introduction
In 2015, the World Health Organization pointed out that people over the age of 60 accounted for 12.24% of the total population, and this proportion is increasing year by year. Stroke is a high incidence among the aging population, and its incidence rate is as high as 11. 2%, which has seriously affected people's daily life.
According to relevant statistics, there are about 2.5 million new stroke patients in my country each year. By 2020, the number of stroke patients in my country will reach 20 million. Among patients with hemiplegia caused by stroke, the majority of people have hand dysfunction.
Scientific research has found that, on average, people perform approximately 1,500 grasping movements a day, which is an important tool for communication between people and the external environment. Therefore, it is of great significance for the rehabilitation of the hands of patients with hemiplegia and the assistance of the patients' daily life.
Existing experimental studies have confirmed that continuous passive reinforcement of periodic training for patients can improve the hand movement function of patients in the later stage of stroke. Hand rehabilitation robots have gradually attracted widespread attention from all walks of life because they can better assist rehabilitation therapists in the rehabilitation of patients with hemiplegia, and at the same time allow patients to perform self-rehabilitation auxiliary training.
The hand rehabilitation robot is a mechatronic system composed of a driver, a force transmission mechanism, an actuator, and a control module. It can drive the fingers to perform a specified movement, and its different control modes can achieve different functions of rehabilitation assistance training.
Research on hand rehabilitation robots originated in Europe and the United States. In the early 21st century, there were commercialized hand rehabilitation robots abroad, such as the hand exoskeleton robot of Carnegie Mellon University in the United States, and the hand exercise assisted rehabilitation developed by Gifu University in Japan. Robots, Hand Mentor rehabilitation device developed by KMI Corporation, etc. The research on hand rehabilitation robots in my country is relatively late. Most hand rehabilitation robots are still in the experimental development stage, such as the trauma finger rehabilitation exoskeleton hand of Harbin Institute of Technology, and the finger exoskeleton of Beihang University.
1. Analysis of human hand characteristics
In the characteristics of the human hand, the size of the bones of the hand and the driving force distance required by each joint will affect the structural design and power selection of the rehabilitation robot. Therefore, this article introduces the standard values of the degrees of freedom, joint motion angles, and grip strength of healthy hands, which can provide references for the structural design and power selection of hand rehabilitation robots.
①Analysis of the biological characteristics of the
human hand The human hand is composed of 27 bones with 20 degrees of freedom, which can realize flexible and delicate movement. The thumb has 3 joints, namely the interphalangeal (IP) joint, metacarpophalangeal (MCP) joint, and carpometacarpal (CMC) joint.
Both IP joints and MCP joints can achieve 1 degree of freedom in flexion/extension, and CMC joints can achieve 2 degrees of freedom in flexion/extension and contraction/extension. The index finger, middle finger, ring finger, and little finger are composed of 3 joints, namely the distal inter-phalangeal (DIP) joint, the proximal interphalangeal (PIP) joint, and the metacarpophalangeal (MCP) joint. Among them, DIP joint and PIP joint can realize 1 degree of freedom of flexion/extension, and MCP joint can realize 2 degrees of freedom of flexion/extension and retraction/extension. Therefore, the human hand has 20 degrees of freedom to complete complex movements. The hand muscles of patients with hemiplegia are in a state of contracture, which makes some joints unable to move autonomously, and the degree of freedom and the range of motion of the joints are greatly reduced.
②Analysis of human hand movement characteristics
Jones et al. of the Illinois Institute of Technology in the United States pointed out that the torque that maintains the normal movement of the three joints of the human hand MCP, PIP, and DIP are 2.0 N·m, 0.75 N·m, and 0.25 N·m, respectively. Lince et al. pointed out the distribution of the grasping force of each finger during the grasping process: thumb 51% ± 0.01%, index finger 25% ± 0.05%, middle finger 12% ± 0.04%, ring finger 7% ± 0.02 %, the little finger 4% ± 0.02%, roughly obey the 1 /2 times descending relationship. Experiments such as Ang found that the maximum motion angles of each joint of a healthy hand, the maximum motion angles of the thumb IP, MCP, and CMC joints are 98.8° ± 4.9°, 23.5° ± 3.1°, 25. 2°±3.6°, the maximum movement angles of the four-finger DIP, PIP, and MCP joints are 69.4°±7.9°, 103.4°±1.1°, 90.4°±2.5, respectively °. Table 1 shows the numerical results of a single fingertip force measurement in daily life activities.
Whether the hand rehabilitation robot can accurately reproduce the degree of freedom of the hand, the angle of joint motion, and the grasping fingertip force of each finger are important parameters to measure the effectiveness of its rehabilitation. In addition, the driving method, mechanical structure, driving force transmission method, and portability of the hand rehabilitation robot will also affect the rehabilitation assistance effect. Among them, the mechanical structure of the hand rehabilitation robot is the key to determining the effect of rehabilitation assistance, and it has a greater impact on the choice of drive mode, portability and control mode selection.
2. The structure of hand rehabilitation robot
The current hand rehabilitation robots are mainly divided into rigid exoskeleton type and flexible wearable type from the structure.
Rigid exoskeleton type
Rigid exoskeleton type refers to a hand rehabilitation robot that transmits driving force to the human hand through rigid components such as connecting rods, gears, and crank sliders and drives the fingers to move. After nearly 30 years of development, the technology has become relatively mature. In recent years, there are still many scientific researchers at home and abroad who are innovating and researching.
In recent years, domestic research has also made some progress, and many universities have conducted research on hand rehabilitation robots. The wearable hand functional rehabilitation robot system designed by Harbin Institute of Technology uses a rack and pinion parallel sliding mechanism, which can realize the flexion and extension of the three joints of the finger and the adduction/abduction of the MCP joint. The rotation center of the exoskeleton joint can be very It coincides well with the rotation center of the human hand joint, and can adapt to palms of different sizes at the same time.
The hand function rehabilitation robot developed by Huazhong University of Science and Technology is driven by an air pump. The air pump is used to inflate and deflate pneumatic muscles to realize the movement of the mechanism. The device has 3 degrees of freedom and is driven by 3 pneumatic muscles to drive the corresponding 3 exoskeletons. The joints use tension springs to control the tension of the wire ropes, which can adapt to different lengths of fingers, and use tension and pressure sensors to measure the angular displacement of the robot joints. Rigid exoskeleton hand rehabilitation robots have the characteristics of accurate motion transmission, precise control, and mechanical protection limit, so there are still many scholars to study this, but its mass, poor adaptability, and rigid impact.
At present, researchers at home and abroad mainly use two ways to reduce this deficiency: one is to reduce the weight of the exoskeleton by choosing lightweight materials, and the other is to use a rigid-flexible coupling method to drive the rigid exoskeleton behind and use elastic members. As a power transmission part, it eliminates the rigid impact of the exoskeleton and reduces the weight of the hand mechanism, thereby reducing the weight of the patient's back.
Flexible wearable
The flexible wearable type refers to a hand rehabilitation robot that transmits driving force to the human hand through elastic elements such as springs, Bowden cables, and pneumatic muscles, and uses the elastic elements to drive the fingers to move. Although the development of the flexible wearable type is not as early as the rigid exoskeleton type, it has developed rapidly in recent years.
Researchers in Malaysia use cooperative rehabilitation training with both hands to realize the movement control of the affected hand by collecting data from the patient's healthy hand. The flexible gloves are driven by a combination of rubber rope and rope, and the rubber rope is used to realize the stretching movement of the fingers. The flexion of the fingers is realized by the motor driving the rope on the back of the hand and overcoming the tension of the rubber rope.
Yang et al. of Northeastern University proposed to realize the associated movement of the three joints of MCP, PIP, and DIP through the method of rope coupling, so that the joint movement trajectory of the flexible rope glove during the driving process is more in line with the law of healthy hand movement.
Compared with the rigid exoskeleton type, the flexible wearable hand rehabilitation robot has the characteristics of lighter weight and high fit with the human hand. In recent years, domestic and foreign researchers have conducted a lot of research on this, but its lack of mechanical structure leads to movement. The transmission is inaccurate and the control is difficult. At present, researchers at home and abroad mainly use sensors to collect motion data to achieve closed-loop control of the system and improve the accuracy of motion. The characteristics of hand rehabilitation robots at home and abroad are summarized in the table below.
3. Hand data collection
In order to ensure the accuracy and effectiveness of the hand rehabilitation robot, it is necessary to detect the joint motion angle and driving torque/force of the human hand in real time, so as to realize the closed-loop control of the system and improve the accuracy and stability of the system. Hand data collection methods at home and abroad generally use sensor elements and motion capture systems.
Joint angle acquisition
The sensors used to measure the angle of joint movement include potentiometers, flexible resistance sensors, and curvature sensors. Among them, the flexible electric group sensor is small in size and easy to install, and is widely used for joint motion angle measurement. There are already camera-based motion angle measurement software. The working principle is to mark the finger joints. During the movement, the camera automatically captures the marked points, and finally displays the angle information of the marked points on the computer in real time.
Driving torque/force acquisition
The sensors for measuring the hand driving torque/force include torque sensors, air pressure sensors, and gasket-type pressure sensors. Among them, gasket-type pressure sensors are widely used to measure the finger tension of rehabilitation robots.
Other data collection The collection of
motion trajectory uses the existing motion capture system, such as MoCap, Prime 13, which can display the real-time motion trajectory of the calibration points of the finger joints. In addition, the sensors used for hand data collection include acceleration sensors and EMG sensors, which are used to measure the position and posture of the hand and collect the electromyographic signal of the forearm of the hand.
Four, control mode
The current control modes of hand rehabilitation robots can be divided into active control and passive control according to the source of the motion signal.
Active control
Active control is mainly used for patients with mild hemiplegia who have residual muscle strength in the forearm. The movement signal comes from the patient's own physiological signal.
Active control includes: master-slave control, EMG control, and EEG control. The master-slave control is to collect the force, angle, torque and other information of the healthy hand of the patient through sensors, and transmit it to the affected hand in real time, and use the normal movement information of the healthy hand to assist the patient to achieve rehabilitation training.
The hand exoskeleton proposed by Leonardis et al. in 2015 controls the flexion/extension movement of the patient's hand by collecting the EMG signal of the healthy hand. The electromyography control is to collect the electromyography signal of the corresponding muscle group of the patient's hand, analyze and process the electromyography signal, obtain the correct movement intention of the affected hand, and then realize the control of the motor.
Italy's Lince et al. controlled the current of the drive motor by collecting the electromyographic signal of the forearm surface, and then controlled the magnitude of the force on each finger. At present, the active control of hand rehabilitation robots is mostly electromyographic control. EEG control is to control the entire device by collecting the patient’s EEG signal and extracting the effective movement intention from the signal.
The mano flexible underactuated rehabilitation robot designed by Randazzo et al. is used to assist patients in daily living activities and neurorehabilitation. The patient needs to wear an EEG helmet to collect EEG signals, and only use EEG to decode the motion image, and then control the exoskeleton hand.
Passive control
Passive control is mainly used for patients with severe hemiplegia without residual muscle strength of the forearm. The source of the motion signal is directly from the controller.
It includes: preset control, motion capture control, trajectory control, etc. Pre-set control is the most widely used passive control method in the existing hand rehabilitation robots. The motion trajectory of the rehabilitation robot is preset by pre-setting the driving parameters of the motor or the air pump, and then the rehabilitation auxiliary training is realized.
The HES exoskeleton hand designed by Italian researchers and the pneumatic exoskeleton hand of the National University of Singapore all use buttons to control the flexion/extension movement of the exoskeleton. The patient only needs to operate the buttons to achieve the expected rehabilitation training.
Motion capture control uses infrared cameras, ultrasonic ranging probes, etc. to capture the object to be grasped. When the human hand is close to the object, the position information of the object is sensed in advance and fed back to the controller, and finally the movement of the motor is realized.
The exoskeleton gloves designed by Popov and others are mainly used for daily life assistance. There is an infrared ranging sensor under the wrist. When the hand is close to an object, the infrared ranging sensor will automatically capture the object and transmit the information to the control module in real time, thereby controlling the motor work and achieving passive control.
5. Summary and outlook
From the research in recent years, it can be seen that flexible wearable hand rehabilitation robots are developing rapidly, especially the rope type and pneumatic muscle type. Traditional rigid exoskeleton type hand rehabilitation robots are also being improved in terms of weight reduction. At present, 3D printing molding technology and rigid-flexible coupling methods are used more to make exoskeleton robots lighter and more flexible. At the same time, the integration of sensor technology makes hand rehabilitation more intelligent.
At present, hand rehabilitation robots have made great progress compared with the 1990s, but there is still room for in-depth research.
①Problems with retention of hand touch. Hand rehabilitation robots all need to be worn on the human hand, but after being worn, the patient will lose the touch of the finger during the rehabilitation assistance process, and the touch is one of the main functions of the human hand in daily life.
② Adaptability of rehabilitation robots. Different people have different finger lengths and thicknesses. However, none of the existing rigid/flexible rehabilitation robots can well adapt to the size of the affected hand, which makes the use of exoskeleton hands limited.
③Portability problem. Although the hand rehabilitation robot has gradually changed from a rigid exoskeleton to a flexible wearable type, its weight has been greatly reduced, but because its drive still uses a motor or air pump, it is not convenient to carry for a long time, which limits the hand rehabilitation robot’s Range of use. In order to enable hand rehabilitation robots to enter homes and communities and better serve humans, portable and flexible hand rehabilitation robots will surely become the mainstream of hand rehabilitation robots. At the same time, many hand rehabilitation robots have been combined with virtual reality technology. It is believed that as virtual reality (VR) technology continues to mature, future hand rehabilitation training will also be more interesting.
references
[1] Chang Ying, Meng Qingyun, Yu Hongliu. Research progress of hand rehabilitation robot technology［J］. Beijing Biomedical Engineering, 2018, Vol. 37 No. 6
"Shh! Quietly insert a small commercial"
The specific parameters of BCIduino 8-channel EEG amplifier are as follows:
Input impedance: 1TΩ
Input bias current: 300pA
Input reference noise: 1μVpp
Sampling rate: 250 Hz/500Hz
Common mode rejection ratio: -110dB
Adjustable gain amplification factor: 1, 2, 4 , 6, 8, 12, 2
resolution: 24-bit ADC, accuracy up to 0.1μV
power consumption: 39mW in normal operation, as low as 10μW in standby,
powered by a rechargeable lithium battery, further reducing external interference.
Size: 50mm*50mm (physical measurement, with slight errors)
Figure 1 BCIduino physical image
Figure 2 BCIduino data waveform in an ordinary noisy environment, suspended state, you can observe no other interference occurs
Figure 3 OpenBCI in an ordinary noisy environment, Data waveform in the floating state (the measurement environment, measurement time, and software filter setting parameters are the same as those in Figure 2BCIduino)
**
Guest officer, please stay
This article is organized or written by the BCIduino brain-computer interface open source community. The BCIduino brain-computer interface community was initiated and established by masters and doctors from Beihang University, Cornell University, Peking University, Capital Medical University, etc. Welcome to join the community by scanning the code and remark "BCI", and also welcome to purchase BCIduino EEG module (some Search for treasure)