Back to results
US 10,588,547 B2Granted
Systems, devices and methods relating to motion data
AssigneeInMotion, Inc.
InventorSrini G. Naidu
Life of this patent
Estimating the term…
Abstract
Systems for studying motion are provided. A computing device [10] has a processor [20], an accelerometer [11], a gyroscope [12], a magnetometer [13] and storage/computer readable media [30] in communication with one another. The computing device [10] can sense, classify, qualify and/or quantify real-time motion data of a moving target against classified initial motion data in a motion library [32]. Motion data is processed as particularized units of motion. The computing device [10] may use machine learning algorithms for “training” and “learning.” The computing device [10] can be used in many industries, including the fitness industry, where computing device [10] can be used with wearable technology.
Claims17
A computing device [10] for analyzing motion of a moving target, the device comprising:
- a) a processor [20], computer-readable media [30], an accelerometer [11], a gyroscope [12], a magnetometer [13], each configured to be in direct or indirect communication with one another;
- and
- b) a computer program [31] including instructions executable by the processor to create a qualitative and quantitative motion analysis, the computer program comprising:
- i) software configured to access the computer-readable media [30] including at least one library [32] of at least one class of initial motion data, such data being processible as units of motion, each such unit of motion being an equation over a predetermined time interval that is a function of separate initial motion data equations for at least acceleration data, angular velocity data, and orientation data in a predetermined number of degrees of freedom, each such unit of motion retaining the quality and quantity of the initial motion data, each such unit of motion having content including a number of data points x equal to the product of the degrees of freedom by the predetermined time interval;
- ii) software configured to receive real-time motion data from sensed signals, the real-time motion data including at least acceleration data from the accelerometer [11], angular velocity data from the gyroscope [12], and orientation data from the magnetometer [13], such real-time motion data being processible in the same units of motion as the initial motion data;
- iii) software configured to compare, by units of motion, the real-time motion data to the initial motion data;
- and
- iv) software configured to qualify and quantify the real-time sensed motion data based at least in part upon the comparison, said unit of motion is an equation reflecting the function of equations at least of acceleration data in an X, Y, and Z direction, angular velocity data in an X, Y, and Z direction, and orientation data in an X, Y and Z directions totaling nine degrees of freedom over the predetermined time interval, the contents of the unit of motion comprising a number of data points x equal to the product of 9 degrees of freedom and the predetermined time interval, each data point having strength information S affiliated therewith, the strength information S indicating how close a content of a unit of motion fits one class compared to another;
- the strength information S is retained after classification for comparison between real time motion data and initial motion data in order to provide information to an algorithm about which motion data to ignore and which motion data to further analyze for qualification.
6 more elements
A method of processing real-time motion data, comprising:
- a) receiving real-time motion data from at least one signal from at least one sensing device [11, 12, 13] configured to sense at least one of acceleration data, angular velocity data and orientation data;
- b) with a processor [20], analyzing real-time motion data in predetermined units of motion, each such unit of motion being an equation over a predetermined time interval that is a function of separate motion data equations for at least acceleration data, angular velocity data and orientation data in a predetermined number of degrees of freedom, each such unit of motion retaining the quality and quantity of the sensed motion data;
- c) with a processor [20], comparing real-time units of motion against initial units of motion classified into classes of motion in computer-readable media [30], such initial units of motion being actual data rather than statistical models of actual data;
- and
- d) with a processor [20], qualifying and quantifying the real-time motion data based upon, at least in part, the comparison, said unit of motion is an equation reflecting the function of separate equations of acceleration data in X, Y, and Z directions, angular velocity data in X, Y, and Z directions and orientation data in X, Y, and Z directions, totaling nine degrees of freedom over the predetermined time interval;
- the contents of the unit of motion, the contents of the unit of motion comprising a number of data points x equal to the product of 9 degrees of freedom and the predetermined time interval, each data point having a strength information affiliated therewith, the strength information S indicating how close a content of a unit of motion fits one class compared to another, the strength information is retained after classification for comparison between real-time motion data and initial motion data in order to provide information to an algorithm about which motion data to ignore and which motion data to further analyze for quantification.
3 more elements
Dependencies are inferred from the claim text.