Skip to main content

Piloting Automated Driving on European Roads

Periodic Reporting for period 2 - L3Pilot (Piloting Automated Driving on European Roads)

Reporting period: 2019-03-01 to 2020-08-31

Recent work indicates how driver assistance systems and automated driving functions (ADFs) can be best validated by means of extensive road tests, with a sufficiently long operation time, to allow extensive interaction with the driver and testable functions. The project will use large-scale testing and piloting of AD with developed SAE Level 3 functions exposed to different users, mixed traffic environments, including conventional vehicles and vulnerable road users (VRUs), along different road networks.

The data collected in these extensive pilots will support the main aims of the project to:

• Lay the foundation for the design of future, user-accepted L3 and L4 systems, to ensure their commercial success. This will be achieved by assessing user reactions, experiences and preferences of the AD systems’ functionalities.

• Enable non-automotive stakeholders, such as authorities and certification bodies, to prepare measures that will support the uptake of AD, including updated regulations for the certification of vehicle functions with a higher degree of automation, as well as incentives for the user.

• Create unified de-facto standardised methods to ensure further development of AD applications (Code of Practice).

• Create a large databank to enable simulation studies of the performance of AD over time which can not be investigated in road tests, due to the time and effort needed. The data will be one product of the pilots.

Pan-European tests are needed in order to collect data from different European conditions and generalise the results to most typical driving conditions and traffic situations in urban and motorway traffic. The results also provide information on how to develop automated systems, infrastructure enhancement needs for automation, expected impacts of automation on the transport system, how the general public feels about automated driving after having hands-on experience on it. Overall, the aim is to provide with new and usable information on automated systems’ potential for the industry, decision makers and users.
After the piloting methodology was created, the work has been focussing on consolidating the methodology. This has consisted of meetings and discussions with the test sites about carrying out pilots and observing the key principles of test design, especially the importance of having both baseline and treatment measurements in place across all the test sites. In parallel, test subjects were recruited, their training was planned and executed. 14 vehicle owners conducted pre-tests to assess the methodology and use feedback from the tests to make the needed enhancements. In this period, statistical data analysis was performed at a preliminary level by using pre-test data with the main purpose to validate the approach. Several aspects were addressed to improve the data processing schemes, ranging from the detection of driving scenarios to the use of the consolidated data base for objective and subjective data. Procedures for the data access and for granting confidentiality were implemented. From the obtained results, the performance of the whole tool chain was considered well in line with the planned objectives. The pilots are running and  delivering data from both the baseline and treatment conditions. Further processing allows to obtain derived measures, performance indicators and upload aggregated data in the consolidated data base.

Advanced evaluation methods are now in place for the different concepts under study, in particular technical operation, traffic, user acceptance as well as impacts on mobility, safety and the environment.
L3Pilot features the first large-scale holistic project on SAE L3 automated driving. The data acquisition is well under way and despite the Covid-19, the expected negative impacts are addressed with manageable delays in piloting.

Consequently, we expect the project providing with new insights and results in several areas such as:

•   Technical functionality of automated driving systems on the SAE L3 level,

•   Automated vehicle-driver interaction,

•   Impacts on safety, efficiency and traffic flow,

•   First data-led guesses on socio-economic impacts,

•   Users’ attitudes and preferences on automated driving both from the test drivers and also globally through extensive surveys,

•   Creation of the main business scenarios for automated driving based on the European experts’ brain storming at dedicated workshops.

Furthermore, the on-going work on a Code of Practise will provide the automated driving community with tools and methods for developing automated systems. Methodology-wise, L3Pilot offers a tool-kit to prepare future road tests without starting from a clean sheet of paper and thus saving time and money in executing road tests. This includes also data management tools and procedures also taking into account the proprietary needs of vehicles owners and releasing public data for wider scientific use.
Figure: project structure and application