Nine-degree-of-freedom upper limb rehabilitation robot

Abstract

The present invention provides a nine-degree-of-freedom upper limb rehabilitation robot, which adds two active degrees of freedom of the sternoclavicular joint to simulate the movement of the shoulder girdle, so that the rehabilitation robot can better match the movement of the human upper limb. The redundant control algorithm classifies tasks based on the null space projection of the Jacobian matrix. Low-priority tasks do not affect the execution of high-priority tasks. The redundant degrees of freedom of the upper limb exoskeleton can be fully utilized to ensure coordinated movement between the joints of the robotic arm, thereby improving the treatment effect.

Description

Technical Field

[0001] The present invention relates to the technical field of rehabilitation robots, and in particular to a nine-degree-of-freedom upper limb rehabilitation robot. Background Art

[0002] Existing upper limb rehabilitation robots typically simplify the upper limb into six or seven active degrees of freedom (DOFs), and the shoulder joint (glenohumeral joint) into a 3-DOF ball-and-spherical joint. It is assumed that shoulder movement is entirely performed by the glenohumeral joint. However, shoulder movement is actually a function of both the glenohumeral joint and the shoulder girdle. For example, when the shoulder is abducted 180 degrees, the glenohumeral joint moves 120 degrees, and the shoulder girdle rotates upward 60 degrees. Together, these two joints complete the 180-degree abduction, with a motion ratio of approximately 2:1 (scapulohumeral rhythm). This results in a mismatch between the robot and the user's upper limb movements.

[0003] Existing upper limb rehabilitation robots do not match the user's upper limb movements, which can easily cause compensatory movements that affect the treatment effect and even cause secondary injuries. Summary of the Invention

[0004] The problem solved by the present invention is the mismatch between the existing upper limb rehabilitation robot and the upper limb movements of the user.

[0005] To solve the above problems, the present invention provides a nine-degree-of-freedom upper limb rehabilitation robot, which includes a rehabilitation training module, a priority-based redundant control algorithm module, an upper limb exoskeleton and an exoskeleton controller; the upper limb exoskeleton includes nine active joint degrees of freedom, namely, sternoclavicular joint extension / retraction, sternoclavicular joint elevation / depression, glenohumeral joint abduction / adduction, glenohumeral joint flexion / extension, glenohumeral joint internal rotation / external rotation, elbow flexion / extension, forearm pronation / supination, wrist dorsiflexion / palmar flexion, and wrist ulnar flexion / radial flexion; the rehabilitation training module is used to generate a rehabilitation training scenario, which includes multiple motion targets; the redundant control algorithm module is used to classify the tasks obtained by decomposing the motion targets based on the null space projection of the Jacobian matrix and determine the control parameters of the classified tasks; the control parameters include the angular velocity of each joint of the upper limb exoskeleton; and the exoskeleton controller is used to adjust the angular velocity of each joint of the upper limb exoskeleton in real time according to the control parameters.

[0006] Optionally, the redundant control algorithm module is used to decompose the motion target into multiple subtasks, assign different priorities to each subtask according to its importance, and calculate the joint angular velocity required for the upper limb exoskeleton to complete each subtask; wherein, the joint limit constraint is set as the task with the highest priority; the shoulder-humeral rhythm constraint is set as the task with the second priority; the end position task and the end posture task related to the scene are set as the task with the third priority and the task with the fourth priority; and the operability constraint is set as the task with the lowest priority.

[0007] Optionally, the joint limit constraints are as follows:

[0008] Define the distance function from the joint limit position:

[0009]

[0010] Where q = [q1,q2,q3,q4,q5,q6,q7,q8,q9] T represents the angles of each joint of the rehabilitation robot, q1 represents the extension / retraction angle of the sternoclavicular joint, q2 represents the elevation / depression angle of the sternoclavicular joint, q3 represents the abduction / adduction angle of the glenohumeral joint, q4 represents the flexion / extension angle of the glenohumeral joint, q5 represents the internal / external rotation angle of the glenohumeral joint, q6 represents the flexion / extension angle of the elbow joint, q7 represents the pronation / supination angle of the forearm, q8 represents the dorsiflexion / palmar flexion angle of the wrist joint, and q9 represents the ulnar / radial flexion angle of the wrist joint; q i represents the angle of the robot's i-th joint, q iM represents the maximum angle allowed for the robot's i-th joint, q im represents the minimum angle allowed for the robot's i-th joint, represents the median value of the range of the robot's i-th joint;

[0011] Calculate the angular velocity of each joint when the rehabilitation robot completes Task 1:

[0012]

[0013] J1=I∈R 9×9 Represents the Jacobian matrix of task 1; J1 # represents the Moore-Penrose pseudoinverse of J1; represents the feedback control item of task 1; KP1 is the artificially set proportional control coefficient; represents the partial derivative of the distance function d(q) with respect to the joint angle.

[0014] Optionally, the scapulohumeral rhythm constraint satisfies the following equality constraint:

[0015]

[0016] f2(q)=q2+0.49q4sin(q3)-20=0

[0017] Differentiation yields

[0018]

[0019]

[0020] in represents the angular velocity of each joint of the rehabilitation robot;

[0021] Calculate the angular velocity of each joint when the robot completes Task 2:

[0022]

[0023] in Denotes the Jacobian of task 2, P1 = (I - J1 # J1) represents the null space matrix of task 1, I∈R 9×9 represents the identity matrix, J2P1 represents the matrix obtained by projecting the Jacobian matrix of task 2 onto the null space of task 1, (J2P1) # represents the Moore-Penrose pseudoinverse of J2P1; represents the feedback control item of task 2, KP f1 ,KP f2 These are two artificially set proportional control coefficients.

[0024] Optionally, the end position task requires the palm center of the arm to be trained to reach position p d =[x d ,y d ,z d ] T ;

[0025] From differential kinematics, we can know the mapping relationship between the end linear velocity and the joint angular velocity in represents the angular velocity of each joint of the rehabilitation robot, Indicates the linear velocity of the palm center of the arm to be trained in the inertial coordinate system, J v ∈R 3×9 represents the linear velocity Jacobian matrix of the rehabilitation robot;

[0026] Calculate the angular velocity of each joint when the robot completes Task 3 after completing Tasks 1 and 2:

[0027]

[0028] Where J3 = J vThe Jacobian matrix of task 3 is P2 = P1(I-(J2P1) # J2P1) represents the null space matrix of task 2, J3P2 represents the matrix obtained by projecting the Jacobian matrix of task 3 onto the null space of task 2, (J3P2) # represents the Moore-Penrose pseudoinverse of J3P2; represents the feedback control item of task 3; KP3 is the artificially set proportional control coefficient; KD3 is the artificially set differential control coefficient; p = [x, y,] T Indicates the actual position of the palm center of the arm to be trained; Indicates the target linear velocity of the palm center of the arm to be trained, which can be obtained by d Differentiation yields; It represents the actual linear velocity of the palm center of the arm to be trained, which can be obtained by differentiating p.

[0029] Optionally, the end posture task requires the palm center of the arm to be trained to reach the posture θ d =[α d ,β d ,γ d ] T ;

[0030] From differential kinematics, we can know the mapping relationship between the end angular velocity and the joint angular velocity in represents the angular velocity of each joint of the rehabilitation robot, It represents the angular velocity of the palm center of the arm to be trained in the inertial coordinate system, J w ∈R 3×9 represents the angular velocity Jacobian matrix of the rehabilitation robot;

[0031] Calculate the angular velocity of each joint when the robot completes task 4 after completing tasks 1, 2, and 3:

[0032]

[0033] Where J4 = J w Denotes the Jacobian of task 4, P3 = P2(I - (J3P2) # J3P2) represents the null space matrix of task 3, and J4P3 represents the matrix obtained by projecting the Jacobian matrix of task 4 onto the null space of task 3. (J4P3) # represents the Moore-Penrose pseudoinverse of J4P3; Represents the feedback control item of task 4; KP ４ Is the artificially set proportional control coefficient; KD ４is the artificially set differential control coefficient; θ=[α,β,γ] 在 Indicates the actual posture of the palm center of the arm to be trained; The target angular velocity of the palm center of the arm to be trained can be obtained by d Differentiation yields; represents the actual angular velocity of the palm center of the arm to be trained, which can be obtained by differentiating θ.

[0034] Optionally, the maneuverability constraint requires the robot to stay away from singular configurations, defining a maneuverability function

[0035]

[0036] in represents the Jacobian matrix of the robot, det* represents the determinant of the matrix*;

[0037] Calculate the angular velocity of each joint when the robot completes task 5 after completing tasks 1, 2, 3, and 4

[0038]

[0039] Where J5=I∈R 9×9 The Jacobian matrix of task 5 is P4 = P3(I-(J4P3) # J4P3) represents the null space matrix of task 4, and J5P4 represents the matrix obtained by projecting the Jacobian matrix of task 5 onto the null space of task 4. (J5P4) # represents the Moore-Penrose pseudoinverse of J5P4; represents the feedback control item of task 5; KP5 is the artificially set proportional control coefficient; represents the partial derivative of the maneuverability function w(q) with respect to the joint angle.

[0040] Optionally, the upper limb exoskeleton is composed of nine joint degrees of freedom, namely, sternoclavicular joint extension / retraction degree of freedom, sternoclavicular joint elevation / depression degree of freedom, glenohumeral joint abduction / adduction degree of freedom, glenohumeral joint flexion / extension degree of freedom, glenohumeral joint internal / external rotation degree of freedom, elbow joint flexion / extension degree of freedom, forearm pronation / supination degree of freedom, wrist joint dorsiflexion / palmar flexion degree of freedom, and wrist joint ulnar / radial flexion degree of freedom, in series; the rotation axis of the sternoclavicular joint extension / retraction degree of freedom of the upper limb exoskeleton is coaxial with the extension / retraction rotation axis of the sternoclavicular joint on the hemiplegic side of the patient; the rotation axis of the sternoclavicular joint elevation / depression degree of freedom of the upper limb exoskeleton is coaxial with the elevation / depression rotation axis of the sternoclavicular joint on the hemiplegic side of the patient; the rotation axis of the glenohumeral joint abduction / adduction degree of freedom of the upper limb exoskeleton is coaxial with the abduction / internal rotation axis of the glenohumeral joint on the hemiplegic side of the patient The rotation axis of the flexion / extension degree of freedom of the glenohumeral joint of the upper limb exoskeleton is coaxial with the flexion / extension rotation axis of the glenohumeral joint on the hemiplegic side of the patient; the rotation axis of the internal rotation / external rotation degree of freedom of the glenohumeral joint of the upper limb exoskeleton is coaxial with the internal rotation / external rotation axis of the glenohumeral joint on the hemiplegic side of the patient; the rotation axis of the flexion / extension degree of freedom of the elbow joint of the upper limb exoskeleton is coaxial with the flexion / extension rotation axis of the elbow joint on the hemiplegic side of the patient; the rotation axis of the pronation / supination degree of freedom of the forearm of the upper limb exoskeleton is coaxial with the pronation / supination rotation axis of the forearm on the hemiplegic side of the patient; the rotation axis of the dorsiflexion / palmar flexion degree of freedom of the wrist joint of the upper limb exoskeleton is coaxial with the dorsiflexion / palmar flexion axis of the wrist joint on the hemiplegic side of the patient; the rotation axis of the ulnar flexion / radial flexion degree of freedom of the wrist joint of the upper limb exoskeleton is coaxial with the ulnar flexion / radial flexion axis of the wrist joint on the hemiplegic side of the patient.

[0041] Optionally, the exoskeleton controller is used to adjust the angular velocity of each joint of the upper limb exoskeleton in real time according to a periodic synchronous speed mode.

[0042] Optionally, a scene display module is further included, and the scene display module is used to display the rehabilitation training scene.

[0043] The present invention provides a nine-degree-of-freedom upper limb rehabilitation robot, which adds two active degrees of freedom of the sternoclavicular joint to simulate the movement of the shoulder girdle, so that the rehabilitation robot can better match the movement of the human upper limb. The redundant control algorithm classifies tasks based on the null space projection of the Jacobian matrix. Low-priority tasks do not affect the execution of high-priority tasks. The redundant degrees of freedom of the upper limb exoskeleton can be fully utilized to ensure coordinated movement between the joints of the robotic arm, thereby improving the treatment effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0045] Figure 1 This is a schematic structural diagram of a nine-degree-of-freedom upper limb rehabilitation robot according to an embodiment of the present invention;

[0046] Figure 2 Schematic diagram of the working process of a nine-degree-of-freedom upper limb rehabilitation robot according to an embodiment of the present invention;

[0047] Figure 3 Schematic diagram of an upper limb exoskeleton structure in an embodiment of the present invention.

[0048] Description of reference numerals:

[0049] 10-Scene display module; 20-Industrial control host; 30-Upper limb exoskeleton; 40-Exoskeleton controller; 301-Sternoclavicular joint extension / retraction degree of freedom; 302-Sternoclavicular joint elevation / depression degree of freedom; 303-Glenoid humeral joint abduction / adduction degree of freedom; 304-Glenoid humeral joint flexion / extension degree of freedom; 305-Glenoid humeral joint internal / external rotation degree of freedom; 306-Elbow joint flexion / extension degree of freedom; 307-Forearm pronation / supination degree of freedom; 308-Wrist joint dorsiflexion / palmar flexion degree of freedom; 309-Wrist joint ulnar / radial flexion degree of freedom; 310-Sternoclavicular joint rotation center; 311-Glenoid humeral joint rotation center; 312-Elbow joint rotation center; 313-Wrist joint rotation center. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0051] In recent years, rehabilitation robots have become a research hotspot in the field of neurological rehabilitation. They can assist or even replace physicians in providing more sustained, effective, and targeted rehabilitation training and treatment, alleviating the shortage of rehabilitation medical personnel. They can also record patient treatment data in real time, providing an objective basis for condition assessment and program improvement. Upper limb rehabilitation robots are primarily divided into two types: end-effector and exoskeleton. Wearable upper limb rehabilitation exoskeletons can significantly reduce the workload of rehabilitation physicians. Rehabilitation exoskeletons are typical human-robot integrated systems. When the user wears the exoskeleton, the human body and the exoskeleton form a spatial multi-loop closed chain. During training, the upper limb and the exoskeleton achieve coordinated motion through the interaction of the connecting mechanisms. This requires that the human-robot closed chain avoids the generation of internal forces unrelated to the rehabilitation training, thereby preventing a poor patient experience and secondary injury to the affected limb. Therefore, it is of great significance to design an upper limb rehabilitation exoskeleton that can effectively address the issue of human-robot motion compatibility and enable a more natural and comfortable rehabilitation training experience.

[0052] An existing seven-degree-of-freedom upper limb rehabilitation robot includes a shoulder adduction and abduction joint, a shoulder flexion and extension joint, a shoulder internal and external rotation joint, an elbow flexion and extension joint, a wrist ulnar and radial deviation joint, a wrist flexion and extension joint, a wrist internal and external rotation joint, an upper arm telescopic portion, a forearm telescopic portion, an upper arm support portion, a forearm support portion, a finger and wrist fixing portion, and a control portion. The seven degrees of freedom are connected in series. The invention is easy to operate. It only requires placing the patient's upper arm and forearm into the upper arm support portion and the forearm support portion, and fixing the patient's wrist and fingers to the wrist internal and external rotation joints through the finger and wrist fixing portion to provide the patient with seven degrees of freedom rehabilitation training: upper arm adduction and abduction movement, upper arm flexion and extension movement, upper arm internal and external rotation movement, elbow flexion and extension movement, wrist ulnar and radial deviation movement, wrist flexion and extension movement, and wrist internal and external rotation movement.

[0053] A seven-degree-of-freedom upper limb exoskeleton rehabilitation robot belongs to the technical field of rehabilitation machinery. It comprises a support mechanism, a shoulder training mechanism, an upper arm adjustment mechanism, an elbow training mechanism, a wrist training mechanism, and a wrist adjustment mechanism, all of which are connected in sequence. By combining seven active degrees of freedom with three passive degrees of freedom, it ensures smooth, comfortable, and precise training, adapting to different users and demonstrating strong applicability.

[0054] An existing upper limb exoskeleton rehabilitation robot with ten degrees of freedom and interchangeable hands has a modular design, comprising a support module, a shoulder joint module, an upper arm module, an elbow joint module, a forearm module, a wrist joint module, and a handle module, connected in sequence. The support module comprises a frame assembly, a bending assembly, and a support assembly. The shoulder joint module comprises a shoulder adduction / abduction assembly, a shoulder internal / external rotation assembly, and a shoulder flexion / extension assembly; the wrist joint module comprises a wrist flexion / extension assembly and a wrist internal / external rotation assembly, for a total of ten degrees of freedom. Six of these active degrees of freedom meet the needs of rehabilitation training, and the support module is equipped with two passive rotational degrees of freedom to increase the flexibility of the mechanism. The stepless adjustment mechanism of the upper arm module and forearm module adds two passive degrees of freedom of movement. By rotating the shoulder internal / external rotation assembly to the other side of the sagittal plane and reversing the motor of the elbow joint module, the left and right hands can be interchanged.

[0055] All of the aforementioned upper limb rehabilitation robots simplify the upper limb to six or seven active degrees of freedom (DOFs) and the shoulder (glenohumeral) joint to a three-DOF ball-and-spherical joint. They assume that shoulder movement is entirely controlled by the glenohumeral joint. In reality, shoulder movement is a function of both the glenohumeral joint and the shoulder girdle. For example, when the shoulder is abducted 180 degrees, the glenohumeral joint moves 120 degrees, while the shoulder girdle rotates upward 60 degrees. Together, these two joints complete 180 degrees of abduction, with a motion ratio of approximately 2:1 (scapulohumeral rhythm). This results in a mismatch between the robot and the user's upper limb movements. While some of these solutions incorporate two passive rotational joints, one for upward and downward rotation of the acromioclavicular joint and one for clavicle extension and retraction in the sternoclavicular joint, the movements of these passive joints are uncontrollable, making it impossible to ensure that the robot's movements match those of the human upper limb. This mismatch between the robot and the user's upper limb movements can easily lead to compensatory movements that compromise treatment effectiveness and even cause secondary injury.

[0056] To solve the above problems, an embodiment of the present invention proposes a nine-degree-of-freedom upper limb rehabilitation robot. In addition to the usual seven degrees of freedom of the upper limb, two active degrees of freedom of the sternoclavicular joint are added to simulate the movement of the shoulder girdle, so that the rehabilitation robot can better match the movement of the human upper limb.

[0057] An embodiment of the present invention provides a nine-degree-of-freedom upper limb rehabilitation robot, which includes a rehabilitation training module, a priority-based redundant control algorithm module, an upper limb exoskeleton, and an exoskeleton controller.

[0058] Among them, the upper limb exoskeleton includes nine active joint degrees of freedom: sternoclavicular joint extension / retraction, sternoclavicular joint elevation / depression, glenohumeral joint abduction / adduction, glenohumeral joint flexion / extension, glenohumeral joint internal / external rotation, elbow flexion / extension, forearm pronation / supination, wrist dorsiflexion / palmar flexion, and wrist ulnar / radial flexion. It can apply auxiliary force / torque to the patient's hemiplegic arm to drive the hemiplegic arm to move.

[0059] The rehabilitation training module is used to generate a rehabilitation training scenario, which includes multiple motion targets. For example, the rehabilitation training module may store rehabilitation training software that can generate a rehabilitation training scenario based on the patient's stroke level. The rehabilitation training scenario consists of a series of motion targets, requiring the patient to move the distal end of the hemiplegic arm (i.e., the center of the palm) to the target position and posture.

[0060] The redundant control algorithm module is used to classify the tasks derived from the motion target decomposition based on the null-space projection of the Jacobian matrix and determine the control parameters for the classified tasks; optionally, the control parameters include the angular velocities of the joints of the upper limb exoskeleton. The redundant control algorithm can classify the tasks based on the null-space projection of the Jacobian matrix, fully utilizing the robot's redundant degrees of freedom to ensure coordinated motion among the joints of the robotic arm.

[0061] The exoskeleton controller is used to adjust the angular velocity of each joint of the upper limb exoskeleton in real time according to the control parameters. For example, the exoskeleton controller can adjust the angular velocity of each joint of the upper limb exoskeleton in real time according to a periodic synchronous speed mode.

[0062] Furthermore, the nine-degree-of-freedom upper limb rehabilitation robot further includes a scene display module, which is used to display the rehabilitation training scene.

[0063] Figure 1 The figure shows a schematic structural diagram of a nine-degree-of-freedom upper limb rehabilitation robot according to an embodiment of the present invention. The nine-degree-of-freedom upper limb rehabilitation robot comprises a scene display module 10 , an industrial control host 20 , an upper limb exoskeleton 30 and an exoskeleton controller 40 .

[0064] The scene display module 10 is used to display the rehabilitation training scene interface.

[0065] The industrial control host 20 runs the rehabilitation training software, and the generation and update of the rehabilitation training scenes and all algorithms are implemented by the rehabilitation training software.

[0066] The upper limb exoskeleton 30 applies an auxiliary torque to the patient's hemiplegic arm to drive the hemiplegic arm to move.

[0067] The exoskeleton controller 40 provides a cyclic synchronous speed mode, which can adjust the angular velocity of each joint of the exoskeleton in real time.

[0068] For example, Figure 2 The following is a schematic diagram showing the workflow of a nine-degree-of-freedom upper limb rehabilitation robot according to an embodiment of the present invention. The workflow of the nine-degree-of-freedom upper limb rehabilitation robot is as follows:

[0069] In S201, the patient sits in the middle position behind the scene display module. The upper arm and forearm of the patient's hemiplegic upper limb are fixed to the exoskeleton robotic arm through adjustable Velcro, and the rehabilitation training software is started.

[0070] S202, the rehabilitation training software generates a rehabilitation training scene. The rehabilitation training scene consists of a series of motion targets. The motion targets require the patient's hemiplegic arm to move to the target position and posture. For example, the rehabilitation training scene gives the motion target, that is, the position and posture of the palm center of the arm to be trained in the inertial coordinate system. The target position is expressed in Cartesian coordinates p d =[x,y,z], the attitude is given in Euler angles θ d =[α,β,γ].

[0071] S203, the redundant control algorithm decomposes the scene goal into a series of subtasks, assigns different priorities to each task according to its importance, and calculates the joint angular velocity required for the upper limb exoskeleton to complete each task.

[0072] In step S204 , the joints of the exoskeleton operate in a periodic synchronous speed mode, and under the control of the exoskeleton controller, the patient's hemiplegic arm is driven to move to the target position and posture in the scene at the angular velocity.

[0073] S205: When the error between the position and posture of the patient's hemiplegic arm end and the scene target position and posture is within the preset range, the current goal is considered to be completed. After completing the current goal, the next goal is entered in sequence until all goals are completed or the training time ends.

[0074] The nine-degree-of-freedom upper limb rehabilitation robot provided by an embodiment of the present invention adds two active degrees of freedom of the sternoclavicular joint to simulate the movement of the shoulder girdle, so that the rehabilitation robot can better match the movement of the human upper limb. The redundant control algorithm classifies tasks based on the null space projection of the Jacobian matrix. Low-priority tasks do not affect the execution of high-priority tasks. It can fully utilize the redundant degrees of freedom of the upper limb exoskeleton, ensure coordinated movement between the joints of the robotic arm, and improve the treatment effect.

[0075] A rigid body in space has six degrees of freedom, meaning that describing its position in space requires at least six variables. Therefore, a robotic arm with six joints in series can control a rigid body to a specified position (assuming the specified position is within the manipulator's workspace). When a robotic arm has more than six degrees of freedom, it can still move after manipulating the rigid body to the specified position because it has redundant degrees of freedom in addition to the six degrees of freedom required for the task (manipulating the rigid body to the specified position). To fully utilize the robot's redundant degrees of freedom, this scheme implements priority-based redundant control. Sometimes the position of the robot end effector is more important than its orientation. For example, in tasks such as object grasping, the end effector's position accuracy must be maintained to ensure grasping the object. Sometimes the posture of the robot end effector is more important than its position. For example, when observing very distant astronomical phenomena, the position error of the robotic arm is not important, but a slight error in its orientation may prevent it from focusing on the observed object. Therefore, a comprehensive robotic control task can be decomposed into different subtasks based on their significance and prioritized according to their importance. Low-priority tasks cannot affect the execution of high-priority tasks.

[0076] In this embodiment, the redundant control algorithm decomposes the scenario objective into a series of subtasks, assigning different priorities to each task based on importance, and then calculating the joint angular velocities required for the upper-limb exoskeleton to complete each task. A null space projection method projects the task vectors of low-priority tasks onto the null space of high-priority tasks. This allows the high-priority tasks to be completed smoothly while also partially completing the low-priority tasks. This projection method also avoids conflicts between tasks.

[0077] For example, the redundant control algorithm implements rehabilitation training by hierarchically assigning tasks based on the null space projection of the Jacobian matrix. The specific method is as follows:

[0078] To ensure safety, joint limit constraints are set as the highest priority task; to ensure that the two active degrees of freedom of the sternoclavicular joint conform to the scapulohumeral rhythm, the scapulohumeral rhythm constraint is set as the second priority task; then the scenario-related end position task and end posture task are set as the third / fourth priority tasks respectively; finally, to prevent the exoskeleton from reaching a singular configuration, the operability constraint is set as the lowest priority task.

[0079] Task 1, joint limit constraint, requires the robot's joints to be away from the limit position, and defines the distance function from the joint limit position

[0080]

[0081] Where q = [q1,q2,q3,q4,q5,q6,q7,q8,q9] Trepresents the angles of each joint of the rehabilitation robot, q1 represents the extension / retraction angle of the sternoclavicular joint, q2 represents the elevation / depression angle of the sternoclavicular joint, q3 represents the abduction / adduction angle of the glenohumeral joint, q4 represents the flexion / extension angle of the glenohumeral joint, q5 represents the internal / external rotation angle of the glenohumeral joint, q6 represents the flexion / extension angle of the elbow joint, q7 represents the pronation / supination angle of the forearm, q8 represents the dorsiflexion / palmar flexion angle of the wrist joint, and q9 represents the ulnar / radial flexion angle of the wrist joint; q i represents the angle of the robot's i-th joint, q iM represents the maximum angle allowed for the robot's i-th joint, q im represents the minimum angle allowed for the robot's i-th joint, represents the median value of the range of the robot's i-th joint.

[0082] Calculate the angular velocity of each joint when the robot completes Task 1

[0083]

[0084] J1=I∈R 9×9 Represents the Jacobian matrix (identity matrix) of task 1; J1 # represents the Moore-Penrose pseudoinverse of J1; represents the feedback control item of task 1; KP1 is the artificially set proportional control coefficient; represents the partial derivative of the distance function d(q) with respect to the joint angle.

[0085] Task 2, scapulohumeral rhythm constraint. The nine-degree-of-freedom upper limb rehabilitation robot must conform to the scapulohumeral rhythm. According to relevant research, the scapulohumeral rhythm must satisfy the following equation constraint:

[0086]

[0087] f2(q)=q2+0.49q4sin(q3)-20=0

[0088] Differentiation yields

[0089]

[0090]

[0091] in Represents the angular velocity of each joint of the rehabilitation robot.

[0092] Calculate the angular velocity of each joint when the robot completes Task 2

[0093]

[0094] in Denotes the Jacobian of task 2, P1 = (I - J1 # J1) represents the null space matrix of task 1, I∈R 9×9 represents the identity matrix, J2P1 represents the matrix obtained by projecting the Jacobian matrix of task 2 onto the null space of task 1, (J2P1) # represents the Moore-Penrose pseudoinverse of J2P1; represents the feedback control item of task 2, KP f1 ,KP f2 are two artificially set proportional control coefficients;

[0095] Task 3, the end position task, requires the center of the palm of the arm to be trained to reach position p d =[x d ,y d ,z d ] T , from differential kinematics, we can know the mapping relationship between the end linear velocity and the joint angular velocity in represents the angular velocity of each joint of the rehabilitation robot, Indicates the linear velocity of the palm center of the arm to be trained in the inertial coordinate system, J v ∈R 3×9 Represents the linear velocity Jacobian matrix of the rehabilitation robot.

[0096] Calculate the angular velocity of each joint when the robot completes task 3 after completing tasks 1 and 2

[0097]

[0098] Where J3 = J v The Jacobian matrix of task 3 is P2 = P1(I-(J2P1) # J2P1) represents the null space matrix of task 2, J3P2 represents the matrix obtained by projecting the Jacobian matrix of task 3 onto the null space of task 2, (J3P2) # represents the Moore-Penrose pseudoinverse of J3P2; v r3 =KP3(o d -p)+ Represents the feedback control item of task 3; KP3 is the manually set proportional control coefficient; KD3 is the manually set differential control coefficient; p = [x, y, z] T Indicates the actual position of the palm center of the arm to be trained; Indicates the target linear velocity of the palm center of the arm to be trained, which can be obtained by d Differentiation yields; It represents the actual linear velocity of the palm center of the arm to be trained, which can be obtained by differentiating p.

[0099] Task 4, the end posture task, requires the center of the palm of the arm to be trained to reach the posture θ d =[α d ,β d ,γ d ] T From differential kinematics, we can know the mapping relationship between the end angular velocity and the joint angular velocity in represents the angular velocity of each joint of the rehabilitation robot, It represents the angular velocity of the palm center of the arm to be trained in the inertial coordinate system, J w ∈R 3×9 Represents the angular velocity Jacobian matrix of the rehabilitation robot.

[0100] Calculate the angular velocity of each joint when the robot completes task 4 after completing tasks 1, 2, and 3

[0101]

[0102] Where J4 = J w Denotes the Jacobian of task 4, P3 = P2(I - (J3P2) # J3P2) represents the null space matrix of task 3, and J4P3 represents the matrix obtained by projecting the Jacobian matrix of task 4 onto the null space of task 3. (J4P3) # represents the Moore-Penrose pseudoinverse of J4P3; represents the feedback control item of task 4; KP4 is the manually set proportional control coefficient; KD4 is the manually set differential control coefficient; θ = [α, β, γ] T Indicates the actual posture of the palm center of the arm to be trained; The target angular velocity of the palm center of the arm to be trained can be obtained by d Differentiation yields; represents the actual angular velocity of the palm center of the arm to be trained, which can be obtained by differentiating θ.

[0103] Task 5, maneuverability constraint, requires the robot to stay away from singular configurations and define the maneuverability function

[0104]

[0105] in represents the Jacobian matrix of the robot, and det* represents the determinant of the matrix*.

[0106] Calculate the angular velocity of each joint when the robot completes task 5 after completing tasks 1, 2, 3, and 4

[0107]

[0108] Where J5=I∈R 9×9 The Jacobian matrix (identity matrix) for task 5 is P4 = P3 (I - (J4P3) # J4P3) represents the null space matrix of task 4, and J5P4 represents the matrix obtained by projecting the Jacobian matrix of task 5 onto the null space of task 4. (J5P4) # represents the Moore-Penrose pseudoinverse of J5P4; represents the feedback control item of task 5; KP5 is the artificially set proportional control coefficient; represents the partial derivative of the maneuverability function w(q) with respect to the joint angle.

[0109] Figure 3 A schematic diagram of the structure of an upper limb exoskeleton according to an embodiment of the present invention is shown. The upper limb exoskeleton is composed of nine joint degrees of freedom, namely, sternoclavicular joint extension / retraction degree of freedom 301, sternoclavicular joint elevation / depression degree of freedom 302, glenohumeral joint abduction / adduction degree of freedom 303, glenohumeral joint flexion / extension degree of freedom 304, glenohumeral joint internal / external rotation degree of freedom 305, elbow joint flexion / extension degree of freedom 306, forearm pronation / supination degree of freedom 307, wrist joint dorsiflexion / palmar flexion degree of freedom 308, and wrist joint ulnar / radial flexion degree of freedom 309, connected in series. Figure 3 Also shown are the sternoclavicular joint rotation center 310, the glenohumeral joint rotation center 311, the elbow joint rotation center 312, and the wrist joint rotation center 313 of the patient's hemiplegic upper limb.

[0110] The rotation axis of the extension / retraction degree of freedom 301 of the upper limb exoskeleton sternoclavicular joint is coaxial with the extension / retraction rotation axis of the sternoclavicular joint on the hemiplegic side of the patient;

[0111] The rotation axis of the sternoclavicular joint elevation / depression degree of freedom 302 of the upper limb exoskeleton is coaxial with the elevation / depression rotation axis of the sternoclavicular joint of the patient's hemiplegic side;

[0112] The rotation axis of the abduction / adduction degree of freedom 303 of the glenohumeral joint of the upper limb exoskeleton is coaxial with the abduction / adduction rotation axis of the glenohumeral joint on the hemiplegic side of the patient;

[0113] The rotation axis of the glenohumeral joint flexion / extension degree of freedom 304 of the upper limb exoskeleton is coaxial with the flexion / extension rotation axis of the glenohumeral joint on the hemiplegic side of the patient;

[0114] The rotation axis of the glenohumeral joint internal rotation / external rotation degree of freedom 305 of the upper limb exoskeleton is coaxial with the internal rotation / external rotation axis of the glenohumeral joint on the hemiplegic side of the patient;

[0115] The rotation axis of the flexion / extension degree of freedom 306 of the elbow joint of the upper limb exoskeleton is coaxial with the flexion / extension rotation axis of the elbow joint of the hemiplegic side of the patient;

[0116] The rotation axis of the upper limb exoskeleton forearm pronation / supination degree of freedom 307 is coaxial with the pronation / supination rotation axis of the patient's hemiplegic forearm;

[0117] The rotation axis of the dorsiflexion / palmar flexion degree of freedom 308 of the upper limb exoskeleton wrist joint is coaxial with the dorsiflexion / palmar flexion rotation axis of the patient's hemiplegic wrist joint;

[0118] The rotation axis of the ulnar / radial flexion degree of freedom 309 of the upper limb exoskeleton wrist joint is coaxial with the ulnar / radial flexion rotation axis of the patient's hemiplegic wrist joint.

[0119] The nine degrees of freedom of the upper limb exoskeleton are all equipped with drive motors, which generate auxiliary torque to drive the exoskeleton itself and the patient's hemiplegic arm to move.

[0120] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0121] 1. In addition to the usual seven active degrees of freedom of the upper limbs, this embodiment adds two active degrees of freedom of the sternoclavicular joint to simulate the movement of the shoulder girdle, so that the rehabilitation robot can better match the movement of the human upper limbs.

[0122] 2. The redundant control algorithm classifies tasks based on the null space projection of the Jacobian matrix. Low-priority tasks do not affect the execution of high-priority tasks. It can fully utilize the redundant degrees of freedom of the upper limb exoskeleton and ensure coordinated movement between the joints of the robotic arm.

[0123] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the various steps of the aforementioned control method for a nine-degree-of-freedom upper limb rehabilitation robot and achieves the same technical effects. To avoid repetition, the details are omitted here. The computer-readable storage medium may be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0124] Of course, those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the control device through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it may include the processes of the above-mentioned method embodiments, wherein the storage medium may be a memory, a disk, an optical disk, etc.

[0125] In this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0126] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0127] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A nine-degree-of-freedom upper limb rehabilitation robot, characterized in that: The nine-degree-of-freedom upper limb rehabilitation robot includes a rehabilitation training module, a priority-based redundant control algorithm module, an upper limb exoskeleton, and an exoskeleton controller; The upper limb exoskeleton includes nine active joint degrees of freedom: sternoclavicular extension / retraction, sternoclavicular joint elevation / depression, glenohumeral joint abduction / adduction, glenohumeral joint flexion / extension, glenohumeral joint internal / external rotation, elbow flexion / extension, forearm pronation / supination, wrist dorsiflexion / palmar flexion, and wrist ulnar / radial flexion. The rehabilitation training module is used to generate a rehabilitation training scene, and the rehabilitation training scene includes multiple motion targets; The redundant control algorithm module is used to classify the tasks obtained by decomposing the motion target based on the null space projection of the Jacobian matrix and determine the control parameters of the classified tasks; the control parameters include the angular velocity of each joint of the upper limb exoskeleton; The exoskeleton controller is used to adjust the angular velocity of each joint of the upper limb exoskeleton in real time according to the control parameters; The redundant control algorithm module is used to decompose the motion target into multiple subtasks, assign different priorities to each subtask according to its importance, and calculate the joint angular velocity required for the upper limb exoskeleton to complete each subtask; Among them, the joint limit constraint is set as the highest priority task 1; the shoulder-humeral rhythm constraint is set as the second priority task 2; the end position task and end posture task related to the scene are set as the third priority task 3 and the fourth priority task 4; the operability constraint is set as the lowest priority task 5; The joint limit constraints are as follows: Define the distance function from the joint limit position: Where q = [q1,q2,q3,q4,q5,q6,q7,q8,q9] T represents the angles of each joint of the rehabilitation robot, q1 represents the extension / retraction angle of the sternoclavicular joint, q2 represents the elevation / depression angle of the sternoclavicular joint, q3 represents the abduction / adduction angle of the glenohumeral joint, q4 represents the flexion / extension angle of the glenohumeral joint, q5 represents the internal / external rotation angle of the glenohumeral joint, q6 represents the flexion / extension angle of the elbow joint, q7 represents the pronation / supination angle of the forearm, q8 represents the dorsiflexion / palmar flexion angle of the wrist joint, and q9 represents the ulnar / radial flexion angle of the wrist joint; q i represents the angle of the robot's i-th joint, q iM represents the maximum angle allowed for the robot's i-th joint, q im represents the minimum angle allowed for the robot's i-th joint, represents the median value of the range of the robot's i-th joint; Calculate the angular velocity of each joint when the rehabilitation robot completes Task 1: J1=I∈R 9×9 Represents the Jacobian matrix of task 1; J1 # represents the Moore-Penrose pseudoinverse of J1; represents the feedback control item of task 1; KP1 is the artificially set proportional control coefficient; represents the partial derivative of the distance function d(q) with respect to the joint angle.

2. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 1, characterized in that: The scapulohumeral rhythm constraint satisfies the following equality constraint: f2(q)=q2+0.49q4sin(q3)-20=0 Differentiation yields in represents the angular velocity of each joint of the rehabilitation robot; Calculate the angular velocity of each joint when the robot completes Task 2: in Denotes the Jacobian of task 2, P1 = (I - J1 # J1) represents the null space matrix of task 1, I∈R 9×9 represents the identity matrix, J2P1 represents the matrix obtained by projecting the Jacobian matrix of task 2 onto the null space of task 1, (J2P1) # represents the Moore-Penrose pseudoinverse of J2P1; represents the feedback control item of task 2, KP f1 ,KP f2 These are two artificially set proportional control coefficients.

3. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 2, characterized in that: The end position task requires the center of the palm of the arm to be trained to reach position p d =[x d ,y d ,z d ] T ; From differential kinematics, we can know the mapping relationship between the end linear velocity and the joint angular velocity in represents the angular velocity of each joint of the rehabilitation robot, Indicates the linear velocity of the palm center of the arm to be trained in the inertial coordinate system, J v ∈R 3×9 represents the linear velocity Jacobian matrix of the rehabilitation robot; Calculate the angular velocity of each joint when the robot completes Task 3 after completing Task 1 and Task 2: Where J3 = J v The Jacobian matrix of task 3 is P2 = P1(I-(J2P1) # J2P1) represents the null space matrix of task 2, J3P2 represents the matrix obtained by projecting the Jacobian matrix of task 3 onto the null space of task 2, (J3P2) # represents the Moore-Penrose pseudoinverse of J3P2; Represents the feedback control item of task 3; KP3 is the manually set proportional control coefficient; KD3 is the manually set differential control coefficient; p = [x, y, z] T Indicates the actual position of the palm center of the arm to be trained; Indicates the target linear velocity of the palm center of the arm to be trained, which can be obtained by d Differentiation yields; It represents the actual linear velocity of the palm center of the arm to be trained, which can be obtained by differentiating p.

4. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 3, characterized in that: The end posture task requires the palm center of the arm to be trained to reach the posture θ d =[α d ,β d ,γ d ] T ; From differential kinematics, we can know the mapping relationship between the end angular velocity and the joint angular velocity in represents the angular velocity of each joint of the rehabilitation robot, It represents the angular velocity of the palm center of the arm to be trained in the inertial coordinate system, J w ∈R 3×9 represents the angular velocity Jacobian matrix of the rehabilitation robot; Calculate the angular velocity of each joint when the robot completes Task 4 after completing Task 1, Task 2, and Task 3: Where J4 = J w Denotes the Jacobian of task 4, P3 = P2(I - (J3P2) # J3P2) represents the null space matrix of task 3, and J4P3 represents the matrix obtained by projecting the Jacobian matrix of task 4 onto the null space of task 3. (J4P3) # represents the Moore-Penrose pseudoinverse of J4P3; represents the feedback control item of task 4; KP4 is the manually set proportional control coefficient; KD4 is the manually set differential control coefficient; θ=[ɑ,β,γ] T Indicates the actual posture of the palm center of the arm to be trained; The target angular velocity of the palm center of the arm to be trained can be obtained by d Differentiation yields; represents the actual angular velocity of the palm center of the arm to be trained, which can be obtained by differentiating θ.

5. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 4, characterized in that: The maneuverability constraint requires the robot to stay away from singular configurations, and the maneuverability function is defined as in represents the Jacobian matrix of the robot, det* represents the determinant of the matrix*; Calculate the angular velocity of each joint when the robot completes Task 5 after completing Task 1, Task 2, Task 3, and Task 4 Where J5=I∈R 9×9 The Jacobian matrix of task 5 is P4 = P3(I-(J4P3) # J4P3) represents the null space matrix of task 4, and J5P4 represents the matrix obtained by projecting the Jacobian matrix of task 5 onto the null space of task 4. (J5P4) # represents the Moore-Penrose pseudoinverse of J5P4; represents the feedback control item of task 5; KP5 is the artificially set proportional control coefficient; represents the partial derivative of the maneuverability function w(q) with respect to the joint angle.

6. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 1, characterized in that: The upper limb exoskeleton is composed of nine joint degrees of freedom connected in series: sternoclavicular joint extension / retraction degree of freedom, sternoclavicular joint elevation / depression degree of freedom, glenohumeral joint abduction / adduction degree of freedom, glenohumeral joint flexion / extension degree of freedom, glenohumeral joint internal / external rotation degree of freedom, elbow joint flexion / extension degree of freedom, forearm pronation / supination degree of freedom, wrist joint dorsiflexion / palmar flexion degree of freedom, and wrist joint ulnar / radial flexion degree of freedom; The rotation axis of the extension / retraction degree of freedom of the sternoclavicular joint of the upper limb exoskeleton is coaxial with the extension / retraction rotation axis of the sternoclavicular joint on the hemiplegic side of the patient; The rotation axis of the sternoclavicular joint elevation / depression degree of freedom of the upper limb exoskeleton is coaxial with the elevation / depression rotation axis of the sternoclavicular joint on the hemiplegic side of the patient; The rotation axis of the glenohumeral joint abduction / adduction degree of freedom of the upper limb exoskeleton is coaxial with the abduction / adduction rotation axis of the glenohumeral joint on the hemiplegic side of the patient; The rotation axis of the glenohumeral joint flexion / extension degree of freedom of the upper limb exoskeleton is coaxial with the flexion / extension rotation axis of the glenohumeral joint on the hemiplegic side of the patient; The rotation axis of the internal rotation / external rotation degree of freedom of the glenohumeral joint of the upper limb exoskeleton is coaxial with the internal rotation / external rotation axis of the glenohumeral joint of the hemiplegic side of the patient; The rotation axis of the flexion / extension degree of freedom of the elbow joint of the upper limb exoskeleton is coaxial with the flexion / extension rotation axis of the elbow joint of the hemiplegic side of the patient; The rotation axis of the pronation / supination degree of freedom of the forearm of the upper limb exoskeleton is coaxial with the pronation / supination rotation axis of the forearm on the hemiplegic side of the patient; The rotation axis of the dorsiflexion / palmar flexion degree of freedom of the upper limb exoskeleton wrist joint is coaxial with the dorsiflexion / palmar flexion rotation axis of the patient's hemiplegic wrist joint; The rotation axis of the ulnar flexion / radial flexion degree of freedom of the upper limb exoskeleton wrist joint is coaxial with the ulnar flexion / radial flexion rotation axis of the patient's hemiplegic wrist joint.

7. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 1, characterized in that: The exoskeleton controller is used to adjust the angular velocity of each joint of the upper limb exoskeleton in real time according to a periodic synchronous speed mode.

8. The nine-degree-of-freedom upper limb rehabilitation robot according to claim 1, characterized in that: It also includes a scene display module, which is used to display the rehabilitation training scene.

Images