An intelligent active-passive hybrid training control method for rehabilitation robots

Abstract

This invention discloses a method for intelligent active-passive hybrid training control of a rehabilitation robot, comprising the following steps: calibrating the patient's joint range of motion and maximum power; obtaining the difference in human-machine interaction force; calculating the patient's active participation in rehabilitation training; calculating a velocity offset; and adjusting the joint movement speed during adaptive passive training. This method can adjust the hip, knee, and ankle joint training speed based on changes in the interaction force between the patient's lower limbs and the rehabilitation robot, effectively ensuring patient safety, achieving excellent human-machine interaction, and increasing the patient's active participation in passive training mode.

Description

Technical Field

[0001] The present invention relates to the technical field of robot human-machine interaction, and in particular to an intelligent active-passive hybrid training control method for a rehabilitation robot. Background Art

[0002] The lower limbs include the hips, thighs, knees, calves, and feet. Injury to any of these parts can affect a person's ability to walk, reducing their ability to carry out daily activities and placing a burden on their families and society. Rehabilitation medicine can help people with limb disabilities and nerve damage regain their mobility and resume a normal life through rehabilitation therapy.

[0003] With the development of my country's economy, the improvement of its comprehensive national strength, and the improvement of people's living standards, medical technology has developed rapidly, gradually approaching world standards. However, rehabilitation treatment effectiveness still lags behind that of overseas countries. This is mainly due to the lack of effective post-illness rehabilitation training, especially for neurological and elderly patients, which lags behind the development of medical technology. With the increasing demand for rehabilitation physicians, more and more scientific researchers are gradually recognizing the importance of rehabilitation medicine and actively supporting its development. They are integrating rehabilitation with artificial intelligence and developing scientific and efficient rehabilitation robots to free up the hands of rehabilitation practitioners and promote the development of rehabilitation in my country.

[0004] Through research on the needs of individuals with lower limb dysfunction and reviewing relevant literature on current lower limb rehabilitation robotics research, it was found that existing lower limb rehabilitation training devices, both domestically and internationally, offer suboptimal results when evaluated based on the patient's metabolic cost and muscle torque, failing to meet personalized training needs. Clinically used lower limb rehabilitation robots primarily focus on joint passive motion machines (CPMs), and most CPM training devices target only a single joint, lacking active patient participation and requiring enhanced training effectiveness. Only by integrating individuals with functional disabilities with rehabilitation training equipment through intelligent control technology, increasing patient engagement and confidence, making rehabilitation robots fully intelligent, and formulating scientific training prescriptions based on therapists' rehabilitation techniques and human movement characteristics, can rapid and effective rehabilitation training be achieved.

[0005] Therefore, technical personnel in this field are committed to providing an intelligent active-passive hybrid training control method for rehabilitation robots to improve the current problems of lower limb rehabilitation robots for orthopedic postoperative patients and nerve injury patients, such as their inability to effectively reduce the user's metabolic cost and the average torque generated by the muscles, the lack of human-computer interaction functions, and the single training mode, so as to improve the training effect of rehabilitation robots. Summary of the Invention

[0006] In view of the defects in the prior art, the technical problem to be solved by the present invention is how to provide an intelligent active-passive hybrid training control method for improving the training effect of a rehabilitation robot.

[0007] To achieve the above object, the present invention provides a rehabilitation robot intelligent active-passive hybrid training control method, comprising the following steps:

[0008] Calibrate the patient's joint range of motion and maximum power;

[0009] Obtain the difference in human-computer interaction force;

[0010] Calculate the patient's active participation in rehabilitation training;

[0011] Calculate velocity offset;

[0012] Adjust joint movement speed for adaptive passive training.

[0013] Furthermore, the calibrating of the patient's joint mobility and maximum power includes: calibrating the patient's joint mobility before the start of training, using the calibrated joint mobility as the training angle, and calibrating the patient's maximum power according to the Hill muscle mechanics model.

[0014] Furthermore, the human-computer interaction force is the difference between the patient pressure collected by the pressure sensor and the pressure obtained by the human body dynamics model.

[0015] Furthermore, obtaining the difference in human-computer interaction force includes: controlling the rehabilitation robot joints to train for two cycles according to calibrated joint range of motion and speed, obtaining the human-computer interaction force in the two cycles, and calculating the difference.

[0016] Furthermore, the active participation degree is the ratio of the patient's current power to the maximum power.

[0017] Furthermore, the calculating of the patient's active participation in rehabilitation training includes: calculating the power of all sampling points in the two cycles, and calculating the mathematical expectation value of the patient's active participation in this training process according to the difference in active participation.

[0018] Furthermore, the calculating of the patient's active participation in rehabilitation training further includes: calculating the mathematical expectation value of the human-computer interaction force of all sampling points in the two cycles.

[0019] Furthermore, the calculating of the speed offset includes: calculating the speed offset according to a mathematical expectation value of the active participation degree and a mathematical expectation value of the human-computer interaction force.

[0020] Furthermore, the adjustment of the adaptive passive training joint movement speed includes: adjusting the passive training joint movement speed of the third cycle according to the speed offset to achieve the calibrated training angle; and feeding back the passive training joint movement speed of the third cycle to the fourth cycle so that the training modes of the third cycle and the fourth cycle are the same.

[0021] Furthermore, the adjusting of the adaptive passive training joint movement speed also includes: calculating the passive training joint movement speed of the fifth and sixth cycles according to the active participation and human-computer interaction force of the third and fourth cycles, and repeating this cycle until the training is completed.

[0022] The present invention has at least the following beneficial technical effects:

[0023] The intelligent active-passive hybrid training control method for a rehabilitation robot provided by the present invention can adjust the training speed of the hip, knee, and ankle joints according to the changes in the interaction force between the patient's lower limbs and the rehabilitation robot, simulate the techniques of rehabilitation therapists, effectively ensure the safety of the patient, and can well realize human-computer interaction, thereby increasing the patient's active participation in the passive training mode.

[0024] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 1 is a flow chart of an intelligent active-passive hybrid training control method for a rehabilitation robot provided by an embodiment of the present invention;

[0026] Figure 2 This is a schematic diagram of a cycle for calculating the active participation of a patient in rehabilitation training according to the intelligent active-passive hybrid training control method for a rehabilitation robot provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following describes several preferred embodiments of the present invention with reference to the accompanying drawings to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0028] In the drawings, components with identical structures are denoted by the same reference numerals, and components with similar structures or functions are denoted by similar reference numerals. The size and thickness of each component shown in the drawings are arbitrary and are not limited by the present invention. For clarity, the thickness of components in some places in the drawings is appropriately exaggerated.

[0029] The present invention provides an intelligent active-passive hybrid training control method for a rehabilitation robot. Before the start of training, the patient's joint range of motion is calibrated, and the calibration angle is used as the training angle. Then, the relationship between force, speed and power is obtained according to the Hill muscle mechanics model, and the human-machine interaction force is obtained according to the human-machine coupling control model. The mathematical expectation value of the interaction force difference of all sampling points in the first two cycles and the power difference of all sampling points are calculated. The change in the patient's active participation in the training process is calculated based on the power difference, and then the speed offset is calculated. The passive training joint movement speed of the third cycle is adjusted to reach the calibrated training angle, and the joint movement speed of the third cycle is fed back and sent to the fourth training cycle to make the training modes of the third and fourth cycles the same. The interaction force difference is then calculated, and this cycle is repeated until the end of the training.

[0030] like Figure 1 As shown, taking single joint rehabilitation training as an example, the training control method of this embodiment is as follows.

[0031] Step 1: Calibrate the patient's joint range of motion and maximum power.

[0032] Power refers to the power that human joints can provide. Before different patients undergo rehabilitation training, the lower limb rehabilitation robot is controlled to train the patient at a joint movement speed V suitable for the patient's current rehabilitation stage, and V is equivalent to the patient's muscle contraction speed, which is fed back to the patient's joint range of motion Q and the maximum human-machine interaction force F. max , Q is the training angle of this rehabilitation training, F max is the patient's maximum muscle contraction force.

[0033] From Hill's muscle mechanics model:

[0034]

[0035] It can be seen that the maximum power is 1 / 6 of the ideal value, which is equal to the product of 1 / 2 of the maximum muscle contraction force and 1 / 3 of the maximum contraction speed. max The patient's maximum power P can be obtained by max for:

[0036]

[0037] Step 2: Obtain the difference in human-computer interaction force.

[0038] The robot joints are controlled to train according to the joint range of motion Q and speed V for two cycles, that is, in the Kth cycle and the K+1th cycle, the equipment performs rehabilitation training according to the passive training speed V and training angle Q of the robot joints.

[0039] Set the hardware system sampling rate to 30ms and obtain all interaction forces F1, F2...F within two cycles. n The interaction force is the difference between the patient pressure measured by the pressure sensor and the pressure calculated by the human body dynamics model. The interaction force is positive when the direction is consistent with the joint motion velocity and negative when it is opposite. That is, the tension is positive during upward motion, and the pressure is positive during downward motion. The difference between the interaction forces in the Kth cycle and the K+1th cycle is then calculated.

[0040] Step 3: Calculate the patient's active participation in rehabilitation training.

[0041] During patient rehabilitation training, the active participation at the current moment is the ratio of the current power to the calibrated maximum power, and the human-computer interaction force is considered to be the patient's muscle strength at the current moment.

[0042] During a single cycle of rehabilitation training, the motor rotates at a constant speed. This means the current speed is the velocity of the passively trained joint. The power at that moment is the product of the interaction force and the passive training degree. The active participation degree M and the interaction force are in the same direction. If the patient is highly active during rehabilitation training, the interaction force in the positive direction is greater, resulting in more work being done in that direction. Therefore, the passive training joint velocity should be increased in the next cycle. Conversely, the passive training joint velocity should be decreased.

[0043] For the force at all sampling moments and the power at all sampling moments in the Kth period and the K+1th period, it satisfies:

[0044]

[0045]

[0046] arrive are the active participation of patients in the Kth cycle, arrive are the active participation of patients in the K+1th cycle, satisfying:

[0047]

[0048]

[0049] Step 4: Calculate the speed offset.

[0050] Calculate the active participation rate in the Kth period and the active participation rate in the K+1th cycle The difference ΔM1 is calculated until the Kth cycle and the K+1th period The difference ΔM n , we get ΔM1, ΔM2···ΔM n , first use the cyclic function to ΔM1, ΔM2···ΔM n Perform data integration, remove duplicate data, and filter out ΔM1, ΔM2···ΔM n All the values x1,x2···x n (m≤n), and use recursive function to find x1,x2···x n The probability of all values appearing in: x1→P1,x2→P2···x m →P m , and then calculate the mathematical expectation value μ1 of active participation:

[0051] μ1=x1P1+x2P2+···+x m P m

[0052] From the above calculation formula, we can know that active participation is the ratio of power, and power is the product of speed and force. According to the active participation ΔM1, ΔM2···ΔM n The same method is used to calculate ΔF1, ΔF2···ΔF n , first use the cyclic function to ΔF1, ΔF2···ΔF n Perform data integration and filter out ΔF1, ΔF2···ΔF n All values y1,y2···y that appear in l (l≤n), and use recursive function to find y1,y2···y l The probability of all values appearing in: y1→p1,y2→p2···y l →p l , and then find the mathematical expectation value μ2 of the force:

[0053] μ2=y1p1+y2p2+···+y l p l

[0054] Then the velocity offset ΔV can be calculated:

[0055]

[0056] Step 5: Adaptive passive training joint movement speed adjustment.

[0057] According to the calculated speed offset, the rehabilitation training speed of the K+2th cycle is compensated for the speed offset based on the passive training joint movement speed of the K+1th cycle:

[0058] V K+2 =V K +ΔV

[0059] At the same time, in the K+3th cycle, V K+2 Rehabilitation training is carried out at a certain speed, and the training angle is the initially calibrated joint range of motion Q. The rehabilitation training speed in the K+4th cycle is calculated according to the above calculation method, and this cycle is repeated until the rehabilitation training time is completed.

[0060] The cyclic calculation process of patient rehabilitation training active participation calculation, speed offset calculation and adaptive passive training joint movement speed adjustment is as follows Figure 2 shown.

[0061] The preferred embodiments of the present invention have been described in detail above. It should be understood that numerous modifications and variations based on the concepts of the present invention are possible without inventive effort by those skilled in the art. Therefore, any technical solution that can be derived by one skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A rehabilitation robot intelligent active-passive hybrid training control method, characterized in that: The following steps are involved: Calibrate the patient's joint range of motion and maximum power: Calibrate the patient's joint range of motion before training begins, use the calibrated joint range of motion as the training angle, and calibrate the patient's maximum power according to the Hill muscle mechanics model; Maximum power P max for: Where V is the joint movement speed, F max is the maximum muscle contraction force; Obtaining the difference in human-machine interaction force: Control the rehabilitation robot joints to train for two cycles according to the calibrated joint range of motion and speed, obtain the human-machine interaction force within the two cycles, and calculate the difference; The human-computer interaction force is the difference between the patient pressure acquired by the pressure sensor and the pressure obtained by the human body dynamics model; Calculating the patient's active participation in rehabilitation training: calculating the power of all sampling points in the two cycles, calculating the mathematical expectation value of the patient's active participation in this training process based on the difference in active participation, and calculating the mathematical expectation value of the human-computer interaction force at all sampling points in the two cycles; The active participation degree is the ratio of the patient's current power to the maximum power; Calculate the speed offset: Calculate the speed offset based on the mathematical expectation value of the active participation and the mathematical expectation value of the human-computer interaction force; the speed offset ΔV is: In the formula, μ1 is the mathematical expectation value of active participation, and μ2 is the mathematical expectation value of human-computer interaction; Adjust the adaptive passive training joint movement speed: adjust the passive training joint movement speed of the third cycle according to the speed offset to achieve the calibrated training angle, and feed back the passive training joint movement speed of the third cycle to the fourth cycle so that the training modes of the third and fourth cycles are the same.

2. The intelligent active-passive hybrid training control method for a rehabilitation robot according to claim 1, characterized in that: The adjusting of the adaptive passive training joint movement speed further includes: calculating the passive training joint movement speed of the fifth and sixth cycles according to the active participation and human-computer interaction force of the third and fourth cycles, and repeating this cycle until the training is completed.

Images