Pioneering Human-Like Driver Behaviour for Safer Automated Mobility

The BERTHA project has reached a significant milestone in its mission to advance Connected, Cooperative, and Automated Mobility (CCAM) across Europe. With the successful release and validation of its advanced driver behavioural modules, BERTHA is setting new standards for human-centric, data-driven mobility solutions.

What’s New? Advanced Modules for Realistic Driver Modelling

Perception Module:

Trained on over nine hours of driving data from 21 users, the Perception module delivers real-time, 360° attention mapping within the CARLA simulation environment. This breakthrough enables automated vehicles to better understand and anticipate human attention patterns, improving safety and reliability in complex traffic scenarios.

Affective Module:

Leveraging Bayesian Belief Networks, the Affective module predicts key mental states such as fatigue, drowsiness, and stress. By modelling these human factors, BERTHA’s approach ensures that automated driving systems can respond more naturally and safely to real-world conditions.

Motor Control Module:

Calibrated with data from 30 participants, the Motor Control module outputs probability distributions for realistic driver actions. This allows for nuanced, human-like responses in a variety of driving situations, bridging the gap between simulation and reality.

Open Science and Community Engagement

In line with BERTHA’s commitment to open science, these modules are now available on GitHub. This transparency not only accelerates innovation within the CCAM ecosystem but also invites collaboration from researchers, developers, and industry stakeholders across Europe and beyond.

Real-World Validation with Field Operational Test Data

Validation planning is well underway, with Field Operational Test (FOT) data set to play a central role. By grounding module performance in real-world driving conditions, BERTHA ensures that its solutions are robust, reliable, and ready for deployment in next-generation automated mobility systems.

Why This Matters for CCAM

The release and validation of these advanced driver behavioural modules position BERTHA as a leader in developing and sharing robust, human-like driver models. These achievements directly support the project’s vision of safer, more efficient, and more human-centric automated mobility for all.