Simplifying Commute by Predicting Obstacles

Prototype for SmartCity Project about detecting obstacles in people’s commute to reduce uncertainty. We used computer vision and other APIs to notify users at the right time. For example, when there are too many people at the subway station and trains are too crowded, we may alert a user who might be thinking about taking the subway. This way users can think about alternative ways like taking an Uber or bike. The project may have a strong use case for mobility impaired individuals for whom even minor incidents can have a significant impact on their commute [1,2].

GUI app onboarding screen
Overview of the commute
Push notification for anomalies during commute

Process

Our design process started from understanding the notion of Smart Cities. We talked to experts about existing projects to learn more about the shortcomings and advantages of using technologies in this space. While narrowing down the scope, we developed multiple prototypes and simultaneously conducted user research to build empathy with our end users (commuters). Finally, I leveraged my technical expertise with Javascript and Machine Learning to develop working prototypes of our ideas that we presented to our stakeholders.

System Design

System Architecture of GUI App


Computer Vision to detect anomalies

<figure st****yle="display: inline-block">
Computer vision to detect anomalies such as over-crowded buses and trains
</figure>


Team


Code

The code can be found here.