Autonomous vehicle safety evaluation through a high-fidelity simulation approach
The autonomous vehicle (AV) industry aims to design strategic plans to ensure the safety of the developed systems before their mass deployment. Real-road testing is shown to be impractical for validating these systems as it requires many years if not decades of testing in different environmental con...
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Estonian Academy Publishers
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/2e4b7039ed7444a0b5b1724e68cd4f53 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | The autonomous vehicle (AV) industry aims to design strategic plans to ensure the safety of the developed systems before their mass deployment. Real-road testing is shown to be impractical for validating these systems as it requires many years if not decades of testing in different environmental conditions. For solving this issue, the method should be complemented with simulation. The primary goal of this research was to develop advanced techniques in the safety validation area by using end-to-end simulation technologies. In this study, we present a simulation approach for safety evaluation of an AV shuttle, iseAuto, currently operating at the Tallinn University of Technology campus. We created a virtual environment by using geospatial data from the specified path on the university campus that includes all relevant features. Then, we converted the map to a 3D format applicable for the SVL simulator. Also, we provided the AV 3D model to use in the simulation and equipped it with the SVL virtual sensors to provide data for the Autoware perception algorithms, which is the control software of the shuttle. To show the efficiency of the proposed method, we designed two overtaking scenarios and observed the AV behaviour under the test. Finally, we demonstrate how the system enables us to evaluate AVʼs decision-making performance and safety in different situations. |
|---|