Method and apparatus for determining sportsman jumps using fuzzy logic
A method and apparatus for detecting a jump of a moving sportsman and separating the jump from other sportsman motion is described. Accelerometer data generated by a sportsman's motion is received in a computing device. The computing device is used to apply fuzzy logic membership functions to a plurality of parameters associated with the accelerometer data to detect a pattern associated with jumps. A subset of the data is identified as representing a jump based upon the detection of the pattern. The determination is used in the computing device to transform data, whereby jumps are separated from other sportsman motion.
1. A method comprising:
receiving, by a computing device, accelerometer data generated by a sportsman's motion from an accelerometer coupled to the sportsman and in communication with the computing device;
applying, by the computing device, a plurality of fuzzy logic membership functions to a plurality of parameters associated with the accelerometer data to detect a pattern associated with a jump, wherein detection of the pattern associated with the jump includes:
determining a fuzzy logic initial jump probability membership function;
determining a fuzzy logic jump start probability membership function;
determining a fuzzy logic landing shock probability membership function;
determining a fuzzy logic false jump probability membership function; and
combining the jump start probability and landing shock probability membership functions together using fuzzy logic rules;
identifying a subset of the accelerometer data representing the jump based upon the detected pattern; and
transforming the accelerometer data, by the computing device, by separating the identified subset of accelerometer data representing the jump from accelerometer data associated with other motion by the sportsman.
2. The method of claim 1 , further comprising:
receiving, by the computing device, global positioning sensor (GPS) measurements from a GPS device coupled to the sportsman and in communication with the computing device; and
identifying the subset of the accelerometer data representing the jump based at least in part on the received GPS measurements.
3. The method of claim 2 , wherein the fuzzy logic membership functions are applied to the GPS measurements to detect the pattern associated with the jump.
4. The method of claim 1 , wherein a time duration of the jump of the sportsman is further determined.
5. The method of claim 3 , wherein acceptance and rejection are mapped as 2D regions on a membership function vs. a jumptime plane.
6. The method of claim 4 , wherein a total norm of a 3x axis accelerometer is used.
7. The method of claim 4 , wherein a projection of an accelerometer vector to the magnetic vector is used.
8. The method of claim 4 , wherein accelerometer data is collected only in response to the computing device measuring a velocity of the sportsman above some predetermined threshold.
9. The method of claim 6 , wherein the accelerometer data is corrected by an approximate rotational acceleration computed from gyro sensors coupled to the sportsman and in communication with the computing device.
10. The method of claim 1 , wherein the accelerometer data is time tagged.
11. The method of claim 1 , wherein the computing device comprises a smartphone with an accelerometer therein.
12. A non-transitory computer readable storage medium storing instructions that, when executed by a computing device, cause the computing device to:
receive accelerometer data generated by a sportsman's motion from an accelerometer coupled to the sportsman and in communication with the computing device;
apply a plurality of fuzzy logic membership functions to a plurality of parameters associated with the accelerometer data to detect a pattern associated with a jump, wherein detection of the pattern associated with the jump includes:
determining a fuzzy logic initial jump probability membership function;
determining a fuzzy logic jump start probability membership function;
determining a fuzzy logic landing shock probability membership function;
determining a fuzzy logic false jump probability membership function; and
combining the jump start probability and landing shock probability membership functions together using fuzzy logic rules;
identify a subset of the accelerometer data representing the jump based upon the detected pattern; and
transform the accelerometer data by separating the identified subset of accelerometer data representing the jump from accelerometer data associated with other motion by the sportsman.