Secure collaboration between multi-digital twins in electric vehicles

On 27 October 2026, I presented our work on Security-by-Design for collaborative digital twins at ICSOB 2026 in Gothenburg. I co-authored the paper with my RISE colleagues Efi Papatheocharous, Rodi Jolak and Pontus Svenson.

Imagine an electric vehicle preparing for its next journey. Deciding when and how to charge it involves several questions: How healthy is the battery? How much energy will the journey require? How does temperature affect the vehicle’s performance?

Digital twins can support these decisions. A digital twin is a digital representation connected to its real-world counterpart through data and interaction during operation. It can help us understand the current state, explore possible changes and predict how something such as a battery will behave.

In vehicles where more and more functions are controlled and updated through software, this offers opportunities to investigate how changes may affect the vehicle throughout its lifetime. In electric vehicles, the relationships between battery condition, thermal management and charging are particularly important.

One vehicle needs several perspectives

In our example, twins and models inside and outside the vehicle contribute different kinds of information. Measurements from the vehicle help track what is happening in the moment. Models outside the vehicle combine these measurements with historical battery data, previous charging sessions, route information and weather conditions.

A twin working in isolation may lack the information needed to understand how its recommendations affect other parts of the vehicle. A charging decision, for example, needs to account for both the next journey and the battery’s condition.

By collaborating, twins can support coordinated decisions. The aim is to improve the vehicle’s overall performance while accounting for how its different parts affect one another.

Collaboration requires clear security responsibilities

The information needed may come from several organisations. Vehicle manufacturers, battery suppliers, charging operators and software providers may control different models, services and data.

This raises a central question in our research at RISE: How can we trust and govern decisions that depend on digital twins controlled by different organisations?

Suppose a model update has been tampered with and produces misleading battery predictions. If another service uses those predictions, the effects may reach charging or energy-use decisions. How far the effects spread depends on the connections, permissions and safeguards in place.

Protecting each twin is therefore part of the task. We also need to determine who verifies shared information, who can reject an unreliable result, who monitors the interactions and who has the authority to intervene.

Security-by-Design means addressing these questions when the collaboration is designed and maintaining protection as the system changes.

Four questions guide our research

In our research-in-progress paper, we identify four gaps in the research areas we examined:

• Which parts are connected? We need a clear description of the twins, services, physical components and organisations involved, including who controls them.
• Who is responsible for cybersecurity? Security requirements, monitoring and incident-response responsibilities need to be allocated across organisational boundaries.
• How can an attack affect other parts of the system? We need to analyse how a compromised model or service can affect other participants and, ultimately, the vehicle’s operation.
• How do we assess the system as a whole? Existing frameworks offer useful perspectives, but these need to be combined to analyse collaborative digital twins.

To structure these questions, we outline a preliminary Multi-DT Security Analysis Stack. It organises security analysis into six layers covering infrastructure, data, models, software development and updates, services, and organisational governance.

This structure is an initial research contribution whose practical value still needs to be evaluated. The next step is to develop a repeatable analysis method and investigate it in further vehicle scenarios.

For organisations developing such systems, a practical starting point is to trace a decision through the twins and services that contribute to it, making security responsibilities explicit at every connection.

About the research

This blog post is based on the work Security-by-Design for Collaborative Multi-Digital-Twin Systems-of-Systems in Software-Defined Electric Vehicles: A Research-in-Progress Problem Landscape, by Iraklis Symeonidis, Efi Papatheocharous, Rodi Jolak and Pontus Svenson at RISE.

The work was presented at ICSOB 2026 in Gothenburg on 27 October 2026. The paper has not yet been published. A link to the publication will be added here when it becomes available.

The research forms part of TWIN-LOOP, funded by the European Union’s Horizon Europe research and innovation programme under grant agreement No. 101192649.

Do you work with digital twins, vehicle software or cybersecurity? I would welcome your experiences: Where is security responsibility hardest to define when systems collaborate?