Automated driving is nowadays considered as a solution to solve traffic congestion and increase driving comfort. However, it will be commercialized only until its functionality and safety are validated. Due to the complexity of traffic situations billons of stochastic test kilometres are required for the validation, which makes the validation in real world expensive. Simulation-based test is a way to reduce the cost. Traffic flow models can generate the various test scenarios required in simulation-based test. However, the traditional modelling methods are normally aimed at realistic representation of a real traffic flow and can hardly be proven that the generated traffic flow can have a high coverage over the corner cases required by automated driving validation. This paper introduces a method for microscopic modelling of stochastic traffic flow for automated driving validation. Its quick coverage over corner cases for automated driving is proven.
Stochastic microscopic modelling of traffic flow for automated driving validation
4th International Conference on Internet of Things and Smart City (IoTSC 2024) ; 2024 ; Hangzhou, China
Proc. SPIE ; 13224
2024-08-07
Conference paper
Electronic Resource
English
Calibration and Validation of Microscopic Traffic Flow Models
German Aerospace Center (DLR) | 2004
|Calibration and Validation of Microscopic Traffic Flow Models
Transportation Research Record | 2004
|Calibration and Validation of Microscopic Traffic Flow Models
British Library Conference Proceedings | 2005
|Calibration and Validation of Microscopic Traffic Flow Models
German Aerospace Center (DLR) | 2003
|