A method and device for identifying dynamic parameters of upper limb rehabilitation training

Abstract

The present invention proposes a method and device for identifying dynamic parameters of upper limb rehabilitation training, which belongs to the field of rehabilitation technology. In the process of using a mechanical arm to pull the patient's upper limb, the system collects the angle, angular velocity, angular acceleration information of the patient's shoulder and elbow and the interaction force information between the patient's upper limb and the end of the cooperative arm in real time, and uses the proposed upper limb dynamic parameter identification method to process the collected data to obtain the dynamic parameters of the patient's upper limb, and then substitutes them into the dynamic model of the human upper limb to calculate the shoulder and elbow torque values, so as to realize the prediction of the torque curve of the patient's upper limb during the entire rehabilitation process, control the traction force of the mechanical arm during the rehabilitation training, and ensure the safety and comfort of the patient's upper limb during the rehabilitation training.

Description

Technical Field

[0001] The invention belongs to the technical field of rehabilitation, and in particular relates to a rehabilitation training system for upper limb motor functions of stroke patients to be rehabilitated. Background Art

[0002] In recent years, the aging of the population has become a major problem facing my country, and the number of stroke patients has increased. Although the development of medical technology has reduced the mortality rate of stroke patients, the probability of hemiplegia in stroke patients remains high. About 70% of patients have varying degrees of upper limb dysfunction, and the trend of younger strokes is becoming more and more obvious. The dysfunction of the upper limbs has brought great inconvenience to the patients' lives, and has brought economic and psychological impacts to the patients and their families. The recovery of upper limb motor function is very important and has become one of the hot spots in modern rehabilitation medicine and medical engineering. The traditional method of upper limb rehabilitation training is for rehabilitation physicians to hold the patient's upper limbs for exercise, or to guide the patient's upper limb movement through some simple instruments. This method has obvious problems: only qualitative observation can be carried out during the rehabilitation process, and the rehabilitation plan is difficult to continuously improve; the experience and level of rehabilitation physicians limit the effect of rehabilitation training; the workload of rehabilitation physicians is heavy, and the efficiency of rehabilitation training is not high.

[0003] With the development of robotics technology, rehabilitation robot technology, which organically combines robotics technology with rehabilitation medicine, has gradually become an effective means of upper limb rehabilitation. Using commercial robotic arms and adopting the end traction method to conduct rehabilitation training on the upper limbs of the human body is low-cost, which is conducive to the industrialization of upper limb rehabilitation robots. It can not only meet the increasing rehabilitation training needs of hemiplegic patients, but also greatly reduce the workload of rehabilitation medical staff, improve the efficiency of upper limb rehabilitation training, and promote the development of rehabilitation medicine. When the robotic arm pulls the patient's upper limb for rehabilitation training, there is a certain interaction force between the end of the robotic arm and the patient's upper limb. If the interaction force is too large at a certain moment in the rehabilitation process, it is likely to cause secondary damage to the patient's upper limb, so it is necessary to monitor and control the interaction force in real time. Adding a force sensor between the end of the robotic arm and the patient's upper limb can obtain the interaction force value in real time, but whether the interaction force value is reasonable needs to be calculated and compared with the dynamic parameters of the patient's upper limb. Obviously, the upper limb dynamic parameters are difficult to measure directly, and the upper limb dynamic parameters of different patients are different. Therefore, the demand for online identification of the dynamic parameters of the patient's upper limb to obtain a reasonable interaction force prediction curve in the entire rehabilitation process is becoming more and more urgent. Summary of the invention

[0004] Based on the above application background, the present invention proposes a method and device for identifying dynamic parameters of upper limb rehabilitation training. In the process of using a robotic arm to pull the patient's upper limb, the system collects the angle, angular velocity, angular acceleration information of the patient's shoulder and elbow and the interaction force information between the patient's upper limb and the end of the collaborative arm in real time. The collected data is processed using the proposed upper limb dynamic parameter identification method to obtain the dynamic parameters of the patient's upper limb, and the dynamic parameters are substituted into the dynamic model of the upper limb of the body, and then the shoulder and elbow torque values ​​are calculated to realize the prediction of the torque curve of the patient's upper limb during the entire rehabilitation process, and the traction force of the robotic arm during the rehabilitation training is controlled to ensure the safety and comfort of the patient's upper limb during the rehabilitation training.

[0005] The present invention proposes a method and device for identifying dynamic parameters of upper limb rehabilitation training, which is characterized by comprising the following steps:

[0006] Step 1: Connect the human arm wrist to the end of the robotic arm through the arm tray, control the robotic arm to pull the human upper limb to move according to the set trajectory in the XOY, XOZ, and YOZ planes, and collect N groups of interaction force between the human upper limb and the end of the robotic arm, the angles of the human shoulder joint and elbow joint, angular velocity, and angular acceleration data;

[0007] Step 2: Filter the collected data, and then use the upper limb dynamics parameter identification method to identify the upper limb dynamics parameters of the human body, including mass, center of mass position and moment of inertia;

[0008] Step 3: Use the upper limb dynamics model to predict the human shoulder and elbow joint torques during upper limb rehabilitation training. During the actual rehabilitation training, obtain the interactive force data between the human upper limb and the end of the robotic arm in real time to monitor the human upper limb joint torques, adjust the robotic arm control strategy in real time, and ensure the comfort and safety of the upper limbs during rehabilitation training.

[0009] The upper limb dynamics parameter identification method is characterized by:

[0010] The N groups of interaction force data between the human upper limbs and the end of the robotic arm after filtering are converted into N groups of equivalent moments at the human shoulder joint and elbow joint through the principle of virtual work;

[0011] According to the N groups of filtered data and the N groups of equivalent moments obtained in 1), the statically indeterminate equation can be obtained:

[0012]

[0013] Where τ N is the moment vector, whose dimension is 2N×1; is the measurement matrix, which is composed of the filtered shoulder and elbow angles, angular velocities and angular accelerations of the human body, and its dimension is 2N×M; X is the matrix composed of the parameters to be identified, including the mass, center of mass position and moment of inertia of the upper arm and forearm of the human body, and its dimension is M×1. The value of M is determined according to the maximum linearly independent group of the column vectors of the measurement matrix;

[0014] Solve the hyperstatic equations on the three planes XOY, XOZ, and YOZ respectively, and average the mass and center of mass positions of the upper arm and forearm of the human upper limbs identified on the three planes. The moment of inertia identified on the XOY, XOZ, and YOZ planes is the moment of inertia of the upper arm and forearm of the human upper limbs on the planes;

[0015] The rehabilitation training device comprises:

[0016] A mobile chassis (1), a horizontal guide rail (2), a lifting column (3), a mechanical arm (4), a force sensor (5), an arm tray (6), and an angle sensor (7); the mobile chassis (1) is provided with a brake device (1-1) which can realize the fixing of the mobile chassis at any position; the lifting column (3) is fixedly connected to the horizontal guide rail (2) and can move along the horizontal guide rail; the mechanical arm (4) is fixedly connected to the lifting column (3) and can move up and down along the lifting column; one end of the force sensor (5) is fixedly connected to the end of the mechanical arm (4) and the other end is fixedly connected to the arm tray (6); the arm tray (6) is provided with a strap to connect to the wrist of a human arm; the angle sensor (7) is respectively installed at the shoulder joint and the elbow joint of the upper limb of a human body.

[0017] Preferably, in the above-mentioned rehabilitation training device, the robotic arm is a 6-DOF collaborative robotic arm. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a schematic diagram of the rehabilitation training device of the present invention (excluding the angle sensor part);

[0019] Figure 2 is a schematic diagram of the angle sensor portion of the rehabilitation training device of the present invention fixed to a human arm; DETAILED DESCRIPTION

[0020] The present invention is further described in detail below in conjunction with embodiments so that those skilled in the art can implement the invention with reference to the description.

[0021] The present invention requires the rehabilitation personnel to complete the dynamic parameter identification and rehabilitation training of the patient's upper limbs with the cooperation of the upper limb rehabilitation training device:

[0022] The upper limb rehabilitation training device comprises: a mobile chassis (1), a horizontal guide rail (2), a lifting column (3), a mechanical arm (4), a force sensor (5), an arm tray (6), and an angle sensor (7); the mobile chassis (1) is equipped with a brake device (1-1) to realize the fixing of the mobile chassis at any position; the lifting column (3) is fixedly connected to the horizontal guide rail (2) and can move along the horizontal guide rail; the mechanical arm (4) is fixedly connected to the lifting column (3) and can move up and down along the lifting column; one end of the force sensor (5) is fixedly connected to the end of the mechanical arm (4), and the other end is fixedly connected to the arm tray (6); the arm tray (6) is equipped with a strap to connect to the wrist of a human arm; the angle sensor (7) is respectively installed at the shoulder joint and elbow joint of the upper limb of the human body.

[0023] Push the mobile chassis (1) to the side of the person to be rehabilitated and fix it by means of a brake device (1-1);

[0024] An angle sensor (7) is installed at the shoulder joint and elbow joint of the upper limb of the person to be rehabilitated;

[0025] According to the position of the upper limbs of the person to be rehabilitated and the rehabilitation needs, the horizontal guide rail (2) and the lifting column (3) are adjusted so that the mechanical arm (4) and the upper limbs of the person to be rehabilitated are in a suitable relative position;

[0026] The upper limb of the person to be rehabilitated is securely connected to the end of the mechanical arm (4) via the arm tray (6).

[0027] Next, the upper limb dynamics parameters of the person to be rehabilitated are identified. Taking the identification in the XOZ plane as an example, a method and device for identifying dynamics parameters of upper limb rehabilitation training in this embodiment include the following steps:

[0028] Step 1: Connect the wrist of the person to be rehabilitated to the end of the robotic arm through the arm tray, control the robotic arm to pull the upper limb of the person to be rehabilitated to move according to the set trajectory in the XOY, XOZ, and YOZ planes, and collect N groups of interaction force between the upper limb of the person to be rehabilitated and the end of the robotic arm, the angle of the shoulder joint and elbow joint of the person to be rehabilitated, the angular velocity and angular acceleration data;

[0029] Step 2: Filter the collected data, and then use the upper limb dynamics parameter identification method to identify the upper limb dynamics parameters, including mass, center of mass position and moment of inertia:

[0030] 1) The upper limb dynamic parameters that need to be identified include upper arm dynamic parameters and forearm dynamic parameters, namely upper arm mass, upper arm center of mass position, upper arm moment of inertia and forearm mass, forearm center of mass position, forearm moment of inertia.

[0031] 2) The N groups of filtered interaction force data between the upper limbs of the rehabilitation personnel and the end of the robotic arm are converted into N groups of equivalent moments at the shoulder joints and elbow joints of the rehabilitation personnel through the principle of virtual work:

[0032]

[0033] Where τ1 and τ4 are the shoulder and elbow joint moments of the patient to be rehabilitated; F x 、F z is the interaction force collected by the force sensor; l1 and l2 are the lengths of the upper arm and forearm of the person to be rehabilitated (obtained by direct measurement); θ1 and θ4 are the angles of the shoulder joint and elbow joint of the person to be rehabilitated on the XOZ plane.

[0034] 3) Based on the N groups of filtered data and the N groups of equivalent moments obtained in 1), the statically indeterminate equation can be obtained:

[0035]

[0036] Where τ N is the moment vector, whose dimension is 2N×1; is the observation matrix, which is composed of the shoulder and elbow angles, angular velocities and angular accelerations of the person to be rehabilitated after filtering, and its dimension is 2N×M; X is the matrix composed of the parameters to be identified, including the upper arm and forearm mass, center of mass position and moment of inertia of the person to be rehabilitated, and its dimension is M×1. The value of M is determined according to the maximum linearly independent group of the column vectors of the observation matrix.

[0037] After following the above processing:

[0038]

[0039]

[0040]

[0041]

[0042] 4) Solve the hyperstatic equation on the XOZ plane, identify the mass and center of mass position of the upper arm and forearm of the person to be rehabilitated, and identify the moment of inertia of the upper arm and forearm of the person to be rehabilitated on the XOZ plane.

[0043] After completing the identification of the upper limb dynamic parameters, the upper limb dynamic model is used to predict the shoulder and elbow joint torques of the person being rehabilitated during the upper limb rehabilitation training. During the actual rehabilitation training, the interaction force data between the upper limbs of the person being rehabilitated and the end of the robotic arm are obtained in real time to monitor the upper limb joint torques of the person being rehabilitated, and the robotic arm control strategy is adjusted in real time to ensure the comfort and safety of the upper limbs of the person being rehabilitated during the rehabilitation training.

[0044] 1) In the above-mentioned rehabilitation training device, the robotic arm is a 6-DOF collaborative robotic arm.

Claims

1. A method for identifying dynamic parameters of upper limb rehabilitation training, characterized in that The following steps are involved: (1) Connect the human arm wrist to the end of the robotic arm through the arm tray, control the robotic arm to pull the human upper limb to move according to the set trajectory in the XOY, XOZ, and YOZ planes, and collect N groups of interaction force between the human upper limb and the end of the robotic arm, the angles of the human shoulder joint and elbow joint, angular velocity, and angular acceleration data; (2) filtering the collected data, and then using the upper limb dynamics parameter identification method to identify the upper limb dynamics parameters of the human body, including mass, center of mass position and moment of inertia; (3) Using the upper limb dynamics model, the human shoulder and elbow joint torques during upper limb rehabilitation training are predicted. During the actual rehabilitation training, the interaction force data between the human upper limb and the end of the robotic arm is obtained in real time to monitor the human upper limb joint torques, and the robotic arm control strategy is adjusted in real time to ensure the comfort and safety of the upper limbs during rehabilitation training. The kinetic parameter identification method is characterized by: 1) The N groups of interaction force data between the human upper limbs and the end of the robotic arm after filtering are converted into N groups of equivalent moments at the human shoulder joint and elbow joint through the principle of virtual work; 2) Based on the N groups of filtered data and the N groups of equivalent moments obtained in 1), the statically indeterminate equation can be obtained: Where τ N is the moment vector, whose dimension is 2N×1; is the observation matrix, which is composed of the filtered shoulder and elbow angles, angular velocities and angular accelerations of the human body, and its dimension is 2N×M; X is the matrix composed of the parameters to be identified, including the mass, center of mass position and moment of inertia of the upper arm and forearm of the human body, and its dimension is M×1, and the value of M is determined according to the maximum linearly independent group of the column vectors of the observation matrix; 3) Solve the hyperstatic equations on the three planes XOY, XOZ, and YOZ respectively, and average the mass and center of mass position of the upper arm and forearm of the human upper limbs identified on the three planes. The moment of inertia identified on the XOY, XOZ, and YOZ planes is the moment of inertia of the upper arm and forearm of the human upper limbs on the planes; The equipment for upper limb rehabilitation training includes: A mobile chassis (1), a horizontal guide rail (2), a lifting column (3), a mechanical arm (4), a force sensor (5), an arm tray (6), and an angle sensor (7); the mobile chassis (1) is provided with a brake device (1-1) which can realize the fixing of the mobile chassis at any position; the lifting column (3) is fixedly connected to the horizontal guide rail (2) and can move along the horizontal guide rail; the mechanical arm (4) is fixedly connected to the lifting column (3) and can move up and down along the lifting column; one end of the force sensor (5) is fixedly connected to the end of the mechanical arm (4) and the other end is fixedly connected to the arm tray (6); the arm tray (6) is provided with a strap to connect to the wrist of a human arm; the angle sensor (7) is respectively installed at the shoulder joint and the elbow joint of the upper limb of a human body.

2. A method for identifying dynamic parameters of upper limb rehabilitation training according to claim 1, characterized in that: The robotic arm is a 6-DOF collaborative robotic arm.

Images