This research aims to investigate the effectiveness of scenario-based testing as a comprehensive and robust approach for evaluating ASs' performance while enhancing their safety and reliability with respect to regulations and standards.
The primary objective of this thesis will be to investigate the benefits of scenario-based testing for automated systems and its compliance with regulations and standards. The following sub-objectives will be pursued:
- Conduct an in-depth review of relevant literature and industry practices on ASs testing methodologies, with a focus on the unique challenges posed by ASs' complex decision-making algorithms and interaction with the dynamic environment.
- Develop a scenario-based testing framework for systematic identification, generation, selection, and execution of realistic and diverse scenarios, for automated systems
- Analyze gaps or areas of non-compliance about the key concepts and requirements outlined in regulations and standards and examine their applicability, implications and improvments for scenario-based testing.
- Conduct extensive validation of the proposed scenario-based testing framework through different real-world and near-realistic transportation applications (various types, varying levels of automation, and diverse real-world scenarios) to evaluate the practical applicability and benefits of the methodology.
The findings and recommendations from this research will ultimately guide AV manufacturers, regulators, and stakeholders in developing and validating ASs that comply with the regulatory framework, fostering the safe and responsible deployment of automated systems in the future.