SciTransfer
SimuSafe · Project

Driving Simulators That Predict and Prevent Road Crashes Before They Happen

transportTestedTRL 5

Imagine being able to recreate any dangerous traffic situation — a drunk driver, a distracted pedestrian, a cyclist in a blind spot — without anyone getting hurt. That's what this project built: realistic driving simulators for cars, motorcycles, and bicycles, plus virtual reality for pedestrians, all interacting in the same scene. They collected data from these simulations and compared it with real-world driving to figure out exactly what causes crashes. The result is a detailed map of human behavior on the road that can be used to design better safety systems, smarter regulations, and more effective driver training.

By the numbers
90%
of road-traffic crashes caused by driver error
EUR 7,991,600
EU funding invested in the research
21
consortium partners across 13 countries
10
industry partners in the consortium
6
SMEs involved in the project
15
total deliverables produced
The business problem

What needed solving

Road crashes kill and injure millions globally, and 90% are caused by human error — yet testing safety systems in genuinely dangerous conditions is impossible on real roads. Companies developing ADAS, insurance risk models, or driver training programs lack a controlled way to study how drivers, cyclists, and pedestrians actually behave in high-risk situations. Without realistic behavioral data, safety products are designed on assumptions rather than evidence.

The solution

What was built

The project built working simulator cockpits for cars, motorcycles, and bicycles, plus a VR system for pedestrians, all capable of running multi-agent scenarios simultaneously. These were backed by behavioral models validated against real-world naturalistic driving data, producing 15 deliverables over 4 years.

Audience

Who needs this

Automotive OEMs and Tier 1 suppliers developing ADAS and autonomous driving systemsMotor insurance companies building usage-based or behavioral risk modelsProfessional driver training and fleet safety companiesRoad infrastructure planners and traffic management authoritiesVR/simulation companies looking for validated transport behavior models
Business applications

Who can put this to work

Automotive safety systems
enterprise
Target: ADAS and autonomous driving technology companies

If you are an automotive safety company dealing with the challenge of testing advanced driver assistance systems in dangerous scenarios — this project developed multi-vehicle driving simulators with realistic behavioral models that let you test crash scenarios with cars, motorcycles, bicycles, and pedestrians all at once, without real-world risk. The simulator cockpits and behavioral data from 21 partners across 13 countries give you validated models closest to real-world driving data.

Insurance and risk assessment
enterprise
Target: Motor insurance companies and actuarial firms

If you are an insurance company trying to better understand why 90% of road crashes are caused by driver error — this project quantified the specific behavioral factors that cause crashes, near-collisions, and traffic violations. The data collected across multiple transport modes (cars, motorcycles, bicycles, pedestrians) can help you build more accurate risk models and develop usage-based insurance products tied to actual driver behavior patterns.

Driver training and education
any
Target: Professional driving schools and fleet training providers

If you are a driver training company looking to move beyond classroom theory and basic on-road practice — this project built simulator cockpits for cars, motorcycles, and bicycles plus VR for pedestrians that recreate scenarios impossible to practice safely in real life. The behavioral models identify which factors most often lead to dangerous situations, letting you design targeted training programs that address the actual causes of 90% of road crashes.

Frequently asked

Quick answers

What would it cost to license or access the simulator technology?

The project was funded with EUR 7,991,600 in EU contribution across 21 partners. Licensing terms would need to be negotiated directly with the coordinator (FUNDACION INSTITUTO TECNOLOGICO DE CASTILLA Y LEON) and relevant consortium members. As an RIA project, results are typically available for licensing.

Can this work at industrial scale for large vehicle fleets?

The project built simulator cockpits covering multiple vehicle types (cars, motorcycles, bicycles) and VR for pedestrians. With 10 industry partners already in the consortium, the technology was designed with scalability in mind. Adapting it for fleet-wide driver assessment would require customization beyond the research scope.

What is the IP situation — who owns what?

IP is shared among the 21-partner consortium across 13 countries according to Horizon 2020 rules. The coordinator in Spain would be the first point of contact for licensing discussions. With 10 industry partners and 6 SMEs involved, commercial exploitation pathways were built into the project structure.

How validated is the simulation data compared to real driving?

The project explicitly correlated simulation data with naturalistic driving tests to ensure the models are the closest possible to real-world conditions. This follows the FESTA-V methodology, which is an established European standard for field operational tests in transport research.

What vehicle types does the simulator support?

Based on the project data, simulators were built for cars, motorcycles, and bicycles, with Virtual Reality used for pedestrian scenarios. The system supports multiple monitored actors interacting in the same simulated scene simultaneously.

How long would integration into existing testing workflows take?

The project ran for 4 years (2017-2021) and produced 15 deliverables including demo simulator cockpits. Integration timeline would depend on your specific use case. Based on available project data, the behavioral models and simulator architecture are documented but would require technical adaptation for commercial deployment.

Consortium

Who built it

SimuSafe brought together 21 partners from 13 countries with a strong industry presence — 10 industry partners (48% of the consortium) including 6 SMEs. This near-equal split between industry (10) and academia/research (10) signals that the technology was developed with commercial application in mind, not just academic publication. The geographic spread across Europe, Israel, and the US suggests broad applicability and diverse driving culture inputs. The EUR 7,991,600 investment and 8 university partners provide a solid scientific foundation, while the industry partners ensured the simulator technology stayed grounded in real market needs.

How to reach the team

FUNDACION INSTITUTO TECNOLOGICO DE CASTILLA Y LEON (Spain) — search for SimuSafe project coordinator contact via their institutional page

Order a report on this consortiumSee FUNDACION INSTITUTO TECNOLOGICO DE CASTILLA Y LEON profile →
Next steps

Talk to the team behind this work.

Want to explore how SimuSafe's driving simulation and behavioral modeling technology could improve your safety testing, insurance risk models, or driver training programs? SciTransfer can arrange a direct introduction to the research team.

Order a report on this topicMore in Transport & Mobility
More in Transport & Mobility
See all Transport & Mobility projects →
Modular Rail Wagons That Cut Steel Industry Raw Material Transport Costs
ITECCO Demo
Open →
Dual-Line Road Marking Inspection Device That Cuts Survey Costs in Half
DuaLine
Open →
Ready-to-Buy Electric In-Wheel Motor That Cuts Weight 25% and Fits Any Vehicle
GEM in-wheel
Open →