RoadWise

Inspiration

Road safety analysis is often reactive, relying on crash data after incidents occur. We wanted to explore whether risk can be identified earlier by combining driver behavior, vehicle dynamics, and roadway conditions. This inspired us to build a proactive, scenario-based risk assessment tool.

What it does

RoadWise is an interactive platform that evaluates driving risk based on user-defined scenarios. It provides a real-time risk score, highlights key contributing factors, and suggests improvements. Users can also compare scenarios to understand how design changes affect safety.

How we built it

We built a web-based interface using a modular structure for driver, vehicle, and road inputs. A weighted risk model combines these components into a unified score. Additional features like explainable insights, comparison mode, and community feedback enhance usability and interpretation.

Challenges we ran into

One major challenge was defining a meaningful risk baseline without real-world crash labels. We addressed this by designing a structured risk classification based on control stability and recovery capability. Balancing simplicity and interpretability within a short development time was also challenging.

Accomplishments that we’re proud of

We successfully transformed a complex transportation concept into an intuitive and interactive tool. The system not only outputs a risk score but also explains why the risk occurs and how to reduce it. The comparison and visualization features make the platform feel like a real decision-support system.

What we learned

We learned how to translate engineering concepts into user-friendly interfaces. We also explored how weighting, visualization, and feedback can improve interpretability. This project showed us the importance of explainability in risk assessment systems.

What’s next for RoadWise

We plan to integrate real-world data such as trajectory and simulator outputs to refine the model. Future versions could include adaptive learning and scenario-based calibration. We also aim to expand this into a more comprehensive transportation safety analysis platform.