SYNTHESES | Data-driven digital twins for safer urban mobility

Digital twins for proactive urban mobility safety

Urban mobility systems are becoming increasingly complex, with growing pressure on infrastructure, safety, and sustainability. Cities need better tools to understand how infrastructure is used in practice and how risks emerge. SYNTHESES develops data-driven digital twins combining multi-modal data and AI to support safer, more efficient and more proactive urban mobility management.
 

Target group

The project is relevant for:

• Cities and public authorities responsible for mobility planning
• Mobility solution providers (traffic management, smart mobility platforms)
• Infrastructure operators and engineering consultancies

Context

Urban mobility systems continuously evolve due to changes in infrastructure, mobility policies and user behaviour. However, current monitoring approaches remain largely reactive and fragmented, limiting the ability to assess infrastructure performance and identify safety risks early.

Critical issues such as poor visibility, infrastructure degradation and unsafe interactions between road users often remain unnoticed until incidents occur. This impacts safety, traffic efficiency and quality of life, while increasing operational and societal costs.

There is a clear need for scalable and data-driven approaches that provide continuous insight into infrastructure usage, mobility flows and emerging safety risks to support more informed urban mobility decisions.
 

Objectives & results

SYNTHESES aims to develop data-driven digital twins of urban mobility systems to improve safety, efficiency and infrastructure management.

The project will:

• Develop methods for multi-modal data fusion combining camera, LiDAR, satellite and floating car data
• Build high-fidelity digital representations of urban infrastructure and mobility flows
• Detect and analyse safety-critical situations and risky interactions between road users
• Develop scenario simulation and impact analysis tools
• Support proactive infrastructure management and mobility optimisation

Key expected results:

• Digital twin demonstrators supporting simulation and mobility planning
• Safety scoring methods to identify and prioritise high-risk infrastructure zones
• Scalable data fusion pipelines integrating heterogeneous mobility data
• AI models for scene understanding and risk detection
• Decision-support tools translating mobility insights into actionable recommendations for cities
   

Approach

The project is structured around four main workstreams.

1. Multi-modal data acquisition and fusion

• Integration of heterogeneous data sources including cameras, LiDAR, satellite imagery and floating car data
• Development of scalable data pipelines and fusion strategies

2. Scene understanding and digital twin generation

• AI-based detection and classification of infrastructure elements and road users
• Creation of high-fidelity digital twins of urban mobility environments

3. Risk detection and scenario analysis

• Identification of safety-critical situations and risky interactions
• Development of safety scoring and impact assessment methodologies
• Simulation of infrastructure changes and mobility scenarios

4. Decision support and real-world validation

• Translation of results into actionable insights for mobility planning
• Validation through real-world use cases with cities and project partners
• Co-creation with stakeholders to ensure operational relevance and applicability 
   

Funding

• Funding agencies: Innoviris and Vlaio

With the support of