Particle filter based heading correction
A method for computing a correction to a compass heading for a portable device worn or carried by a user is described. The method involves determining a heading for the device based on a compass reading, collecting data from one or more sensors, determining if the device is indoors or outdoors based on the collected data, and correcting the heading based on the determination of whether the device is indoors or outdoors.
1. A computer-implemented method for correcting a heading of a tracked device, comprising:
receiving sensor data from one or more sensors associated with the tracked device, the sensor data representing an observation of the tracked device's heading and location;
generating a particle distribution from the sensor data, wherein each particle in the particle distribution represents a possible location and heading of the tracked device;
applying a particle filter to the particle distribution to reduce noise associated with building structure and other sources of magnetic fields, comprising:
determining, based on the filtered particle distribution, a tracked device heading and a tracked device location, the tracked device location being in or in close proximity to a building having an associated building outline and a plurality of grid directions indicated within the building outline
weighting one or more particles in the filtered particle distribution based on a heading confidence determination comprising:
determining a probable building grid direction; and
increasing a weight of a particle associated with the tracked device heading when a heading for the particle is along the probable grid direction; and
updating the tracked device location and the tracked device heading based on the weighted particles.
2. The method of claim 1 , wherein the sensor data further comprises one or more of compass data, magnetic sensor data, location data, gyroscope sensor data, and inertial sensor data.
3. The method of claim 2 , wherein the heading confidence is further based on a magnetic sensor confidence.
4. The method of claim 3 , wherein the magnetic sensor confidence is based on one or more of expected magnitude, inclination, and data from at least one additional sensor.
5. The method of claim 3 , wherein the magnetic sensor confidence is increased when the magnetic sensor receives consistently high reliability heading information and decreased when the magnetic sensor receives inconsistent heading information.
6. The method of claim 1 , wherein the probable building grid direction varies as a function of the tracked device location.
7. The method of claim 1 , wherein the probable building grid direction decreases with a distance the tracked device location is from a wall associated with the building outline.
8. The method of claim 1 , wherein the weight of particles increase as a function of time that the particles move along the probable building grid direction.
9. The method of claim 1 , wherein the weight of a particle is affected by a location of the particle inside or outside of the building outline.
10. A computing system for correcting a heading of a tracked device, comprising:
a processor;
a memory communicatively coupled to the processor, the memory bearing instructions that, when executed on the processor, cause the computing system to at least:
receive sensor data from one or more sensors associated with the tracked device, the sensor data representing an observation of the tracked device's heading and location;
generate a particle distribution from the sensor data, wherein each particle in the particle distribution represents a possible location and heading of the tracked device;
apply a particle filter to the particle distribution to reduce noise associated with building structure and other sources of magnetic fields, by further causing the computing system to at least:
determine, based on the filtered particle distribution, a tracked device heading and a tracked device location, the tracked device location being in or in close proximity to a building having an associated building outline and a plurality of grid directions indicated within the building outline; and
weight one or more particles in the filtered particle distribution based on a heading confidence determination comprising:
determining a probable building grid direction; and
increasing a weight of a particle associated with the tracked device heading when a heading for the particle is along the probable grid direction; and
update the tracked device location and the tracked device heading based on the weighted particles.
11. The computing system of claim 10 , wherein the sensor data further comprises at least one of compass data, magnetic sensor data, location data, gyroscope sensor data, and inertial sensor data.
12. The computing system of claim 11 , wherein the heading confidence is further based on a magnetic sensor confidence.
13. The computing system of claim 12 , wherein the magnetic sensor confidence is increased when the magnetic sensor receives consistently high reliability heading information and decreased when the magnetic sensor receives inconsistent heading information.
14. The computing system of claim 10 , wherein the probable building grid direction varies as a function of the tracked device location.
15. The computing system of claim 10 , wherein the probable building grid direction decreases with a distance the tracked device location is from a wall associated with the building outline.
16. The computing system of claim 10 , wherein the weight of particles increase as a function of time that particles move along the probable building grid direction.
17. The computing system of claim 10 , wherein the particles located outside of the building outline receive a lower weight than the particles located inside the building outline.
18. A non-transitory computer readable storage medium comprising instructions that, when executed on a computing system of a tracked device, cause the computing system to at least:
receive sensor data from one or more sensors associated with the tracked device, the sensor data representing an observation of the tracked device's heading and location;
generate a particle distribution from the sensor data, wherein each particle in the particle distribution represents a possible location and heading of the tracked device;
apply a particle filter to the particle distribution to reduce noise associated with building structure and other sources of magnetic fields, by further causing the computing system to at least:
determine, based on the filtered particle distribution, a tracked device heading and a tracked device location, the tracked device location being in or in close proximity to a building having an associated building outline and a plurality of grid directions indicated within the building outline; and
weight one or more particles in the filtered particle distribution based on a heading confidence determination comprising:
determining a probable building grid direction; and
increasing a weight of a particle associated with the tracked device heading when a heading for the particle is along the probable grid direction; and
update the tracked device location and the tracked device heading based on the weighted particles.
19. The non-transitory computer readable storage medium of claim 18 , wherein the weight of particles increase as a function of time that particles move along the probable building grid direction.
20. The non-transitory computer readable storage medium of claim 18 , wherein the particles located outside of the building outline receive a lower weight than the particles located inside the building outline.