Medical robotic system with sliding mode control

• US 7,453,227 B2
• Filed: 12/20/2006
• Issued: 11/18/2008
• Est. Priority Date: 12/20/2005
• Status: Active Grant

• Current Assignee: Intuitive Surgical Operations, Inc. (Intuitive Surgical Holdings, LLC)
• Original Assignee: Intuitive Surgical Incorporated (Intuitive Surgical Holdings, LLC)
• Inventors: Prisco, Giuseppe M., Larkin, David Q
• Primary Examiner(s): Ro; Bentsu
• Application Number:
  US11/613,962
• Publication Number:
  US 20070138992A1
• Time in Patent Office: 699 Days
• Field of Search: 318/568.11, 318/568.2, 318/568.21, 318/568.22, 318/573, 318/609, 318/610, 318/619, 318/621, 318/632, 700/245, 700/250, 700/257, 700/264, 600/101, 600/160, 606/1, 606/130
• US Class Current: 318/568.11
• CPC Class Codes: A61B 2034/305 Details of wrist mechanisms at distal ends of robotic arms
• CPC Class Codes: A61B 34/30 Surgical robots
• CPC Class Codes: A61B 34/37 Master-slave robots A61B34/35 takes precedence
• CPC Class Codes: A61B 34/70 Manipulators specially adapted for use in surgery
• CPC Class Codes: A61B 90/361 Image-producing devices, e.g. surgical cameras
• CPC Class Codes: B25J 9/1646 variable structure system, sliding mode control
• CPC Class Codes: B25J 9/1689 Teleoperation
• CPC Class Codes: G05B 2219/42347 Switch to a saturation control signal if deviation from switch line is too large
• CPC Class Codes: G05B 2219/42351 PIVSC proportional integral compensated vsc
• CPC Class Codes: G05B 2219/42352 Sliding mode controller SMC, select other gain
• CPC Class Codes: G05B 2219/45123 Electrogoniometer, neuronavigator, medical robot used by surgeon to operate

Abstract

A medical robotic system has a joint coupled to medical device or a slave manipulator or robotic arm adapted to hold and/or move the medical device for performing a medical procedure, and a control system for controlling movement of the joint according to user manipulation of a master manipulator. The control system includes at least one joint controller having a sliding mode control for reducing stick-slip behavior on its controlled joint during fine motions of the joint. The sliding mode control computes a distance to a sliding surface, computes a reaching law gain, and processes the distance and reaching law gain to generate a sliding mode control action that is in absolute value less that a maximum desired feedback control action. The sliding mode control action is then further processed to generate a feedback torque command for the joint motor.

Claims

1. A method for controlling a joint in a medical robotic system in response to user manipulation of a master input device, comprising:

computing a distance from a sliding surface using at least a position error associated with commanded movement of the joint;

computing a reaching law gain as a function of the absolute value of the distance;

computing a current state for an integrator using a prior state of the integrator and the distance;

computing a product of the reaching law gain and the distance;

limiting the product to be less than a maximum control action;

modifying the current state of the integrator using the limited product so as to limit a sum of the product and the current state of the integrator to be less than the maximum control action;

computing a sliding mode control action by summing the product and the modified current state of the integrator, and

processing the sliding mode control action to generate one or more command signals to drive the joint according to the user manipulation of the master input device.

2. The method according to claim 1, wherein the computation of the distance from the sliding surface also uses a velocity error associated with the commanded movement of the robotic arm.

3. The method according to claim 2, wherein the computation of the distance from the sliding surface involves a linear function of the position error and the velocity error.

4. The method according to claim 2, wherein the computation of the distance from the sliding surface involves a quadratic function of the position error and the velocity error.

5. The method according to claim 1, wherein the computation of the reaching law gain uses a piecewise linear function of the absolute value of the distance.

6. The method according to claim 1, wherein the computation of the reaching law gain uses a power rate function of the absolute value of the distance.

7. The method according to claim 1, wherein the modification of the current state of the integrator comprises modifying the current state of the integrator so as to avoid exceeding a specification for a motor used to move the joint.

8. The method according to claim 1, wherein the processing of the sliding mode control action includes filtering the sliding mode control action to avoid abrupt changes.

9. The method according to claim 1, wherein the processing of the sliding mode control action further includes dynamically limiting such control action using an acceleration factor derived from the commanded movement of the joint.

10. A medical robotic system having a master input device, a plurality of joints coupled to corresponding of a medical device or a robotic arm for moving the medical device, and a processor responsive to user manipulation of the master input device to control movement of one of the plurality of joints, wherein the improvement comprises configuration of the processor to control such movement by:

computing a distance from a sliding surface using at least a position error associated with commanded movement of the joint;

computing a reaching law gain as a function of the absolute value of the distance;

computing a current state for an integrator using a prior state of the integrator and the distance;

computing a product of the reaching law gain and the distance;

limiting the product to be less than a maximum control action;

modifying the current state of the integrator using the limited product value so as to limit a sum of the product and the current state of the integrator to be less than the maximum control action;

computing a sliding mode control action by summing the product and the modified current state of the integrator; and

processing the sliding mode control action to generate one or more command signals to drive the joint according to the user manipulation of the master input device.

11. The medical robotic system according to claim 10, wherein the processor is configured to compute the distance from the sliding surface by also using a velocity error associated with the commanded movement of the joint.

12. The medical robotic system according to claim 11, wherein the processor is configured to compute the distance from the sliding surface using the following function:

σ

=λ

·

(QC−

QL)+(VC−

VL),

wherein σ

represents the distance, λ

is a parameter value, (QC−

QL) is the position error, and (VC−

VL) is the velocity error.

13. The medical robotic system according to claim 11, wherein the processor is configured to compute the distance from the sliding surface using the following function:

σ

=λ

·

(QC−

QL)+(VC−

VL)²,

wherein σ

represents the distance, λ

is a parameter value, (QC−

QL) is the position error, and (VC−

VL) is the velocity error.

14. The medical robotic system according to claim 10, wherein the processor is configured to compute the reaching law gain using the following reaching law:

15. The medical robotic system according to claim 10, wherein the processor is configured to compute the reaching law gain using the following reaching law:

16. The medical robotic system according to claim 10, wherein the processor is configured to limit the modified current state of the integrator according to the following function:

INT(kT_s)=Sat(KI·

T_s[σ

+INT((k−

1)·

Ts)]+Sat(GAIN_RL·

σ

))−

Sat(GAIN _RL·

σ

)

wherein INT((k−

1)·

Ts) represents the prior state of the integrator, T_MAX is a maximum torque for a motor being used to move the joint, GAIN_RL is the reaching law gain, σ

is the distance, and KI is an integrator gain for the integrator.

17. The medical robotic system according to claim 10, wherein the processor is configured to calculate the current state of the integrator using a sum of the prior state of the integrator and the distance, and multiplying the sum by an integrator gain value.

18. The medical robotic system according to claim 10, wherein the processor is configured to process the sliding mode control action by:

filtering the sliding mode control action;

computing a summed value by summing the filtered sliding mode control action and a velocity error factor derived from the commanded movement of the joint; and

dynamically limiting the summed value using an acceleration factor derived from the commanded movement of the joint.

19. The medical robotic system according to claim 18, wherein the processor is configured to filter the sliding mode control action using a low pass filter function.

20. The medical robotic system according to claim 18, wherein the processor is configured to dynamically limit the absolute value of the summed value using the following function:

PDMAX=PDMAX_STATIC+|ACC·

AVG_INERTIAL_GAIN|,

wherein PDMAX represents a maximum value allowed for the summed value, PDMAX_STATIC is a maximum static value allowed for the summed value, and |ACC·

AVG_INERTIAL_GAIN| is an absolute value for the acceleration factor.

21. A method for controlling a joint coupled to a medical device or a robotic arm adapted to move the medical device, in response to user manipulation of a master input device, comprising;

computing a distance from a sliding surface using at least a position error associated with commanded movement of the joint;

computing a reaching law gain as a function of the absolute value of the distance;

computing a sliding mode control action by multiplying the reaching law gain and the distance;

limiting the sliding mode control action to be less than a maximum control action; and

processing the limited sliding mode control action to generate one or more command signals to drive the joint according to the user manipulation of the master input device.

22. The method according to claim 21, wherein the processing of the sliding mode control action comprises:

filtering the sliding mode control action;

computing a summed value by summing the filtered sliding mode control action and a velocity error factor derived from the commanded movement of the joint; and

dynamically limiting the summed value using an acceleration factor derived from the commanded movement of the joint.

23. The method according to claim 22, wherein the dynamic limiting of the summed value uses the following function:

PDMAX=PDMAX_STATIC+|ACC·

AVG_INERTIAL_GAIN|,

wherein PDMAX represents a maximum value allowed for the summed value, PDMAX_STATIC is a maximum static value allowed for the summed value, and |ACC·

AVG_INERTIAL_GAIN| is an absolute value for the acceleration factor.

24. The method according to claim 21, wherein the computation of the distance from the sliding surface also uses a velocity error associated with the commanded movement of the joint.

25. The method according to claim 23, wherein the computation of the distance from the sliding surface involves a linear function of the position error and the velocity error.

26. The method according to claim 24, wherein the linear function is as follows:

σ

=λ

·

(QC−

QL)+(VC−

VL),

wherein σ

represents the distance, λ

is a parameter value, (QC−

QL) is the position error, and (VC−

VL) is the velocity error.

27. The method according to claim 24, wherein the computation of the distance from the sliding surface involves a quadratic function of the position error and the velocity error.

28. The method according to claim 27, wherein the quadratic function is as follows:

σ

=λ

·

(QC−

QL)+(VC−

VL)²,

wherein σ

represents the distance, λ

is a parameter value, (QC−

QL) is the position error, and (VC−

VL) is the velocity error.

29. The method according to claim 21, wherein the computation of the reaching law gain uses a piecewise linear function of the absolute value of the distance.

30. The method according to claim 29, wherein the linear reaching law is as follows:

31. The method according to claim 21, wherein the computation of the reaching law gain uses a power rate function of the absolute value of the distance.

32. The method according to claim 31, wherein the power rate reaching law is as follows:

33. A medical robotic system having a master input device, a joint coupled to a medical device or a robotic arm adapted to move the medical device, and a processor responsive to user manipulation of the master input device to control movement of the joint, wherein the improvement comprises the processor being configured to control such movement by:

computing a distance from a sliding surface using at least a position error associated with commanded movement of the joint;

computing a reaching law gain using the distance;

computing a sliding mode control action by multiplying the reaching law gain and the distance;

limiting the sliding mode control action to be less than a maximum control action,

processing the limited sliding mode control action to generate one or more command signals to drive the joint according to the user manipulation of the master input device.

34. The medical robotic system according to claim 33, wherein the processor is configured to process the sliding mode control action by:

filtering the sliding mode control action;

computing a summed value by summing the filtered sliding mode control action and a velocity error factor derived from the commanded movement of the joint; and

dynamically limiting the summed value using an acceleration factor derived from the commanded movement of the joint.

35. The medical robotic system according to claim 34, wherein the processor is configured to dynamically limit the summed value using the following function:

PDMAX=PDMAX_STATIC+|ACC·

AVG_INERTIAL_GAIN|,

wherein PDMAX represents a maximum value allowed for the summed valuer PDMAX_STATIC is a maximum static value allowed for the summed value, and |ACC·

AVG_INERTIAL_GAIN|, is an absolute value for the acceleration factor.

36. The medical robotic system according to claim 33, wherein the processor is configured to compute the distance from the sliding surface by also using a velocity error associated with the commanded movement of the joint.

37. The medical robotic system according to claim 36, wherein the processor is configured to compute the distance from the sliding surface using a linear function of the position error and the velocity error.

38. The medical robotic system according to claim 36, wherein the processor is configured to compute the distance from the sliding surface using a quadratic function of the position error and the velocity error.

39. The medical robotic system according to claim 36, wherein the processor is configured to compute the reaching law gain using a piecewise linear function of the absolute value of the distance.

40. The medical robotic system according to claim 36, wherein the processor is configured to compute the reaching law gain using a power rate function of the absolute value of the distance.

41. A medical robotic system comprising:

a joint coupled to a medical device or a robotic arm adapted to move the medical device; and

a controller configured to control movement of the joint by;

computing a distance from a sliding surface using the following equation;

σ

=λ

·

(QC−

QL)+(VC−

VL)²,

wherein σ

represents the distance, λ

is a parameter value, (QC−

QL) is a position error and (VC−

VL) is a velocity error associated with commanded movement of the joint;

computing a sliding mode control action based at least in part on the distance to the sliding surface; and

causing the joint to be moved at least partially in response to the sliding mode control action.

42. A medical robotic system comprising:

a joint coupled to a medical device or a robotic arm adapted to move the medical device; and

a controller configured to control movement of the joint by;

computing a distance from a sliding surface;

computing a reaching law gain using the following equation;