Imagine trying to teach a self-driving car how to handle a chaotic human driver who might be distracted or unpredictable. Instead of risking real crashes, this project creates a highly realistic digital 'twin' of human drivers. It acts like a sophisticated flight simulator for cars, allowing developers to test safety in thousands of virtual scenarios before the car ever hits the road.
Autonomous vehicle developers lack a credible, standardized way to prove their systems are as safe as human drivers. Current simulations often fail to capture the unpredictable and diverse nature of real human driving behavior.
A modular simulation library of human driving behavior and an evaluation framework used across 5 target applications to establish safety metrics.
If you are an OEM dealing with the high cost of physical crash tests — this project developed a simulation library that provides a realistic human road safety baseline. This allows you to virtually assess CCAM systems against human behavior to speed up safety certification.
If you are an insurance provider dealing with uncertainty in autonomous vehicle risk profiles — this project developed a methodology to account for uncertainty in human behaviors. This helps in establishing more accurate risk baselines for AV-related claims.
If you are a regulator dealing with the lack of standards for AV licensing — this project developed building blocks that pave the way for a driving license for AVs. It provides a credible methodology to validate if an AI driver is as safe as a human.
Based on available project data, no specific pricing or cost structures are mentioned as this is a HORIZON-RIA research project.
Yes, the project aims to lay the foundation for a new industry-standard methodology and includes 5 industrial partners, including OEMs and Tier 1 suppliers, to ensure industrial relevance.
Based on available project data, specific licensing terms are not provided, but the project involves a consortium of 17 partners across 10 countries.
It provides a modular simulation library and an evaluation framework that can be integrated with target applications to establish safety thresholds and metrics.
The project period runs from 2022-10-01 to 2026-03-31, suggesting the methodology will be finalized by early 2026.
The consortium is well-balanced for technology transfer, featuring a 29% industry ratio with 5 industrial partners (including OEMs and Tier 1s) and 2 SMEs. The presence of 8 universities and 3 research centers across 10 countries, including key hubs in the US, China, and Australia, indicates a strong global validation strategy using diverse driving simulators and field labs.
Contact PANTEIA BV in the Netherlands
Contact us to explore licensing opportunities for the i4Driving simulation library.
