BEhavioural ReplicaTion of Human drivers for CCAM. BERTHA.

The main objetive of this European project is to develop a Driver Behavioural Model (DMB) to make autonomous vehicles safer and more human-like.

This validated - and scientifically based - driver behavior model (DBM) will be able to cover the aspects of human driving performance which is one of the current gaps for the development of connected, cooperative and automated mobility (CCAM). The DBM will make possible to understand and test the interaction of the CCAM with other cars in a safer and more predictable way from a human perspective. DBM is the cornerstone for CCAM component development. It will enable the design and analysis of CCAM components, digitally and with a common language between TIER and OEM. It will ensure they digital validation and, if incorporated into ECU software, will generate a more humane response from autonomous vehicles (at any level) and will increase their acceptance.

Thanks to this, Europe will be able to take advantage of the opportunities offered by connected, cooperative and automated mobility (CCAM).

Aims

BERTHA will develop a scalable probabilistic behavioural model that covers the full spectrum of drivers, covering physical, cognitive, and emotional domains, including personal, cultural, and contextual factors; that determine what, how, and why drivers react in real life and complex situations. This will allow more human-like vehicles and allow CAV to really understand other road users and anticipate unexpected risks. The model will be embedded on an open, transparent, and scalable simulation platform (HUB) for OEM and TIERs

to incorporate new uses cases according to driver diversity and situational and environmental aspects that are almost impossible with FOT (Field Operational Test). This will speed up their developments, reduce costs, and increase user acceptance, enhancing society acceptance.

The Specific Objectives of BERTHA project are:


  • Development of an affective module for drivers

  • Motor control for wheel drive and pedal activation influenced by the affective state of the driver

  • Development of two computational modules aiming to simulate risk assessment and decision making, as performed by real human drivers

  • Development of a perceptual module

  • Methodology for DBM scalable development

  • A scalable integrated probabilistic model of the human driver

  • AI-driven traffic scenario generator

  • Evaluate the interaction of the human-driver behavioural aspects on the performance of Cooperative, Connected & Automated Mobility (CCAM) Systems

  • Identification of a Typology of Driver Reaction Patterns

  • Enabling an Open multicollaborative space for the development and testing of an ADAS autopilot, considering the current legal and business constraints based on a human centred approach and lean cycle of develop tests

  • Community-focused approach (tooling and SiL-infrastructure) for collaborative HUB-assisted cross-stakeholder sharing, testing, and evaluation of trafficparticipant model, ADAS test cases (e.g., cities, traffic), Artificial Intelligence (AI) models, and formatted data

  • To train and test autonomous driving solutions under more realistic behaviour simulations


 

berthaproject.eu

Collaborating companies

CONSORTIUM


-  Instituto de Biomecánica (IBV)


-  Institut VEDECOM


- University Gustave Eiffel


- German Research Centre for Artificial Intelligence


- Computer Vision Center


- Capgemini Engineering


- VORTEX-Colab


- Continental Engineering


- Fundación CIDAUT


- Austrian Institute of Technology


- Universitat de València


- Europcar Mobility Group


-  FI Group

Duration and reference no.

Start date: 01-11-2023
End date: 31-10-2026
Project reference no.: Project number: 101076360

Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Climate, Infrastructure and Environment Executive Agency (CINEA). Neither the European Union nor the granting authority can be held responsible for them.