SMART Positional Systems

Inspiration

Movement patterns can tell a lot about the current state or potential state of a situation. We decided to build an application that uses this idea in congruence with an IOT platform to help minimize the costs of these situations.

What it does

The SMART Positioning System uses data sent from IOT devices attached to home residents to minimize risks. This could be a phone with a GPS enabled or some kind of smart watch that allows for tracking of residence within the home. Using data from possible risky scenarios and possible high risk situations, alerts can be sent to residents or emergency contacts.

How I built it

Our implementation uses the idea that we are able to track the location of an individual inside in environment. Additionally, we can sample the state of an open window or unattended stove through these features. Certain behaviours of the position tracker can alert the program to potential risks or to minimize damages that have already occurred. For example, if the position of a person is not within the radius of an open flame, there is clearly a risk of fire. Our program can detect this and send a text message alert to residents of the household. Using Java Interfaces and visualizations, we can see how our program reacts in a simulated environment.

Challenges I ran into

One of the biggest issues was finding a way to send data over JSON through A6 that would allow for easy data retrieval on the client side. Additionally, we had trouble using a Python client to interact with the A6 platform but were able to remedy this issue by switching to Java.

Accomplishments that I'm proud of

We are proud of the fact that we are able to use remote sensor inputs to plot the location of residents in the home to potentially hazardous situations. Our three clients, the sensor-client, server interface, and Graphical User Interface are able to interact together with the WebHooks platform to minimize risks.

What I learned

As our first hackathon, our team learned how to work under extreme physical and time constraints. Additionally, we learned how to use different modules to interact with a web based platform. This also allowed us to think about how connected technology can be used to improve future living experiences inside a connected home.

What's next for SMART Positional Systems

SMART Positional System could integrate data analysis to perform calculations that lead to high-risk situations. Machine-learning is an obvious candidate as we can take data from situations and apply it to some kind of polynomial regression algorithm to further predict risky situations.