Device and method to decode volitional motor commands using a biomechanical model for controlling a prosthetic limb

1. A system for decoding volitional motor commands for a prosthetic device, the system comprising:

a prosthetic limb and an actuator connected with the prosthetic limb;

a controller in connection with the prosthetic limb, wherein the controller is configured to implement a control feedback loop that specifies a joint torque for the prosthetic limb; and

one or more processors and a non-transitory computer-readable medium having executable instructions encoded thereon such that when executed, the one or more processors perform operations of:

using a biomechanical model and physical action data, relating biological signal data to kinetic data, the physical action data including at least one of position, joint angle, speed, and acceleration of at least part of a limb, the biological signal data including at least one of recorded neural signals and recorded muscle signals, the kinetic data including at least one of force, power, torque, and stress;

mapping from inferred joint angles onto the biological signal data using a Hidden Markov Model (HMM);

generating joint angle commands to achieve at least one of an intended position and movement of the prosthetic limb, wherein the biomechanical model constrains the joint angle commands to those that are biomechanically feasible within the biomechanical model;

using the HMM to constrain the joint angle commands to be continuous; and

sending the joint angle commands to the controller to cause the joint angle commands to be applied by the actuator connected with the prosthetic limb, wherein the actuator causes mechanical motion of the prosthetic limb based on the at least one of the intended position and movement of the prosthetic limb.

2. The system as set forth in claim 1 , wherein the biomechanical model is used for prediction of muscle activity related to movement of a missing limb.

3. The system as set forth in claim 1 , wherein the one or more processors further perform an operation of using an output of a biomechanical simulation to determine the joint angle commands.

4. The system as set forth in claim 1 , wherein a biomechanical simulation provides training data for a machine learning method that maps a set of recorded muscle signals onto the joint angle commands.

5. The system as set forth in claim 1 , wherein the biomechanical model is personalized for an individual user having the prosthetic limb.

6. The system as set forth in claim 1 , wherein the prosthetic limb further comprises at least one sensor configured to provide feedback of a limb posture of the prosthetic limb in the form of a set of joint angles measured by the at least one sensor.

7. The system as set forth in claim 6 , wherein the feedback of the limb posture is sent to the controller which is configured to compensate for any error between the joint angle commands and the measured set of joint angles.

8. A computer-implemented method for decoding volitional motor commands for a prosthetic device, comprising:

an act of causing one or more processors to execute instructions stored on a non-transitory memory such that upon execution, the one or more processors perform operations of:

using a biomechanical model and physical action data, relating biological signal data to kinetic data, the physical action data including at least one of position, joint angle, speed, and acceleration of at least part of a limb, the biological signal data including at least one of recorded neural signals and recorded muscle signals, the kinetic data including at least one of force, power, torque, and stress;

mapping from inferred joint angles onto the biological signal data using a Hidden Markov Model (HMM);

generating joint angle commands to achieve at least one of an intended position and movement of a prosthetic limb having an actuator connected therewith generating joint angle commands to achieve at least one of an intended position and movement of the prosthetic limb, wherein the biomechanical model constrains the joint angle commands to those that are biomechanically feasible within the biomechanical model;

using the HMM to constrain the joint angle commands to be continuous; and

sending the joint angle commands to a controller in connection with the prosthetic limb, wherein the controller is configured to implement a control feedback loop that specifies a joint torque for the actuator, causing the joint angle commands to be applied by the actuator connected with the prosthetic limb, wherein the actuator causes mechanical motion of the prosthetic limb based on the at least one of the intended position and movement of the prosthetic limb.

9. The method as set forth in claim 8 , wherein the biomechanical model is used for prediction of muscle activity related to movement of a missing limb.

10. The method as set forth in claim 8 , wherein the one or more processors further perform an operation of using an output of a biomechanical simulation to determine the joint angle commands.

11. The method as set forth in claim 8 , wherein a biomechanical simulation provides training data for a machine learning method that maps a set of recorded muscle signals onto the joint angle commands.

12. The method as set forth in claim 8 , wherein the biomechanical model is personalized for an individual user having the prosthetic limb.

13. A computer program product for decoding volitional motor commands for a prosthetic device, the computer program product comprising:

computer-readable instructions stored on a non-transitory computer-readable medium that are executable by a computer having one or more processors for causing the processor to perform operations of:

using a biomechanical model and physical action data, relating biological signal data to kinetic data, the physical action data including at least one of position, joint angle, speed, and acceleration of at least part of a limb, the biological signal data including at least one of recorded neural signals and recorded muscle signals, the kinetic data including at least one of force, power, torque, and stress;

mapping from inferred joint angles onto the biological signal data using a Hidden Markov Model (HMM);

generating joint angle commands to achieve at least one of an intended position and movement of a prosthetic limb having an actuator connected therewith generating joint angle commands to achieve at least one of an intended position and movement of the prosthetic limb, wherein the biomechanical model constrains the joint angle commands to those that are biomechanically feasible within the biomechanical model;

using the HMM to constrain the joint angle commands to be continuous; and

sending the joint angle commands to a controller in connection with the prosthetic limb, wherein the controller is configured to implement a control feedback loop that specifies a joint torque for the actuator, causing the joint angle commands to be applied by the actuator connected with the prosthetic limb, wherein the actuator causes mechanical motion of the prosthetic limb based on the at least one of the intended position and movement of the prosthetic limb.

14. The computer program product as set forth in claim 13 , wherein the biomechanical model is used for prediction of muscle activity related to movement of a missing limb.

15. The computer program product as set forth in claim 13 , further comprising instructions for causing the one or more processors to perform an operation of using an output of a biomechanical simulation to determine the joint angle commands.

16. The computer program product as set forth in claim 13 , wherein a biomechanical simulation provides training data for a machine learning method that maps a set of recorded muscle signals onto the joint angle commands.

17. The computer program product as set forth in claim 13 , wherein the biomechanical model is personalized for an individual user having the prosthetic limb.