Systems and methods for constructing motion models based on sensor data

Abstract

This disclosure relates to systems, media, and methods for updating motion models using sensor data. In an embodiment, the system may perform operations including receiving sensor data from at least one motion sensor; generating training data based on at least one annotation associated with the sensor data and at least one data manipulation; receiving at least one experiment parameter; performing a first experiment using the training data and the at least one experiment parameter to generate experiment results; and performing at least one of: update or validate a first motion model based on the experiment results.

1 - 20 . (canceled) 
     
     
         21 . A computing device for using a motion model, the computing device comprising:
 a memory storing instructions; and   at least one processor configured to execute the instructions to perform operations comprising:
 receiving motion data comprising sensor data and at least one motion classification or annotation associated with the sensor data; 
 performing an experiment by inputting the motion data and at least one experiment parameter to the motion model, the motion model generating experiment results; 
 annotating, based on the experiment results, a segment of the sensor data; and 
 transmitting the annotated segment of the sensor data to a remote device. 
   
     
     
         22 . The computing device of  claim 21 , wherein the sensor data is generated by an inertial measurement unit (IMU). 
     
     
         23 . The computing device of  claim 22 , wherein the IMU is part of a wearable device. 
     
     
         24 . The computing device of  claim 21 , wherein the at least one motion classification or annotation indicates a motion associated with the sensor data as at least one of favorable, unfavorable, of a particular motion type, or not of a particular motion type. 
     
     
         25 . The computing device of  claim 21 , wherein the operations further comprise receiving an indication of whether the experiment correctly identified an event. 
     
     
         26 . The computing device of  claim 21 , wherein:
 the remote device is a user device; and   at least a portion of the motion data is received from the user device.   
     
     
         27 . The computing device of  claim 21 , wherein:
 at least a portion of the motion data is received from a user device; and   the user device received the sensor data from a sensor device when the user device was communicably connected to the user device.   
     
     
         28 . The computing device of  claim 21 , wherein the at least one motion classification or annotation was determined in response to a user interaction at a user device. 
     
     
         29 . The computing device of  claim 28 , wherein the user interaction at the user device comprises a user interaction with data visually displayed at the user device. 
     
     
         30 . The computing device of  claim 21 , wherein the operations further comprise performing at least one data manipulation to the motion data prior to performing the experiment. 
     
     
         31 . The computing device of  claim 30 , wherein the at least one data manipulation is based on the at least one motion classification or annotation. 
     
     
         32 . The computing device of  claim 30 , wherein the at least one data manipulation includes at least one of:
 vertically scaling at least a portion of the sensor data;   time scaling at least a portion of the sensor data;   time shifting at least a portion of the sensor data;   integrating white noise into at least a portion of the sensor data;   applying random nonlinear skewing to at least a portion of the sensor data; or   generating a plurality of data streams based on the sensor data, the data streams having differing smoothing intensities.   
     
     
         33 . The computing device of  claim 21 , wherein the annotated segment of the sensor data is viewable, together with video data associated with the sensor data, at a display of the remote device. 
     
     
         34 . The computing device of  claim 21 , wherein:
 the motion data comprises at least one annotation associated with the sensor data; and   the operations further comprise altering the at least one annotation prior to performing the experiment.   
     
     
         35 . The computing device of  claim 21 , wherein the motion model is configured for a particular type of motion associated with the motion data. 
     
     
         36 . The computing device of  claim 21 , wherein the operations further comprise identifying the segment of sensor data as indicative of a motion type. 
     
     
         37 . The computing device of  claim 36 , wherein annotating the segment of the sensor data comprises generating an annotation associated with the motion type. 
     
     
         38 . A method for updating motion models using a motion model, comprising:
 receiving motion data comprising sensor data and at least one motion classification or annotation associated with the sensor data;   performing an experiment by inputting the motion data and at least one experiment parameter to the motion model, the motion model generating experiment results;   annotating, based on the experiment results, a segment of the sensor data; and   transmitting the annotated segment of the sensor data to a remote device.   
     
     
         39 . The method of  claim 38 , wherein the at least one motion classification or annotation was determined in response to a user interaction at a user device. 
     
     
         40 . The method of  claim 38 , wherein the annotated segment of the sensor data is viewable at a display of the remote device.