Scenario-Based Simulation Testing of Autonomous Vehicle using Driver-in-the-loop Simulation: Verification and Validation Framework

The deployment of autonomous vehicle has become one of the major interests in most of developing countries such as Malaysia to reduce the road accidents and minimize traffic congestions. This has increased the need for conducting safety testing for autonomous vehicle to ensure the road safety before actual deployment on mixed traffic roads. The main challenges for deployment of autonomous vehicles in developing countries is due to road and traffic conditions which is heavily involves with pedestrians, motorist and commercial vehicles. Moreover, a standardized safety testing process coupled with driving scenario from developing countries is required to establish the verification and validation framework. Therefore, scenario-based testing has been adopted in this study as the safety testing procedure for autonomous vehicle based on ISO standards. The safety assessments are designed based on Operational Design Domain (ODD) referring to Malaysian traffic conditions. Besides, Malaysian Road Scenario Database (MaRSeD) has been developed in this study to conduct the scenario-based safety testing. MaRSeD is used for identification and classification of the scenarios based on the ISO Standards for developing countries. This is mainly because no accurate dataset has been gathered and critically analyzed to conduct the safety assessment. Moreover, a motion platform called AMoDS has been developed in this study to conduct the driver-in-the-loop simulation testing for the validation process.