As mobility technologies evolve, ensuring safety in mixed traffic environments, where human-driven and autonomous vehicles coexist, becomes increasingly critical. The FlexCrash project aims to address this challenge by developing innovative tools to better understand how drivers and autonomous systems interact in potentially risky situations. One of its most promising outputs is the FlexCrash online simulation platform.
Simulating real-world traffic risks in a virtual environment
The FlexCrash platform, developed by the IMC University of Applied Sciences Krems, enables researchers, engineers, and policy stakeholders to simulate complex traffic scenarios involving both human-driven and autonomous vehicles. By offering a user-friendly online interface, the platform allows users to explore how different driver behaviours, vehicle types, and decision-making algorithms affect crash risks and road safety outcomes.
How the FlexCrash platform works
The recently released demo video gives an overview of how users can interact with the platform. Here is how it works:
Scenario selection: Users can choose from a range of pre-defined traffic scenarios—such as intersections, pedestrian crossings, or merging situations—or customise their own setups. Each scenario includes a mix of vehicle types (autonomous or human-driven), road geometries, and environmental conditions.
Parameter configuration: Key variables can be adjusted, including:
-
- Driver attentiveness or reaction times
- Autonomous system behaviour models
- Vehicle dynamics and braking capabilities
- Traffic flow and density levels
This enables tailored simulations to study very specific use cases.
Simulation and visualisation: The platform runs dynamic simulations that show how vehicles move and interact in the selected environment. Visual outputs include:
-
- Real-time animations of traffic scenarios
- Risk heatmaps indicating potential collision zones
- Behavioural data charts for driver and system responses
Analysis tools: After running a scenario, users can access detailed analytics:
-
- Collision probability metrics
- Time-to-collision measurements
- Comparative data between human and AV responses
These insights support deeper research into adaptive vehicle control strategies, driver assistance systems, and urban safety planning.
The FlexCrash platform offers a new approach to understanding and mitigating crash risks in the mixed traffic era. It empowers stakeholders to ask, “what if?” questions and get data-driven answers, paving the way for safer road interactions between humans and machines.
Authors: Eurecat and IMC University of Applied Sciences Krems
