusion
Details
4.3.4 Comparison
Details
4.4 Geometrical Fusion using the Object Model
Details
4.4.1 Dimension Estimation
Details
4.4.2 Extraction of Fused Coordinates
Details
93–115
5 Existence Probability
93–115
Details
5.1 Sensor-Level Processing
Details
5.1.1 Existence Prediction
Details
5.1.2 Existence Update
Details
5.1.3 Generalized Bayes Extension
Details
5.1.4 Modeling the Parameters
Details
5.1.5 Object Management
Details
5.2 Fusion
Details
5.2.1 Architecture
Details
5.2.2 Modeling with Dempster-Shafer Evidence Theory
Details
5.2.3 Extension for Occlusion Modeling
Details
5.2.4 Modeling the Trust Probability
Details
116–132
6 Classification
116–132
Details
6.1 Sensor-Level Processing
Details
6.1.1 Measurement Classification
Details
6.1.2 Temporal Filtering
Details
6.2 Fusion
Details
6.2.1 Modeling with the Dempster-Shafer Evidence Theory
Details
6.2.2 Modeling the Trust Probability
Details
133–151
7 Evaluation
133–151
Details
7.1 Test Vehicle and Sensor Configuration
Details
7.2 OvertakingManeuver with Ground Truth
Details
7.2.1 Ground Truth Calculation
Details
7.2.2 State Estimation
Details
7.2.3 Existence
Details
7.2.4 Classification
Details
7.3 Performance in Real Traffic Scenarios
Details
7.3.1 Detection Rate
Details
7.3.2 Classification Performance
Details
7.3.3 Integration in Automated Driving and ADAS Projects
Details
152–154
8 Conclusion and Discussion
152–154
Details
155–161
A Synchronous Track-to-Track Fusion Algorithms
155–161
Details
A.1 Simple Weighted Fusion
Details
A.2 Use of Cross-Covariance
Details
A.3 Covariance Intersection
Details
A.4 Comparison
Details
162–163
B Information Matrix Fusion Derivation
162–163
Details
164–166
C Determining the Trust Probability
164–166
Details
C.1 Existence
Details
C.2 Classification
Details
167–171
D Evaluation Scenario Descriptions
167–171
Details
D.1 Training Data
Details
D.2 Evaluation Data
Details
D.2.1 Test Track
Details
D.2.2 Real Traffic
Details
172–202
References
172–202
Details
Durchsuchen Sie das Werk
Geben Sie ein Keyword in die Suchleiste ein
Partial access
Object-Level Fusion for Surround Environment Perception in Automated Driving Applications
Titelei/Inhaltsverzeichnis
Autoren
Michael Aeberhard
DOI
doi.org/10.51202/9783186804129-I
ISBN print: 978-3-18-380412-2
ISBN online: 978-3-18-680412-9
Chapter Preview
Share
Download PDF
Download citation
RIS
BibTeX
Copy DOI link
doi.org/10.51202/9783186804129-I
Share by email
Video schließen
Share by email VDI Verlag eLibrary
Recipient*
Sender*
Message*
Your name
Send message
This site is protected by reCAPTCHA and the Google
Privacy Policy
and
Terms of Service
apply.