RouteSafe

We’re turning navigation from fastest route → safest route.

Comment

\section{RouteSafe}

\subsection{Inspiration}

The idea for RouteSafe came from a simple thought during the hackathon. When people evacuate during wildfires, they often rely on navigation systems to guide them to safety. Applications such as Google Maps or Apple Maps are excellent at finding the fastest route or the shortest distance. However, they do not account for something that becomes extremely dangerous during wildfire events, which is smoke exposure.

Wildfire smoke can travel far beyond the fire itself and can quickly spread across large regions. A route that appears fast on a map may guide someone directly through areas with hazardous air quality. When we realized this, we started asking ourselves a different question. Instead of asking what the fastest route is, what if navigation systems asked what the safest route is?

This idea became the foundation of RouteSafe. We wanted to build a routing system that considers smoke exposure as a key factor when guiding drivers.

\subsection{What it does}

RouteSafe is a wildfire aware navigation system that calculates the safest route for drivers by considering predicted smoke exposure along the path.

Traditional navigation systems focus on optimizing travel time, distance, or traffic conditions. RouteSafe introduces a new factor into the routing process, which is smoke exposure risk. For every possible route, the system estimates how much smoke a driver might encounter and uses that information when choosing a path.

A key feature of RouteSafe is that it does not only look at current conditions. Smoke moves and spreads over time. Because of this, the system predicts what the smoke conditions will be when a driver actually reaches each part of the route.

If a driver arrives at road segment $i$ at time $t_i$, the system estimates the smoke concentration at that time and incorporates it into the routing decision. In simplified form, we modeled this with the cost function

[ \text{Cost} = \sum_{i=1}^{n} \left( \alpha T_i + \beta S(t_i) \right) ]

where $T_i$ represents the travel time for segment $i$, and $S(t_i)$ represents the predicted smoke exposure when the driver arrives there. The parameters $\alpha$ and $\beta$ allow us to balance travel time and safety.

This allows RouteSafe to recommend routes that reduce health risks even if they are not the absolute fastest option.

\subsection{How we built it}

We started by modeling wildfire hotspots as hazard sources that produce smoke which spreads outward over time. This allowed us to estimate smoke levels at different locations.

Next, we built a system that predicts how smoke concentration changes as time passes. This allowed us to estimate smoke exposure not just at the present moment but also in the near future.

After that, we integrated this prediction system into a routing algorithm. For each possible route segment, we estimated when the driver would arrive and calculated the expected smoke exposure at that time.

Finally, we modified the path optimization process so that it minimizes both travel time and smoke exposure. This produced routes that naturally avoided areas where smoke levels were predicted to be high.

\subsection{Challenges we ran into}

One of the challenges we encountered was pressure to include artificial intelligence or machine learning in the project. Many people suggested that we train a model or add AI features. However, after thinking carefully about the problem, we realized that it did not truly require machine learning. The core challenge was better solved with predictive modeling and routing algorithms, so we decided to stay focused on that approach.

Another challenge was realizing that other teams might be working on similar wildfire related ideas. For a moment this made us question whether our project was unique. Instead of changing directions, we focused on strengthening the technical aspects of our routing system.

The path optimization process was also more difficult than we initially expected. Small changes in our parameters could dramatically affect the routes produced by the system. We spent a significant amount of time tuning the optimizer so that routes balanced safety and practicality.

We also ran into technical difficulties as we added new features. At one point the system became unstable and parts of the routing logic stopped working properly. We eventually had to revert to an earlier working version of the project and rebuild features more carefully.

\subsection{Accomplishments that we're proud of}

One accomplishment we are proud of is building a routing system that considers safety in a meaningful way. Instead of simply visualizing wildfire data, RouteSafe actively changes navigation decisions based on predicted smoke exposure.

We are also proud of successfully implementing a predictive approach. The system estimates future smoke conditions when a driver reaches each part of a route rather than only reacting to current data.

Finally, we are proud that we stayed focused on solving the problem in a practical way. Even when people encouraged us to add unnecessary complexity, we kept the system grounded in algorithms and optimization.

\subsection{What we learned}

Working on RouteSafe taught us that not every problem requires artificial intelligence. In some cases, careful modeling and well designed algorithms can produce effective solutions.

We also learned how sensitive optimization systems can be. Small parameter changes can significantly affect results, which means experimentation and iteration are essential.

Another important lesson was the importance of building systems incrementally. When our code broke after adding several features, we realized how important it is to maintain stable versions and introduce changes carefully.

\subsection{What's next for RouteSafe}

In the future, RouteSafe could be expanded by integrating real wildfire data sources and more advanced environmental models. Adding wind data and real time air quality measurements could improve the accuracy of smoke predictions.

Another possible direction is integrating the system with existing navigation platforms so that users can easily access safer routing options during wildfire events.

Our long term vision is to create navigation systems that prioritize not only efficiency but also health and safety during environmental emergencies.

Built With

Share this project:

Updates