L3Pilot – Analysing driving patterns of road users for digital "prediction models"

In the EU-funded L3Pilot project, DLR is investigating how automated driving can be implemented as a safe and efficient transport solution in the future. The focus here is on the presentation of systems of the so-called L(evel)3 automation functions, whereby some L4 automation functions are also being tested in real traffic.

In general, a distinction is made between five levels of automation. Level 0 stands for "no automation", i.e. no form of automation at all. Level 4 stands for "full self-driving automation". The driver can completely relinquish control of the vehicle and becomes a passenger. Level 3, which is the focus of the L3Pilot project, is highly automated driving. Vehicles with this level of automation can perform certain driving tasks independently and without human intervention, but only for a limited period of time and under suitable conditions specified by the manufacturer.

The scope of the situations tested in L3Pilot ranges from parking and overtaking to driving at urban junctions. With around 1,000 test drivers and 100 vehicles in 11 European countries, data is being collected to evaluate technical aspects, user acceptance, driving and travel behaviour and the impact of automation functionalities on traffic and society.

Aims of the project:

Create the basis for future, user-accepted L3 and L4 automation functions.
Develop measures that can be used by authorities and certification bodies to authorise automated vehicles.
Develop a generally applicable code of practice for the safe design of driver assistance systems for automated and connected driving systems.
Create a database with real driving data in order to be able to test situations in the simulation that cannot be analysed in real traffic due to time and costs, for example.

As part of the project, DLR is supporting the testing of L3 and L4 functions in urban areas. With the help of mobile sensor systems from DLR's Application Platform for Intelligent Mobility (AIM), data is collected at an urban roundabout and at an urban intersection. The aim is to use the collected trajectory and video data to investigate the manoeuvres of the automated vehicle and the influence of automation on the interaction with other motorised and non-motorised road users and to derive the impact on road safety.