SUaaVE · Project

AI System That Reads Passenger Emotions to Make Self-Driving Cars Feel Safe

transportTestedTRL 5

Imagine getting into a self-driving car and it actually notices you're nervous — then adjusts the lighting, seat position, and driving style to calm you down. That's what SUaaVE built: an AI called ALFRED that reads your emotional and cognitive state during autonomous rides and adapts the experience in real time. They tested this with over 4,000 real users and 100 experts using an immersive simulation platform, because the biggest barrier to self-driving cars isn't the technology — it's that people don't trust them yet.

By the numbers

4,000+

Users involved in participatory design and testing

100

Domain experts and specialists consulted

Consortium partners across the project

European countries represented in the consortium

Level 4+

Target automation level for the ALFRED system

The business problem

What needed solving

Self-driving cars are technically getting safer, but consumers still don't trust them. Surveys consistently show that most people feel uncomfortable riding in an autonomous vehicle, threatening billions in automotive R&D investment. The core issue isn't engineering — it's the emotional gap between what the car can do and what the passenger believes it can do.

The solution

What was built

SUaaVE built ALFRED, an AI system combining an EMpathY Unit (reads passenger emotional and cognitive states) with an Adaptive Interface (adjusts vehicle dynamics, ambient conditions, and seat comfort in real time). They also created the V-HCD platform — an immersive simulation environment for testing human reactions to autonomous driving before building physical prototypes.

Audience

Who needs this

Automotive OEMs developing Level 4+ autonomous vehiclesAutonomous shuttle and robotaxi operatorsVehicle interior and HMI (human-machine interface) suppliersMobility simulation and virtual testing companiesPublic transport authorities planning autonomous fleet deployment

Business applications

Who can put this to work

Automotive OEMs & Tier-1 Suppliers

enterprise

Target: Vehicle manufacturers developing Level 4+ autonomous vehicles

If you are an automotive OEM struggling with consumer hesitation toward autonomous vehicles — this project developed ALFRED, an AI system that detects passenger emotions and cognitive state, then adapts vehicle dynamics, ambient comfort, and interfaces in real time. Tested with over 4,000 users across 5 European countries, it directly addresses the acceptance gap that threatens your autonomous vehicle rollout.

Mobility Simulation & Testing

mid-size

Target: Companies providing virtual testing environments for autonomous vehicles

If you are a simulation company looking to expand into human-factors testing — SUaaVE created the V-HCD platform, an immersive virtual environment specifically designed to test how humans react to autonomous driving scenarios. This platform lets you assess user acceptance before building physical prototypes, validated with 100 domain experts, cutting development costs significantly.

Public Transport & Ride-Hailing

any

Target: Autonomous shuttle operators and mobility-as-a-service providers

If you are a mobility service provider deploying autonomous shuttles and facing passenger dropout due to discomfort — SUaaVE's EMpathY Unit monitors passengers' emotional state and adapts the ride experience through comfort services. The system was designed with input from over 4,000 passengers, traditional drivers, and vulnerable road users to maximize acceptance across diverse user groups.

Frequently asked

Quick answers

What would it cost to license or integrate this emotion-sensing AI into our vehicles?

The project has not published specific licensing fees or integration costs. The coordinator (Instituto de Biomecánica de Valencia, an SME) and the 5 industry partners in the consortium are positioned to exploit ALFRED and the V-HCD platform commercially. Contact through SciTransfer for pricing discussions.

Can this scale to mass-production vehicles or is it lab-only?

ALFRED was demonstrated through simulation on the V-HCD platform, not yet integrated into production vehicles. However, the consortium includes 5 industry partners across automotive and technology sectors in 5 countries, suggesting a clear path toward integration. The system was validated with over 4,000 users, indicating readiness for larger-scale pilot programs.

Who owns the IP and can we license the technology?

The consortium of 13 partners across Germany, Spain, France, Italy, and the Netherlands holds the IP. The ALFRED algorithms and V-HCD software are marked as confidential, while the demonstrator and video demonstrations are public. Licensing would need to be negotiated with the consortium, led by the Spanish coordinator.

Does this meet automotive safety regulations?

SUaaVE produced guidelines specifically for public authorities on autonomous vehicle acceptance. The project addresses ethical principles in its design approach, including data ownership models where users control their personal data. Based on available project data, specific regulatory certifications (e.g., ISO 26262) are not mentioned.

How long before this could be deployed in real vehicles?

The project ended in October 2022 with a working demonstrator in simulation. Moving from simulated demonstrator to in-vehicle deployment would likely require 2-3 additional development cycles. The consortium's exploitation plans indicate intent to bring ALFRED and V-HCD to market.

Can this integrate with our existing vehicle control systems?

ALFRED is designed as a modular system with an EMpathY Unit for emotion detection and an Adaptive Cognitive and Emotional Interface for service delivery (vehicle dynamics, ambient comfort, postural comfort). Based on available project data, it was built for Level 4+ automated vehicles, so integration depends on your automation architecture.

Consortium

Who built it

The SUaaVE consortium brings together 13 partners from 5 countries (Germany, Spain, France, Italy, Netherlands), with a strong industry presence: 5 industry partners (38% ratio) alongside 4 universities and 4 research organizations. The coordinator, Instituto de Biomecánica de Valencia, is a Spanish SME specializing in biomechanics — an unusual but strategic choice that brings human-body expertise to the vehicle comfort problem. Having 2 SMEs in the consortium suggests agility in commercializing results, while the geographic spread across major European automotive markets (Germany, France, Italy) provides direct access to the continent's biggest car manufacturers and their supply chains.

INSTITUTO DE BIOMECANICA DE VALENCIACoordinator · ES
CENTRE AQUITAIN DES TECHNOLOGIES DEL'INFORMATION ET ELECTRONIQUES - CATIEthirdparty · FR
CENTRO RICERCHE FIAT SCPAparticipant · IT
FICOSA AUTOMOTIVE SLparticipant · ES
INSTITUT VEDECOMparticipant · FR
TECHNISCHE UNIVERSITAET MUENCHENparticipant · DE
KEYSIGHT TECHNOLOGIES FRANCE S.A.S.participant · FR
IDIADA AUTOMOTIVE TECHNOLOGY SAparticipant · ES
INSTITUT POLYTECHNIQUE DE BORDEAUXparticipant · FR
UNIVERSITE GUSTAVE EIFFELparticipant · FR
IDNEO TECHNOLOGIES SAUparticipant · ES
RIJKSUNIVERSITEIT GRONINGENparticipant · NL

How to reach the team

Instituto de Biomecánica de Valencia (Spain) — a biomechanics SME coordinating the consortium. Contact via SciTransfer for introductions.

Order a report on this consortium See INSTITUTO DE BIOMECANICA DE VALENCIA profile →

Next steps

Talk to the team behind this work.

Want to explore licensing ALFRED's emotion-sensing AI or the V-HCD simulation platform for your autonomous vehicle program? SciTransfer can arrange a direct introduction to the consortium.

Order a report on this topic More in Transport & Mobility