Improving neural networks for automated driving using corner cases and sensorfusionsion / submitted by Kamil Peter Kowol, M.Sc. Wuppertal, [2023]
Inhalt
- Acknowledgments
- Foreword
- Contents
- Introduction
- State of the Art and Theoretical Foundation
- Feedforward Neural Networks
- Learning Process
- Convolutional Neural Networks
- Image Segmentation
- Object Detection
- Evaluation Metrics
- Model Development Process
- Driving Simulator
- Introduction
- Software
- Hardware Components
- Code Adaptation and Acceleration
- Error Occurrence and Correction
- Corner Cases in Autonomous Driving
- Levels of Driving Automation
- Corner Case
- YOdar: Uncertainty-based Sensor Fusion for Vehicle Detection with Camera and Radar Sensors kowol21yodar
- Characteristics
- Object Detection via Radar
- Object Detection via Sensor Fusion
- Fusion Metrics and Methods
- Numerical Experiments
- Conclusion
- Datasets for Tracking and Retrieval of Out-of-Distribution Objects maag2022oodtracking
- Datasets for Anomaly Detection bogdoll2023AnomalySurvey
- Applications for Driving Simulator
- A-Eye: Driving with the Eyes of AI for Corner Case Generation kowol22simulator
- survAIval: Survival Analysis with the Eyes of AI Kowol2023survAIval
- A Taxonomy of Corner Cases in Trajectory Datasets for Automated Driving Roesch2022CCTrajectory
- Conclusion & Outlook
- List of Figures
- List of Tables
- List of Notations and Abbreviations
- Bibliography