Scalable indoor navigation and positioning systems and methods
Embodiments of the present disclosure related to methods and systems of determining the position of a device in indoor space. A map importing module takes a floor map of an indoor space and processes it for further use. A calibration module creates a fingerprint map of the indoor space described by the floor map. A positioning module uses the calibration data embodied in a fingerprint map as well as device signal and sensor readings in order to estimate a trajectory for the device, then position the device within the indoor space described by the floor map using a probability density function.
1. A method of determining a position of a device within an indoor location, comprising:
gathering device signal data, the device signal data including at least one of Wi-Fi data, Bluetooth data, GPS data and cellular data;
gathering device sensor data from at least one of a set of device sensors including an accelerometer, a gyroscope and a magnetometer;
estimating a plurality of potential trajectories describing a movement of the device through the indoor location as a sequence of positions based on the device signal data, a fingerprint map calibrated for the indoor location, and by determining, from the device sensor data, a device direction of travel angle subject to a misalignment correction based on monitoring at least one of a device sensor drift and a device sensor bias factors, wherein the fingerprint map comprises a plurality of fingerprint locations, each of the plurality of fingerprint locations having at least one associated device signal data value or device sensor data value; and
determining the position of the device based on a data fusion of the estimated plurality of potential trajectories, the data fusion based at least in part upon a probabilistic weighting for respective ones of the estimated plurality of potential trajectories.
2. The method of claim 1 , wherein the estimating the plurality of potential trajectories comprises a lightweight local optimization.
3. The method of claim 1 , wherein the fingerprint map is constructed by inferring the plurality of signal data values based on actual signal data values.
4. The method of claim 3 , wherein the actual signal data values are recorded by a calibration module.
5. The method of claim 1 , further comprising estimating a step length and a heading based on the device sensor data.
6. The method of claim 1 , further comprising updating the fingerprint map with the position of the device.
7. The method of claim 1 , further comprising orienting the device to the global coordinate system.
8. The method of claim 7 , wherein the orienting the device to the global coordinate system is performed using a rotation matrix.
9. The method of claim 1 , wherein the device signal data comprises a signal strength measurement value.
10. A system for determining a position of a device in an indoor location, the system comprising:
a memory;
a processor;
a module that is stored in the memory and, when executed by the processor, determines the position of a device in an indoor location by performing the steps of: gathering device signal data, the device signal data including at least one of Wi-Fi data, Bluetooth data, GPS data and cellular data;
gathering device sensor data from at least one of a set of device sensors including an accelerometer, a gyroscope and a magnetometer;
estimating a plurality of potential trajectories describing a movement of the device through the indoor location as a sequence of positions based on the device signal data, a fingerprint map calibrated for the indoor location, and by determining, from the device sensor data, a device direction of travel angle subject to a misalignment correction based on monitoring at least one of a device sensor drift and a device sensor bias factors, wherein the fingerprint map comprises a plurality of fingerprint locations, each of the plurality of fingerprint locations having at least one associated device signal data value or device sensor data value; and
determining the position of the device based on a data fusion of the estimated plurality of potential trajectories, the data fusion based at least in part upon a probabilistic weighting for respective ones of the estimated plurality of potential trajectories.
11. The system of claim 10 , wherein the estimating the plurality of potential trajectories comprises a lightweight local optimization.
12. The system of claim 10 , further comprising estimating a step length and a heading based on the device sensor data.
13. The system of claim 12 , wherein the actual signal data values are recorded by a calibration module.
14. The system of claim 13 , wherein the fingerprint map is constructed by inferring the plurality of signal data values based on actual signal data values.
15. The system of claim 10 , wherein the device signal data comprises a signal strength measurement value.
16. A non-transitory computer-readable medium disposed in an audio device, the non-transitory computer-readable medium containing a set of instructions that causes a processor of a device to perform a process comprising:
gathering device signal data, the device signal data including at least one of Wi-Fi data, Bluetooth data, GPS data and cellular data;
gathering device sensor data from at least one of a set of device sensors including an accelerometer, a gyroscope and a magnetometer;
estimating a plurality of potential trajectories describing a movement of the device through an indoor location as a sequence of positions from the device signal data, a fingerprint map calibrated for the indoor location, and by determining, from the device sensor data, a device direction of travel angle subject to a misalignment correction based on monitoring at least one of a device sensor drift and a device sensor bias factors, wherein the fingerprint map comprises a plurality of fingerprint locations, each of the plurality of fingerprint locations having at least one associated device signal data value or device sensor data value; and
determining a position of the device relative to the indoor location based on a data fusion of the estimated plurality of potential trajectories, the data fusion based at least in part upon a probabilistic weighting for respective ones of the estimated plurality of potential trajectories.