263results about How to "Improve the effect of rehabilitation training" patented technology

The invention discloses an exoskeletal rehabilitation robot for upper limbs. The exoskeletal rehabilitation robot comprises a base, two mechanical arm assemblies and six motor driving components, wherein the base comprises a moving base, an electrical cabinet, an electric lifting post, a base platform, a motor mounting corner bracket, a base rotating motor, a coupling, a base main bearing block, a base and ball screw nut assembly, a base nut seat, a base shared secondary bearing block, a right support, a base linear guide rail assembly and a left support; the mechanical arm assembly comprises a mechanical shoulder girdle assembly, a mechanical shoulder joint assembly, a mechanical elbow joint assembly, a mechanical front arm assembly, a mechanical wrist joint assembly and a mechanical hand part assembly; each motor driving assembly comprises a motor and speed reducing component, a torque sensor component and a driving component which are respectively arranged on the same motor driving base frame. The exoskeletal rehabilitation robot can be worn on the upper limbs of the human body and can be used for assisting the upper limbs of the human body in moving in a three-dimensional space and performing rehabilitation training.

The invention provides a wearable upper limb exoskeleton rehabilitation device, which includes a fixed backboard served as a base; a driver module can make torque force transfer to each joint by Bowden cable; an exoskeleton module of elbow join is used for coupled rehabilitation movement. rehabilitative training of the elbow, coupled rehabilitation of forearm and exoskeleton of the upper arm and the exoskeleton; a mechanical hard limit device for joint, can achieve mechanical hard limit protection for a forearm connecting rod and a upper arm connecting rod; a three direction adjustable adaptive module, the position of joint winch in the shoulder adduction is adjusted to adopt different somatotype of the patients; a shoulder joint module, the shoulder joint module can achieve rehabilitation training for abduction degree of freedom in the shoulder and rehabilitation training of shoulder flexion and extension degree. The Wearable upper limb exoskeleton rehabilitation device makes use of the wearable structure to realize the free movement of patients during rehabilitation training.

The invention relates to a movable parallel wire driven lower limb rehabilitation robot and an implementation method thereof. The robot comprises a suspension weight reducing unit, a tension sensor, safety straps, a safety handrail, a spring, a door fence, wire driving units, universal driven wheels, a leg wearing device, a treadmill, rear driving wheels, wires, a supporting frame and a liquid crystal display screen. The whole robot is of a truss structure formed by the supporting frame, and the suspension weight reducing unit on the upper portion is connected with the safety straps through the wires; the movement of the waist of a patient is constrained through four groups of springs between four vertical beams on the left and on the right and between the safety straps; the liquid crystalscreen is arranged above the safety handrail on the front portion of the robot; the patient and a wheelchair enter and go out through the left door fence; four cross beams at corresponding positionsof an affected limb sagittal plane on each side are provided with the four wire driving units; the affected limb is driven for rehabilitation training by pulling the leg wearing device through the wires; the universal driven wheels are installed on the front portion of the bottom of the robot, and the rear driving wheels are arranged on the rear portion of the robot. The robot can achieve gait training and walking training in passive, power-assisted and active modes.

The invention discloses a novel lower extremity exoskeleton rehabilitation training device and a training method; in accordance with problems on efficiently and safely recovering the walking capacity of lower extremities of a patient, enhancing the adaptability of the training device and other problems in a human lower extremity rehabilitation training process, the rehabilitation training device disclosed by the invention can achieve all-directional rehabilitation training with five freedom degrees of motion on the human lower extremities. The rehabilitation device is added with angle, displacement and pressure sensors, and by detecting related motion parameters of the patient, self-adaptive adjustment is achieved; and since lower extremity exoskeleton, width regulating and height regulating mechanisms and an anti-gravity treadmill are subjected to modular design, training intensity and amplitude can be regulated in accordance with the figure of the rehabilitation patient as well as a rehabilitation progress, so that adaptability is enhanced. By integrating the designed lower extremity exoskeleton on the anti-gravity treadmill, the comfort of the patient in the rehabilitation training is enhanced, and meanwhile, the effect of the rehabilitation training is improved as well.

The invention discloses a rehabilitation and walking exoskeleton robot based on master-slave control. The rehabilitation and walking exoskeleton robot based on master-slave control comprises a wheelchair, an exoskeleton, a central controller and a battery pack, wherein the wheelchair comprises a wheelchair frame, an operation panel, a wheelchair motor driver and a wheelchair motor; the exoskeleton comprises an exoskeleton framework, an exoskeleton motor driver, an exoskeleton motor and a sensor system; and the central controller is capable of receiving a function control signal from the operation panel and a feedback signal from the sensor system, setting a movement working mode and a rehabilitation working mode of the robot, and outputting control signals to the wheelchair and the exoskeleton, thus the motion of the exoskeleton and the movement of the wheelchair can be cooperatively controlled. The rehabilitation and walking exoskeleton robot based on master-slave control, which is disclosed by the invention, adequately considers the subjective motion intention of a user, and is high in functionality, simple to control, safe and reliable, and remarkable in rehabilitation effect.

The invention relates to a robot device provided with an adjustable training plane for hemiplegic rehabilitation of an upper limb, and belongs to the technical field of medical rehabilitation appliance. In order to meet space training requirements provided by a medical robot in the hemiplegic rehabilitation training, the invention provides the rehabilitation robot device provided with the training plane, which comprises a seat, the training plane, an adjusting device for the training plane, a mechanical arm arranged on the training plane, a wrist fixing bracket and an elbow bracket. The adjusting device for the training plane can adjust the height of the training plane relative to the ground and the included angle between the training plane and the horizontal plane, so as to adapt to the requirements of different patients to finish a plurality of rehabilitation training actions in a plurality of planes. The robot device has the advantages that the robot device has a simple structure, can meet the requirements of the patient on large active and passive training on various inclined planes through adjusting the pose of a working face, and can do special training for an affected joint and a corresponding muscle group through target-oriented adjustment of the inclined planes, so as to achieve better rehabilitation effect.

The invention provides a soft-bodied robot glove for hand movement function recovery. The soft-bodied robot glove comprises a decision processing module, a driving module, a flexible exoskeleton glove and a Bowden cable. The decision processing module is electrically connected with the driving module. One end of the Bowden cable is connected with the finger end of the flexible exoskeleton glove, and the other end of the Bowden cable is connected to a winch of the driving module. The decision processing module drives the Bowden cable through the winch of the driving module. By the adoption of the flexible exoskeleton glove, compared with a rigid exoskeleton, the wear comfort can be improved. The driving module and the flexible exoskeleton glove are separated, so that the wear burden on a user is reduced.

The invention discloses a multi-mode fusion hand function rehabilitation training and intelligent evaluation system, which comprises a rehabilitation training mechanism, a sensor data collecting device, a rehabilitation evaluation system and a scene interaction game system, wherein the rehabilitation training mechanism is used for performing rehabilitation training on the fingers and the wrist; the sensor data collecting device is used for collecting one or various kinds of signals from a surface electromyographic signal, a force signal and a posture signal in real time; the rehabilitation evaluation system performs quantitative evaluation on the rehabilitation training condition of a patient according to a multi-mode signal collected by the sensor data collecting device; and the scene interaction game system is used for providing information feedback such as rehabilitation training, effect evaluation and vision for the patient. The multi-mode fusion hand function rehabilitation training and intelligent evaluation system disclosed by the invention has the advantages that by aiming at the rehabilitation training of a cerebral stroke patient, real-time feedback regulation and control, quantitative evaluation and game feedback can be performed in the rehabilitation training process; the training enthusiasm of the patient is effectively aroused; and the rehabilitation training effect is enhanced.

The invention discloses a multi-pose lower-limb rehabilitation robot and a use method thereof, and belongs to the field of medical apparatuses. The rehabilitation robot comprises a pedestal, a liftingmechanism and a bed body which are arranged sequentially from bottom to top; a pair of thigh training mechanisms and a pair of leg training mechanisms are symmetrically arranged on the two sides of the bed body correspondingly; the thigh training mechanisms and the leg training mechanisms are used for performing lower-limb rehabilitation training on a patient whose lower limbs have dyskinesia; the patient is subjected to lower-limb rehabilitation training by multiple modes; furthermore, the bed body and a seat cushion are overturned, so that training of three poses of sitting, lying and standing can be realized, the training mode is comprehensive and the training effect is good. By the method adopting the multi-pose lower-limb rehabilitation robot, the training effect is excellent.

The invention discloses a double-lasso driving flexible joint used for an exoskeleton robot, which comprises a drive mechanism, a pre-tightening mechanism, a lasso transmission mechanism, a flexible mechanism and an exoskeleton joint. A motor drives a drilling roller to rotate; the two ends of a rope are respectively fixedly connected with the driving roller and a load roller; a driving torque is delivered to the load roller through the lasso transmission mechanism; the load roller is fixedly connected with the exoskeleton joint to drive the joint to move; a positioning hole and a positioning groove both capable of limiting the motion range of the joint are formed in the load roller; the pre-tightening mechanism adjusts a pre-tightening force of a rope by changing the screwing depth of a pre-tightening bolt so as to avoid looseness of a drive system; the flexible mechanism enables the joint to achieve flexibility and can measure an output torque based on a displacement offset of the motor and the joint. The double-lasso driving flexible joint used for the exoskeleton robot can realize separation of the motor and the exoskeleton robot, overcomes the problems of the quality of each joint of the exoskeleton and too great inertia, and can perform precise position control and force control on the drive joint.

The invention discloses an intelligent limb rehabilitation training system and method. The system comprises a rehabilitation teacher module, a patient training module, an intelligent analysis decisionmodule and a remote network module. According to the intelligent limb rehabilitation training method, a rehabilitation goal is set by a rehabilitation teacher according to the patient condition, a rehabilitation training plan is made, and the rehabilitation situation is intelligently generated by the system according to setting of the rehabilitation teacher; rehabilitation training is performed by the patient according to the system guidance prompt in the rehabilitation situation, training data of the patient is obtained and recorded by the system, intelligent analysis of the data is performed, the recovery condition of the patient is evaluated, according to the evaluation result, the training content, training difficulty and the training progress are automatically adjusted, the trainingprocess is made to be adaptive to the patient, intelligent and personalized rehabilitation training is achieved, efficiency of rehabilitation training is improved, and effectiveness of rehabilitationtraining is improved.

The invention belongs to the field of an upper limb exoskeleton rehabilitation robot, and discloses a flexible exoskeleton glove system for hand rehabilitation training. The flexible exoskeleton glove system comprises a finger data collecting glove, a decision driving module and a flexible exoskeleton glove; the finger data collecting glove comprises a first glove main body, a flexible sensor and a wireless sending module; the decision driving module comprises a decision module and a driving module; the decision module comprises a microcontroller and a wireless receiving module connected with the microcontroller; the driving module comprises a module supporting bracket, an air pump and an electrical proportional valve installed on the module supporting bracket commonly; the flexible exoskeleton glove comprises a second glove main body and a flexible bending driver; the flexible bending driver comprises a Kevlar fiber wire, a silicone tube, a glass fiber cloth and a pneumatic sealing joint. In the invention, fingers can be independently moved or coordinately moved, thus the hand rehabilitation training effect and wearing comfort of patients are improved.

The invention discloses a multi-person interactive virtual reality rehabilitation training and evaluation system comprising a multi-source sensor module, a signal receiver, a computer-side and a virtual reality processing unit. Multiple patients carry out group rehabilitation training in rehabilitation training fields or in rehabilitation training groups by means of the Internet; aiming at rehabilitation training on patient body parts, the wearable multi-source sensor module based on acceleration, gyroscopes, magnetometers, force sensors, and bending sensors is developed for data collecting, signal processing and feature extracting for rehabilitation training; multiple daily life situation interactive games are developed and models and interfaces of rehabilitation training movement characteristics and daily life situation interactive games are established so that patients carry out multiple-person interactive feedback rehabilitation training and rehabilitation assessment is carried out on the rehabilitation training effects of patients. The system of the invention is good for increasing the activeness of patients in rehabilitation training, enhancing the rehabilitation training effects and increasing the daily life self caring ability of patients.

The invention discloses an autism cognitive education reinforcement, intelligent evaluation and rehabilitation training system which comprises a rehabilitation device, an evaluation device and an auxiliary device. Settings of basic information and parameters of a patient are input through an input module in the rehabilitation device, rehabilitation treatment and collection of physiological data and facial expression images are conducted through a sensor module, and operation data of the patient is recorded in real time when the patient carries on operation training according to the content designed in a man-machine interaction image; the evaluation device conducts processing according to the data obtained by the rehabilitation device to generate an evaluation report and a control command,the rehabilitation device adjusts the training parameters and the control command according to the rehabilitation training scheme sent by an external doctor terminal to drive an electrode to work, andmeanwhile the auxiliary device plays a brain wave music according to the evaluation report to reinforce the rehabilitation training. The system has the advantages that learning training and evaluating are integrated, the parameters in the rehabilitation device are adjusted according to the evaluation report, and therefore more scientific service is provided with the autism patient.

The invention discloses a direct-driven rehabilitation exoskeleton. The direct-driven rehabilitation exoskeleton comprises a machine frame, hip joint torque motors, thigh exoskeleton bodies, knee joint torque motors and shank exoskeleton bodies. The invention further discloses a training method. The training method comprises the steps of obtaining the angles of hip joints / knee joints, setting a gait curve and exoskeleton output torque, obtaining the torque applied to the hip joints / knee joints, calculating the matching degree difference between the exoskeleton and a patient, setting the matching sensitivity, adjusting the assistance coefficients of the joints and performing rehabilitation training. The direct-driven rehabilitation exoskeleton adopts the torque motors to directly drive lower limb joints of the patient to move, no extra sensor is needed, the rehabilitation torque applied to each joint of the patient is accurately measured and monitored, the training method is designed according to the joint rehabilitation torque of the patient, so that the assistance magnitude of the rehabilitation exoskeleton can be matched with the rehabilitation requirement of the patient, the rehabilitation training effect of the exoskeleton can be improved, and meanwhile the rehabilitation effect evaluation can be made according to the rehabilitation torque.

The invention belongs to the field of medical rehabilitation technology, and discloses a virtual reality cerebral apoplexy upper limb rehabilitation training system and a control method thereof. The system comprises a scene constructing module, an action data acquisition module, a parameter configuring module, a central control module, an expert diagnosis module, a rehabilitation reminding module,a massage module, a VR displaying module and a training data generating module. According to the system of the invention, a rehabilitation plan which is made by medical caring personnel is compared with a stored rehabilitation plan through the rehabilitation reminding module; on the condition of no conflict, the rehabilitation plan is stored and preset reminding time is set, thereby facilitatingaccuracy in making the rehabilitation plan; furthermore through reminding a patient at preset reminding time, executing efficiency and accuracy of the rehabilitation plan can be improved; and furthermore virtual training can be realized through the scene constructing module, the action data acquisition module and the VR displaying module. The virtual reality cerebral apoplexy upper limb rehabilitation training system and the control method thereof have advantages of reducing training space occupation space, enriching training modes and improving rehabilitation training effect.

The invention discloses an adult hearing and speaking rehabilitation system comprising an information management module, a parameter setting module, a single character pronouncing assessment module, a single character pronouncement training module, a word reading training module, a speech recording and playing module, an automatic speaking assessment module and a random sequence producing module. The system is applied to speaking and reading training for aphasia patients with repetition impediment or loud reading impediment and dysarthria during speaking rehabilitation, the speaking feedback ability of the aphasia patient is improved, the pronouncing accuracy level of the dysarthria patient is raised, the continuous pronouncing coordinate ability is improved, and the work intensity is reduce significantly for speaking therapists. Multisensory reinforcing stimulation is performed on the patient in multimedia interacting manners of graphs, texts, dynamic images and voices, the subjective initiation of the patient for the rehabilitation training is improved, and the rehabilitation training effect is improved; the simple, boring and repeated work can be reduced for the therapists.

The invention relates to a multifunctional limb intelligent rehabilitation machine and a computer auxiliary control method thereof. The multifunctional limb intelligent rehabilitation machine comprises an upper limb rehabilitation mechanism, a weight reduction protection squatting rehabilitation mechanism and an induction type track walking machine. The induction type track walking machine is located below the upper limb rehabilitation mechanism, and the weight reduction protection squatting rehabilitation mechanism is located above the induction type track walking machine and is opposite to the upper limb rehabilitation mechanism. The upper limb rehabilitation mechanism, the induction type track walking machine and the weight reduction protection squatting rehabilitation mechanism are connected with a computer. The rehabilitation training motions can be led through corresponding training mechanisms; according to remaining functions of limbs of patients, a computer auxiliary control technology is utilized to generate a rehabilitation training plan, life signals and motion states of the patients are monitored in real time in training, the running amplitude and speed of the training mechanisms are adjusted, and the safety and effectiveness in rehabilitation training are guaranteed.

The invention provides a motion evaluation method of shadowboxing based rehabilitation training. The motion evaluation method comprises the steps of S1, building a standard shadowboxing animation model library; S2, building a trainer model; S3, building virtual training scenes and setting training modes and training motion levels; S4, registering or carrying out account password login by a trainer; S5, selecting a suitable virtual training scene, a suitable training mode and a training motion level; S6, training, including, carrying out shadowboxing training by imitating the standard shadowboxing animation model library of the training mode and the training motion level selected in the step S5 by the trainer, collecting limb motion data of the trainer, mapping the limb motion of the trainer to a trainer model in real time, and shooting a training video; S7, carrying out training scoring and evaluation; and S8, querying historical training data. The motion evaluation method of the shadowboxing based rehabilitation training can well combine the traditional shadowboxing with the virtual reality rehabilitation training, and is good for improving the rehabilitation training effect of the trainer.

The invention discloses a dual-arm support rehabilitation training robot and an operation method of the dual-arm support rehabilitation training robot. The dual-arm support rehabilitation training robot comprises a crawler type walking machine, a base, a lifting screw mechanism, a manipulator part and a counterweight mechanism, wherein the manipulator part comprises shoulder joints, rotary joints, elbow joints and wrist joints, wherein each joint comprises a motor, a harmonic reducer and an output end. By adopting the dual-arm support rehabilitation training robot, the rehabilitation training of three-linear degree-of-freedom and two-rotational degree-of-freedom on hips of a human body can be realized in a gait training process; physiological movements of the hips in the gait training process can be highly simulated to effectively enhance the rehabilitation training effect.

The invention provides a pelvis posture control mechanism for assisting walking rehabilitation training. The posture control mechanism comprises waist supports, load bearing hinges, push rods, electric cylinders, extensible motors, extension arms, fixed arms, a base, a sliding platform, a guide rail motor, a load bearing box and an inclination motor. The base can move up and down along a guide rail in the vertical direction of a robot to realize moving degree of freedom of pelvis in the vertical direction; a load bearing plate is fixed to the sliding platform, and the guide rail motor drives the sliding platform to move while moving degree of freedom of the pelvis in the left-right direction is realized; the inclination motor drives the fixed arms to realize inclination motion of the pelvis; the push rods extend and contract forwards and backwards to realize twisting motion of the pelvis; the waist supports can rotate along the load bearing hinges to realize pitching motion of the pelvis. The pelvis posture control mechanism can actively help a trainee to complete posture needed by pelvis natural gait, reduce constraint on degree of freedom of the trainee and improve training comfort of a patient, balance control ability of the patient in the process of lower limb walking and lower limb rehabilitation training effect.

The invention relates to the field of augmented reality, and relates to a gait rehabilitation training evaluation method and system based on the augmented reality. The method comprises selecting a walking training environment of lower limbs according to an evaluation result of a walking function of a rehabilitation training patient; selecting a lower limb walking training mode; projecting the patient footprints onto a treadmill conveyor belt to build a walking training augmented reality environment; guiding the patient to perform walking training, judging the hit rate of the feet of the patient on the augmented reality footprints in the walking training augmented reality environment in the preset time period and giving the feedback; after the end of the walking training, processing the rehabilitation training data and outputting a score loss analysis and rehabilitation evaluation report of the walking training. According to the method and system, the walking training augmented realityenvironment is established to guide the patient to perform targeted walking training, a rich walking training scheme is provided, the interest of the patient walking training is improved, and the lower limb walking rehabilitation training effect is improved.

The invention discloses an active rehabilitation training method for the bilateral limb coordinated movement of a hemiplegic patient. The active rehabilitation training method comprises the steps that active movement electroencephalogram, healthy and affected side muscle signals, healthy and affected side muscle signals and healthy and affected side movement trajectory signals of the hemiplegic patient are acquired; movement prediction is performed by using a deep learning method according to the acquired signals, and a movement prediction result is obtained; and the obtained movement prediction result is coordinated and controlled and then input to an affected side joint movement control system. The active rehabilitation training method adopts the deep learning method to perform movement prediction, and the movement prediction result is coordinated and controlled and then input into the affected side joint movement control system, so that the affected side joint movement control system controls the affected side joint and the healthy side joint to maintain coordinate and unified movement, and the effect of rehabilitation training for the hemiplegic patient is greatly improved. The active rehabilitation training method for the bilateral limb coordinated movement of the hemiplegic patient can be widely used in the field of computer medical application

The invention discloses an autonomous rehabilitation training system and method. The system comprises: an unaffected side ankle motion detection device, an affected side ankle rehabilitation training device, an electromyograph (EMG) detection device, and an functional electrical stimulation (FES) device; the unaffected side ankle motion detection device is used mainly for detecting patient's unaffected side data of unaffected side ankle motion, such as angle, speed and exerting power; the affected side ankle rehabilitation training device is used mainly for training the patient's affected side for rehabilitation and detecting the patient's unaffected side data of affected side ankle motion, such as angle, speed and exerting power; the EMG detection device is used mainly for detecting EMG signals of unaffected side muscles when the patient is in motion; the FES device is used mainly for performing electrical simulation auxiliary training for affected side muscles. The patient's subjective activity and passive training are combined fully to improve rehabilitation effect; the autonomous rehabilitation training system combining patient's unaffected side motion and affected side training is established using multi-information fusion technology.

The invention discloses a multifunctional double-supporting rehabilitation exercising robot. The multifunctional double-supporting rehabilitation exercising robot comprises a crawler type walking machine, a base, a lifting screw rod mechanism, a manipulator part and a counterweight mechanism, wherein the manipulator part comprises shoulder joints, revolving joints, elbow joints and wrist joints; each joint comprises a motor, a harmonic speed reducer, and an output end. The multifunctional double-supporting rehabilitation exercising robot has the advantages that the rehabilitation exercising on three linear degrees of freedom and two revolving degrees of freedom of the hips of human body is realized in the step training process; the physiological movement of the hips can be highly simulated in the step training process, and the effect of rehabilitation training is effectively improved.

The invention discloses soft-bodied hand rehabilitation gloves through combination of diverse rehabilitation training modes. Each of the soft-bodied hand rehabilitation gloves comprises a training glove body, soft-bodied exoskeleton fingers which can realize bi-directional bending and is convenient to disassemble, and a pneumatic control system. The invention further discloses two hand rehabilitation training modes which can be finished by using the soft-bodied hand rehabilitation gloves. Through bi-directional bending of the soft-bodied exoskeleton fingers, different action force can be provided for finger movement of users so as to complete two training modes. The invention aims to provide novel rehabilitation training equipment for hand function obstacle crowds. In accordance with the problem that the hand rehabilitation equipment is single in training modes, the invention provides a solving scheme, and besides, the soft-bodied hand rehabilitation gloves have the advantages of beinghigh in safety, low in price, favorable in adaptability and the like, and are convenient to launch.

The invention relates to a compound rehabilitation system. The rehabilitation system comprises a rehabilitation training bed, a lower-limb pulling muscle joint training module in the accommodating cavity and a control system, wherein the rear half part of the rehabilitation training bed is provided with an accommodating cavity; a lower limb blood circulation rehabilitation module is fixed on the lower-limb pulling muscle joint training module; and the control system comprises a microprocessor, a memory, an input module, an output module, a rotary coder, a pulling motor speed-regulation controller, an overload torque sensor, an air pressure sensor, a temperature sensor and a communication module. The compound rehabilitation system integrates joint motion, muscle extension, blood vessel function and blood circulation into a whole, is used for controlling motion of the rehabilitation system by means of the control system, and is intelligent since spontaneous determination can be carried out according to the rehabilitation training condition of a patient; and rehabilitation training condition of a patient can be carries safety and efficiency.

The invention discloses an upper limb rehabilitation training exoskeleton mechanism with human-machine motion compatibility and passive gravity balance characteristics. The upper limb rehabilitation training exoskeleton mechanism is mainly in a series-connected structural form, and comprises an exoskeleton mechanism, a human-machine connecting mechanism and a supporting piece; the exoskeleton mechanism is fixedly connected with the supporting piece on the back; the human-machine connecting mechanism is closely connected with an upper limb of a human body to form a human-machine closed chain; the human-machine connecting mechanism is divided into an upper arm connecting mechanism and a front arm connecting mechanism, so that the human-machine closed chain has exact constraint. An exoskeleton comprises a shoulder joint equivalent mechanism, an elbow joint equivalent mechanism, a front arm equivalent component and an upper arm equivalent component, wherein the shoulder joint equivalent mechanism and the upper arm equivalent component assist the upper arm to realize the motion of bending / extending, folding / unfolding, and inwards rotating / outwards rotating; the elbow joint equivalent mechanism and the front arm equivalent component assist a front arm to realize the motion of bending / extending; the front arm connecting mechanism is connected with a front arm component to assist the front arm to realize the motion of inwards rotating / outwards rotating; the existence of the deviation of a human-machine axis is completely allowed; acting force which has nothing to do with training cannot be produced in the human-machine closed chain.

The invention discloses a mutual-assistance walking device for upper limbs and lower limbs. The device comprises a leg outer framework (1) and a fixing belt (2) which is connected with the leg outer framework (1), the upper end of the leg outer framework (1) is connected with a mutual-assistance hand grip (3), and the lower end of the leg outer framework (1) is connected with a shoe support (4); the leg outer framework (1) comprises an upper leg framework (5) and a lower leg framework (6), and the upper leg framework (5) is movably connected with the lower leg framework (6). The mutual-assistance walking device for the upper limbs and lower limbs is simple in structure and low in equipment cost, a patient with lower limb paralysis can use the device separately in the rehabilitation training process, the coordination of the patient can be trained, and the rehabilitation training effect is improved.

Embodiments of the invention provide an evaluation and handling method and device for rehabilitation training, and equipment. The method comprises the following steps: acquiring rehabilitation training video data corresponding to a to-be-rehabilitated user and acquired by video acquisition equipment, and determining a motion image set according to the rehabilitation training video data; identifying the moving image set by using a motion function evaluation model, and determining an action parameter value corresponding to each moving image in the moving image set, wherein the motion function evaluation model is obtained by conducting training according to the motion training image set; and based on a pre-stored rehabilitation medical knowledge graph, analyzing and processing the motion parameter values corresponding to the motion images in the motion image set to obtain a training result, and sending the training result to terminal equipment corresponding to the to-be-rehabilitated user. Thus, rehabilitation training effect is improved, and then the rehabilitation experience of the user to be rehabilitated is improved.