System and method for tracking movement
A fitness tracking system includes a shoe, a magnetometer, and a controller. The magnetometer is mounted on the shoe and is configured to generate three-axis direction data in response to movement of the shoe during a predetermined time period. The controller is operably connected to the magnetometer and is configured to generate two-axis calibrated direction data based on the three-axis direction data after the predetermined time period. The two-axis calibrated direction data corresponds to an orientation of the shoe during the predetermined time period.
1. A method of determining an orientation of a shoe worn by a user with a magnetometer mounted on the shoe, the method comprising:
generating three-axis direction data with the magnetometer during a predetermined time period;
fitting a plane to the three-axis direction data;
converting the three-axis direction data to two-axis uncalibrated direction data located in the plane;
fitting an ellipse to the two-axis uncalibrated direction data; and
generating two-axis calibrated direction data based on the three-axis direction data after the predetermined time period, the two-axis calibrated direction data corresponding to the orientation of the shoe during the predetermined time period, wherein generating two-axis calibrated direction data includes changing a position of the two-axis uncalibrated direction data by converting the fitted ellipse to a circle and positioning the circle at an origin of a coordinate system.
2. The method of claim 1 , further comprising:
identifying outlier data points of the two-axis uncalibrated direction data; and
deleting the outlier data points from the two-axis uncalibrated direction data.
3. The method of claim 2 , further comprising:
fitting a confidence band to the two-axis uncalibrated direction data; and
identifying the outlier data points as the two-axis uncalibrated direction data located outside of the confidence band.
4. A method of determining an orientation of a shoe worn by a user with a magnetometer mounted on the shoe, the method comprising:
generating three-axis direction data with the magnetometer during a predetermined time period;
fitting a plane to the three-axis direction data;
converting the three-axis direction data to two-axis uncalibrated direction data located in the plane;
fitting data points of the two-axis uncalibrated direction data into a plurality of bins;
deleting the data points of the two-axis uncalibrated direction data that are fitted into a bin of the plurality of bins having fewer than a predetermined number of data points; and
generating two-axis calibrated direction data based on the three-axis direction data after the predetermined time period, the two-axis calibrated direction data corresponding to the orientation of the shoe during the predetermined time period.
5. The method of claim 1 , further comprising:
detecting when a shoe of the user is in contact with the ground using an accelerometer mounted on the shoe; and
generating the three-axis direction data only when the shoe is in contact with the ground.
6. The method of claim 1 , further comprising:
applying a data smoothing process to the three-axis direction data; and
generating the two-axis calibrated direction data based on the three-axis direction data after the application of the data smoothing process.
7. The method of claim 1 , wherein the three-axis direction data are a measure of magnetic flux density.
8. The method of claim 1 further comprising:
fitting data points of the two-axis uncalibrated direction data into a plurality of bins; and
deleting the data points of the two-axis uncalibrated direction data that are fitted into a bin of the plurality of bins having fewer than a predetermined number of data points.
9. The method of claim 4 further comprising:
identifying outlier data points of the two-axis uncalibrated direction data; and
deleting the outlier data points from the two-axis uncalibrated direction data.
10. The method of claim 9 further comprising:
fitting a confidence band to the two-axis uncalibrated direction data; and
identifying the outlier data points as the two-axis uncalibrated direction data located outside of the confidence band.
11. The method of claim 4 further comprising:
detecting when a shoe of the user is in contact with the ground using an accelerometer mounted on the shoe; and
generating the three-axis direction data only when the shoe is in contact with the ground.
12. The method of claim 4 , further comprising:
applying a data smoothing process to the three-axis direction data; and
generating the two-axis calibrated direction data based on the three-axis direction data after the application of the data smoothing process.
13. The method of claim 4 , wherein the three-axis direction data are a measure of magnetic flux density.