Unsupervised indoor localization and heading directions estimation
Systems and methods for unsupervised indoor localization are provided. Sensor data obtained from a device carried by a user can be used to simultaneously estimate the indoor location of a user and identify landmarks within the indoor environment based on their signatures. The landmarks can be used to reset the location estimate of the user, and the location estimate of the user can be used to improve the learned location of the landmarks. This recursive process leads to excellent accuracy in indoor localization. Systems and methods for estimating the heading direction of a user are also provided. Sensor data obtained from the user can be used to analyze the forces acting on the user in order to give an accurate heading direction estimate.
1. A method of unsupervised indoor localization, the method comprising:
estimating a location of a user within an indoor environment, wherein estimating the location of the user comprises:
estimating a movement trace of the user within the indoor environment using first sensor data obtained from at least one sensor, wherein the first sensor data comprises accelerometer data; and
identifying at least one landmark within the indoor environment using second sensor data obtained from the at least one sensor;
wherein estimating the movement trace of the user comprises:
estimating a number of steps taken by the user over a first period of time by analyzing the accelerometer data;
estimating a distance traveled by the user during the first period of time by multiplying the number of steps with an estimated step size;
estimating a heading direction of the user; and
calculating a movement trace vector based on the distance traveled and the heading direction.
2. The method of claim 1 ,
wherein the second sensor data comprises at least one selected from the group consisting of: accelerometer data; compass data; gyroscope data; magnetometer data; barometer data; microphone data; light sensor data; temperature sensor data; chemical sensor data; humidity sensor data; Bluetooth signal strength data; WiFi signal strength data; and cellular signal strength data.
3. The method according to claim 1 , wherein at least a portion of the first sensor data is the same as at least a portion of the second sensor data.
4. The method according to claim 1 , wherein the at least one landmark does not include a label.
5. The method according to claim 1 , wherein the at least one landmark comprises at least one structure selected from the group consisting of: a building entrance; a staircase; an elevator; and an escalator.
6. The method according to claim 1 , further comprising:
identifying a plurality of landmarks within the indoor environment using the second sensor data; and
resetting the estimate of the movement trace of the user each time a landmark is identified, wherein the estimate of the movement trace of the user is reset to the location of the identified landmark.
7. The method according to claim 1 , further comprising:
obtaining sensor data from a plurality of users within the indoor environment; and
analyzing the sensor data from the plurality of users to obtain a weighted-average estimated location for each of a plurality of landmarks within the indoor environment.
8. The method of claim 1 , wherein the entire method is performed locally by a device of the user, wherein the device comprises the at least one sensor.
9. The method according to claim 8 , wherein the first sensor data and the second sensor data remain locally stored on the device and are not uploaded to any other device during the method.
10. A method of unsupervised indoor localization, the method comprising:
estimating a location of a user within an indoor environment, wherein estimating the location of the user comprises:
estimating a movement trace of the user within the indoor environment using first sensor data obtained from at least one sensor, wherein the first sensor data comprises accelerometer data, gyroscopic data, and compass data;
identifying at least one landmark within the indoor environment using second sensor data obtained from the at least one sensor;
estimating a heading direction of the user, wherein estimating the heading direction of the user comprises:
determining a motion mode of the user by analyzing the accelerometer data;
determining a number of steps taken by the user by analyzing the accelerometer data in view of the motion mode;
determining an initial heading direction of the user in three-dimensional space using the gyroscope data and the number of steps;
determining a heading direction vector of the user by estimating the heading angle of the user relative to a reference direction; and
projecting the heading direction vector onto a plane on which the user is moving to estimate the heading direction of the user.
11. The method according to claim 10 , wherein projecting the heading direction vector onto a plane on which the user is moving comprises projecting the heading direction vector onto a horizontal plane perpendicular to a gravity vector.