Preliminary studies on criticalities and opportunities for virtual testing of driving automation

摘要

This paper presents some preliminary analyses on the development of new tools for the transport industry, able to deal with the introduction of increasing levels of vehicle automation. Driving assistance is aimed at increasing road safety, but it needs a renewed approach to the research and development process. Indeed, vehicle automation spans several scientific disciplines and it's becoming exceedingly difficult and too costly for a single research innovation team to "go deep" across all technologies and solutions. The only way to ensure an easy, fast, efficient, and scalable introduction of the required innovation is to adopt integrated and complex testing platform for the simulation of automation solutions. To this aim virtual-testing platforms should be conceived to allow different actors to work on different components, possibly at different levels of detail, any of the actors being allowed to sophisticate with a particular simulation issue (e.g. the driver behaviour in presence of Advanced Driving Assistance Systems) knowing that the other components (e.g. the vehicle dynamics) have been (or will be) simulated at the required sophistication level, possibly by another actor of the innovation process. In this paper the authors wishes to contribute to the development of these new-generation tools. Analyses will be carried out in order to identify the key opportunities and criticalities in the development of virtual testing platform for testing driving automation. From these analyses research perspectives will be identified and proposed for future developments.