129 results about "Gait (human)" patented technology

The invention relates to the technical field of rehabilitation medical treatment, and discloses a muscle group stress analysis method and device, a terminal and a storage medium, and the method comprises the steps: respectively obtaining motion data collected by inertial measurement units arranged on the body of a user, and recognizing gait events based on the motion data corresponding to the inertial measurement units of the lower limbs, dividing the continuous motion data into a plurality of gait cycles according to the gait event; fusing the motion data acquired by each inertial measurement unit to obtain a three-dimensional posture of each preset segment of the body, and constructing a three-dimensional motion track of the spine area of the user in combination with a human kinematics chain model; processing the three-dimensional motion track and the known muscle group stress data by adopting a singular value decomposition embedding regression method to obtain a preliminary target muscle group stress curve of each gait cycle; and correcting the initial target muscle group stress curve based on a human body multi-rigid-body dynamic model in combination with a control cost function to obtain a final target muscle group stress curve.

The invention discloses a double-clutch variable-stiffness knee joint exoskeleton, and relates to the technical field of wearable lower limb exoskeletons.The double-clutch variable-stiffness knee joint exoskeleton comprises a gait sensing module arranged outside and a limb connecting arm module used for being connected with a human body, and an energy storage module is installed on one side of the limb connecting arm module; a sensor module is mounted on the other side of the limb connecting arm module; the limb connecting arm module comprises a limb connecting arm assembly I used for being connected with the thigh of the human body and a limb connecting arm assembly II used for being connected with the shank of the human body, the limb connecting arm assembly I and the limb connecting arm assembly II are each provided with a clutch unit, and the core speed increaser is installed between the limb connecting arm assembly I and the limb connecting arm assembly II. The energy storage module and the sensor module are arranged on the two sides of the core speed increaser, and the core speed increaser is connected with the energy storage module. The device has the advantages of compact structure, variable rigidity bandwidth and good motion adaptability, can realize high energy recovery efficiency in a multi-gait mode, and reduces the metabolic consumption of a human body.

The invention belongs to the technical field of gait recognition, and particularly relates to a human gait recognition method and system in a complex scene. And sequentially performing illumination correction, human body posture angle alignment and shielding processing on the extracted human body gait data, and performing human body gait recognition based on the gait data obtained after illumination correction, human body posture angle alignment and shielding processing. According to the occlusion processing, a comprehensive and reliable occlusion area judgment standard is set based on the connection relation between skeleton nodes, the human anatomy reference distance, kinematics constraint and the time continuity requirement, and then a missing human body area is reconstructed through optical flow compensation or bilinear interpolation based on information of visible pixels around the occlusion area. The accuracy of gait recognition in a complex scene is improved, the problem of adaptability of the complex scene is solved in a breakthrough mode, high-precision gait identity recognition is achieved, and the security and protection monitoring requirement of a real scene is met.

The invention discloses a quasi-passive knee joint exoskeleton and a design method thereof, the knee joint exoskeleton comprises a transmission energy storage module, a power-assisted regulation and control module and a wearing module, and the power-assisted state is switched according to the gait stage of a human body. The transmission energy storage module transmits joint torque through an input gear, an intermediate gear and an output gear, and a pressing rod on an output shaft is used for driving a flexible hinge to deform, so that efficient energy storage and release in a small space are realized; the assistance regulation and control module controls a cam shaft to rotate, drives a swing shaft to push an intermediate gear to move and enables the gears to be meshed in the supporting phase, motion limiting is formed through contour matching of a cam shaft body and a swing shaft matching cambered surface, mechanical self-locking is achieved, locking force does not need to be additionally provided, and walking assistance is achieved. In the swing phase, the camshaft is controlled to rotate to enable the unlocking cambered surface to turn to the swing shaft, the swing shaft is unlocked, a driving pin of the camshaft drives the swing shaft to rotate, the gear is disengaged, and free swing is achieved. Compared with the prior art, the invention has the advantages of light weight and low power consumption.

The invention relates to the technical field of medical data analysis, in particular to a spatio-temporal characteristic quantitative evaluation method for Parkinson's disease motion symptoms, which comprises the following steps: deploying an inertial measurement unit to collect three-dimensional acceleration angular velocity magnetic field data, constructing a human body connection structure to generate a connection neural network topological structure, the method comprises the following steps: calculating trajectory direction angle and angular velocity change adaptive weighting to judge stability, aggregating multiple rounds of convolution propagation of adjacent features to form a space-time fusion motion feature set, identifying tremor gait amplitude according to time sequence multi-head attention to extract a time sequence feature mode, calculating tremor gait coordination to generate a Parkinson's disease motion symptom quantitative evaluation result, and calculating a Parkinson's disease motion symptom quantitative evaluation result. According to the method, the limb coordination is captured by constructing sensor network motion data topological connection and fusing multi-dimensional part information, the remote association is captured by keeping the time sequence stable through adaptive weight attenuation and connection convolution depth aggregation features, and the multi-head attention fine recognition tremor frequency and gait change are combined. And the Parkinson's disease symptom identification and evaluation consistency is improved.

The invention relates to the technical field of artificial intelligence and medical detection, and particularly provides a human body gait variability detection system based on an improved TCN model. The system comprises a multi-modal sensor array, a signal preprocessing module, a self-adaptive time sequence feature extraction module, a bidirectional time sequence context fusion module, a progressive depth feature enhancement module and a variability quantization output module, a receptive field is dynamically adjusted through a learnable expansion coefficient, the bidirectional time sequence context is fused to improve the period segmentation precision, and the precision of the period segmentation is improved. And gradual feature enhancement is adopted to relieve deep network feature degradation, so that accurate quantification of five indexes such as a gait cycle variation coefficient and a stride time standard deviation is realized, and the sensitivity and reliability of clinical gait variation detection are improved.

The invention discloses a gait recognition method fusing skeleton point action information, and belongs to the technical field of human gait recognition. Comprising the steps of gait data acquisition, skeleton point action relation modeling, dynamic spatio-temporal feature extraction and identity feature library self-evolution updating. Modeling is carried out on human skeleton actions and semantic relationships thereof through a structured semantic action graph, and identification features of key gait action fragments are adaptively extracted in combination with a variable structure space-time fusion network, so that high-precision gait identification under a complex environment and a variable view angle is realized. Meanwhile, a continuous self-evolution identity feature library mechanism is provided, gait change and new identity access can be automatically adapted, and the manual maintenance cost is reduced. According to the method, the accuracy, robustness and expansibility of gait recognition are improved, and the method has wide practical application value.

The invention relates to the technical field of intelligent control, and discloses a power-assisted control method of a lower limb pneumatic flexible exoskeleton and a lower limb pneumatic flexible exoskeleton system.The method comprises the steps that healthy side plantar pressure data are obtained, healthy side key gait event nodes are recognized, and the power-assisted control method of the lower limb pneumatic flexible exoskeleton is obtained according to the healthy side key gait event nodes and ankle joint angle data; estimating an uninjured side continuous gait phase, and calculating an affected side target gait phase according to the uninjured side continuous gait phase; according to the target gait phase of the affected side, auxiliary torque and tension are determined; according to the auxiliary torque and the pulling force, target air pressure for supporting the airbag and shrinking the airbag is determined, an electromagnetic valve is controlled to execute deflation or inflation operation according to the target air pressure, and the internal pressure of the airbag is controlled within a target range. The invention aims to provide flexible assistance conforming to the biomechanical characteristics of the ankle joint of a human body for people with ankle joint dysfunction, improve the walking ability and improve the safety and comfort of rehabilitation training.

The invention discloses a multi-dimensional lower limb parameter detection method and system fusing spatial-temporal characteristics, and the method comprises the steps: collecting the lower limb movement information of a wearer through a wearable device; the lower limb movement information comprises position information, posture information and sole multi-dimensional stress information; estimating lower limb space-time parameters and gait events according to the lower limb movement mechanical model and the lower limb movement information; based on the plantar multi-dimensional force measurement model and the lower limb movement information, plantar multi-dimensional force data fused with the spatio-temporal characteristics are obtained. The method can accurately estimate lower limb space-time parameters and human body tracks, and realizes plantar three-dimensional force space resolving and mapping.

The invention provides a self-adaptive assistance system and method for adjusting an exoskeleton based on a sweat sensor, and the system comprises a sweat amount detection chip which is embedded into the surface of wearable flexible clothes, is in contact with the skin of a human body, and is used for converting the detected sweat amount into an electric signal, and further converting the electric signal into a control signal; the knapsack comprises a control system, a driving system and a sensing system, and is worn on the trunk of the human body after being connected with the wearable flexible clothing; the power assisting device comprises a power assisting boot which is worn on the foot of the human body and connected with the driving system through a Bowden cable; and the specific output value of the power assisting device is controlled through the control signal. Dynamic matching of exercise intensity and mechanical assistance is achieved through an amplification control mechanism driven by sweat amount, gait phase compensation fed back by plantar pressure is combined, the energy utilization efficiency is improved by 20% or above, the problem that a traditional exoskeleton physiological-physical interface is disjointed is effectively solved, and the energy utilization rate is increased by 20% or above. The assisting synchronism is improved, and meanwhile the wearing comfort degree and the motion freedom degree are remarkably improved.

The invention discloses a method and system for driving and controlling a lower limb exoskeleton through brain-computer interface intention confidence, and relates to the technical field of medical rehabilitation. The method comprises the following steps: firstly, identifying an original electroencephalogram signal to obtain intention confidence, and mapping an impedance parameter group containing joint stiffness and a damping coefficient in combination with gait phase information; secondly, extracting a motion position and an interaction force error, comparing an intention confidence coefficient with an actual execution state by using a sliding time window to obtain a nerve matching error, and fusing the three into a total composite error; meanwhile, the human-machine coupling compliance is evaluated based on the human body joint angle and the exoskeleton torque variation, and a safety gain coefficient is discriminated and generated; and finally, performing gain operation on the total composite error based on the impedance parameter group, and performing safety gain coefficient correction to obtain a target control torque to be output to a driving actuator. In this way, a nerve-force-motion three-closed-loop framework is constructed, and compliant self-adaptive on-demand assistance and high-safety man-machine collaborative rehabilitation with defense protection are achieved.

The invention relates to the technical field of exoskeleton gait classification, in particular to an exoskeleton auxiliary gait recognition method and system based on neuromuscular signals, and the method comprises the steps: carrying out the signal segmentation of an electroencephalogram signal and an electromyographic signal according to a periodic gait time period set, and obtaining an electroencephalogram signal segment set and an electromyographic signal segment set, and calculating a power spectral density feature value set of the electroencephalogram signal fragments, calculating a muscle activation time period set and a root mean square value set of the electromyographic signal fragments, and inputting the exoskeleton torque data, the power spectral density feature value set and the muscle activation time periods into a gait classification model for gait classification to obtain a predicted gait. The human gait recognition accuracy of the exoskeleton gait assisting technology can be improved.

The invention provides an electrical stimulation and lower limb exoskeleton device fusion control method and system, and the method comprises the steps: driving an exoskeleton to move through bottom admittance control based on the actual human-computer interaction torque and expected torque of a knee joint and reference trajectory data of the knee joint in a gait cycle, and triggering ankle joint electrical stimulation through a fixed stimulation parameter group; after the gait cycle is finished, torque errors are calculated, and periodic iteration updating is conducted on the knee joint reference trajectory data through an iterative learning control algorithm till convergence is conducted to obtain target knee joint reference trajectory data; and then determining a target moment of a gait phase based on the target knee joint reference trajectory data, and obtaining a corresponding target stimulation parameter group from the set stimulation parameter time mapping library to replace the fixed stimulation parameter group for subsequent stimulation. According to the method, through single-cycle and cross-cycle trajectory iteration, the adaptation defect of a traditional fixed trajectory is avoided, the stimulation parameters are synchronously optimized, and the dorsiflexion assisting effect is indirectly improved.

The invention provides a gait evaluation method and system based on a human body nonlinear system analysis technology, and the method comprises the steps: collecting a gait cycle six-channel high-precision time sequence signal outputted by a wearable inertial measurement unit, eliminating noise and gait difference through wavelet threshold denoising and Z-score standardization, extracting a chaotic feature vector composed of a Lyapunov index spectrum and a Kolmogorov entropy value, and carrying out the recognition of a gait signal, a wavelet neural network is adopted to realize nonlinear mapping of chaotic features and phase-space reconstruction parameters, and a self-adaptive feedback mechanism is introduced to dynamically optimize modeling parameters, so that the accuracy and personalized matching capability of gait pattern recognition and stability evaluation are effectively improved; quantitative characterization of the gait chaos level and real-time online model optimization can be achieved, and high-robustness support is provided for rehabilitation training and exercise aided decision making.

The invention relates to an embedded human body gait recognition system and method. The system comprises a pressure sensing array which comprises a plurality of pressure sensing units distributed on a human body sole fabric and is used for collecting and outputting a plurality of paths of sole pressure signals; the filtering module is used for filtering the input plantar pressure signal; the data processing module is used for sequentially gating the pressure sensing units to obtain filtered plantar pressure signals and converting the plantar pressure signals into digital signals to obtain plantar pressure data; and the classification and recognition module is used for converting the plantar pressure data into a plantar pressure cloud picture, and then recognizing the plantar pressure cloud picture by using a neural network model to obtain a human body gait classification result. According to the method, efficient processing of the plantar pressure data and rapid output of the gait classification result can be achieved, the requirement for real-time gait recognition in an actual scene is met, and the method has wide application prospects in the fields of gait anomaly monitoring, falling risk prevention and the like.

The invention belongs to the technical field of target tracking, and particularly relates to a human body gait data extraction method and system in a multi-target scene. The method comprises the following steps: correcting a mahalanobis distance formula for calculating spatial feature similarity in an original DeepSORT algorithm by utilizing a displacement field obtained by calculating dense optical streams of adjacent frames to obtain an improved DeepSORT algorithm, performing target tracking in a multi-target scene by utilizing the improved DeepSORT algorithm, and when a target is temporarily shielded or illumination suddenly changes, performing target tracking by utilizing the improved DeepSORT algorithm. The optical flow field can still speculate the target position through background pixel displacement, correct prediction deviation caused by target shielding or sudden illumination change, avoid tracking target loss or target ID jump, and realize stable tracking of multiple targets of pedestrians in a multi-target complex scene, so that the quality of gait data extraction is improved, and the precision of a gait recognition result is improved.

The application provides a contralateral movement direction prediction method for heterolateral lower limb myoelectric signals and a related device, and belongs to the technical field of human biological signal processing and pattern recognition. The application uses the combined features of the surface myoelectric signals of the left and right lower limbs, the left foot plantar pressure and the right foot three-dimensional acceleration to construct a training set and a test set; a three-class gait prediction model is constructed, the three-class gait prediction model is trained based on the constructed training set, the three-class gait prediction model training result is tested based on the constructed test set, and a trained three-class gait prediction model is obtained; the combined features of the surface myoelectric signals of the left and right lower limbs, the left foot plantar pressure and the right foot three-dimensional acceleration are input into the trained three-class gait prediction model for right foot movement direction prediction, and a right foot movement direction prediction result is obtained. The application solves the problem of low accuracy of contralateral movement direction prediction for heterolateral lower limb myoelectric signals.

The invention discloses a multi-mode rehabilitation training system for a patient with limb disability in bed, and relates to the technical field of medical rehabilitation instruments. The system comprises an adjustable movable portal frame fixed above a sickbed, a core host installed on the portal frame, at least four mechanical arms capable of being independently programmed and controlled, a broadband vibration output end and a flexible sling. The core is that the control unit integrates three rehabilitation modes: a bionic gait motion mode, which simulates human body walking through cooperative linkage of mechanical arms, and pulls four limbs of a patient to passively swing; in the broadband vibration physiotherapy mode, mechanical vibration of 0.5-100 Hz is conducted to the trunk or limbs of the patient through a vibration output end; in the composite rehabilitation exercise mode, targeted vibration is applied to four limbs during gait exercise. According to the system, physiological signals of a patient are collected in real time through an integrated sensor, motion and vibration parameters are adjusted in a self-adaptive mode based on an intelligent algorithm, safe and efficient on-bed in-situ rehabilitation is achieved, and various complications caused by long-term bed lying are systematically prevented and relieved.

The application discloses a lower limb rehabilitation training exoskeleton control method, an electronic device and a storage medium. The method comprises the following steps: collecting standard gait information in the process of human walking; generating a hip joint angle change curve and a knee joint angle change curve according to the standard gait information; performing smoothing processing on the hip / knee joint angle change curve; generating a variable step frequency step length hip joint curve according to the hip joint angle change curve and a hip joint amplitude parameter and a step frequency parameter; generating a variable step frequency step length knee joint curve according to the knee joint angle change curve and a knee joint amplitude parameter and a step frequency parameter; and driving a hip / knee joint motor according to the variable step frequency step length hip / knee joint curve, and driving a human lower limb through an exoskeleton. The method is based on normal motion data of the human body for gait planning, is closer to the human motion mode, has strong adaptability, requires a small amount of data, has strong practicability, and solves the problem of speed mutation caused by non-smooth adjacent period curves.

The application discloses a flexible variable stiffness knee joint rehabilitation exoskeleton and a motion control method thereof. The flexible variable stiffness knee joint rehabilitation exoskeleton comprises a back fixing assembly, a thigh fixing assembly and a shank fixing assembly. The thigh fixing assembly and the shank fixing assembly are hingedly connected through a knee joint shaft group and are provided with a pose detection assembly at the hinge connection. An output wheel and an input wheel are connected through a Bowden cable group. The back fixing assembly is respectively provided with a controller, a driving mechanism, a nonlinear variable stiffness mechanism and an energy recovery mechanism. The device uses magnetic repulsion between a permanent magnet and a power coil to construct a nonlinear stiffness adjusting mechanism. The actuator is fixed on the human back through a bandage. The Bowden cable is used to realize remote transmission of the torque to the knee joint, thereby effectively reducing the load bearing burden of the lower limbs. In the swing phase of the gait cycle, the system uses the principle of permanent magnet cutting magnetic induction lines to realize energy feedback, thereby improving the overall energy efficiency of the assisting process.

The invention discloses a gait recognition method based on a Pluecker straight line, and belongs to the technical field of computer vision and biological feature recognition. The core of the method is to provide a double-layer gait feature modeling framework. The method comprises the following steps: firstly, modeling human skeletons into spatial straight lines by utilizing Pluecker coordinates, and calculating absolute rigid body motion characteristics of each skeleton between adjacent frames through dual quaternions; furthermore, by combining the absolute motion characteristics of adjacent bones, the relative motion characteristics of the joints, which can better reflect the physiological characteristics of individuals, are calculated. And finally, aggregating the two complementary features to construct a gait feature sequence, and completing identity recognition by using a time sequence deep learning model. According to the method, the overall motion information and the local joint motion information are separated and fused, so that the defects that an existing method is incomplete in feature representation and is greatly interfered by the overall motion are effectively overcome, and the accuracy, the discrimination capability and the robustness of gait recognition are remarkably improved.

This application discloses a method and system for controlling a lower limb exoskeleton using a brain-computer interface based on intent confidence, relating to the field of medical rehabilitation technology. First, the system identifies the original electroencephalogram (EEG) signal to obtain intent confidence, and maps an impedance parameter set including joint stiffness and damping coefficients by combining gait phase information. Second, it extracts motion position and interaction force errors, compares the intent confidence with the actual execution state using a sliding time window to obtain neural matching errors, and merges these three into a total composite error. Simultaneously, it assesses human-machine coupling compliance based on changes in human joint angles and exoskeleton torque, and generates a safety gain coefficient. Finally, it performs gain calculations on the total composite error based on the impedance parameter set, and corrects it with the safety gain coefficient to obtain the target control torque for output to the actuator. This constructs a neural-force-motor three-loop architecture, achieving compliant and adaptive on-demand assistance and highly safe human-machine collaborative rehabilitation with defensive protection.

The application discloses a control method of an autonomous navigation lower limb robot, which comprises the following steps: collecting human gait data to obtain an original gait curve, optimizing the gait in combination with sensor data, and controlling an exoskeleton to perform rehabilitation training; based on lower limb motion parameters and navigation instructions, the speed of a driving wheel is controlled, so that the platform advances to a target direction in cooperation with the lower limb motion parameters, thereby realizing the functions of gait training and autonomous navigation, and the comfort during lower limb rehabilitation training and the experience of a patient are improved, and real-time regulation and control of lower limb motion and autonomous control of a walking direction of the patient during training are realized.

Bodywear apparatus configured to be worn by a human user and attachable to footwear of the human user in order to augment the abilities of the lower limbs of the user is disclosed. In another aspect, footwear apparatus configured to be worn by a human user in order to augment the abilities of the lower limbs of the user is disclosed. Such bodywear apparatus and footwear apparatus are configured to reduce the effort a user must exert and improve the user's performance during walking, running, hiking, marching, and various other gaits as well as jumping, hopping, and other activities. In an aspect, bodywear apparatus further comprises a supplemental power device configured to further augment the user's abilities by activating during portions of the user's gait cycle.

The application discloses a gait recognition method and device based on an event camera and equipment, belongs to the technical field of human gait recognition, and particularly relates to a gait recognition method based on an event camera. The method solves the problem that event stream noise cannot be effectively processed and fine space-time characteristics cannot be fully mined to improve recognition accuracy in the prior art. The method comprises the following steps: acquiring a human gait event data set collected by an event camera; and performing denoising processing on a training set by adopting a two-stage denoising method based on event density to filter background noise. The gait recognition method, device and equipment based on the event camera are suitable for scenes such as security monitoring and judicial identification, and have high requirements for long-distance, non-contact identity recognition and privacy protection.

The invention discloses a decision-making-level multi-mode human body gait prediction method, and belongs to the technical field of neuroscience and rehabilitation medicine. The invention provides a decision-level fusion scheme for solving the problem that the prediction precision is limited due to the fact that an optimal mode cannot be dynamically selected according to gait characteristics in an existing multi-mode gait prediction method based on data-level or feature-level fusion. According to the technical scheme, the method is characterized by comprising the following steps: constructing an independent expert prediction model for respectively processing an inertial measurement unit signal and a surface electromyogram signal, and training a gait mode decision maker for judging whether a gait belongs to a periodic or non-periodic type in real time; and during online prediction, according to a judgment result of the decision maker, adaptively selecting the output of the myoelectricity expert model corresponding to the periodic gait or the inertial expert model corresponding to the non-periodic gait as a final prediction result. According to the method, through decision-level modal selection, complementary advantages of different sensors are effectively fused, and the gait prediction precision and the system robustness are improved.

This application provides an adaptive multi-terrain lower limb exoskeleton assistive control method and device. The device includes a backplate, a drive structure, a lower limb structure, sensors, a hip and knee joint transmission system, and a control system. The backplate provides stable support and reduces the device's weight; the lower limb structure enables a natural gait; sensors provide data for precise assistance through real-time monitoring; the hip and knee joint transmission system achieves dynamic torque distribution through pulleys and cams, improving assistance efficiency; the control system includes an electrically connected structure-adaptive allocation layer and a predictive adjustment layer. The predictive adjustment layer incorporates an offline-trained temporal convolutional network (TCN) torque prediction model, and the structure-adaptive allocation layer is electrically connected to the hip and knee joint transmission system. This application enables the lower limb exoskeleton assistive device to better adapt to the human body's natural gait, achieve dynamic torque distribution at the hip and knee joints, improve assistance efficiency and comfort, and enhance adaptability and practicality in complex sports scenarios.

The invention provides an intelligent wearable health monitoring system, and relates to the technical field of sports biomechanics, and the method comprises the steps: collecting original sensing data streams of a triaxial accelerometer and a gyroscope in real time, and converting the original sensing data streams to a human skeleton coordinate system through a rotation matrix; calculating the Euler angle of the limb joint based on the converted data, and extracting gait phase parameters in combination with the curvature features of the motion trail; calculating a real-time stride frequency according to the gait phase parameter; when a plurality of continuous stride frequency sampling values exceed a preset medium-high intensity motion threshold value, generating a medium-high intensity motion state identifier; synchronously executing according to the medium-high intensity motion state identification, constructing a blood glucose dynamic evaluation window based on a continuous sampling sequence of a blood glucose sensor, calculating a time sequence gradient of blood glucose reading in the window, and generating a dynamic compensation coefficient according to the time sequence gradient. According to the invention, dynamic and accurate health monitoring and risk early warning in a motion scene are realized.

The application discloses a wearable human gait analysis system, comprising a synchronous data acquisition module, a plantar pressure topology graph module, a pre-training model module, a real scene model module and a pathological gait classification model module; the synchronous data acquisition module collects real force platform data, insole pressure time series data and inertial measurement unit data, and constructs a gait data set; the plantar pressure topology graph module topologically expresses the insole pressure time series data, and outputs a plantar pressure topology graph corresponding to a time sequence based on the insole pressure time series data; the pre-training model module constructs a pre-training model based on a gait data set and a space-time feature fusion network to output a gait dynamics prediction result; the real scene model module performs target domain semi-supervised self-adaptive fine-tuning on the pre-training model to generate a real scene model; and the pathological gait classification model module extracts multi-domain classification features based on the prediction results of the real scene model of each subject in a pathological state, and outputs a final classification result.

The invention discloses a knee joint orthopedic rehabilitation brain-computer interface exoskeleton power-assisted control method and system, and relates to the technical field of knee joint rehabilitation. The method comprises the following steps: based on an electroencephalogram electrode cap and a knee joint angle sensor, analyzing a corresponding relation between an electroencephalogram main peak and a joint action, combining plantar pressure change, screening gait characteristics, judging an intention signal and an action trend, proofreading exoskeleton execution, and obtaining a closed-loop action offset index. According to multi-dimensional trend analysis of the action execution process, electroencephalogram intention judgment is dynamically associated with the knee joint action process, intention signals and action state changes are comprehensively analyzed through trend synchronous judgment and continuous behavior cooperation, and layer-by-layer linkage of intentions, behaviors and the execution process is achieved; the control output has real-time adaptation and compliant coordination ability, the training stage can adapt to multi-source signal changes, the false triggering probability is effectively reduced, the action continuity is guaranteed, and the man-machine coordination consistency and the rehabilitation training safety guarantee ability are improved.