Method and system to activate limb movement in amputees

1 . A limb activation system comprising:

at least one sensor configured to be coupled to a prosthetic arm of a user and configured to monitor movement of the user;

at least one motor configured to be coupled to the prosthetic arm, the at least one motor configured to cause the prosthetic arm to move in a cyclical motion;

a clutch configured to disengage the at least one motor; and

a computing device operatively coupled to the at least one sensor, the at least one motor, and the clutch, the computing device comprising a memory and at least one processor, wherein the memory is configured to store computer-readable instructions and predetermined arm movement data patterns, wherein the predetermined arm movement data patterns include a kinematic relationship between walking speeds and elbow joint range-of-motion, wherein upon execution of the computer-readable instructions, the at least one processor is configured to:

cause a motion classifier to determine, based on data from one or more sensors of the at least one sensor, whether the user is engaged in walking, whether the user is engage in running, and whether the user is engaged in neither walking nor running, wherein the motion classifier is configured to generate an on signal when the motion classifier determines the user is engaged in walking or running, and wherein the motion classifier is configured to generate an off signal when the motion classifier determines the user is engaged in neither walking nor running;

cause a cadence detector to determine, based on data from one or more sensors of the at least one sensor, a step cadence of the user;

cause a trajectory generator to determine, based on the step cadence of the user and the predetermined arm movement data patterns, a cyclical trajectory of the prosthetic arm worn by the user when the motion classifier generates the on signal;

not cause the trajectory generator to determine the cyclical trajectory when the motion classifier generates the off signal,

cause the motor to move the prosthetic arm of the user in the cyclical motion based on the cyclical trajectory,

determine a difference between a desired prosthetic position based on the cyclical trajectory and an actual prosthetic position based on data from one or more sensors of the at least one sensor,

cause the motor to move the prosthetic arm of the user in the cyclical motion based on the determined difference between the desired prosthetic position and the actual prosthetic position, and

cause the clutch to disengage the motor when the motion classifier generates the off signal.

2 . The system of claim 1 , further comprising a sensory stimulator configured to be mounted to the user, wherein the on signal triggers the sensory stimulator to alert the user to perform movement during the walking or running.

3 . The system of claim 2 , wherein the sensory stimulator comprises a vibration stimulator, a pressure stimulator, a skin stretching stimulator, or an auditory stimulator.

4 . The system of claim 1 , wherein the computing device determines the step cadence of the walking or running based on the determined movement of the user, and wherein the on signal is generated based at least in part on the cadence of the walking or running.

5 . The system of claim 4 , wherein the at least one sensor generates a plurality of instantaneous acceleration signals, and wherein the computing device is configured to identify repetitive features in the instantaneous acceleration signals and a timing in between the repetitive features to determine the cadence.

6 . The system of claim 1 , wherein the at least one sensor comprises at least one of an accelerometer or a gyroscope.

7 . The system of claim 1 , wherein the at least one sensor is configured to provide frequency data, and wherein the determination that the user is engaged in walking or running is based on the frequency data.

8 . The system of claim 1 , wherein the computing device determines a window signal power based on the monitored movement, and wherein the determination that the user is engaged in walking or running is based on the window signal power.

9 . The system of claim 1 , wherein the at least one sensor comprises an accelerometer that generates an acceleration signal, and wherein the computing device is configured to process the acceleration signal to determine a single-sided frequency spectrum, wherein the computing device is configured to process the acceleration signal using a fast Fourier transform (FFT) or a machine learning algorithm.

10 . The system of claim 9 , wherein the computing device determines a proportion of the single-sided frequency spectrum that occurs in a time window by generating a first sum of magnitudes of frequency components in the time window and dividing the first sum by a second sum of magnitudes of frequency components across all frequencies.

11 . A limb activation system comprising:

at least one sensor configured to be coupled to a prosthetic arm of a user and configured to monitor movement of the user wearing the prosthetic arm; and

a computing device operatively coupled to the at least one sensor, the computing device comprising a memory and at least one processor, wherein the memory is configured to store computer-readable instructions and predetermined arm movement data patterns, wherein the predetermined arm movement data patterns includes a kinematic relationship between walking speeds and elbow joint range-of-motion, wherein upon execution of the computer-readable instructions, the at least one processor is configured to:

cause a motion classifier to determine, based on data from one or more sensors of the at least one sensor, whether the user is engaged in walking, whether the user is engaged in running, and whether the user is engaged in neither walking nor running, wherein the motion classifier is configured to generate an on signal when the motion classifier determines the user is engaged in walking or running, and wherein the motion classifier is configured to generate an off signal when the motion classifier determines the user is engaged in neither walking nor running;

cause a cadence detector to determine, based on data from one or more sensors of the at least one sensor, a step cadence of the user;

cause a trajectory generator to determine, based on the step cadence of the user and the predetermined arm movement data patterns, a cyclical trajectory of the prosthetic worn by the user when the motion classifier generates the on signal;

not cause the trajectory generator to determine the cyclical trajectory when the motion classifier generates the off signal, and

cause movement of the prosthetic of the user in the cyclical motion based on the cyclical trajectory,

determine a difference between a desired prosthetic position based on the cyclical trajectory and an actual prosthetic position based on data from one or more sensors of the at least one sensor, and

cause the prosthetic arm of the user to move in the cyclical motion based on the difference.

12 . The system of claim 11 , further comprising a sensory stimulator configured to be mounted to the user, wherein the on signal triggers the sensory stimulator to alert the user to perform movement during the walking or running.

13 . The system of claim 11 , wherein the computing device determines the step cadence of the walking or running based on the determined movement of the user, and wherein the on signal is generated based at least in part on the cadence of the walking or running.

14 . The system of claim 13 , wherein the sensor generates a plurality of instantaneous acceleration signals, and wherein the computing device is configured to identify repetitive features in the instantaneous acceleration signals and a timing in between the repetitive features to determine the cadence.

15 . The system of claim 11 , wherein the sensor comprises at least one of an accelerometer or a gyroscope.

16 . The system of claim 11 , wherein the at least one sensor provides frequency data, and wherein the determination that the user is engaged in walking or running is based on the frequency data.

17 . The system of claim 11 , wherein the computing device determines a window signal power based on the monitored movement, and wherein the determination that the user is engaged in walking or running is based on the window signal power.

18 . The system of claim 11 , wherein the at least one sensor comprises an accelerometer that generates an acceleration signal, and wherein the computing device is configured to process the acceleration signal to determine a single-sided frequency spectrum, wherein the computing device is configured to process the acceleration signal using a fast Fourier transform (FFT) or a machine learning algorithm.

19 . The system of claim 18 , wherein the computing device determines a proportion of the single-sided frequency spectrum that occurs in a time window by generating a first sum of magnitudes of frequency components in the time window and dividing the first sum by a second sum of magnitudes of frequency components across all frequencies.