Two students from Egypt’s Alexandria University have developed a mobile app that “ allows for crowdsourcing of data from smartphone sensors to create indoor floorplans automatically”.
CrowdInside, developed by Moustafa Alzantot and Moustafa Youssef, gives accurate information of indoor locations. It also facilitates easy tracking of locations should the global satellite positioning system (GSPS), which maps out places such as road names, streets and shopping malls, fail.
The founders say the system can be used for complex tasks such as coding higher-level data as it configures shapes of rooms by monitoring the spread of traces or position of doors through intersection between corridors and rooms.
According to Alzantot and Youssef the system is superior in crowdsourcing where many users are involved. Through crowdsourcing, it sharpens floorplans, giving more accurate and concise data.
The sensor data is used to spot the users when they are in an elevator or walking up or down stairs. These movements produce a unique pattern of acceleration, which is different from walking, therefore making it easy to spot, the students said.
Use of GPS data is insufficient to determine the entrance to a building. The app’s dead-reckoning is useful in such a scenario.
Smartphones have a wide variety of sensors such as magnetometers, GPS devices, accelerometers and WiFi signals strength meters, that give a rough estimation of the distance to be covered to the nearest hotspot.
According to the developers, use of dead-reckoning, using an accelerometer as a pedometer and a magnetometer as a direction finder, can be used to determine the number of steps in a specific direction, thus estimating the distance to be covered.
When calculating the current position of a place or a person, CrowdInside uses accelerometer as a pedometer and the magnetometer as a locator that determines the number of steps in a specific direction, in addition to the distance to be covered.
The system can also recalibrate using points at a known location to overcome dead-reckoning. It is however prone to errors that develop rapidly within a short period.
The two have not yet decided on whether to commercialise CrowdInside.
