Common approaches for the safety validation of CCAM systems (CCAM Partnership)

A decisive factor for the successful implementation of innovative CCAM technologies and for their acceptance and adoption in society will be assuring the effective safety of CCAM systems.

While different assessment methods for automated driving functions have been developed, common standard methodologies meeting all the requirements for testing, validation and certification of all levels and use cases of automated driving do not yet exist. Therefore, consensus building between all stakeholders is urgently needed to establish common, validated methodologies and tools. Existing approaches are currently analysed in Horizon 2020, an experts’ network has been set up and validation concepts are demonstrated for selected use cases e.g. in the HEADSTART project[[https://www.headstart-project.eu/]].

Proposed actions should move the development of common verification and validation methodologies to a new level by widening substantially the scope of use cases addressed and by preparing the required tools to enable the comprehensive safety verification and validation of CCAM systems. This should take into account mixed traffic situations and include functional safety issues and cybersecurity. Such methodologies and tools should allow for their further development and adaptation with future technological evolution. Scenario-based approaches combining virtual and physical testing are needed, as conventional verification and validation approaches would require hundreds of millions of test kilometres for higher levels of CCAM.

Proposed actions are expected to develop a commonly accepted and harmonised simulation environment with standardised, open interfaces and quality controlled data exchange to enable the virtual testing of CCAM functions and systems in a multitude of relevant test cases and to enable the efficient and seamless use of validated models from different sources.

The validation of CCAM systems depends on the definition of relevant safety-critical scenarios and test cases. Several national and European projects have started to collect such scenarios and store them in databases. There is, however, no European database of relevant scenarios nor an agreed database structure. Scenario descriptions also need to be harmonised and not all relevant scenarios are known. Therefore, proposals need to define and develop processes and tools to continuously identify relevant events and convert them into detailed scenarios from various sources (including accidents), complemented by an ontology-based tool to define relevant future/theoretical scenarios. Diverse weather, lighting and road conditions, a broad spectrum of behaviour of other road users as well as edge cases should be considered. Following the collection of such scenarios, which can partly be derived from other projects and collaborations, scenarios need to be shared and centrally stored in a European database. This database should be established by the proposed actions based on an agreed structure and a harmonised scenario description (ontology layer) and metadata framework, in line with the FAIR data principles.

Proposed actions are expected to develop recommendations for harmonisation, standardisation and homologation including the conceptual description of an approval scheme for CCAM systems considering all types of vehicles and fed into on-going discussions regarding EU type vehicle approval rules as well as in the framework of the UNECE.

This topic requires the effective contribution of SSH disciplines and the involvement of SSH experts, institutions, as well as the inclusion of relevant SSH expertise, in order to produce meaningful and significant effects enhancing the societal impact of the related research activities.

In order to achieve the expected outcomes, international cooperation is advised, in particular with projects or partners from the US, Japan, Canada, South Korea, Singapore, Australia.

This topic implements the co-programmed European Partnership on ‘Connected, Cooperative and Automated Mobility’ (CCAM).