DETROID
Dependable Software-Defined Vehicles with Resilient Powertrain and Chassis Dynamics
A four-year staff exchange bringing together fifteen universities, research institutes and companies across ten countries to make the chassis and powertrain of software-defined vehicles safer, more efficient and more resilient — and to train the people who will build them.
DETROID
Dependable by design, resilient on the road.
Why software-defined vehicles need dependable chassis and powertrain design
Software-defined vehicles (SDVs) are widely seen as the third pillar of future road mobility, alongside electrification and automated driving. In an SDV, a car's core functions — powertrain and chassis control, driver-assistance systems, infotainment — are run and continuously updated in software. Hardware and software are decoupled, so features can be added, and performance tuned, throughout the vehicle's life.
Seven objectives spanning research & innovation and training & networking, delivered across five work packages.
Interdisciplinary knowledge
An SDV sits between automotive engineering, systems design and IT. Early design choices must balance hardware constraints — real-time performance, thermal limits — against how flexible the software can be.
Multi-level safety
SDVs must satisfy functional safety (ISO 26262), the intended-functionality standard SOTIF (ISO 21448) and cybersecurity (ISO/SAE 21434), all at once — a higher bar than earlier vehicle generations.
Complex verification & validation
Regular over-the-air updates and AI-based decision-making mean testing — simulation, hardware-in-the-loop, proving-ground trials — needs a new approach that keeps validating a vehicle throughout its life.
Research & innovation objectives
Identify and prioritise the technical challenges of building chassis and powertrain systems into scalable SDV architectures — hardware/software decoupling, real-time constraints, cybersecurity.
Design dependable, trustworthy SDV systems that meet the safety and cybersecurity standards outlined above, across the vehicle's life.
Develop AI- and OTA-enabled vehicle-dynamics control that improves comfort, energy efficiency and resilience, including eco-driving and wheel-corner architectures.
Build real-time adaptive control algorithms that respond to driving conditions and road previews using data-driven methods.
Evaluate and validate project outcomes through lab and simulated prototypes, assessing dependability, comfort, performance and lifecycle sustainability.
Training & networking objectives
Promote interdisciplinary knowledge exchange through secondments, workshops and joint research, including open-science practice.
Strengthen participants' professional skills through international, intersectoral secondments that support career progression and new collaborations.
Four phases, from concept to validated prototype
DETROID's methodology follows one development chain across the project's five work packages.
Design & conceptualisation
Define a baseline SDV architecture and its functional, safety and security requirements, using model-based design and interdisciplinary workshops.
Development & algorithms
Develop and benchmark control algorithms in simulation, sharing models across the consortium and using the EMR formalism to scale results across vehicle types.
Integration & prototyping
Build prototypes of at least two SDV topologies — a conventional single-motor layout and a wheel-corner EV with in-wheel motors — using DevOps and containerised development.
Verification & iteration
Validate designs in software, on hardware-in-the-loop rigs and on real test vehicles across the consortium, then iterate on scalability and replicability.
Five work packages
| WP | Title | Lead | Months |
|---|---|---|---|
| WP1 | Project management | TU Ilmenau | 1–48 |
| WP2 | Design methods for dependable SDV systems | THI | 1–24 |
| WP3 | Advanced control systems and algorithms within the SDV framework | POLITO | 10–42 |
| WP4 | Prototyping and evaluation | ELAPHE | 22–48 |
| WP5 | Dissemination, exploitation and communication | VUB | 1–48 |
Secondments & networking
Staff exchanges are DETROID's core mechanism. Researchers and engineers move to a partner institution, work on the project there, and bring what they've learned back home.
person-months of secondments planned in total
person-months of interdisciplinary exchange within Europe
Secondments run between DETROID's academic and industrial partners across Europe, Japan, Australia and Vietnam, following the training needs set out for each work package — from EMR-based modelling and hardware-in-the-loop testing to safety-critical controller design and on-road data collection.
Beyond secondments
-
Collaborative workshops
Three planned so far: SDV architecture & safety-critical processes (TUIL, THI); data-driven & AI-based chassis/powertrain control (UNIBZ, HERON, ELAPHE); KPIs for dependable SDV systems (POLITO, VUB).
-
Thematic webinars
Two per year, open to remote participants, led by TU Ilmenau and VUB.
-
EMR International Summer School
An annual hybrid event on the Energetic Macroscopic Representation formalism, with a dedicated DETROID session in later project years.
-
Shared online platform
A collaboration space for papers, vehicle models, datasets and control-algorithm code across the consortium.
Fifteen partners, ten countries
Eleven research organisations and four industrial companies, coordinated from Germany.
Technische Universität Ilmenau (TUIL) — Germany
Coordinates the project and leads on overall SDV architecture, verification & validation methodology, and fail-safe, redundant system design.
Beneficiaries
Elaphe Pogonske Tehnologije (ELAPHE)
In-wheel motor specialist contributing powertrain hardware, test vehicles and prototype integration.
SMEHeron Sports (HERON)
Chassis-systems developer working on cloud-connected, over-the-air control functions.
SMEPolitecnico di Torino (POLITO)
Leads work on individualised motion-control algorithms, validated on its PIXKIT platform.
Technische Hochschule Ingolstadt (THI)
Brings its CARISSMA vehicle-safety centre and ANTON test platform to dependable-systems design.
Libera Università di Bolzano (UNIBZ)
Designs and validates low-level, safety-critical controllers on robotic ground vehicles.
Université de Lille (ULille)
Develops digital twins and energy-management strategies with its Energetic Macroscopic Representation (EMR) method.
Vrije Universiteit Brussel (VUB)
Works on eco-driving and eco-charging strategies, and leads dissemination, exploitation and communication.
Cranfield University (CU)
Contributes real-time model-predictive control expertise for ride comfort and vehicle dynamics.
Università Politecnica delle Marche (UNIVPM)
Adds nonlinear optimal control, robotics and multi-sensor fusion expertise on scaled autonomous platforms.
Partner organisations
Meccanica 42
Supplies by-wire hardware and rapid-prototyping tools for chassis-control testing.
SMERimac Technology
High-performance EV powertrain and ECU specialist supporting real-world validation.
University of Tokyo
Trains partners in motion control for electric-vehicle mechatronic systems using its own test vehicles.
RMIT University
Hosts researchers at its Autonomous Systems Lab for control-algorithm development and validation.
Hanoi University of Science & Technology (HUST)
Runs on-road experiments and traffic-data collection under Vietnamese driving conditions.
Scientific, economic and societal
Scientific
- Model-based, AI-enhanced SDV design methods and a physically-grounded way to organise digital twins, control algorithms and hardware-in-the-loop testing.
- New verification & validation approaches combining remote testbeds, AI decision-making and hardware replication.
Measured by joint publications and citations within three years.
Economic & technological
- Faster, more dependable development chains for SDV components and plug-in-compliant testing tools.
- Scalable architectures — including wheel-corner, in-wheel-motor layouts — with stronger fail-safe redundancy.
Measured by energy-efficiency gains and shorter design/validation cycles.
Societal
- Fewer accidents through safety-critical, fail-operational control and predictive emergency handling.
- Lower energy use via eco-driving and eco-charging, and a new generation trained in dependable SDV design.
Measured by safety/performance KPIs and researchers trained.
News & events
DETROID officially starts on 1 October 2026. News, workshop announcements and event dates will be posted here as the project gets under way.
Publications & results
All DETROID papers will be freely available under the gold Open Access model, with data and code shared via Zenodo. Publications will be listed here as they are released.
Contact
Project contact
Project email — to be added
LinkedIn — coming soon