System and method of biomechanical posture detection and feedback including sensor normalization

Abstract

A system and method are described herein for a sensor device which biomechanically detects in real-time a user's movement state and posture and then provides real-time feedback to the user based on the user's real-time posture. The feedback is provided through immediate sensory feedback through the sensor device (e.g., a sound or vibration) as well as through an avatar within an associated application with which the sensor device communicates. The sensor device detects the user's movement state and posture by capturing data from a tri-axial accelerometer in the sensor device. Streamed data from the accelerometer is normalized to correct for sensor errors as well as variations in sensor placement and orientation. Normalization is based on accelerometer data collected while the user is wearing the device and performing specific actions.

1 .- 2 . (canceled) 
     
     
         3 . An apparatus comprising:
 a sensor system configured to be worn by a user, the sensor system comprising:
 at least one movement sensor comprising an accelerometer and a gyroscope; 
 a feedback apparatus; and 
 a microprocessor configured to:
 receive data from the at least one movement sensor regarding movement of the user; 
 normalize the received data; 
 use an activity detection algorithm to ascertain a postural description of the user based on the normalized received data, the postural description comprising at least one movement state; and 
 trigger the feedback apparatus based on the postural description of the user. 
 
   
     
     
         4 . The apparatus of  claim 3 , wherein the processor is operative to:
 compute a sequence of movement data features from the normalized movement data; and   analyze the sequence of movement data features according to a rule-based model to determine, in part, the postural description.   
     
     
         5 . The apparatus of  claim 3 , wherein the processor is operative to
 compute a sequence of movement data features from the normalized movement data; and   compare the sequence of movement data features according to a movement library to determine, in part, the postural description.   
     
     
         6 . The apparatus of  claim 3 , wherein the feedback apparatus comprises at least one motor, and wherein the motor is activated based on the postural description of the user. 
     
     
         7 . The apparatus of  claim 3 , wherein the at least one movement state comprises standing, sitting, lying, running, or walking. 
     
     
         8 . The apparatus of  claim 3 , wherein the microprocessor is operative to use the activity detection algorithm to detect a transition between a first movement state and a second movement state. 
     
     
         9 . The apparatus of  claim 3 , wherein the microprocessor is operative to use the activity detection algorithm to detect a transition between any two of a set of movement states including at least a standing movement state and a sitting movement state. 
     
     
         10 . The apparatus of  claim 3 , wherein the microprocessor is operative to use the activity detection algorithm to detect at least one transition between any two of a set of the movement states including at least a walking movement state, a standing movement state, and a sitting movement state. 
     
     
         11 . The apparatus of  claim 3 , wherein the postural description comprises a posture quality for each the at least one movement state. 
     
     
         12 . The apparatus of  claim 11 , wherein the feedback apparatus is triggered based on the posture quality. 
     
     
         13 . The apparatus of  claim 3 , wherein the sensor system further comprises a communications interface, wherein the processor is operative to process user inputs received via the communications interface. 
     
     
         14 . The apparatus of  claim 13 , wherein the processed user inputs comprise control of the feedback apparatus. 
     
     
         15 . The apparatus of  claim 13 , wherein the processed user inputs comprise calibration events. 
     
     
         16 . The apparatus of  claim 3 , wherein the sensor system further comprising a non-transitory storage device for storing data, wherein the microprocessor is operative to store movement data of the user, user history of reaction to triggered feedback, or user usage patterns in the non-transitory storage device. 
     
     
         17 . A method for using a system worn by a user, the system comprising a microprocessor, a feedback apparatus, and at least one movement sensor comprising an accelerometer and a gyroscope, the method comprising:
 receiving data from the at least one movement sensor regarding movement of the user;   normalizing the received data;   using an activity detection algorithm to ascertain a postural description of the user based on the normalized received data, the postural description comprising at least one movement state; and   triggering the feedback apparatus based on the postural description of the user.   
     
     
         18 . The method of  claim 17 , wherein using the activity detection algorithm comprises:
 computing a sequence of movement data features from the normalized movement data; and   analyzing the sequence of movement data features according to a rule-based model to determine, in part, the postural description.   
     
     
         19 . The method of  claim 17 , wherein using the activity detection algorithm comprises:
 computing a sequence of movement data features from the normalized movement data; and   comparing the sequence of movement data features according to a movement library to determine, in part, the postural description.   
     
     
         20 . The method of  claim 17 , wherein the feedback apparatus comprises at least one motor, and wherein the motor is activated based on the postural description of the user. 
     
     
         21 . The method of  claim 17 , wherein the at least one movement state comprises standing, sitting, lying, running, or walking. 
     
     
         22 . The method of  claim 17 , wherein using the activity detection algorithm comprises detecting a transition between a first movement state and a second movement state.