PRYSTINE · Project

Fail-Safe Sensor Fusion Technology That Makes Autonomous Driving Work in Real Cities

transportPilotedTRL 6

Imagine a self-driving car that needs to "see" everything around it — pedestrians, cyclists, other cars — and never make a mistake, even if one of its sensors breaks. PRYSTINE built the brain behind that: a system that combines radar, LiDAR, and cameras so they back each other up, like having three pairs of eyes that double-check each other. They tested this on real vehicles, including a Maserati and heavy-duty trucks, proving it works not just in the lab but on actual roads. The key breakthrough is making the system keep working safely even when something goes wrong — which is the biggest hurdle before autonomous vehicles can legally hit city streets.

By the numbers

industrial demonstrators validated

project partners in the consortium

countries represented

industry partners

SME partners

63%

industry participation ratio

The business problem

What needed solving

Autonomous vehicles cannot operate safely in cities because current sensor systems have single points of failure — if one camera or radar unit malfunctions, the car cannot safely continue driving. This is the biggest regulatory and engineering barrier preventing Level 4+ autonomous vehicles from reaching European roads. Companies developing self-driving technology need sensor fusion systems that keep working even when individual components fail.

The solution

What was built

PRYSTINE built and demonstrated fail-operational sensor fusion systems combining radar, LiDAR, and cameras with AI-powered object recognition and decision-making. Concrete outputs include a Maserati prototype with full autonomous driving sensor integration, a heavy-duty truck fleet with complete E/E architecture, embedded perception software with real-time data fusion, and fail-operational control software — all validated across 22 industrial demonstrators.

Audience

Who needs this

Automotive OEMs developing Level 3-5 autonomous driving featuresTier-1 suppliers building ADAS sensor modules and control unitsTruck and commercial vehicle manufacturers automating freight transportSemiconductor companies designing automotive-grade AI chipsAutonomous shuttle and robotaxi operators needing safety-certified perception systems

Business applications

Who can put this to work

Automotive OEMs & Tier-1 Suppliers

enterprise

Target: Vehicle manufacturers or component suppliers developing ADAS and autonomous driving features

If you are an automotive OEM or Tier-1 supplier struggling to make your autonomous driving stack reliable enough for urban conditions — this project developed fail-operational sensor fusion architectures validated across 22 industrial demonstrators, including passenger cars and heavy-duty trucks. The embedded hardware/software reference designs let you integrate radar, LiDAR, and camera fusion with built-in redundancy, cutting your path to safety certification.

Commercial Vehicle & Logistics

enterprise

Target: Truck manufacturers or fleet operators investing in automated freight transport

If you are a truck manufacturer or logistics company looking to automate highway and urban freight — PRYSTINE equipped a heavy-duty vehicle and trailer fleet with full sensor arrays and fail-operational control software. The demonstrated E/E architecture communication across the fleet means you get a tested blueprint for deploying automated commercial vehicles with real redundancy, not just lab prototypes.

Semiconductor & Embedded Systems

any

Target: Chip makers or embedded computing companies serving the automotive sector

If you are a semiconductor or embedded systems company building components for the autonomous vehicle market — PRYSTINE, led by Infineon, created reference architectures that co-integrate AI and conventional signal processing into dependable embedded platforms. With 38 industry partners already using these designs, adopting the PRYSTINE architecture positions your components in a validated, industry-backed ecosystem of 60 partners across 14 countries.

Frequently asked

Quick answers

What would it cost to license or adopt this technology?

PRYSTINE was a publicly funded ECSEL project led by Infineon with 60 partners. Licensing terms would depend on individual consortium members who own specific IP. Contact the coordinator or relevant partner directly to discuss commercial terms for specific components or reference architectures.

Can this scale to production vehicles, not just prototypes?

The project validated its technology across 22 industrial demonstrators, including a Maserati passenger vehicle and a heavy-duty truck fleet with full sensor integration. The reference hardware/software architectures were designed for production E/E architectures, making the step from demonstrator to series production significantly shorter than starting from scratch.

Who owns the IP and how can I access it?

IP is distributed among 60 consortium partners across 14 countries, with Infineon Technologies AG as coordinator. Given the ECSEL funding structure, partners typically retain IP on their contributions. Specific licensing discussions should be directed to the partner holding the relevant component IP.

Does this meet automotive safety regulations?

PRYSTINE specifically targeted safety-compliant integration of AI for autonomous driving. The fail-operational design approach and dependable embedded architectures were built to address functional safety requirements. The 22 demonstrators provided validation evidence that would support safety certification processes.

How long would integration take for our existing vehicle platform?

The project delivered reference FUSION hardware/software architectures and embedded control software designed for integration into existing automotive E/E architectures. Based on available project data, the demonstrators achieved integration on real vehicles including production-class platforms, suggesting integration timelines measured in months rather than years for companies with existing ADAS infrastructure.

What sensors and hardware does this require?

The system uses radar, LiDAR, camera systems, and safety controllers for perception, combined with embedded computing boards for AI processing. Based on deliverable descriptions, the architecture supports multiple sensor configurations with fail-operational redundancy, so it can adapt to different hardware setups rather than requiring a single fixed sensor suite.

Is there ongoing support or follow-up development?

The project closed in October 2021. However, with 38 industry partners including major automotive OEMs and semiconductor companies, many partners continue developing commercial products based on PRYSTINE results. The project website at prystine.eu and individual partner companies are the best sources for current status of commercialization efforts.

Consortium

Who built it

PRYSTINE assembled one of the largest automotive R&D consortia in European research with 60 partners across 14 countries. What makes this particularly credible for business adoption is the 63% industry ratio — 38 of the partners are companies, not just universities. The consortium is led by Infineon Technologies, a global semiconductor leader, and includes 14 SMEs alongside major automotive OEMs and technology firms. With 12 universities and 9 research institutes providing the science, but industry clearly driving the application, this is a consortium built to produce commercially viable results, not just academic papers. The geographic spread across 14 countries also means the technology has been tested against diverse European road conditions and regulatory environments.

INFINEON TECHNOLOGIES AGCoordinator · DE
FICOSA ADAS, S.L.participant · ES
AVL ARASTIRMA VE MUHENDISLIK SANAYI VE TICARET LIMITED SIRKETIparticipant · TR
SCANIA CV ABparticipant · SE
DATIK INFORMACION INTELIGENTE S.L.thirdparty · ES
FRIEDRICH-ALEXANDER-UNIVERSITAET ERLANGEN-NUERNBERGparticipant · DE
MURATA ELECTRONICS OYparticipant · FI
CISC SEMICONDUCTOR GMBHparticipant · AT
TECHNISCHE UNIVERSITAET GRAZparticipant · AT
OKMETIC OYparticipant · FI
UAB METIS BALTICparticipant · LT
INFINEON TECHNOLOGIES AUSTRIA AGparticipant · AT
KUNGLIGA TEKNISKA HOEGSKOLANparticipant · SE
IDEAS & MOTION SRLparticipant · IT
ROBERT BOSCH GMBHparticipant · DE
ANYWI TECHNOLOGY BVparticipant · NL
CENTRO RICERCHE FIAT SCPAparticipant · IT
NEDERLANDSE ORGANISATIE VOOR TOEGEPAST NATUURWETENSCHAPPELIJK ONDERZOEK TNOparticipant · NL
TEKNOLOGIAN TUTKIMUSKESKUS VTT OYparticipant · FI
FUNDACION TECNALIA RESEARCH & INNOVATIONparticipant · ES
INTERUNIVERSITAIR MICRO-ELECTRONICA CENTRUMparticipant · BE
MASERATI SPAparticipant · IT
HOCHSCHULE OFFENBURGparticipant · DE
AITEK SPAparticipant · IT
RE:LAB SRLparticipant · IT
INNOLUCE BVparticipant · NL
AGENCIA ESTATAL CONSEJO SUPERIOR DE INVESTIGACIONES CIENTIFICASparticipant · ES
G.N.T. SYSTIMATA PLIROFORIKIS AEparticipant · EL
OSTBAYERISCHE TECHNISCHE HOCHSCHULEAMBERG-WEIDENparticipant · DE
IDIADA AUTOMOTIVE TECHNOLOGY SAparticipant · ES
VIRTUAL VEHICLE RESEARCH GMBHparticipant · AT
TTS KEHITYS OYparticipant · FI
TAMPEREEN KORKEAKOULUSAATIO SRparticipant · FI
NXP SEMICONDUCTORS NETHERLANDS BVparticipant · NL
POLITECNICO DI TORINOparticipant · IT
ROVIMATICA SLparticipant · ES
NOKIA SOLUTIONS AND NETWORKS OYparticipant · FI
AVL LIST GMBHparticipant · AT
UNIVERSITATEA POLITEHNICA DIN BUCURESTIparticipant · RO
TTTECH AUTO GMBHparticipant · AT
TTTECH COMPUTERTECHNIK AGparticipant · AT
INSTITUTO TECNOLOGICO DE INFORMATICAparticipant · ES
NOORD-BRABANT PROVINCIEparticipant · NL
VIDEANTIS GMBHparticipant · DE
ELEKTRONIKAS UN DATORZINATNU INSTITUTSparticipant · LV
FORD OTOMOTIV SANAYI ANONIM SIRKETIparticipant · TR
MATTERSOFT OYparticipant · FI
TECHNISCHE UNIVERSITEIT EINDHOVENparticipant · NL
TENNECO AUTOMOTIVE EUROPE BVBAparticipant · BE
UNIVERSITA DEGLI STUDI DI MODENA E REGGIO EMILIAparticipant · IT
TURUN YLIOPISTOparticipant · FI
TECHNISCHE UNIVERSITEIT DELFTparticipant · NL
BAYERISCHE MOTOREN WERKE AKTIENGESELLSCHAFTparticipant · DE
IRIZAR S COOPparticipant · ES

How to reach the team

Infineon Technologies AG (Germany) coordinated this 60-partner consortium. SciTransfer can facilitate a direct introduction to the right technical contact.

Order a report on this consortium See INFINEON TECHNOLOGIES AG profile →

Next steps

Talk to the team behind this work.

Want to explore how PRYSTINE's fail-operational sensor fusion technology fits your autonomous vehicle roadmap? SciTransfer provides one-page technology briefs and arranges direct introductions to the consortium partners who built what you need.

Order a report on this topic More in Transport & Mobility