Decision

Estimote Sensors

This week marked the first week for our project, and consisted of finding a team and choosing a type of sensor to evaluate over the duration of the project. Upon forming our team, we decided upon the ‘Estimote Location Beacon’ (https://estimote.com) sensors to analyse over the course of our project. The aim of this project is to gather data from our sensor technology, then contextualize and portray the results in an informative way, by performing tests and analysing the results, we will be able to collect informative data on the accuracy and precision of the devices we are studying. All the readings are to be gathered by programming a small Android application.

The devices themselves, are somewhat interesting in nature, they are intended to provide real time tracking information of a specific Bluetooth enabled device inside a room, using a form of Bluetooth tracking technology. The sensors must be attached to walls about chest height around the room, and after some setup and information about the room they are in, they will all blast out signals into the room until they receive location feedback. We find this device interesting, as we’re not really sure yet what the genuine market use for this product would be, but the further we get into the project, the more we (hopefully) will learn about them and how they fit into the world of ubiquitous computing.

Potential Room 1, HCI Meeting Room

Our team ended up with 6 location beacons (you can use as many as you want to get readings), that each have to be individually programmed with room dimensions and where the devices are positioned in that room. Another team doing the project has chosen a set of 4 location beacons, however these 4 all communicate with each other and relay information about their own position, whereas ours could be defined as their manually programmed counterpart. We have chosen our initial control room to be a meeting room we have 24/7 access to and our lounge as a potential comparison room.

Potential Room 2, Flat lounge

Now that we have our feet off the ground, it is time to start getting into it. Our first task is to get the sensors up and running, so we can get some initial readings and figure out how they work and what values they give us. After receiving the values, we have to interpret them so as to make sense of random data, then once we understand what we’re looking at, we can start work on developing a testing suite.
The testing suite will be one of the most important parts of the project, it will enable us to perform repeated checks on the accuracy and precision of the tracking devices for whatever room we choose to measure from. With this data we can provide statistics on the devices and use those findings to perform a final evaluation on their effectiveness.

Screenshot of the Android app
showing sensor data

To facilitate the sensor data, we have written a custom Android app that uses Estimotes SDK to connect to the different beacons and pull their sensor data. Currently, we do this by using the phones built in Bluetooth device to scan for nearby beacons and then connect and pull its sensor data. We then display it in a list on screen. This will be useful for measuring the accuracy of the different sensors in preparation for the first assignment. Developing this custom app was useful in other ways too as we now know how to trigger certain actions based on the proximity of the phone to a beacon. More specifically, we now know how to monitor the range to beacons and when we go within a certain range, we can trigger certain events. This could be useful for many applications such as automatically starting the coffee machine when you walk near the beacon that is at the entrance to a kitchen.










 

Measuring sensor data against real world truths