CN-121987218-A - Exoskeleton power-assisted control method and system for knee joint orthopedic rehabilitation brain-computer interface

CN121987218ACN 121987218 ACN121987218 ACN 121987218ACN-121987218-A

Abstract

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 of analyzing the corresponding relation between the main peak of the brain electricity and the joint action based on the brain electrode cap and the knee joint angle sensor, combining the plantar pressure change, screening gait characteristics, judging intention signals and action trends, and checking exoskeleton execution to obtain a closed-loop action offset index. According to multidimensional trend analysis in the action execution process, the brain electricity intention is dynamically associated with the knee joint action process, intention signals and action state changes are comprehensively distinguished through trend synchronous judgment and continuous behavior coordination, layer-by-layer linkage of the intention, the behavior and the execution process is realized, control output has real-time adaptation and compliance coordination capability, the multi-source signal changes can be self-adapted in the training stage, false triggering probability is effectively reduced, action continuity is ensured, and man-machine coordination consistency and rehabilitation training safety guarantee capability are improved.

Inventors

LI XIAO
LI JIAHANG
SHI XIUXIU
CUI CHENGWEN
DENG YONGPING

Assignees

中国人民解放军总医院第四医学中心

Dates

Publication Date: 20260508
Application Date: 20251230

Claims (10)

1. The knee joint orthopaedics rehabilitation brain-computer interface exoskeleton assistance control method is characterized by comprising the following steps: s1, analyzing the corresponding relation between the main peak variation of the brain electricity and the joint action record based on an brain electrode cap and a knee joint angle sensor, synchronously comparing the variation time periods of the two types of signals, and judging whether the main peak fluctuation and the joint action occur simultaneously or not to obtain the brain electricity cooperative discrimination characteristic; S2, based on the electroencephalogram cooperative discrimination characteristics, monitoring synchronous trends of joint action change and plantar pressure distribution change in the walking process, comparing the track change direction with the stress change direction, and identifying a continuous action interval to obtain gait behavior cooperative parameters; S3, judging the consistency degree of the brain electrical signal intention characteristic and the knee joint action trend based on the gait action cooperative parameter, analyzing the change direction of data in the same time period, and checking the exoskeleton execution state to obtain a movement instruction fusion factor; S4, starting the exoskeleton knee joint to execute based on the motion instruction fusion factor, comparing the condition that the running direction of the motor is consistent with the joint action direction, and recording the change process of the action track along with time to obtain a knee joint execution track interval; And S5, comparing the action state fed back by the knee joint angle sensor with a target action based on the knee joint execution track section, analyzing the difference trend of the feedback position and the action end point, and performing compensation adjustment on the offset action to obtain a closed-loop action offset index.
2. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton-assisted control method according to claim 1, wherein the electroencephalogram cooperative discrimination features comprise a main peak response feature, a synchronization correlation feature and a time sequence coordination feature, the gait behavior cooperative parameters comprise a stride feature, a stride frequency feature and a pressure distribution feature, the motion instruction fusion factor comprises a motion synchronization feature, a trend consistency feature and an instruction coordination feature, the knee joint execution track interval comprises a motion continuity feature, a track morphology feature and a response amplitude feature, and the closed loop motion offset index comprises a motion offset feature, a compensation correction feature and a continuous stability feature.
3. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton helping hand control method according to claim 1, wherein the step of acquiring the electroencephalogram cooperative discrimination characteristics is specifically as follows: S101, analyzing brain electrical signals and joint action data based on brain electrical electrode caps and knee joint angle sensors, corresponding main peak time points of brain electrical signal motion associated frequency bands and change time of knee joint actions, and judging synchronous expression of the data on a time axis to obtain a brain movement synchronous event group; S102, based on the brain movement synchronization event group, comparing the change direction of the brain electric main peak with the change direction of the knee joint movement, judging whether the change trends of the brain electric main peak and the knee joint movement are synchronous, and identifying the data combination with the consistent change trend to obtain a nerve joint trend cluster; And S103, based on the nerve joint trend cluster, aggregating brain wave dynamic time sequence characteristics, action response rhythm characteristics and data superposition intervals, and classifying and integrating data according to the characteristics to obtain brain wave cooperative discrimination characteristics.
4. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton assistance control method of claim 1, wherein the step of acquiring gait behavior cooperative parameters specifically comprises the following steps: S201, analyzing a data sequence output by an angle sensor of the knee joint based on the electroencephalogram cooperative discrimination characteristics, grouping the changing directions of continuous angle rotation segments, judging the sequence of the start-stop position and the direction change of each segment of rotation, identifying the segments with the changing characteristics in the walking cycle, and aggregating the rotation cycles according to the changing time sequence to obtain a joint gait structure segment group; S202, based on the joint gait structure segment group, comparing the joint gait structure segment group with data acquired by a sole pressure sensor, analyzing the movement track of stress centers of all partitions of the sole, judging the corresponding relation between the track change direction and the knee joint rotation direction, and identifying synchronously changed data segments to obtain a corresponding segment set of the step-pressing joint; S203, based on the corresponding segment set of the step-pressing joint, judging segments with consistent continuous trends, counting time overlapping intervals of the knee joint movement state and the plantar stress state, classifying the state switching processes, and marking gait phases according to the joint movement state types to obtain gait behavior cooperative parameters.
5. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton helping hand control method of claim 1, wherein the step of obtaining the motion instruction fusion factor specifically comprises the following steps: S301, analyzing action data of an electroencephalogram electrode cap signal and a knee joint angle sensor in a corresponding period based on the gait action coordination parameter, comparing the change direction of the data in each gait cycle, judging the corresponding relation between the change form of a nerve signal and the rotation trend of joint action, and counting a section with the same change trend to obtain a brain joint trend coordination sequence; S302, judging the action direction of an exoskeleton execution unit corresponding to each trend consistent section based on the brain joint trend cooperative sequence, comparing the synchronous relation between the exoskeleton unit action and the knee joint angle change trend, analyzing the correspondence between the signal direction and the action direction in each section, and identifying the synchronous section to obtain an execution trend synchronous sheet group; And S303, calculating the coordination of the electroencephalogram signal trend, the knee joint rotation trend and the exoskeleton execution direction in the same time interval based on the execution trend synchronization slice group, and analyzing the synchronization and the direction consistency of each type of data to obtain a motion instruction fusion factor.
6. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton assistance control method of claim 1, wherein the knee joint execution track interval obtaining step specifically comprises: S401, analyzing the action sequence of the motor driving device based on the motion instruction fusion factor, judging whether the action direction required by the action instruction is matched with the action direction output by the driving device, and identifying an action section with the same direction to obtain an action direction matching section; S402, based on the action direction matching segment, comparing action processes executed by the exoskeleton knee joint, monitoring the corresponding situation of each motor output action section and the joint action direction, judging whether the change process of the action path shows a consistent trend with the action executed by the knee joint, and obtaining a joint path collaborative mapping set; S403, judging time coincidence intervals of motor driving action paths and knee joint action sequences based on the joint path collaborative mapping set, analyzing morphological characteristics and amplitude coordination relations of paths in each interval, classifying and summarizing track ranges in each action execution period, and obtaining knee joint execution track intervals.
7. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton assistance control method of claim 1, wherein the step of obtaining the closed-loop motion offset index specifically comprises the following steps: s501, comparing the knee joint execution track section with a target action state fed back by a knee joint angle sensor action, analyzing a spatial correspondence between the position of the feedback action and the end point of the motor track section, and judging the offset condition among all groups of positions to obtain an action space offset parameter group; S502, based on the motion space offset parameter groups, judging offset phenomena of feedback positions of the groups and motor track interval end points, analyzing distribution of offset sections, adjusting motor output motion sequences, and correcting joint motions aiming at the detected offset motion sections to obtain offset compensation execution sequences; And S503, based on the offset compensation execution sequence, monitoring the performance in continuous motion feedback, judging the offset change of each motion sequence after compensation, analyzing the sustainability and stability of the correction motion, and summarizing the parameters of each motion adjustment to obtain a closed-loop motion offset index.
8. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton-assisted control method according to claim 1, wherein the main peak change represents an amplitude point of an electroencephalogram waveform, which occurs in a motion-related frequency band, and is a motion intention characteristic signal, and the change period represents a time point when the main peak change of the electroencephalogram signal and the knee joint angle change occur respectively.
9. The knee joint orthopedic rehabilitation brain-computer interface exoskeleton-assisted control method according to claim 1, wherein the intention feature represents an electroencephalogram feature acquired by an electroencephalogram electrode cap and reflecting an active motor imagination or intention of a patient, and the joint motion trend represents a motion direction represented by knee joint angle data, including a continuous bending or stretching trend.
10. A knee joint orthopedic rehabilitation brain-computer interface exoskeleton assistance control system for implementing a knee joint orthopedic rehabilitation brain-computer interface exoskeleton assistance control method according to any one of claims 1 to 9, characterized in that the system comprises: The electroencephalogram judgment module is used for analyzing the corresponding relation between the main peak variation of the electroencephalogram and the joint action record based on the electroencephalogram electrode cap and the knee joint angle sensor, synchronously comparing the variation time periods of the two types of signals, and judging whether the main peak fluctuation and the joint action occur simultaneously or not to obtain electroencephalogram cooperative judgment characteristics; The gait synchronous identification module is used for monitoring synchronous trends of joint action change and plantar pressure distribution change in the walking process based on the electroencephalogram cooperative discrimination characteristics, comparing the track change direction with the stress change direction, and identifying a continuous action interval to obtain gait behavior cooperative parameters; The intention trend fusion module is used for judging the consistency degree of the intention characteristics of the electroencephalogram signals and the action trend of the knee joint based on the gait behavior cooperative parameters, analyzing the change direction of data in the same time period, and correcting the exoskeleton execution state to obtain a movement instruction fusion factor; The track generation module starts the exoskeleton knee joint to execute based on the motion instruction fusion factor, compares the condition of the consistency of the running direction of the motor and the action direction of the joint, records the change process of the action track along with time, and obtains a knee joint execution track interval; And the closed loop compensation module is used for comparing the action state fed back by the knee joint angle sensor with the target action based on the knee joint execution track interval, analyzing the difference trend of the feedback position and the action end point, and compensating and adjusting the offset action to obtain the closed loop action offset index.

Description

Exoskeleton power-assisted control method and system for knee joint orthopedic rehabilitation brain-computer interface Technical Field The invention relates to the technical field of knee joint rehabilitation, in particular to a knee joint orthopedic rehabilitation brain machine interface exoskeleton power-assisted control method and system. Background Knee joint rehabilitation relates to a series of techniques and methods for restoring knee joint function, relieving pain and improving the quality of life of patients, and the knee joint rehabilitation technique mainly comprises physical treatment, rehabilitation after surgery, treatment by using external auxiliary equipment and the like. The traditional knee joint orthopedic rehabilitation exoskeleton power-assisted control method refers to a technology for performing auxiliary control on knee joints through exoskeleton equipment, generally comprises a control system of the exoskeleton equipment, monitors actions of a patient in real time by using a sensor and an algorithm, and adjusts a movement mode of the exoskeleton according to requirements of the patient, so that accurate rehabilitation training support is provided for the patient. The existing method relies on single action data as an execution basis, action judgment only pays attention to surface action signals and sensor changes, depth cooperative identification between brain movement intention and action actual occurrence process is lacking, auxiliary control is difficult to accurately match with movement intention actively participated by a user, the training process is easily affected by environment interference, signal drift and nonstandard action, false triggering and mutation risks exist, safety and smoothness in the action process are insufficient, and reliable man-machine cooperative stable output cannot be maintained under complex change of training scene. Disclosure of Invention In order to solve the technical problems in the prior art, the embodiment of the invention provides a knee joint orthopedic rehabilitation brain-computer interface exoskeleton power-assisted control method and system. The technical scheme is as follows: in one aspect, a knee joint orthopaedics rehabilitation brain-computer interface exoskeleton assistance control method is provided, which comprises the following steps: s1, analyzing the corresponding relation between the main peak variation of the brain electricity and the joint action record based on an brain electrode cap and a knee joint angle sensor, synchronously comparing the variation time periods of the two types of signals, and judging whether the main peak fluctuation and the joint action occur simultaneously or not to obtain the brain electricity cooperative discrimination characteristic; S2, based on the electroencephalogram cooperative discrimination characteristics, monitoring synchronous trends of joint action change and plantar pressure distribution change in the walking process, comparing the track change direction with the stress change direction, and identifying a continuous action interval to obtain gait behavior cooperative parameters; S3, judging the consistency degree of the brain electrical signal intention characteristic and the knee joint action trend based on the gait action cooperative parameter, analyzing the change direction of data in the same time period, and checking the exoskeleton execution state to obtain a movement instruction fusion factor; S4, starting the exoskeleton knee joint to execute based on the motion instruction fusion factor, comparing the condition that the running direction of the motor is consistent with the joint action direction, and recording the change process of the action track along with time to obtain a knee joint execution track interval; And S5, comparing the action state fed back by the knee joint angle sensor with a target action based on the knee joint execution track section, analyzing the difference trend of the feedback position and the action end point, and performing compensation adjustment on the offset action to obtain a closed-loop action offset index. Optionally, the electroencephalogram cooperative discrimination feature comprises a main peak response feature, a synchronization association feature and a time sequence cooperation feature, the gait behavior cooperative parameter comprises a stride feature, a stride frequency feature and a pressure distribution feature, the motion instruction fusion factor comprises an action synchronization feature, a trend consistency feature and an instruction coordination feature, the knee joint execution track interval comprises an action continuity feature, a track morphology feature and a response amplitude feature, and the closed loop action offset index comprises an action offset feature, a compensation correction feature and a continuous stability feature. Optionally, the step of acquiring the electroencephalogram cooperative discrimination feature s