Neuromuscular Model-Based Sensing And Control Paradigm For A Robotic Leg

• US 20130310979A1
• Filed: 04/18/2013
• Published: 11/21/2013
• Est. Priority Date: 04/18/2012
• Status: Active Grant

• Current Assignee: Massachusetts Institute of Technology
• Original Assignee: Massachusetts Institute of Technology
• Inventors: Herr, Hugh M., Endo, Ken, Krishnaswamy, Pavitra, Markowitz, Jared, Eilenberg, Michael Frederick, Wang, Jing
• Granted Patent: US 9,221,177 B2
• Time in Patent Office: Days
• US Class Current: 700/258
• CPC Class Codes: B25J 9/1694 characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
• CPC Class Codes: B62D 57/032 with alternately or sequentially lifted supporting base and legs; with alternately or sequentially lifted feet or skid B62D57/024 takes precedence
• CPC Class Codes: G05B 2219/40324 Simulation, modeling of muscle, musculoskeletal dynamical system
• CPC Class Codes: Y10S 901/01 Mobile robot

Abstract

A neuromuscular model-based controller for a robotic limb having at least one joint includes a neuromuscular model having a muscle model, muscle geometry and reflex feedback loop to determine at least one torque or impedance command to be sent to the robotic limb. One or more parameters that determine relation between feedback data and activation of the muscle model are adjusted consequent to sensory data from at least one of an intrinsic sensor and an extrinsic sensor. A controller in communication with the neuromuscular model is configured to receive the at least one torque or impedance command and controls at least one of position, torque and impedance of the robotic limb joint.

Claims

1. A neuromuscular model-based controller for controlling at least one robotic limb joint of a robotic limb, the controller comprising:

a) a neuromuscular model including a muscle model, muscle geometry and a reflex feedback loop, wherein the reflex feedback loop conveys feedback data of at least one of muscle force, muscle length and muscle contractile velocity of the muscle model, to thereby adjust activation of the muscle model, and wherein one or more parameters that determine the relation between the feedback data and activation of the muscle model are adjusted consequent to sensory data from at least one of an intrinsic sensor and an extrinsic sensor, the neuromuscular model employing the muscle model and the muscle geometry to determine at least one of a torque command and an impedance command; and

b) a control system in communication with the neuromuscular model, whereby the control system receives at least one of the torque command and the impedance command from the neuromuscular model and controls at least one of position, torque and impedance of the robotic limb joint.

2. The neuromuscular model-based controller of claim 1, wherein the at least one sensor detects at least one of a velocity of the robotic limb and a terrain underlying the robotic limb.

3. The neuromuscular model-based controller of claim 1, wherein the sensor is an extrinsic sensor.

4. The neuromuscular model-based controller of claim 3, wherein the extrinsic sensor detects brain signals.

5. The neuromuscular model-based controller of claim 3, wherein the extrinsic sensor detects electromyographic signals.

6. The neuromuscular model-based controller of claim 5, wherein the extrinsic sensor detects surface electromyographic signals.

7. The neuromuscular model-based controller of claim 6, wherein the extrinsic sensor detects surface electromyographic signals from a residual limb.

8. The neuromuscular model-based controller of claim 1, wherein the sensor is an intrinsic sensor.

9. The neuromuscular model-based controller of claim 8, wherein the intrinsic sensor includes at least one member of the group consisting of a digital encoder, a hall-effect sensor, a torque sensor, and an inertial measurement unit.

10. The neuromuscular model-based controller of claim 1, wherein the feedback data includes muscle force.

11. The neuromuscular model-based controller of claim 1, wherein the feedback data includes muscle length.

12. The neuromuscular model-based controller of claim 1, wherein the feedback data includes muscle contractile velocity.

13. The neuromuscular model-based controller of claim 1, wherein the relation between the feedback data and activation of the muscle model is linear.

14. The neuromuscular model-based controller of claim 1, wherein the relation between the feedback data and activation of the muscle model is non-linear.

15. The neuromuscular model-based controller of claim 1, wherein the parameters include a pre-stim parameter, force and state thresholds, gains, delays, and exponential factors.

16. The neuromuscular model-based controller of claim 1, wherein the control system includes a feed forward gain, a lead compensator and a friction compensator to control at least one of position, torque and impedance of the robotic limb joint.

17. The neuromuscular model-based controller of claim 1, wherein the control system further includes a motor controller for driving an actuator of the robotic limb joint with a current command.

18. The neuromuscular model-based controller of claim 1, further including at least one sensor at the robotic limb, wherein the sensor includes at least one member of the group consisting of an angular joint displacement sensor, a velocity sensor, and an inertial measurement unit, to thereby generate joint state data.

19. The neuromuscular model-based controller of claim 18, wherein the joint state data includes a measured robotic limb joint angle and joint angular velocity measured by the at least one sensor.

20. The neuromuscular model-based controller of claim 19, wherein the control system further includes a parallel spring model that employs the measured robotic limb joint angle to thereby estimate at least one of the torque command and the impedance command.

21. The neuromuscular model-based controller of claim 1, wherein the muscle model includes a contractile element and wherein the muscle model determines muscle force using contractile element length, a contractile element velocity and an activation input from the reflex feedback loop.

22. The neuromuscular model-based controller of claim 21, wherein the muscle model includes at least one of a bilinear muscle model, a Hill-type muscle model and a clutch-spring model.

23. The neuromuscular model-based controller of claim 22, wherein the reflex feedback loop is configured as a local feedback loop, whereby the reflex feedback loop is configured to receive muscle feedback information from the muscle model and to provide the activation input to the muscle model.

24. The neuromuscular model-based controller of claim 23, wherein the reflex feedback loop comprises a positive force feedback.

25. The neuromuscular model-based controller of claim 23, wherein the reflex feedback loop is configured to mimic a stretch reflex of an intact human muscle.

26. The neuromuscular model-based controller of claim 1, wherein the neuromuscular model and the control system are configured to control the robotic limb, wherein the robotic limb is a leg or part of a leg, and wherein the neuromechanical controller further includes a finite state machine synchronized to the leg gait cycle, the finite state machine being configured to receive data from at least one of the sensors and to determine a gait phase of the robotic leg using the data received.

27. The neuromuscular model-based controller of claim 16, wherein the neuromuscular model and the control system are configured to control a robotic leg comprising a knee joint.

28. The neuromuscular model-based controller of claim 26, wherein the neuromuscular model and the control system are configured to control a robotic leg comprising an ankle joint.

29. The neuromuscular model-based controller of claim 28, the robotic leg further comprising a knee joint.

30. The neuromuscular model-based controller of claim 29, the robotic leg further comprising a hip joint.

31. A method for controlling a robotic limb that includes at least one robotic limb joint, the method comprising the steps of:

a) transmitting a measured joint state of the robotic limb to a neuromuscular model, the neuromuscular model including a muscle model, muscle geometry and a reflex feedback loop, whereby at least one of a muscle force, muscle length and muscle contractile velocity is conveyed by the reflex feedback loop as feedback data to thereby adjust activation of the muscle model, and wherein one or more parameters that determine the relation between the feedback data and activation of the muscle model are further adjusted consequent to at least one of velocity of the robotic limb and a terrain underlying the limb, whereby the muscle length, muscle contractile velocity, and the muscle activation generate a muscle force, which, together with the muscle geometry, determines at least one of a torque command and an impedance command; and

b) transmitting at least one of the torque command and the impedance command to a control system,

whereby the control system receives at least one of the torque command and the impedance command from the neuromuscular model and controls at least one of position, torque and impedance of the robotic limb joint.

32. The method of claim 31, wherein the reflex feedback loop applies a delay and a gain to at least one of the muscle force, muscle length and muscle contractile velocity, which is added to an offset pre-stimulation to thereby obtain a neural stimulation signal that is employed to adjust activation of the muscle model.

33. The method of claim 32, wherein at least one of the delay and gain is adjusted consequent to at least one of the velocity of the robotic limb and the underlying terrain.

34. The method of claim 33, wherein the robotic limb includes a robotic ankle, and further including the steps of:

i) switching the reflex feedback loop between an on position and an off position, so that the reflex feedback loop is operating when an individual wearing the robotic limb is in a stance phase; and

ii) applying a shaping filter to the neural stimulation signal.

35. The method of claim 31, further including the step of communicating data from at least one sensor to the reflex feedback loop.

36. The method of claim 35, wherein the sensor measures velocity of the robotic limb.

37. The method of claim 35, wherein the sensor is extrinsic.

38. The method of claim 37, wherein the sensor detects brain signals.

39. The method of claim 37, wherein the sensor detects electromyographic signals.

40. The method of claim 39, wherein the sensor detects surface electromyographic signals.

41. The method of claim 40, wherein the sensor detects surface electromyographic data from a residual limb.

42. The method of claim 35, wherein the sensor is intrinsic.

43. The method of claim 42, wherein the sensor includes at least one member of the group consisting of a digital encoder, a force sensor, a hall-effect sensor, and an inertial measurement unit.

44. The method of claim 31, wherein the feedback data includes muscle force.

45. The method of claim 31, wherein the feedback data includes muscle length.

46. The method of claim 31, wherein the feedback data includes muscle contractile velocity.

47. The method of claim 31, wherein the relation between the feedback data and activation of the muscle model is linear.

48. The method of claim 31, wherein the relation between the feedback data and activation of the muscle model is non-linear.

49. The method of claim 31, wherein the parameters include a pre-stim parameter, force and state thresholds, gains, delays, and exponential factors.

50. The method of claim 31, wherein the control system includes a feed forward gain, a lead compensator and a friction compensator to control at least one of position, torque and impedance of the robotic limb joint.

51. The method of claim 31, wherein the control system includes a motor controller for driving an actuator of the robotic limb joint with a current command.

52. The method of claim 31, further including the step of employing at least one sensor at the robotic limb, wherein the sensor includes at least one member of the group consisting of an angular joint displacement sensor, a velocity sensor, a torque sensor, an impedance sensor and an inertial measurement unit, to thereby generate intrinsic data.

53. The method of claim 52, wherein the joint state data includes a measured robotic limb joint angle and joint angular velocity measured by the at least one sensor.

54. The method of claim 53, wherein the muscle geometry is configured to determine a muscle moment arm and a muscle tendon length using the measured joint angle.

55. The method of claim 54, wherein the muscle model includes at least one muscle-tendon unit having a contractile element and a series elastic element, and wherein the muscle model includes a contractile element, and wherein the muscle model determines muscle force using contractile element length, contractile element velocity and an activation input from the reflex feedback loop.

56. The method of claim 55, wherein the muscle model includes at least one of a bilinear muscle model, a Hill-type muscle model and a clutch-spring model.

57. The method of claim 56, wherein the reflex feedback loop is configured as a local feedback loop, whereby the reflex feedback loop is configured to receive muscle feedback information from the muscle model and to provide the activation input to the muscle model.

58. The method of claim 57, wherein the muscle feedback is a positive force feedback.

59. The method of claim 57, wherein the reflex feedback loop is configured to mimic a stretch reflex of an intact human muscle.

60. The method of claim 52, wherein the at least one sensor is mounted to the robotic limb, and wherein the neuromuscular model and the control system are configured to control the robotic limb, wherein the robotic limb is a leg or a part of a leg, and wherein the neuromechanical controller further includes a finite state machine synchronized to the leg gait cycle, the finite state machine being configured to receive the feedback data from the at least one sensor and to determine a gait phase of the robotic leg using the feedback data received.

61. The method of claim 31, wherein the neuromuscular model and the control system are configured to control a robotic leg comprising a knee joint.

62. The method of claim 31, wherein the neuromuscular model and the control system are configured to control a robotic leg comprising an ankle joint.

63. The method of claim 62, the robotic leg further comprising a knee joint.

64. The method of claim 63, the robotic leg further comprising a hip joint.

65. A neuromuscular model-based controller for controlling at least one robotic limb joint of a robotic limb, the controller comprising:

a) at least one extrinsic sensor that detects at least one extrinsic signal of a subject wearing the robotic limb;

b) a neuromuscular model in communication with the at least one sensor, the neuromuscular model including a reflex feedback loop that acts upon intrinsic sensory data based upon parameters that are adjusted in response to the at least one extrinsic signal from the sensor, the neuromuscular model thereby determining at least one of a torque command and an impedance command; and

c) a control system in communication with the neuromuscular model, whereby the control system receives at least one of the torque command and the impedance command from the neuromuscular model and controls at least one of position, torque and impedance of the robotic limb joint.

66. The neuromuscular model-based controller of claim 65, wherein the neuromuscular model includes a muscle model, whereby parameters that determine the relation between muscle feedback data, including at least one of muscle length, muscle contractile velocity and muscle force, and activation of the muscle model are adjusted consequent to the at least one extrinsic signal of the extrinsic sensor.

67. The neuromuscular model-based controller of claim 65, wherein the at least one extrinsic signal is an electromyographic signal.