A three-dimensional cooperative control method and system for a rehabilitation exoskeleton assistance device

Abstract

This application relates to the technical field of smart devices and provides a three-dimensional collaborative control method and system for a rehabilitation exoskeleton assistive device. The method includes acquiring state data and spatial motion data of the upper limb joints corresponding to the exoskeleton assistive device; identifying the motion stage corresponding to the current operation to determine the target motion stage; determining the corresponding target primary assisting joint based on the target motion stage; and establishing a corresponding joint coordination relationship based on the target motion stage. Based on the target primary assisting joint and the joint coordination relationship, corresponding three-dimensional collaborative assist control parameters are generated to control the exoskeleton assistive device to perform three-dimensional collaborative motion. Through this solution, the exoskeleton assistive device can dynamically adjust the assist coordination state between joints according to different motion stages, improving the motion following accuracy, motion stability, and assist output coordination during complex spatial movements, and solving the problem of difficulty in dynamically switching the primary force-applying joint under different motion stages.

Description

Technical Field

[0001] This invention relates to the field of intelligent devices, and more particularly to a three-dimensional collaborative control method and system for a rehabilitation exoskeleton assistive device. Background Technology

[0002] With the development of rehabilitation assistive devices, wearable robots, and human-machine collaborative control technology, exoskeleton assistive devices are increasingly being applied in fields such as medical rehabilitation, sports assistance, and human strength enhancement. Especially in rehabilitation scenarios related to muscle manipulation, bone setting, and spinal manipulation, these operations typically require complex movements such as continuous pressure, stretching, rotational adjustment, and fine-tuning of the distal end, thus placing high demands on the operator's movement stability, sustained force application ability, and spatial movement control. To reduce the human body's load during prolonged operations and improve the stability of movement output, some exoskeleton assistive devices are beginning to be applied in upper limb support and rehabilitation biomechanical enhancement scenarios.

[0003] Typically, an exoskeleton support structure combined with a joint-driven mechanism is used to provide corresponding movement assistance to the human shoulder, elbow, or wrist joints. Existing exoskeleton assistive devices generally acquire data such as the human body's joint angles, posture, or movement trajectory to control the drive mechanism to output corresponding assistance, thereby assisting the human body in performing actions such as lifting, pushing, or trajectory following. In this way, the muscle load on the human body during continuous movement is reduced, the stability during movement execution is improved, and a human-machine collaborative assistance effect is achieved.

[0004] While acquiring human joint status data and controlling the exoskeleton to output corresponding assistance can assist human movement and reduce movement load, there is a problem in operational scenarios such as bone setting and spinal manipulation that require switching between multiple spatial movements. This presents the challenge of dynamically switching the primary force-applying joints for different movement stages. Since most existing exoskeleton assistive devices use fixed joints, a mismatch between the direction of the assistance output and the actual movement direction can easily occur during different movement stages such as pressing, stretching, rotational adjustment, and end-effector fine-tuning. This affects the continuity of movement and the stability of spatial movement control. Summary of the Invention

[0005] This invention provides a three-dimensional collaborative control method and system for rehabilitation exoskeleton assistive devices to solve the problem of difficulty in dynamically switching the main force-applying joints at different stages of movement.

[0006] In a first aspect, embodiments of this application provide a three-dimensional collaborative control method for a rehabilitation exoskeleton assistive device, comprising: acquiring state data and spatial motion data of human upper limb joints corresponding to the exoskeleton assistive device; wherein the human upper limb joint data includes at least shoulder joints, elbow joints, and wrist joints; determining a target action stage corresponding to the current operation action based on the state data and the spatial motion data; determining a corresponding assisting joint as a target primary assisting joint from the shoulder joint, elbow joint, and wrist joint according to the target action stage, and establishing a corresponding joint collaboration relationship based on the target action stage; determining a corresponding target assisting direction, target assisting path, and target assisting intensity based on the target primary assisting joint and the joint collaboration relationship, to generate corresponding three-dimensional collaborative assisting control parameters; and controlling the target primary assisting joint corresponding to the exoskeleton assistive device to output primary assisting force, and controlling the other collaborative joints to output corresponding collaborative assisting force, to form a three-dimensional collaborative assisting output corresponding to the target action stage, based on the three-dimensional collaborative assisting control parameters.

[0007] Preferably, the step of identifying the action stage corresponding to the current operation based on the state data and the spatial motion data includes: determining the corresponding joint movement direction, joint movement amplitude, and joint movement change trend based on the shoulder joint state data, elbow joint state data, and wrist joint state data of the state data; determining the corresponding end-effector movement direction and spatial motion trajectory based on the spatial motion data; and identifying the action stage corresponding to the current operation based on the joint movement direction, joint movement amplitude, joint movement change trend, end-effector movement direction, and spatial motion trajectory to determine that the current operation is in at least one of the following target action stages: pressing stage, stretching stage, rotation stage, fine-tuning stage, or instantaneous release stage.

[0008] Preferably, the step of determining the corresponding joint movement direction, joint movement amplitude, and joint movement change trend based on the shoulder joint state data, elbow joint state data, and wrist joint state data includes: determining the joint rotation direction and shoulder joint angle change amount corresponding to the shoulder joint based on the shoulder joint state data; determining the joint flexion and extension direction and elbow joint movement change amount corresponding to the elbow joint based on the elbow joint state data; determining the distal end rotation direction and wrist joint posture change amount corresponding to the wrist joint based on the wrist joint state data; using the joint rotation direction, the joint flexion and extension direction, and the distal end rotation direction as the joint movement direction; determining the joint movement amplitude through the shoulder joint angle change amount, elbow joint movement change amount, and wrist joint posture change amount, and determining the corresponding joint movement change trend based on the change of the joint movement amplitude over a continuous time period.

[0009] Preferably, the step of determining the corresponding end-effector movement direction and spatial movement trajectory based on the spatial motion data includes: determining the spatial movement direction of the corresponding end-effector execution position of the exoskeleton assistive device based on the spatial motion data; obtaining corresponding spatial position change data based on the spatial movement direction; determining the corresponding end-effector movement direction based on the spatial position change data; and generating the corresponding spatial movement trajectory.

[0010] Preferably, the step of determining the corresponding assisting joint from the shoulder joint, elbow joint, and wrist joint as the target primary assisting joint based on the target action stage, and establishing the corresponding joint coordination relationship based on the target action stage, includes: when the target action stage corresponds to a large-range spatial movement stage, determining the assisting joint corresponding to the shoulder joint in the shoulder joint state data as the target primary assisting joint; when the target action stage corresponds to a stable pushing stage or a continuous stretching stage, determining the assisting joint corresponding to the elbow joint in the elbow joint state data as the target primary assisting joint; when the target action stage corresponds to a fine-tuning stage of the distal direction or a rotational adjustment stage, determining the assisting joint corresponding to the wrist joint in the wrist joint state data as the target primary assisting joint; analyzing the participation status of the remaining joints other than the target primary assisting joint to determine at least one cooperating assisting joint corresponding to the target primary assisting joint; determining the corresponding assist distribution relationship and action synchronization relationship between the target primary assisting joint and the cooperating assisting joint based on the target action stage; and establishing the corresponding joint coordination relationship based on the assist distribution relationship and the action synchronization relationship.

[0011] Preferably, the step of determining the corresponding target assist direction, target assist path, and target assist intensity based on the target main assist joint and the joint coordination relationship to generate corresponding three-dimensional coordinated assist control parameters includes: determining the corresponding target spatial action direction according to the target action stage; generating the corresponding target assist path according to the target spatial action direction and the spatial action trajectory of the target spatial action direction; and determining the corresponding target assist intensity according to the load state corresponding to the target main assist joint and the mechanical requirements corresponding to the target action stage.

[0012] Preferably, the step of controlling the target primary assist joint of the exoskeleton assist device to output primary assist and controlling the other assist joints to output corresponding assistive forces according to the three-dimensional collaborative assist control parameters includes: controlling the target primary assist joint to output corresponding primary assist along the target assist direction; controlling the other assist joints to output corresponding assistive forces according to the joint collaboration relationship; and forming a three-dimensional collaborative assist output corresponding to the target action stage through the combined output of the primary assist and the assistive forces.

[0013] Secondly, embodiments of this application provide a three-dimensional collaborative control system for a rehabilitation exoskeleton assistive device, comprising: a data acquisition module for acquiring state data and spatial motion data of the human upper limb joints corresponding to the exoskeleton assistive device; wherein the human upper limb joint data includes at least the shoulder joint, elbow joint, and wrist joint; a motion recognition module for determining the target motion stage corresponding to the current operation motion based on the state data and the spatial motion data; a motion establishment module for determining the corresponding assist joint as the target primary assist joint from the shoulder joint, elbow joint, and wrist joint according to the target motion stage, and establishing a corresponding joint coordination relationship according to the target motion stage; a motion determination module for determining the corresponding target assist direction, target assist path, and target assist intensity based on the target primary assist joint and the joint coordination relationship, so as to generate corresponding three-dimensional collaborative assist control parameters; and a motion control module for controlling the target primary assist joint corresponding to the exoskeleton assistive device to output primary assist according to the three-dimensional collaborative assist control parameters, and controlling the other coordination joints to output corresponding coordination assist force, so as to form a three-dimensional collaborative assist output corresponding to the target motion stage.

[0014] In one approach provided by the aforementioned method and system, state data and spatial motion data of the upper limb joints corresponding to the exoskeleton assistive device are acquired. The motion states of the shoulder, elbow, and wrist joints are jointly analyzed. Combined with the end-effector movement direction, spatial motion trajectory, and joint motion change trends, the target action stage corresponding to the current operation is identified. Based on the motion characteristics corresponding to different target action stages, the corresponding primary assist joint and cooperating auxiliary joints are dynamically determined, and corresponding assist distribution and motion synchronization relationships are established to form corresponding joint coordination relationships. Furthermore, based on the target spatial action direction, spatial motion trajectory, and load state of the target primary assist joint corresponding to the target action stage, corresponding target assist direction, target assist path, and target assist intensity are generated, forming corresponding three-dimensional cooperating assist control parameters. Based on these three-dimensional cooperating assist control parameters, the assist output states of the target primary assist joint and cooperating auxiliary joints are jointly controlled to achieve synchronous output of primary and cooperating assist forces. This allows the exoskeleton assistive device to dynamically adjust the assist coordination state between joints according to different action stages, improving motion following accuracy, motion stability, and assist output coordination during complex spatial motion processes, and solving the problem of difficulty in dynamically switching the primary force-applying joint under different action stages. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of a three-dimensional collaborative control method for a rehabilitation exoskeleton assistive device according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a data acquisition structure in one embodiment of the present invention; Figure 3 This is a schematic diagram of a structure for action phase recognition in one embodiment of the present invention; Figure 4 This is a schematic diagram of the calculation process of joint participation weight and assist allocation coefficient in one embodiment of the present invention; Figure 5 This is a schematic diagram of the calculation process for load status values ​​in one embodiment of the present invention; Figure 6 This is a schematic diagram of a piecewise linear mapping of assist strength in one embodiment of the present invention; Figure 7 This is a schematic diagram of the output process of three-dimensional collaborative assistance in one embodiment of the present invention; Figure 8 This is a flowchart illustrating a three-dimensional collaborative control system for a rehabilitation exoskeleton assistive device according to an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof. It should also be understood that, as used in this specification and the appended claims, the term "and / or" refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0019] Furthermore, in the description of this invention and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0021] It should be understood that the sequence number of each step in the following embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0022] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0023] In recent years, with the continuous development of intelligent rehabilitation equipment, human-computer interaction systems, and flexible drive technology, exoskeleton assistive devices have received widespread attention and application in rehabilitation therapy, sports training, strength compensation, and fine motor skills assistance. Especially in rehabilitation procedures such as massage therapy, spinal correction, joint release, and myofascial release, operators often need to perform complex movements such as continuous pressure, directional traction, multi-angle rotation, and precise position adjustment, placing high demands on the coordination and control capabilities, sustained output capacity, and movement precision of the upper limb joints.

[0024] To alleviate muscle fatigue caused by repetitive tasks and improve movement stability and force consistency during operation, exoskeleton technology is increasingly being introduced into rehabilitation and assistance scenarios. Existing technologies typically involve constructing a mechanical support frame that matches the human upper limb, combined with motor drives, elastic compensation, or force feedback control mechanisms to provide corresponding motion assistance and load-sharing functions for the shoulder, elbow, and wrist joints.

[0025] The relevant devices typically collect data on human posture, joint angle changes, electromyographic signals, or end-effector trajectory information to control the drive unit in real time. This allows the exoskeleton to output corresponding assistance based on the user's movement intentions, thereby aiding in actions such as lifting, pushing, traction, or trajectory following. Using this technical solution, the operator's physical exertion can be reduced to some extent, movement execution efficiency can be improved, and the stability of human-machine collaborative work can be enhanced.

[0026] However, in applications such as massage therapy, spinal adjustment, and musculoskeletal rehabilitation that require frequent switching of multi-degree-of-freedom spatial movements, existing exoskeleton assistive devices still have insufficient adaptability.

[0027] Because the main force-generating joints and movement patterns differ significantly across different operational phases—for example, the pressing phase primarily relies on the coordinated output of vertical force from the shoulder and elbow joints, while the rotational adjustment phase depends more on the wrist joint and forearm posture control, and the end-effector fine adjustment phase requires high-precision movement control over a small range—most existing exoskeleton systems employ preset assist modes or fixed joint assist strategies. This makes it difficult to dynamically adjust the assist joints and assist direction according to changes in the operational state, easily leading to inconsistencies between the assist output and the actual force requirements of the human body. This, in turn, affects the smoothness of movement transitions, the accuracy of force application, and the overall operational effect.

[0028] like Figure 1 As shown, to address the existing problems, this application provides a three-dimensional collaborative control method for a rehabilitation exoskeleton assistive device. This method synchronously identifies the joint motion state and spatial movement state of the human upper limbs and dynamically switches the corresponding target primary assistive joint based on different movement stages, thereby achieving three-dimensional collaborative assistive control of the exoskeleton assistive device during different spatial movement processes. The method includes the following steps: Step 10: Obtain the state data and spatial motion data of the human upper limb joints corresponding to the exoskeleton assistive device; wherein, the human upper limb joint data includes at least the shoulder joint, elbow joint and wrist joint.

[0029] The spatial motion data includes triaxial acceleration data, triaxial angular velocity data, and end-effector position coordinate data output by the inertial measurement unit (IMU). The IMU is positioned at the forearm and wrist connection points of the exoskeleton assistive device. The triaxial acceleration data characterizes the acceleration changes of the end-effector position in the X, Y, and Z axes; the triaxial angular velocity data characterizes the attitude changes of the corresponding joint positions; and the end-effector position coordinate data characterizes the real-time coordinates of the end-effector position in three-dimensional space.

[0030] The end-effector position coordinates are calculated using an inertial navigation algorithm. This algorithm is an attitude integral calculation algorithm used to calculate the corresponding spatial position coordinates based on triaxial acceleration and angular velocity data. The data sampling frequency of the inertial measurement unit is set to 100Hz to 200Hz.

[0031] The exoskeleton assistive device is worn on the upper limbs of the human body, with corresponding joint motion acquisition components for the shoulder, elbow, and wrist joints. The joint motion acquisition components include an angle encoder, an inertial measurement unit, and an attitude acquisition unit. The angle encoder collects real-time rotation angle data of the corresponding joints, the inertial measurement unit collects angular velocity and acceleration data, and the attitude acquisition unit collects spatial attitude change data. By synchronously acquiring the angle, angular velocity, acceleration, and spatial attitude change data for each joint, the corresponding state data of the human upper limb joints is formed.

[0032] The shoulder joint status data includes the joint rotation angle data, angular velocity data, and angular acceleration data corresponding to the shoulder joint in the forward-backward, left-right, and up-down directions; the elbow joint status data includes the elbow joint flexion-extension angle data, flexion-extension velocity data, and flexion-extension acceleration data; and the wrist joint status data includes the wrist joint deflection angle data, rotation angle data, and wrist posture change data.

[0033] Spatial motion data is used to characterize the spatial motion state corresponding to the actions performed by the upper limbs and to reflect the changes in the spatial trajectory of therapeutic techniques during rehabilitation. Spatial motion data includes end-effector position data, spatial movement direction data, and spatial trajectory change data. The end-effector position data is acquired through a spatial positioning component located at the wrist, which includes an inertial navigation unit and a spatial coordinate positioning unit. The inertial navigation unit collects the wrist's acceleration and direction change data in three-dimensional space, while the spatial coordinate positioning unit acquires the corresponding spatial coordinate position data of the wrist.

[0034] like Figure 2 As shown, the exoskeleton assistive device is equipped with a synchronous data acquisition module, which synchronizes and aligns shoulder joint state data, elbow joint state data, wrist joint state data, and spatial motion data according to a unified timestamp, forming a motion state data set under a unified time sequence. The synchronous data acquisition module uses a sliding time window method for data synchronization processing, with the sliding time window length set to 0.5 seconds to 2 seconds.

[0035] For example, when an exoskeleton assistive device is worn on the upper limbs of a person performing bone-setting and muscle manipulation, the angle encoder at the shoulder joint acquires a forward shoulder angle of 18°, the angle encoder at the elbow joint acquires an elbow flexion-extension angle of 42°, and the posture acquisition unit at the wrist joint acquires a wrist deflection angle of 12°. Simultaneously, the spatial coordinate positioning unit at the wrist position obtains the end effector position as moving 45mm in the X-axis direction, 12mm in the Y-axis direction, and 8mm in the Z-axis direction. The synchronous data acquisition module synchronizes the above data in chronological order, forming a corresponding set of upper limb movement state data.

[0036] Step 20: Determine the target action stage corresponding to the current operation action based on the status data and spatial motion data.

[0037] The synchronized motion state data set is processed through motion phase analysis to identify the target motion phase in which the human upper limb is currently moving. The motion phase identification process includes motion direction analysis, motion amplitude analysis, spatial trajectory analysis, and motion continuity change analysis.

[0038] Specifically, by analyzing the combined motion state between the shoulder, elbow, and wrist joints, the overall motion characteristics corresponding to the current upper limb movement are determined; by analyzing the spatial trajectory changes of the end-effector position, the spatial motion characteristics corresponding to the current movement are determined; and by performing trend analysis on the motion change data over a continuous time period, it is determined whether the current movement is in a state of continuous force application, rotational adjustment, or instantaneous release.

[0039] The action phase identification process employs a threshold-based action classification method. Specifically, a pre-defined action phase identification rule table is used, which includes thresholds for joint movement direction, joint movement amplitude, spatial trajectory change, and action duration. The current action state data is matched against the corresponding thresholds to determine the corresponding target action phase.

[0040] Among them, the joint movement direction threshold in the action phase recognition rule table is set according to the range of motion of the human upper limb joints; the joint movement amplitude threshold is set according to the range of joint changes corresponding to different operation actions; the spatial trajectory change threshold is set according to the range of spatial displacement changes corresponding to the end effector position; and the action duration threshold is set according to the action duration corresponding to different action phases.

[0041] When the change in elbow flexion and extension angle is greater than 25° within 0.8 seconds, the end-effector position is continuously pressed down along the Z-axis, and the wrist posture change is less than 5°, it corresponds to the treatment action phase of continuously applying treatment pressure during rehabilitation. The current action is identified as the stable pushing phase by the action phase recognition rule table. When the change in wrist rotation angle is greater than 15° within 0.5 seconds, and the end-effector position has an arc-shaped trajectory change, the current action is identified as the rotation adjustment phase by the action phase recognition rule table.

[0042] In one embodiment, the step of identifying the action stage corresponding to the current operation based on state data and spatial motion data may further be preferred to be: Based on the shoulder joint status data, elbow joint status data, and wrist joint status data, determine the corresponding joint movement direction, joint movement range, and joint movement change trend.

[0043] Joint motion analysis was performed on the shoulder joint state data, elbow joint state data, and wrist joint state data to extract the corresponding joint motion direction information, motion change range information, and motion continuity information.

[0044] Among them, the joint movement direction is used to characterize the current direction of the corresponding joint's movement change; the joint movement amplitude is used to characterize the degree of movement change of the corresponding joint within a continuous time period; and the joint movement change trend is used to characterize the corresponding change state of the corresponding joint during continuous movement.

[0045] The joint motion analysis process employs a continuous time window approach, with the window length set to 0.3 to 1 second. Statistical analysis of joint angle changes within the continuous time window determines the corresponding joint motion state changes.

[0046] When the shoulder joint rotates forward continuously for 0.6 seconds and the change in angle continues to increase, the trend of joint movement change corresponding to the shoulder joint is determined to be a continuously increasing trend; when the wrist joint maintains a stable rotation angle for 0.4 seconds, the trend of joint movement change corresponding to the wrist joint is determined to be a stable and maintained trend.

[0047] The direction of shoulder joint rotation and the change in shoulder joint angle are determined based on shoulder joint status data.

[0048] The triaxial angle data corresponding to the shoulder joint position are decomposed to determine the rotational changes of the shoulder joint in the anterior-posterior, lateral, and vertical directions. The changes in shoulder joint angles are calculated by the difference in shoulder joint angles within a continuous time window.

[0049] The change in shoulder joint angle is calculated as follows: The change in shoulder joint angle is obtained by subtracting the shoulder joint angle value at the beginning of the continuous time window from the current shoulder joint angle value.

[0050] When the shoulder joint changes from 12° to 28° within a 0.5-second time window, the change in shoulder joint angle is determined to be 16°; when the direction of the change in shoulder joint angle is forward yaw, the direction of shoulder joint rotation is determined to be forward swing.

[0051] The elbow joint flexion and extension direction and the amount of elbow joint movement change are determined based on elbow joint status data.

[0052] Continuous sampling of the flexion-extension angle data corresponding to the elbow joint is performed to determine whether the elbow joint is currently in a flexed or extended state. The change in elbow joint motion is determined by the change in flexion-extension angle values ​​within a continuous time window.

[0053] Specifically, when the elbow joint angle continuously decreases, the flexion-extension direction corresponding to the elbow joint is determined as the flexion direction; when the elbow joint angle continuously increases, the flexion-extension direction corresponding to the elbow joint is determined as the extension direction.

[0054] When the elbow joint changes from 65° to 38° within 0.4 seconds, the flexion-extension direction of the elbow joint is determined to be the flexion direction, and the change in elbow joint motion is determined to be 27°.

[0055] The direction of wrist joint rotation and the amount of wrist joint posture change are determined based on wrist joint status data.

[0056] Posture analysis is performed on the deflection and rotation angle data corresponding to the wrist joint to determine the spatial rotation direction of the wrist. The change in wrist joint posture is determined by the change in posture angles within a continuous time window.

[0057] The wrist joint attitude angles include yaw, pitch, and roll. The changes in wrist joint attitude are obtained by weighted calculation of the yaw, pitch, and roll angles. The weighted calculation uses Euclidean distance.

[0058] When the wrist joint changes 8° in yaw angle, 5° in pitch angle, and 4° in roll angle within 0.5 seconds, the change in wrist joint posture is calculated to be 10.25° using Euclidean distance. As the yaw angle continues to increase, the direction of end-effector rotation of the wrist joint is determined to be clockwise.

[0059] The direction of joint rotation, the direction of joint flexion and extension, and the direction of end-rotation are taken as the direction of joint movement.

[0060] The rotation direction of the shoulder joint, the flexion and extension direction of the elbow joint, and the distal rotation direction of the wrist joint are uniformly mapped to form a set of corresponding joint movement directions.

[0061] The unified orientation mapping process includes orientation encoding and orientation normalization. Orientation encoding converts the orientation states of different joints into a unified orientation identifier; orientation normalization eliminates the differences in orientation dimensions between different joints.

[0062] For example, the forward swing direction of the shoulder joint is encoded as direction code D1, the flexion direction of the elbow joint is encoded as direction code D2, and the clockwise rotation direction of the wrist joint is encoded as direction code D3, forming the corresponding set of joint movement directions {D1, D2, D3}.

[0063] The range of motion of the joints is determined by the changes in shoulder joint angle, elbow joint movement, and wrist joint posture, and the corresponding trend of joint movement change is determined by the changes in the range of motion of the joints over a continuous period of time.

[0064] The changes in shoulder joint angle, elbow joint movement, and wrist joint posture are calculated together to determine the overall joint movement range of the human upper limb during the current movement.

[0065] Among them, the shoulder joint angle change is used to characterize the large-range swing of the shoulder, the elbow joint motion change is used to characterize the elbow flexion and extension, and the wrist joint posture change is used to characterize the wrist end-position adjustment. Since the motion dimensions corresponding to different joints are different, the shoulder joint angle change, elbow joint motion change, and wrist joint posture change are normalized before the joint amplitude calculation.

[0066] The normalization process employs extreme value normalization, which specifically includes: determining the maximum range of motion of the shoulder joint, the maximum flexion and extension angle of the elbow joint, and the maximum posture change angle of the wrist joint based on the preset range of motion, and proportionally converting the changes of each joint based on the corresponding maximum range of motion to obtain the corresponding normalized motion values.

[0067] Among them, the maximum range of motion of the shoulder joint is set to 180° based on the range of motion of the human shoulder joint, the maximum flexion and extension angle of the elbow joint is set to 150°, and the maximum posture change angle of the wrist joint is set to 90°.

[0068] The range of motion of the joints was obtained by weighting the normalized changes in shoulder joint angle, elbow joint motion, and wrist joint posture. In the weighted calculation, the weight value for the shoulder joint was set to 0.4, the weight value for the elbow joint was set to 0.35, and the weight value for the wrist joint was set to 0.25.

[0069] The trend of joint motion changes is determined by analyzing the changes in the amplitude of joint motion within a continuous time window. The length of the continuous time window is set to 0.5 seconds to 1.5 seconds.

[0070] Specifically, when the joint motion amplitude continuously increases, the corresponding joint motion change trend is determined to be an enhancing trend; when the joint motion amplitude continuously decreases, the corresponding joint motion change trend is determined to be a weakening trend; when the joint motion amplitude maintains a fluctuation range of less than 5% within a continuous time window, the corresponding joint motion change trend is determined to be a stable trend; and when the joint motion amplitude changes rapidly in a short period of time and the rate of change is greater than a preset change threshold, the corresponding joint motion change trend is determined to be an instantaneous change trend.

[0071] The preset change threshold is set based on the change value of joint motion amplitude per unit time within a continuous time window. Specifically, it is set to a change in joint motion amplitude greater than 0.12 per 0.1 seconds.

[0072] For example, when the shoulder joint angle change is 24°, the elbow joint motion change is 36°, and the wrist joint posture change is 15°, the corresponding normalized motion values ​​are 0.133, 0.24, and 0.167, respectively. A weighted calculation yields a joint motion amplitude of 0.177. If the corresponding joint motion amplitude continuously increases from 0.12 to 0.177 within 0.8 seconds, then the corresponding joint motion change trend is determined to be an increasing trend.

[0073] The corresponding end effector direction and spatial motion trajectory are determined based on the spatial motion data.

[0074] By jointly analyzing and processing the spatial coordinate change data, spatial direction change data, and end displacement change data in the spatial motion data, the actual motion direction and corresponding spatial trajectory change state of the corresponding end execution position of the exoskeleton assistive device in three-dimensional space can be determined.

[0075] Among them, the end-effector motion direction is used to characterize the current spatial motion direction of the end-effector execution position, and the spatial motion trajectory is used to characterize the spatial movement path of the end-effector execution position during continuous motion.

[0076] Spatial motion data is stored using a three-dimensional coordinate data structure, which includes X-axis, Y-axis, and Z-axis coordinate values. The X-axis represents displacement in the left-right direction, the Y-axis represents displacement in the forward-backward direction, and the Z-axis represents displacement in the up-down direction.

[0077] During the spatial motion trajectory generation process, trajectory fitting is performed on the spatial coordinate data within a continuous time window. The trajectory fitting process employs a cubic spline interpolation algorithm. The cubic spline interpolation algorithm is a piecewise cubic polynomial interpolation algorithm used to fit continuous trajectories of discrete spatial coordinate points to obtain smooth spatial motion trajectories.

[0078] For example, within a continuous 1-second time window, the spatial coordinates corresponding to the end execution position are (12, 5, 3), (18, 9, 5), (26, 15, 8), and (35, 22, 12), respectively. The spatial coordinates are fitted with the above spatial coordinates using a cubic spline interpolation algorithm to generate the corresponding spatial motion trajectory.

[0079] Based on spatial motion data, determine the spatial movement direction of the corresponding end effector position of the exoskeleton assistive device, and obtain the corresponding spatial position change data based on the spatial movement direction.

[0080] The spatial coordinate change data within a continuous time window is processed by direction vector calculation to determine the current spatial movement direction of the end execution position.

[0081] The direction of spatial movement is determined by the direction vector between the starting and ending coordinates of a continuous time window. The direction vector includes components of change along the X, Y, and Z axes.

[0082] Spatial location change data is obtained by calculating the difference between spatial coordinate values ​​at each moment within a continuous time window.

[0083] Among them, the X-axis direction change component is used to characterize the movement state of the end effector position in the left-right direction; the Y-axis direction change component is used to characterize the movement state of the end effector position in the front-back direction; and the Z-axis direction change component is used to characterize the movement state of the end effector position in the up-down direction.

[0084] For example, if the end-effector moves from coordinates (15, 8, 6) to coordinates (42, 18, 14) within a continuous 0.6-second time window, the corresponding direction vector is (27, 10, 8). Based on the direction vector, the spatial movement direction corresponding to the end-effector's position is determined to be the forward and upward movement direction.

[0085] The corresponding end-effector movement direction is determined based on the spatial position change data, and the corresponding spatial motion trajectory is generated.

[0086] The motion direction analysis is performed on the continuous directional change state in the spatial position change data to determine the current end-effector motion direction corresponding to the end-effector execution position.

[0087] Specifically, when the X-axis direction change component continuously increases, the end motion direction is determined to be the right-side motion direction; when the Y-axis direction change component continuously decreases, the end motion direction is determined to be the backward motion direction; and when the Z-axis direction change component continuously decreases, the end motion direction is determined to be the downward pressing motion direction.

[0088] Spatial motion trajectories are obtained by connecting spatial coordinate points within a continuous time window in chronological order. During the trajectory connection process, the trajectory curvature between adjacent spatial coordinate points is calculated to determine the trajectory change state of the corresponding spatial motion trajectory.

[0089] The trajectory curvature is calculated using the spatial turning angle formed by three consecutive spatial coordinate points. The larger the spatial turning angle, the greater the degree of directional change in the corresponding spatial trajectory.

[0090] For example, within a continuous time window, the spatial coordinates change sequentially to (20, 15, 18), (28, 20, 15), and (36, 24, 10). Since the Z-axis direction change component continuously decreases, the end motion direction is determined to be the downward motion direction. By connecting the above spatial coordinates, the corresponding spatial motion trajectory is generated, and it is determined that the corresponding trajectory has a downward bending change state.

[0091] like Figure 3 As shown, the action stage corresponding to the current operation is identified based on the joint movement direction, joint movement amplitude, joint movement change trend, end-effector movement direction, and spatial movement trajectory, so as to determine that the current operation is in at least one of the target action stages, namely the pressing stage, stretching stage, rotation stage, fine-tuning stage, or instantaneous release stage.

[0092] During the action phase recognition process, the action phase recognition results within a continuous time window are subjected to stable confirmation processing.

[0093] The continuous time window consists of at least three consecutive sampling periods. The duration of each sampling period is set to 20ms.

[0094] When the action stage identification results are consistent within a continuous time window, the current action stage identification result is determined to be a valid target action stage; when the action stage identification results are inconsistent within a continuous time window, the previous target action stage remains unchanged.

[0095] The stability confirmation process during the action phase employs a sliding time window algorithm. This algorithm is a continuous state stability determination algorithm used to determine the target state based on the consistency of states over a continuous time period.

[0096] When the duration of the target action phase is less than 60ms, the target primary assist joint switching process is not triggered.

[0097] The joint movement direction, joint movement amplitude, joint movement change trend, end-effector movement direction, and spatial movement trajectory are combined with motion feature matching to determine the target movement stage corresponding to the current operation.

[0098] During the motion phase recognition process, a set of motion phase recognition rules is established. The set of motion phase recognition rules includes rules for the pressing phase, stretching phase, rotation phase, fine-tuning phase, and instantaneous release phase.

[0099] The rules for identifying the compression phase include: the direction of the end-effector movement corresponds to the downward direction, the amount of change in elbow joint movement continuously increases, and the spatial movement trajectory continuously decreases along the Z-axis; the rules for identifying the stretching phase include: the amount of change in shoulder joint angle continuously increases, the direction of the end-effector movement corresponds to the direction of departure, and the spatial movement trajectory corresponds to a straight stretching trajectory; the rules for identifying the rotation phase include: the amount of change in wrist joint posture is greater than a preset rotation threshold, and the spatial movement trajectory corresponds to an arc-shaped trajectory; the rules for identifying the fine-tuning phase include: the amount of change in wrist joint posture is less than a preset fine-tuning threshold, and the spatial movement trajectory shows continuous small-amplitude deflection changes; the rules for identifying the instantaneous release phase include: the joint movement amplitude rapidly decreases in a short period of time, and the spatial movement trajectory corresponds to a rapid withdrawal trajectory. The stretching phase corresponds to joint range of motion recovery training, soft tissue extension training, or muscle stretching treatment.

[0100] The preset rotation threshold is set to 12°, and the preset fine-tuning threshold is set to 5°.

[0101] Action phase identification is performed using rule matching. When the current action features meet the corresponding action phase identification rules, the corresponding target action phase is determined.

[0102] When the direction of the end-effector movement corresponds to the downward direction, the change in elbow joint movement continuously increases from 18° to 42°, and the spatial movement trajectory decreases by 32mm along the Z-axis within 0.7 seconds, the target movement stage corresponding to the current operation is determined to be the pressing stage. The pressing stage corresponds to the rehabilitation treatment stage of continuously applying therapeutic pressure to the patient's target tissue. When the change in wrist joint posture reaches 16° and the spatial movement trajectory exhibits an arc-shaped change, the target movement stage corresponding to the current operation is determined to be the rotation stage.

[0103] Step 30: Based on the target action stage, identify the corresponding assisting joint from the shoulder joint, elbow joint, and wrist joint as the target main assisting joint, and establish the corresponding joint coordination relationship based on the target action stage.

[0104] Based on the movement characteristics corresponding to the current target movement stage, the main force-generating joints of the upper limb during the current movement are identified, thereby ensuring that the treatment load under different rehabilitation techniques can be accurately applied to the target treatment area, and establishing corresponding joint coordination control relationships based on the main force-generating joints.

[0105] Among them, the target primary assisting joint is used to characterize the assisting joint that undertakes the main motion output in the current motion phase; the joint coordination relationship is used to characterize the corresponding motion linkage relationship and assisting cooperation relationship between the target primary assisting joint and the other coordinating assisting joints.

[0106] In the process of determining the target primary assist joint, the degree of joint involvement corresponding to different target action stages is analyzed and processed. The degree of joint involvement is obtained by jointly calculating the joint's range of motion, change in joint output direction, and duration of action in the current action stage.

[0107] During the establishment of joint coordination relationships, the motion synchronization status between the target primary assisting joint and the cooperating auxiliary joints is matched. The motion synchronization status includes the synchronization status of joint motion direction, motion amplitude, and motion time.

[0108] When the current target action phase is the pressing phase, the elbow joint has the largest range of motion and the longest continuous output time. Therefore, the elbow joint is identified as the target primary assisting joint. The shoulder and wrist joints are identified as assisting joints and corresponding joint coordination relationships are established.

[0109] In another example, the step of identifying the corresponding assisting joint from the shoulder joint, elbow joint, and wrist joint as the target primary assisting joint based on the target action phase, and establishing the corresponding joint coordination relationship based on the target action phase, may preferably be as follows: When the target action phase corresponds to a large-scale spatial movement phase, the assisting joint corresponding to the shoulder joint whose shoulder joint state data is determined is the target main assisting joint.

[0110] The spatial trajectory range corresponding to the current target action phase is analyzed and processed. When the spatial displacement range corresponding to the spatial action trajectory is greater than a preset spatial movement threshold, the current target action phase is determined to be a large-scale spatial movement phase.

[0111] The preset spatial movement threshold is set based on the total spatial displacement length of the end execution position within a continuous time window, specifically set to 120mm.

[0112] Since the shoulder joint corresponds to the large-range swinging motion of the human upper limb, the assisting joint corresponding to the shoulder joint is identified as the target main assisting joint during the large-range spatial movement phase.

[0113] When the total spatial movement length corresponding to the end effector position reaches 185mm within 1 second and the shoulder joint angle change reaches 35°, the large-range spatial movement phase corresponding to the current target action phase is determined, and the assisting joint corresponding to the shoulder joint is determined as the target main assisting joint.

[0114] When the target action phase corresponds to a stable pushing phase or a continuous stretching phase, the assisting joint corresponding to the elbow joint whose elbow joint status data is determined is the target main assisting joint.

[0115] The joint force state, duration of movement, and direction of spatial force application corresponding to the current target movement stage are analyzed and processed together to determine whether the human upper limb is in a stable pushing or continuous stretching stage during the current movement.

[0116] The stable pushing phase corresponds to the action state in which the end-effector continuously outputs pressure in a fixed direction; the continuous stretching phase corresponds to the action state in which the end-effector continuously outputs tensile force in a fixed direction. Since the elbow joint corresponds to the main flexion and extension output position of the human upper limb, the assisting joint corresponding to the elbow joint is identified as the target main assisting joint in both the stable pushing phase and the continuous stretching phase.

[0117] During the identification of the stable pushing phase, the changes in elbow joint motion, the duration of elbow joint output, and the stability of the spatial motion trajectory direction are jointly judged. When the changes in elbow joint motion are greater than a preset pushing threshold, the duration of the action is greater than a preset duration threshold, and the rate of change of the spatial motion trajectory direction is less than a preset direction change threshold, the current target action phase is determined to be the stable pushing phase.

[0118] The preset push threshold is set to 25°, the preset duration threshold is set to 0.6 seconds, and the preset direction change threshold is set to 8°. The rate of change of spatial motion trajectory direction is calculated by the change in the angle between adjacent direction vectors within a continuous time window.

[0119] During the identification of the sustained extension phase, the combined motion state between the shoulder and elbow joints is analyzed and processed. When the elbow joint remains in extension and the spatial trajectory length corresponding to the end-effector position continuously increases, the sustained extension phase corresponding to the current target motion phase is determined.

[0120] When the elbow joint changes continuously from 42° to 78° within 1.2 seconds, the spatial movement trajectory moves forward continuously by 85mm along the Y-axis, and the rate of change of the spatial trajectory direction remains within 5°, the current target movement phase is determined to correspond to the stable pushing phase, and the assisting joint corresponding to the elbow joint is determined to be the target main assisting joint.

[0121] When the target action phase corresponds to the end-effector fine-tuning phase or rotation adjustment phase, the assisting joint corresponding to the wrist joint whose wrist joint state data is determined is the target primary assisting joint.

[0122] The wrist joint posture change, end-effector direction change, and spatial motion trajectory curvature change are jointly analyzed to determine whether the current target motion phase corresponds to the end-effector direction fine-tuning phase or the rotation adjustment phase.

[0123] The distal end fine-tuning phase corresponds to a small-range directional correction of the distal end position; the rotational adjustment phase corresponds to a rotational adjustment of the distal end position around the target position. Since the wrist joint corresponds to the position of fine motor control of the human upper limb, the assisting joint corresponding to the wrist joint is identified as the target primary assisting joint in both the distal end fine-tuning and rotational adjustment phases.

[0124] During the end-effector orientation fine-tuning phase identification process, the changes in wrist joint posture and the spatial motion trajectory offset are jointly judged. When the changes in wrist joint posture are less than a preset fine-tuning angle threshold and the spatial motion trajectory offset is less than a preset offset threshold, the end-effector orientation fine-tuning phase corresponding to the current target motion phase is determined.

[0125] The preset fine-tuning angle threshold is set to 6°, and the preset offset threshold is set to 15mm.

[0126] During the identification of the rotation adjustment phase, the changes in the wrist joint yaw angle and the curvature of the spatial movement trajectory are jointly analyzed and processed. When the change in the wrist joint yaw angle is greater than the preset rotation angle threshold and the spatial movement trajectory corresponds to an arc-shaped trajectory, the current target movement phase is determined to correspond to the rotation adjustment phase. The rotation adjustment phase corresponds to the rotational intervention movements in joint mobilization therapy, spinal correction therapy, and myofascial release therapy.

[0127] The curvature of the spatial motion trajectory is calculated using the spatial angle formed by three consecutive spatial coordinate points. When the change in the spatial angle is consistently greater than 12°, the corresponding circular arc trajectory is determined.

[0128] When the wrist joint yaw angle changes by 18° within 0.5 seconds, and the spatial motion trajectory corresponds to an arc-shaped trajectory with a radius of 65mm, the current target motion phase is determined to be the rotation adjustment phase, and the assisting joint corresponding to the wrist joint is determined to be the target main assisting joint.

[0129] The participation status of joints other than the target primary assist joint is analyzed and processed to determine the cooperative assist joints that form a linkage relationship with the target primary assist joint.

[0130] Coordinating assist joints are used to cooperate with the target primary assist joint to achieve motion stability, motion following, and directional adjustment during the corresponding motion phase. The process of determining coordinating assist joints includes joint motion correlation analysis and motion synchronization analysis.

[0131] Joint motion correlation is determined by the consistency of motion changes among different joints within a continuous time window. The consistency of motion changes is assessed using the Pearson correlation coefficient calculation method. The Pearson correlation coefficient is a linear correlation analysis algorithm used to characterize the degree of correlation between changes in two continuous variables.

[0132] Among them, when the correlation coefficient between different joints is greater than 0.75, it is determined that there is a motion linkage relationship between the corresponding joints.

[0133] When the assisting joint corresponding to the elbow joint is the target primary assisting joint, the correlation coefficient between the change in shoulder joint angle and the change in elbow joint motion reaches 0.82, and the correlation coefficient between the change in wrist joint posture and the change in elbow joint motion reaches 0.77. Therefore, it is determined that the assisting joints corresponding to the shoulder joint and the assisting joints corresponding to the wrist joint are both cooperative assisting joints.

[0134] Based on the target action phase, determine the corresponding assistance distribution relationship and action synchronization relationship between the target primary assist joint and the cooperating assist joint.

[0135] The contribution allocation relationship is represented by the contribution allocation coefficient K.

[0136] like Figure 4 As shown, the assist distribution coefficient K is calculated using the following formula: Where: K i W represents the assist distribution coefficient corresponding to the i-th joint; i represents the joint participation weight corresponding to the i-th joint; m represents the number of joints participating in the collaborative output.

[0137] Joint participation weight W i It is obtained by jointly calculating the range of motion, load value, and duration of the action of the corresponding joint.

[0138] Joint participation weight W i The following formula is used to calculate: W i =aA i +bL i +cD i ; Among them: A i Indicates the range of motion of the corresponding joint; L i Indicates the load state value of the corresponding joint; D iThis represents the duration of the action in the corresponding action phase; a, b, and c represent weight parameters.

[0139] Among them, the value range of 'a' is set to 0.3 to 0.4; the value range of 'b' is set to 0.4 to 0.5; and the value range of 'c' is set to 0.1 to 0.2.

[0140] Based on the motion load requirements and spatial motion requirements corresponding to the current target motion stage, the assist output ratio and motion synchronization status between the target main assist joint and the cooperative assist joint are configured.

[0141] Among them, the assist distribution relationship is used to characterize the corresponding assist output ratio between the target main assist joint and the cooperative assist joint; the motion synchronization relationship is used to characterize the motion execution time synchronization state and motion direction synchronization state between different joints.

[0142] The assist distribution relationship is determined by the degree of joint involvement. The degree of joint involvement is obtained by jointly calculating the range of motion of the corresponding joint, the duration of joint output, and the stability of the movement direction.

[0143] Action synchronization is achieved by matching the start time, end time, and direction change of actions for different joints. During the synchronization matching process, the time difference between actions of different joints is calculated. When the time difference is less than a preset synchronization time threshold, the corresponding joints are determined to satisfy the action synchronization relationship.

[0144] The preset synchronization time threshold is set to 0.15 seconds.

[0145] When the target action phase corresponds to the stable pushing phase, the assist output ratio for the elbow joint is set to 60%, the assist output ratio for the shoulder joint is set to 25%, and the assist output ratio for the wrist joint is set to 15%. When the action time difference between the elbow joint and the shoulder joint is 0.08 seconds, it is determined that the elbow joint and the shoulder joint meet the action synchronization relationship.

[0146] Based on the assist distribution relationship and the motion synchronization relationship, a corresponding joint coordination relationship is established.

[0147] The joint mapping process is performed on the assist output ratio, motion synchronization state, and motion direction linkage state between the target primary assist joint and the cooperative assist joint to establish the corresponding joint coordination relationship.

[0148] Joint coordination relationships are stored using a joint coordination parameter table. This table includes the target primary assist joint identifier, cooperating auxiliary joint identifiers, assist output ratio, motion synchronization time parameters, and motion direction linkage parameters.

[0149] The motion direction linkage parameter is used to characterize the correspondence between the motion directions of different joints. When the motion direction of the target primary assist joint changes, the corresponding synergistic assist joint synchronously adjusts its corresponding motion direction according to the motion direction linkage parameter.

[0150] When the elbow joint is the target primary assisting joint, the joint coordination parameter table records: the assisting output ratio of the shoulder joint is 25%, the assisting output ratio of the wrist joint is 15%, the motion synchronization time parameter is 0.1 seconds, and the motion direction linkage parameter is in the same direction linkage mode, thereby establishing the corresponding joint coordination relationship.

[0151] Step 40: Based on the target primary assist joint and the joint coordination relationship, determine the corresponding target assist direction, target assist path and target assist intensity to generate the corresponding three-dimensional coordinated assist control parameters.

[0152] The three-dimensional collaborative assist control parameters are represented by a parameter set P.

[0153] The parameter set P includes: P = {D, R, S, T}; Where: D represents the target assist direction parameter; R represents the target assist path parameter; S represents the target assist intensity parameter; and T represents the action synchronization time parameter.

[0154] The target assist direction parameter D is represented by a three-dimensional direction vector; the target assist path parameter R is represented by a spatial trajectory node sequence; the target assist intensity parameter S is represented by the target output force value; and the action synchronization time parameter T is represented by the joint drive synchronization time difference.

[0155] The action synchronization time parameter T satisfies: T≤0.03s.

[0156] When the motion synchronization time parameter T is greater than 0.03 seconds, the output state of the corresponding collaborative auxiliary joint is adjusted for synchronization compensation.

[0157] Based on the spatial motion requirements corresponding to the target motion stage and the motion output state corresponding to the target main assist joint, the assist output direction, assist movement path and assist output intensity of the exoskeleton assist device are jointly controlled to generate corresponding three-dimensional collaborative assist control parameters.

[0158] Among them, the target assistance direction is used to characterize the assistance output direction of the exoskeleton assistance device; the target assistance path is used to characterize the spatial motion path during the assistance output process; and the target assistance intensity is used to characterize the corresponding assistance output magnitude.

[0159] The three-dimensional collaborative assist control parameters are stored in the form of parameter data sets, which include direction parameters, path parameters, and intensity parameters. The direction parameters include the output components in the X-axis, Y-axis, and Z-axis directions; the path parameters include the spatial trajectory coordinate sequence and trajectory curvature parameters; and the intensity parameters include the target output force value and the rate of change of the force value.

[0160] When the target action phase corresponds to the stable pushing phase, the Z-axis output component in the direction parameter is set to the maximum value, the path parameter corresponds to the straight downward pressing trajectory, and the strength parameter corresponds to an output force value of 42N.

[0161] In another example, the step of determining the corresponding target assistance direction, target assistance path, and target assistance intensity based on the target primary assist joint and the joint coordination relationship, in order to generate the corresponding three-dimensional coordinated assistance control parameters, can preferably be: The corresponding target space direction of action is determined based on the target action phase.

[0162] The target action phase is analyzed and processed jointly to determine the target spatial action direction of the exoskeleton assistive device by analyzing the action output requirements, end-effector direction, and spatial trajectory change direction.

[0163] The target space direction of action is used to characterize the direction of the exoskeleton assistive device's output assistance during the current target action phase. The target space direction of action is represented by a three-dimensional direction vector. This three-dimensional direction vector includes X-axis, Y-axis, and Z-axis components.

[0164] Among them, the X-axis direction component is used to characterize the left and right direction assist output state; the Y-axis direction component is used to characterize the front and back direction assist output state; and the Z-axis direction component is used to characterize the up and down direction assist output state.

[0165] During the determination of the target space action direction, the direction of the end effector motion and the tangential change direction of the spatial motion trajectory are matched. When the end effector motion direction is consistent with the tangential direction of the spatial motion trajectory, the corresponding direction is determined as the target space action direction; when there is an angular deviation between the end effector motion direction and the tangential direction of the spatial motion trajectory, the target space action direction is corrected according to the angular deviation.

[0166] The angular deviation is determined by calculating the vector angle. This calculation uses an inverse cosine direction calculation algorithm. The inverse cosine direction calculation algorithm is a vector angle solving algorithm used to calculate the angle between corresponding directions based on the dot product relationship of two direction vectors.

[0167] Specifically, when the directional angle is greater than 15°, the direction of action in the target space is corrected; when the directional angle is less than or equal to 15°, the current direction of action in the target space remains unchanged.

[0168] When the direction vector corresponding to the end motion direction is (0.65, 0.32, -0.71) and the direction vector corresponding to the tangential direction of the spatial motion trajectory is (0.61, 0.28, -0.74), the angle between the vectors is calculated to be 6.3°, and the current direction vector is determined as the target spatial action direction; when the angle reaches 21°, the target spatial action direction is corrected according to the tangential direction of the spatial motion trajectory.

[0169] Based on the target space direction of action and the spatial motion trajectory of the target space direction, a corresponding target assistance path is generated.

[0170] A joint trajectory mapping process is performed on the target space action direction and the spatial motion trajectory of the target space action direction to generate the target assistance path corresponding to the exoskeleton assist device.

[0171] The target assist path is used to characterize the continuous assist output trajectory of the exoskeleton assistive device during the current action phase. The target assist path is represented by a spatial path coordinate sequence. The spatial path coordinate sequence includes multiple spatial trajectory nodes arranged in chronological order.

[0172] During the target-assisted path generation process, the trajectory curvature, trajectory direction change state, and trajectory length change state in the spatial motion trajectory are analyzed and processed, and the spatial motion trajectory is fitted with the target spatial action direction to perform assist path fitting.

[0173] The path fitting process employs a Bézier curve fitting algorithm. This algorithm is used to generate smooth, continuous spatial trajectories based on multiple spatial control points.

[0174] Specifically, when the spatial motion trajectory corresponds to a straight-line movement trajectory, a corresponding straight-line target assistance path is generated; when the spatial motion trajectory corresponds to an arc-shaped movement trajectory, a corresponding arc-shaped target assistance path is generated; and when the spatial motion trajectory has continuous changes in multiple directions, a corresponding composite curve-shaped target assistance path is generated.

[0175] During the target-assisted path generation process, the path spacing between spatial trajectory nodes is constrained. The path spacing is set to 5mm to 15mm to ensure the continuous and stable output trajectory of the assist.

[0176] When the spatial motion trajectory corresponds to an arc trajectory with a radius of 80mm and the target spatial action direction corresponds to the forward and downward direction, the corresponding arc-shaped target assistance path is generated by the Bezier curve fitting algorithm; the corresponding target assistance path includes 18 spatial trajectory nodes, and the path spacing between adjacent trajectory nodes is 10mm.

[0177] The target assist intensity is determined based on the load state of the target main assist joint and the mechanical requirements of the target action phase.

[0178] like Figure 5 As shown, the load state corresponding to the target main assist joint is represented by the load state value L.

[0179] The load status value L is calculated using the following formula: L = αT + βF + γV; Where: L represents the load state value, T represents the real-time output torque of the drive motor, F represents the real-time pressure detection value of the joint force sensor, V represents the real-time movement speed of the target main assist joint, and α, β and γ represent the corresponding weighting coefficients.

[0180] The weighting coefficients satisfy: α+β+γ=1; The values ​​of α are set to be 0.35 to 0.45; the values ​​of β are set to be 0.4 to 0.5; and the values ​​of γ are set to be 0.1 to 0.2.

[0181] When the load status value L is greater than the preset high load threshold, the current target main assist joint is determined to be in a high load state; when the load status value L is less than the preset low load threshold, the current target main assist joint is determined to be in a low load state.

[0182] The joint load state, motion output duration, and mechanical requirements corresponding to the target motion stage of the target main assist joint are jointly analyzed and processed to determine the target assist intensity of the exoskeleton assist device.

[0183] Target assist intensity is used to characterize the magnitude of the assist output from the exoskeleton assist device during the current movement phase. Target assist intensity is expressed as a target output force value, with the unit being Newtons.

[0184] The load status is obtained by jointly calculating the joint driving force, joint resistance, and joint motion change of the target main assist joint. The joint driving force is obtained by converting the output torque of the drive motor; the joint resistance is determined by the pressure detection value corresponding to the joint force sensor.

[0185] Mechanical requirements are determined by the type of movement corresponding to the target action stage, and comprehensively determined by combining the therapeutic load requirements of the target tissues during rehabilitation treatment. Specifically, the stable pushing stage corresponds to continuous pressure mechanical requirements; the continuous stretching stage corresponds to continuous stretching mechanical requirements; the rotational adjustment stage corresponds to dynamic rotation mechanical requirements; and the fine-tuning stage corresponds to low-amplitude fine-adjustment mechanical requirements.

[0186] like Figure 6 As shown, in the process of determining the target assist intensity, an intensity mapping process is performed between the load state and the mechanical requirements. This intensity mapping process employs a piecewise linear mapping algorithm. The piecewise linear mapping algorithm is an interval linear mapping algorithm used to adjust the assist intensity according to different output intensity intervals corresponding to different load intervals.

[0187] Specifically, when the load value corresponding to the target main assist joint is greater than the preset high load threshold, the target assist intensity is increased; when the load value corresponding to the target main assist joint is less than the preset low load threshold, the target assist intensity is decreased.

[0188] The preset high load threshold is set to 65N, and the preset low load threshold is set to 20N.

[0189] When the load value of the target main assist joint corresponding to the elbow joint reaches 72N, and the current target action phase corresponds to the stable pushing phase, the target assist intensity is determined to be 48N through a piecewise linear mapping algorithm; when the load value corresponding to the wrist joint is 16N and the current target action phase corresponds to the fine-tuning phase, the target assist intensity is determined to be 12N.

[0190] Step 50: Based on the three-dimensional collaborative assistance control parameters, control the target main assist joint of the exoskeleton assist device to output the main assist, and control the other collaborative joints to output the corresponding collaborative assist force, so as to form a three-dimensional collaborative assistance output corresponding to the target action stage.

[0191] like Figure 7 As shown, based on the direction parameters, path parameters, and intensity parameters in the three-dimensional collaborative assist control parameters, the exoskeleton assist device performs joint control processing on each joint drive unit to form a three-dimensional collaborative assist output corresponding to the current target action stage, so as to assist the therapist in stably completing the corresponding rehabilitation therapy action.

[0192] Among them, the target main assist joint is used to output the main driving force in the current action phase; the cooperative auxiliary joint is used to output the corresponding auxiliary stabilizing force, auxiliary following force or auxiliary adjusting force.

[0193] During the three-dimensional collaborative assist output process, the motion synchronization status between the target main assist joint and the collaborative auxiliary joints is controlled in real time. During motion synchronization control, time deviations between the corresponding drive signals of different joints are compensated for synchronously. The synchronization compensation time is controlled within 0.02 seconds.

[0194] During the drive control process, the output angle and output force of each joint drive motor are dynamically adjusted according to the spatial trajectory node sequence corresponding to the target assist path. Servo drive motors are used. Servo drive motors are closed-loop control drive motors used to adjust the output angle and output torque in real time according to the target control parameters.

[0195] When the current target action phase corresponds to the stable pushing phase, the target main assist joint corresponding to the elbow joint outputs 42N main assist, the corresponding synergistic auxiliary joint corresponding to the shoulder joint outputs 18N auxiliary stabilizing force, and the corresponding synergistic auxiliary joint corresponding to the wrist joint outputs 10N auxiliary adjusting force, thus forming the corresponding three-dimensional synergistic assist output.

[0196] In another example, the step of controlling the target primary assist joint of the exoskeleton assist device to output primary assist and controlling the other synergistic joints to output corresponding synergistic assist forces can preferably be: Control the target main assist joint to output the corresponding main assist along the target assist direction.

[0197] Based on the target assist direction and target assist intensity, the output direction and output force value of the drive motor corresponding to the target main assist joint are controlled.

[0198] During the target main assist joint output process, the three-dimensional direction vector corresponding to the target assist direction is decomposed into components to generate the corresponding X-axis output control quantity, Y-axis output control quantity and Z-axis output control quantity.

[0199] The drive motor adjusts the output torque direction and output angle according to the corresponding axial control quantity, so that the target main assist joint outputs the corresponding main assist along the target assist direction.

[0200] During the main assist output process, closed-loop correction is performed on the real-time position deviation corresponding to the target main assist joint. This closed-loop correction is handled using a PID control algorithm. The PID control algorithm is a proportional-integral-derivative (PID) control algorithm used to dynamically adjust the output control quantity based on the deviation between the target value and the actual value.

[0201] The position deviation is calculated as the difference between the target trajectory position and the real-time detection position. When the position deviation is greater than 3mm, the drive output is corrected for the deviation.

[0202] When the direction vector corresponding to the target assist direction is (0.42, 0.31, -0.85) and the target assist intensity is 45N, the corresponding directional torque is output to the drive motor corresponding to the target main assist joint, and the real-time position deviation is controlled within 2mm through the PID control algorithm.

[0203] Control the remaining synergistic joints to output corresponding synergistic auxiliary forces according to the joint synergistic relationship; among which, the synergistic auxiliary forces include auxiliary stabilizing force, auxiliary following force, or auxiliary adjusting force.

[0204] The auxiliary stabilizing force is used to limit the lateral displacement of the target primary assist joint during movement.

[0205] When the lateral offset of the target main assist joint is greater than the preset offset threshold, the output value of the auxiliary stabilizing force is increased.

[0206] The preset offset threshold is set to 8mm.

[0207] The auxiliary following force is used to maintain the synchronous speed of movement between the coordinating auxiliary joint and the target primary assisting joint.

[0208] When the time difference of the action corresponding to the assisted joint is greater than the preset synchronization threshold, the output value of the assisted following force is increased.

[0209] The preset synchronization threshold is set to 0.03 seconds.

[0210] The auxiliary adjustment force is used to correct the spatial orientation deviation corresponding to the end effector position.

[0211] When the spatial direction deviation angle corresponding to the end execution position is greater than the preset direction deviation threshold, the auxiliary adjustment force output value is increased.

[0212] The preset directional deviation threshold is set to 10°.

[0213] Based on the assist output ratio, motion synchronization time parameters, and motion direction linkage parameters in the joint coordination relationship, the drive output state corresponding to the coordinated assist joint is processed for coordinated control.

[0214] The auxiliary stabilizing force is used to maintain the motion balance of the target primary assist joint during its movement; the auxiliary following force is used to maintain the motion synchronization between the cooperating assist joint and the target primary assist joint; and the auxiliary adjusting force is used to correct the spatial orientation deviation corresponding to the end effector position.

[0215] During the collaborative assist force output process, the output force values ​​corresponding to different collaborative assist joints are proportionally allocated. The proportional allocation is determined based on the assist output ratio in the joint collaboration parameter table.

[0216] Specifically, when the target action phase corresponds to the stable pushing phase, the output ratio of the auxiliary stabilizing force is increased; when the target action phase corresponds to the continuous stretching phase, the output ratio of the auxiliary following force is increased; and when the target action phase corresponds to the fine-tuning phase or the rotation adjustment phase, the output ratio of the auxiliary adjusting force is increased.

[0217] When the target action phase corresponds to the rotation adjustment phase, the target main assist joint corresponding to the wrist joint outputs 30N main assist force, the corresponding auxiliary joint corresponding to the shoulder joint outputs 8N auxiliary stabilizing force, and the corresponding auxiliary joint corresponding to the elbow joint outputs 12N auxiliary following force, thus forming a corresponding auxiliary force output state.

[0218] By combining the main assist and the collaborative assist, a three-dimensional collaborative assist output corresponding to the target action stage is formed.

[0219] The main assist force output by the target main assist joint and the collaborative assist force output by the collaborative assist joint are fused in spatial direction and motion synchronization to form a corresponding three-dimensional collaborative assist output.

[0220] The three-dimensional collaborative assist output includes spatial direction collaborative state, motion trajectory collaborative state, and assist intensity collaborative state.

[0221] Spatial direction coordination state is used to characterize the directional consistency between the output directions of each joint; motion trajectory coordination state is used to characterize the degree of trajectory matching between the motion trajectories of each joint; assist intensity coordination state is used to characterize the balance state of the corresponding assist output between different joints.

[0222] During the collaborative output process, the time difference of movement, spatial orientation deviation, and trajectory offset between different joints are corrected in real time. When the time difference of movement is greater than 0.03 seconds, the spatial orientation deviation is greater than 10°, or the trajectory offset is greater than 12mm, the output state of the corresponding collaborative auxiliary joint is dynamically corrected.

[0223] When the target action phase corresponds to the continuous stretching phase, the elbow joint, corresponding to the target main assist joint, outputs 52N main assist force, the shoulder joint, corresponding to the cooperative auxiliary joint, outputs 20N auxiliary following force, and the wrist joint, corresponding to the cooperative auxiliary joint, outputs 9N auxiliary adjustment force. By performing directional synchronization control and trajectory synchronization control on the output states of different joints, a three-dimensional cooperative assist output corresponding to the continuous stretching phase is formed.

[0224] In some rehabilitation treatment scenarios, the three-dimensional collaborative assistance output is also used to help therapists maintain a preset range of treatment intensity, reduce fluctuations in treatment intensity caused by long-term treatment operations, and improve the patient's treatment experience and the consistency of treatment effects.

[0225] In this embodiment, by acquiring the state data and spatial motion data of the upper limb joints corresponding to the exoskeleton assistive device, the motion states of the shoulder, elbow, and wrist joints are jointly analyzed. Combined with the end-effector movement direction, spatial motion trajectory, and joint motion change trend, the target motion stage corresponding to the current operation is identified. Furthermore, based on the motion characteristics corresponding to different target motion stages, the corresponding target primary assist joint and cooperating auxiliary joints are dynamically determined, and corresponding assist distribution relationships and motion synchronization relationships are established to form corresponding joint coordination relationships. On this basis, according to the target spatial action direction, spatial motion trajectory, and load state corresponding to the target primary assist joint of the target motion stage, the corresponding target assist direction and target assist are generated. The path and target assistance intensity are determined, and corresponding three-dimensional collaborative assistance control parameters are formed. Subsequently, based on the three-dimensional collaborative assistance control parameters, the assistance output state of the target main assistance joint and the collaborative auxiliary joint is jointly controlled to achieve synchronous output of main assistance and collaborative assistance forces. This enables the exoskeleton assistive device to dynamically adjust the assistance coordination state between each joint according to different movement stages, improve the movement following accuracy, movement stability and assistance output coordination in complex spatial movement processes, ensure the consistency of treatment trajectory, treatment intensity and treatment direction in rehabilitation treatment, improve the treatment quality and treatment safety in joint function recovery training, soft tissue release treatment and spinal rehabilitation intervention, and solve the problem of difficulty in dynamically switching the main force-applying joint in different movement stages.

[0226] like Figure 8 As shown, this application provides a three-dimensional collaborative control system 10 for a rehabilitation exoskeleton assistive device, comprising: The data acquisition module 11 is used to acquire the state data and spatial motion data of the human upper limb joints corresponding to the exoskeleton assistive device; wherein, the human upper limb joint data includes at least the shoulder joint, elbow joint and wrist joint.

[0227] The action recognition module 12 is used to determine the target action stage corresponding to the current operation action based on the status data and spatial action data.

[0228] The motion establishment module 13 is used to determine the corresponding assisting joint from the shoulder joint, elbow joint and wrist joint as the target main assisting joint according to the target motion stage, and to establish the corresponding joint coordination relationship according to the target motion stage.

[0229] The motion determination module 14 is used to determine the corresponding target assistance direction, target assistance path and target assistance intensity based on the target main assist joint and the joint coordination relationship, so as to generate the corresponding three-dimensional coordinated assistance control parameters.

[0230] The motion control module 15 is used to control the target main assist joint of the exoskeleton assist device to output the main assist according to the three-dimensional collaborative assist control parameters, and to control the other collaborative joints to output the corresponding collaborative assist force, so as to form a three-dimensional collaborative assist output corresponding to the target motion stage.

[0231] In this embodiment, the data acquisition module 11 acquires the state data and spatial motion data of the human upper limb joints corresponding to the exoskeleton assistive device, and performs joint analysis on the motion state of each joint of the human upper limb based on the state data of the shoulder joint, elbow joint, and wrist joint. Then, the motion recognition module 12 identifies the target motion stage corresponding to the current operation by combining the joint motion direction, joint motion amplitude, joint motion change trend, end-effector motion direction, and spatial motion trajectory. Furthermore, the motion establishment module 13 dynamically determines the corresponding target primary assist joint among the shoulder, elbow, and wrist joints based on the motion characteristics corresponding to different target motion stages, and establishes corresponding assist distribution relationships and motion synchronization relationships, thereby forming corresponding joint coordination relationships. Based on this... Above, the motion determination module 14 combines the load state of the target main assist joint, the target spatial action direction, and the spatial motion trajectory of the target spatial action direction to generate the corresponding target assist direction, target assist path, and target assist intensity, and form the corresponding three-dimensional collaborative assist control parameters. Subsequently, the motion control module 15 performs joint control on the drive output state of the target main assist joint and the collaborative auxiliary joint according to the three-dimensional collaborative assist control parameters, so that the target main assist joint outputs the corresponding main assist, and the collaborative auxiliary joint outputs the corresponding auxiliary stabilizing force, auxiliary following force, and auxiliary adjusting force, thereby realizing the three-dimensional collaborative assist output under different action stages, improving the motion following accuracy, joint motion coordination, and assist output stability of the exoskeleton assist device in complex spatial motion processes.

[0232] It should be noted that the information interaction and execution process between the above systems / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0233] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0234] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A three-dimensional collaborative control method for a rehabilitation exoskeleton assistive device, characterized in that, include: Acquire state data and spatial motion data of the human upper limb joints corresponding to the exoskeleton assistive device; wherein, the human upper limb joint data includes at least the shoulder joint, elbow joint and wrist joint; The target action stage corresponding to the current operation action is determined based on the state data and the spatial motion data. Based on the target action phase, the corresponding assisting joint among the shoulder joint, elbow joint, and wrist joint is identified as the target main assisting joint, and a corresponding joint coordination relationship is established based on the target action phase. Based on the target main assist joint and the joint coordination relationship, the corresponding target assist direction, target assist path and target assist intensity are determined to generate the corresponding three-dimensional coordinated assist control parameters. Based on the three-dimensional collaborative assistance control parameters, the target main assist joint of the exoskeleton assist device is controlled to output main assistance, and the other collaborative joints are controlled to output corresponding collaborative assistance forces, so as to form a three-dimensional collaborative assistance output corresponding to the target action stage.

2. The method as described in claim 1, characterized in that, The step of identifying the action stage corresponding to the current operation based on the state data and the spatial action data includes: Based on the shoulder joint status data, elbow joint status data, and wrist joint status data, determine the corresponding joint movement direction, joint movement amplitude, and joint movement change trend. The corresponding end effector direction and spatial motion trajectory are determined based on the spatial motion data. The action stage corresponding to the current operation is identified based on the joint movement direction, the joint movement amplitude, the joint movement change trend, the end-effector movement direction, and the spatial movement trajectory, so as to determine that the current operation is in at least one of the target action stages: pressing stage, stretching stage, rotation stage, fine-tuning stage, or instantaneous release stage.

3. The method as described in claim 2, characterized in that, The step of determining the corresponding joint movement direction, joint movement range, and joint movement change trend based on the shoulder joint state data, elbow joint state data, and wrist joint state data includes: The shoulder joint rotation direction and shoulder joint angle change are determined based on the shoulder joint status data. The elbow joint flexion and extension direction and the amount of elbow joint movement change are determined based on the elbow joint status data. The wrist joint state data are used to determine the corresponding end rotation direction of the wrist joint and the amount of wrist joint posture change. The joint rotation direction, the joint flexion and extension direction, and the distal rotation direction are taken as the joint movement direction; The range of motion of the joints is determined by the change in shoulder joint angle, the change in elbow joint movement, and the change in wrist joint posture, and the corresponding trend of joint movement change is determined based on the change in the range of motion of the joints over a continuous time period.

4. The method as described in claim 2, characterized in that, The steps of determining the corresponding end effector direction and spatial motion trajectory based on the spatial motion data include: Based on the spatial motion data, determine the spatial movement direction of the corresponding end effector position of the exoskeleton assistive device, and obtain the corresponding spatial position change data based on the spatial movement direction; The corresponding end-effector movement direction is determined based on the spatial position change data, and the corresponding spatial motion trajectory is generated.

5. The method as described in claim 1, characterized in that, The step of determining the corresponding assisting joint from the shoulder joint, elbow joint, and wrist joint as the target primary assisting joint based on the target action phase, and establishing the corresponding joint coordination relationship based on the target action phase, includes: When the target action phase corresponds to a large-scale spatial movement phase, the assisting joint corresponding to the shoulder joint in the shoulder joint state data is determined as the target main assisting joint. When the target action phase corresponds to a stable pushing phase or a continuous stretching phase, the assisting joint corresponding to the elbow joint in the elbow joint state data is determined as the target main assisting joint. When the target action phase corresponds to the end-direction fine-tuning phase or the rotation adjustment phase, the assisting joint corresponding to the wrist joint in the wrist joint state data is determined as the target main assisting joint. The participation status of the joints other than the target main assist joint is analyzed to identify at least one cooperative assist joint corresponding to the target main assist joint. Based on the target action phase, determine the corresponding assistance distribution relationship and action synchronization relationship between the target primary assist joint and the cooperative assist joint; Based on the aforementioned assist distribution relationship and the aforementioned action synchronization relationship, a corresponding joint coordination relationship is established.

6. The method as described in claim 1, characterized in that, The step of determining the corresponding target assist direction, target assist path, and target assist intensity based on the target main assist joint and the joint coordination relationship, in order to generate the corresponding three-dimensional coordinated assist control parameters, includes: Determine the corresponding target spatial action direction based on the target action phase; Based on the target space direction of action and the spatial motion trajectory of the target space direction of action, a corresponding target assistance path is generated; The target assist intensity is determined based on the load state of the target main assist joint and the mechanical requirements of the target action phase.

7. The method as described in claim 1, characterized in that, The step of controlling the target primary assist joint of the exoskeleton assist device to output primary assist based on the three-dimensional collaborative assist control parameters, and controlling the other collaborative joints to output corresponding collaborative assist forces, includes: Control the target main assist joint to output the corresponding main assist along the target assist direction; Control the remaining cooperating joints to output corresponding cooperative auxiliary forces according to the joint cooperation relationship; The combined output of the main assist and the cooperative assist force forms a three-dimensional cooperative assist output corresponding to the target action stage.

8. A three-dimensional collaborative control system for a rehabilitation exoskeleton assistive device, characterized in that, include: The data acquisition module is used to acquire the state data and spatial motion data of the human upper limb joints corresponding to the exoskeleton assistive device; wherein, the human upper limb joint data includes at least the shoulder joint, elbow joint and wrist joint; The action recognition module is used to determine the target action stage corresponding to the current operation action based on the state data and the spatial action data. The motion establishment module is used to determine the corresponding assisting joint from the shoulder joint, the elbow joint and the wrist joint as the target main assisting joint according to the target motion stage, and to establish the corresponding joint coordination relationship according to the target motion stage. The motion determination module is used to determine the corresponding target assist direction, target assist path and target assist intensity based on the target main assist joint and the joint coordination relationship, so as to generate corresponding three-dimensional coordinated assist control parameters. The motion control module is used to control the target main assist joint of the exoskeleton assist device to output main assist according to the three-dimensional collaborative assist control parameters, and to control the other collaborative joints to output corresponding collaborative assist forces, so as to form a three-dimensional collaborative assist output corresponding to the target motion stage.

Images