-
Application logo
-
Environment 1
-
Environment 2
Inspiration
When entering into an unusually cold or warm room, we've always wondered what the room temperature was. However, it is simply impractical to bring an ambient thermometer everywhere. Furthermore, few smartphones on the market currently have a built-in thermometer. The need for accessible and versatile instruments in developing countries is also high, as cost is often a major barrier to procurement. Thus, we were inspired to build this easy-to-use thermometer application that operates at no additional cost to the user.
What it does
It is a well-known phenomenon that batteries operate at different efficiencies based on the temperature of the battery. Thus, the ambient temperature in which a battery operates can often have a major impact on its performance. (This is especially seen with larger batteries, such as in electric vehicles.) So, by measuring the discharge rate of a smartphones battery, this application is able to determine the temperature of the battery's ambient environment.
How we built it
A literature review showed us that there exists a relationship between ambient temperature and percent capacity. To determine percent capacity, we knew that we had to find the ratio between the battery's theoretical capacity and its real capacity as measured by the Android BatteryManager class. However, bench testing showed that capacity decreases as voltage decreases (i.e. as the battery drains). To account for this, we developed a function that expresses the theoretical capacity at a given voltage.
When the user requests a temperature reading, the app uses the battery voltage to calculate the theoretical capacity. Then, the actual capacity is recorded and the ratio of real to theoretical found. This becomes an input into a function relating percent capacity to temperature.
Challenges we ran into
We did not have the luxury of conducting repeated tests at a wide range of battery charges, temperatures, and capacities, which affects the reliability of our mathematical model.
Accomplishments that we're proud of
We were successfully able to detect different ambient temperatures in different environments through the application. Furthermore, we made the UI user-friendly and avoided complications.
What we learned
We learned a lot about how batteries work, and the different variables that affect their life cycles and discharge rates. Furthermore, we learned how to extract quite a bit of raw data from the device such as voltage, current, and battery capacity.
What's next for Tura
Over the next few weeks, we want to improve the algorithm and mathematical model such that the predicted ambient temperature is more accurate and faster to obtain. On top of this, we will be improving AI and learning how to run these processes on an iOS device, thereby developing an iOS-compatible application.
Log in or sign up for Devpost to join the conversation.