CN118211739A - Exoskeleton-based track planning method, storage medium and exoskeleton - Google Patents

Abstract

The invention relates to the technical field of exoskeleton, and discloses a track planning method based on exoskeleton, a storage medium and the exoskeleton. The exoskeleton-based track planning method comprises the following steps: the method comprises the steps of obtaining a target limb health grade, determining a target track model corresponding to the target limb health grade, and planning the joint track of the exoskeleton according to the target track model.

Description

Exoskeleton-based track planning method, storage medium and exoskeleton

Technical Field

The invention relates to the technical field of exoskeleton, in particular to an exoskeleton-based track planning method, a storage medium and an exoskeleton.

Background

The exoskeleton can be worn outside the body of a user, and provides assistance for the limbs of the wearer through an external power source or a mechanical energy storage mechanism so as to assist the wearer to more conveniently complete daily life. In recent years, the world is entering an aging society. Problems with the increased number of elderly people include an increased number of elderly people with difficulty in daily life due to reduced exercise function, and a shortage of caregivers, especially deterioration of lower limb functions, which may lead to a blocked basic movements of standing, sitting, walking, etc., a reduced exercise opportunity for going out, etc., and eventually bedridden, resulting in a vicious circle with further deterioration of functions. Under the circumstance, exoskeleton is increasingly valued by a plurality of scholars and scientific researchers at home and abroad as equipment for realizing the function, and becomes a new research hotspot.

The related art only develops the joint track of the exoskeleton for a single rehabilitation scenario, for example, only develops the joint track for a patient with lower limbs having no exercise ability or rehabilitation to a certain extent, and for a wearer with fully rehabilitated lower limbs, the exoskeleton lacks effective interactive control with such wearer, so that the existing exoskeleton cannot be effectively compatible with multiple rehabilitation control modes.

Disclosure of Invention

An object of an embodiment of the invention is to provide an exoskeleton-based track planning method, a storage medium and an exoskeleton, and aims to solve the technical problem that related technologies cannot be compatible with various lower limb rehabilitation scenes.

In a first aspect, an embodiment of the present invention provides an exoskeleton-based trajectory planning method, including:

Obtaining a target limb health grade;

Determining a target track model corresponding to the target limb health grade;

and planning the joint track of the exoskeleton according to the target track model.

In a second aspect, embodiments of the present invention provide a storage medium storing computer-executable instructions for causing an electronic device to perform the exoskeleton-based trajectory planning method described above.

In a third aspect, embodiments of the present invention provide an exoskeleton comprising:

at least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the exoskeleton-based trajectory planning method described above.

In the exoskeleton-based track planning method provided by the embodiment of the invention, the health grade of the target limb is obtained, the target track model corresponding to the health grade of the target limb is determined, and the joint track of the exoskeleton is planned according to the target track model. According to the embodiment, the corresponding target track model is flexibly selected to plan the joint track which accords with the limb health condition of the user by combining the target limb health grade of the user, so that the embodiment can be compatible with various lower limb rehabilitation scenes, and the application range of the exoskeleton provided by the embodiment is favorably improved.

Drawings

One or more embodiments are illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements, and in which the figures of the drawings are not to be taken in a limiting sense, unless otherwise indicated.

FIG. 1 is a schematic diagram of an exoskeleton control system according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an exoskeleton employing a hierarchical control system according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a three-ring control structure of position, speed and moment based on FOC control algorithm according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of the principle framework of the torque-current control shown in FIG. 3;

FIG. 5 is a schematic flow chart of an exoskeleton-based trajectory planning method according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of the working conditions of the joint muscles of a human body when walking according to the embodiment of the invention;

FIG. 7 is a schematic illustration of a normalized joint moment curve provided by an embodiment of the present invention;

FIG. 8 is a schematic view of standard joint trajectories corresponding to hip joint locations, knee joint locations, and ankle joint locations provided in an embodiment of the present invention;

FIG. 9 is a schematic diagram of generating a trajectory database based on user characteristics according to the embodiment of FIG. 1;

FIG. 10 is a schematic diagram of a principle framework based on an admittance control strategy according to an embodiment of the present invention;

FIG. 11 is a schematic illustration of an ith health belt of an ith IMU sensor provided in an embodiment of the present invention;

FIG. 12 is a schematic diagram of an ith health belt for an ith muscular force transducer provided by an embodiment of the present invention;

Fig. 13 is a schematic circuit diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be noted that, if not in conflict, the features of the embodiments of the present invention may be combined with each other, which is within the protection scope of the present invention. In addition, while functional block division is performed in a device diagram and logical order is shown in a flowchart, in some cases, the steps shown or described may be performed in a different order than the block division in the device, or in the flowchart. Furthermore, the words "first," "second," "third," and the like as used herein do not limit the order of data and execution, but merely distinguish between identical or similar items that have substantially the same function and effect.

The embodiment of the invention provides an exoskeleton control system. Referring to fig. 1, an exoskeleton control system 100 includes a remote server 200 and an exoskeleton 300, where the remote server 200 is communicatively connected to the exoskeleton 300.

The remote server 200 is capable of device management, user management, and storing specific sensor data based on different users. Device management includes the following: the information such as the model, configuration, warranty period, identification code and the like of the management equipment can be maintained and managed by a manufacturer. User management includes the following: the user rights of the login user can be configured, login account numbers can be created, the roles of the login user can be allocated, and the like. The information storage and display module comprises the following contents: the state (fault, normal, etc.), electric quantity information, work statistics information such as walking steps, energy consumption reduction, equipment positioning, etc. of the equipment can be displayed. The motion stored data may be processed by a deep learning algorithm built into the remote server 200. When the exoskeleton 300 is an exoskeleton for assisting rehabilitation of a human limb, the remote server 200 can output a patient's health report, assisting in making a patient rehabilitation plan, etc. by interacting with the exoskeleton 300. When exoskeleton 300 is an exoskeleton that is applied in an industrial field or a military field, remote server 200 can output an evaluation report of the power assisting effect of the wearer.

Exoskeleton 300 employs a hierarchical control system. Referring to fig. 2, the hierarchical control system of the exoskeleton 300 includes a high-level decision layer 301, a middle control and signal conversion layer 302, and a low-level control and driving layer 303. Because the application scene of the exoskeleton is complex, the control requirement of the exoskeleton cannot be met by simple bottom layer control, and the hierarchical control system provided by the embodiment can be simultaneously applied to the exoskeleton system for rehabilitation and assistance, has strong expandability and supports the multi-module sharing function.

The upper decision layer 301 is the core of the hierarchical control system, and the control framework of the upper decision layer 301 is composed of a motion intention recognition module 304, a track planning module 305 and a finite state machine 306.

The movement intention recognition module 304 is configured to recognize a movement intention of the user according to gait motion data of the user. The recognition of the movement intention includes the following 3 meanings:

① Recognition of motion patterns: generally, 8 basic actions frequently involved in daily life are taken as exercise modes to be identified, namely sitting still-SI (Sit), standing still-SD (Stand), walking-WK (Walk), running-RN (Run), ascending stairs-SA (Stair Ascent), descending stairs-SD (Stair Descent), ascending slopes-RA (Ramp Ascent) and descending slopes-RD (Ramp Descent).

② Identification of gait phase: gait phases are further subdivisions of motion in each motion pattern. The degree of subdivision of the phase is different depending on the purpose of the study. For example, the gait phase of a person's walking can be seen as consisting of alternating support and swing phases.

③ Predicting a motion state switching event;

Wherein ①② is to judge the current situation according to the training and learning knowledge, ③ is to predict the impending action intention.

The identification of the movement intention can be completed by adopting a mode based on a physical sensor (sensors such as angle, force/moment and inertia measurement), a mode based on a bioelectric signal (myoelectric meter skin signal, electrocardio, electroencephalogram and the like) or a combination of the two modes.

The trajectory planning module 305 can adjust or generate the desired joint trajectory online, wherein the trajectory planning module 305 can select a trajectory model according to the patient disability level. The trajectory planning module 305 integrates a trajectory database that stores a variety of standard joint trajectories. The selection of the standard joint trajectory is based on the subject's height, age, gait parameters (stride, stride frequency), speed and disability level selection or online generation. If the subject is at a completely disabled level, the corresponding standard joint trajectory is selected from the trajectory database according to the subject's basic biological parameters for generating motor position commands via the intermediate control and signal conversion layer 302. When the height of the subject does not exist in the track database, the standard joint track of the subject can be obtained by interpolation or fitting and the like by utilizing the standard joint track data of the adjacent heights. The height is selected to establish the track database, and information such as the thigh length and the calf length may be selected as the basis for establishing the track database.

The finite state machine 306 is capable of switching from one state to another with the wearer's intent to move as a trigger. Each locomotor mode is a state, each gait phase is a sub-state in the locomotor mode, and the transition between the states is accomplished by a state switching event. In order to improve the robustness of state switching, the state switching event is composed of a plurality of redundant events through AND operation, namely, the state switching is completed when a plurality of conditions are simultaneously established.

The intermediate control and signal conversion layer 302 is configured to invoke a corresponding dynamics model according to each identified motion state, and generate a joint track of each joint motor online in real time. The intermediate control and signal conversion layer 302 includes a variable-impedance motion control module and a dynamics analysis module, where the variable-impedance motion control module is used to generate a joint track, and the dynamics analysis module is used to generate a control type instruction. The joint track and the control type command are used as input of the lower control and driving layer 303, and the corresponding joint module is controlled to work according to the joint track according to the control type command, wherein the exoskeleton can control the FOC algorithm according to the rotation vector to control the joint module.

The lower control and driving layer 303 comprises a sensor acquisition board and a joint module, wherein the sensor acquisition board comprises a plurality of IMU sensors, a plurality of pressure sensors, a plurality of muscle sensors, a plurality of current sensors and a plurality of encoders, and the joint module comprises a position controller, a torque controller, a current controller and a motor arranged at the corresponding joint part.

The plurality of IMU sensors are respectively a waist IMU sensor, a first IMU sensor, a second IMU sensor, a third IMU sensor, a fourth IMU sensor, a fifth IMU sensor and a sixth IMU sensor, wherein the waist IMU sensor is bound on the back of the waist by using a binding belt, the first IMU sensor is bound on the hip joint part of the left leg by using a binding belt, the second IMU sensor is bound on the knee joint part of the left leg by using a binding belt, the third IMU sensor is bound on the ankle joint part of the left leg by using a binding belt, the fourth IMU sensor is bound on the hip joint part of the right leg by using a binding belt, the fifth IMU sensor is bound on the knee joint part of the right leg by using a binding belt, and the sixth IMU sensor is bound on the ankle joint part of the right leg by using a binding belt, and each IMU sensor is used for detecting the joint angle of the corresponding joint part.

The plurality of pressure sensors includes a first pressure sensor, a second pressure sensor, and a third pressure sensor, wherein each pressure sensor is disposed at a designated foot location on each side of the foot.

For each leg, the plurality of muscle sensors are a first muscle sensor, a second muscle sensor, a third muscle sensor, a fourth muscle sensor, a fifth muscle sensor, and a sixth muscle sensor, respectively. The first muscle force sensor and the second muscle force sensor are respectively oppositely arranged on muscles on two opposite sides of the hip joint part. The third muscle force sensor and the fourth muscle force sensor are respectively oppositely arranged on muscles on two opposite sides of the knee joint part. The fifth muscle force sensor and the sixth muscle force sensor are respectively arranged on muscles on two opposite sides of the ankle joint part in a relative way.

The first pressure sensor is arranged at the toe, the second pressure sensor is arranged at the midfoot, the third pressure sensor is arranged at the heel, and the first pressure sensor is used for collecting the pressure applied by the toe of a user to the first pressure sensor and obtaining a pressure signal of the toe. The second pressure sensor is used for collecting the pressure applied to the second pressure sensor by the midfoot part of the user, and obtaining a pressure signal of the midfoot part. The third pressure sensor is used for collecting the pressure applied to the third pressure sensor by the heel of the user, and obtaining a pressure signal of the heel.

Each current sensor is arranged on a current path of the motor and is used for detecting the driving current of the motor. Each encoder is mounted on an output shaft of the motor and is used for detecting the rotating speed of the motor. The motor is a brushless motor, and the embodiment can control the brushless motor to control and switch multiple modes by adopting a proper rotation control FOC algorithm.

Referring to fig. 3, the present embodiment adopts a three-ring control structure of position, speed and moment based on the FOC control algorithm, and in order to conveniently realize the switching of multiple motion modes, the system can freely switch among the multiple motion modes through a preset switch instruction, that is, through a bus preset instruction, the motor can work in any motion mode according to the control requirement.

Referring to fig. 4, the torque-current controller is composed of a torque controller and a d/q axis current controller. The embodiment can input the difference between the expected torque and the feedback torque to the torque controller and output a current reference command. The torque controller can be completed by adopting a PI or PID control structure. The d/q axis current controller may be constituted by a PI or PID control structure. The feedback torque is real-time torque obtained by detecting the permanent magnet synchronous motor, and the feedback torque obtained by detecting is the actual torque of the permanent magnet synchronous motor under the condition of not considering noise. Since it is difficult to directly measure the torque, and the torque satisfies the following relationship with the exciting current and the torque current:

T＝kN_p{ψ+(L_d-L_q)i_sd}i_sd

thus, the feedback torque can be derived from the feedback excitation current and the feedback torque current, specifically, the feedback torque is:

T_fbk＝kN_p{ψ+(L_d-L_q)i_{sd_fbk}}i_{sq_fbk}

Where T _fbk is feedback torque, i _{sd_fbk} is feedback exciting current, i _{sq_fbk} is feedback torque current, and k is a conversion coefficient, which is a constant. N _p is the pole pair number, ψ is the magnetic flux of the permanent magnet, and L _d and L _q are the direct axis winding inductance value and the quadrature axis winding inductance value of the permanent magnet synchronous electrode represented by the two-phase quadrature direct axis coordinate system dq. The feedback exciting current is real-time exciting current obtained by detecting the permanent magnet synchronous motor, the feedback torque current is real-time exciting current obtained by detecting the permanent magnet synchronous motor, the feedback exciting current obtained by detection is the actual exciting current of the permanent magnet synchronous motor under the condition of not considering noise, and the feedback torque current is the actual torque current of the permanent magnet synchronous motor.

The "current sensing and coordinate transformation" performs the following operation, namely, the three-phase current is converted into exciting current and torque current on the dq axis:

In the above formula, i _su、i_sv and i _sw are three-phase currents, and K is a transformation coefficient. When performing an energy-invariant absolute transformation of coordinates, When a relative transformation is performed with unchanged amplitude before and after transformation/>I _sd and i _sq are the excitation current and the torque current in the two-phase direct current component, respectively; θ _r is the synchronous rotation angle of the rotor, which can be obtained by multiplying the mechanical rotation angle calculated by the encoder by the pole pair number Np, or by calculating through a hall sensor.

Therefore, the embodiment can measure the three-phase current of the permanent magnet synchronous motor to obtain the three-phase feedback current, and then the three-phase feedback current is converted through the formula to obtain the feedback exciting current and the feedback torque current. When measuring the three-phase current of the permanent magnet synchronous motor, according to the symmetry of the current, the current of the last phase can be obtained by arbitrarily measuring the two-phase current, and the u-phase current i _su and the v-phase current i _sv are actually detected. The w-phase current i _sw is: i _sw＝-(i_su+i_sv). When the current is measured, the current can be measured by a current sensor, and also can be measured by connecting a shunt resistor in series in a three-phase winding led out by the permanent magnet synchronous motor. In addition to the feedback control described above, the torque controller may also employ a feedforward controller that generates a torque command current using the following equation:

Wherein i1' _{sq_ref} is a first torque command current, K _FF is a gain coefficient of the torque feedforward controller, N _p is a pole pair number of the permanent magnet synchronous motor, ψ is a magnetic flux of a permanent magnet of the permanent magnet synchronous motor, and T _ref is a desired torque.

In some embodiments, referring to fig. 2, the hierarchical control system further includes a man-machine interaction layer 304, where the man-machine interaction layer 304 is configured to provide an interaction interface for a user to interact with the exoskeleton, and the man-machine interaction layer 304 may interact with the exoskeleton in a manner of a man-machine interaction screen or physical buttons.

As another aspect of the embodiments of the present invention, the embodiments of the present invention provide an exoskeleton-based trajectory planning method. Referring to fig. 5, the exoskeleton-based trajectory planning method includes the following steps:

S51: and obtaining the target limb health grade.

In this step, the limb health grade is a grade for representing the health of the limb of the user, wherein the limb may be an upper limb or a lower limb. The limb health grade comprises a normal grade, an adjacent health grade, an auxiliary exercise health grade, a certain exercise health grade and a complete paralysis grade, and the target limb health grade is one of the health grades. Wherein the normal level is a level at which the limb can normally act autonomously. The near health grade is a grade where the limb can act autonomously with less difficulty. The exercise health assistance level is a level that can assist the autonomous action of the limb with little assistance applied to the exoskeleton. A certain sports health grade is a grade that applies most of the assistance to the exoskeleton to assist in the autonomous movement of the limb. The complete paralysis grade is the grade that the limbs can not act at all and the whole process cooperation of the exoskeleton is required to act autonomously. After the user wears the exoskeleton, the exoskeleton acquires limb detection data, and a target limb health grade is determined according to the limb detection data.

S52: a target trajectory model corresponding to the target limb health level is determined.

In this step, the track model is a model for generating a joint track, where the track model includes a passive control model, an adaptive model, and a timely power-assisting model, and the target track model is one of the track models. Wherein, the passive control model is suitable for patients with lower limbs completely unconscious, namely, the lower limbs of the patients are driven by the exoskeleton completely. The self-adaptive model is suitable for patients with lower limbs recovering to a certain stage, the lower limbs of the patients have certain movement capacity, and the exoskeleton can adopt strategies such as impedance control and the like. The timely power-assisted model is suitable for patients with complete rehabilitation of lower limbs, the exoskeleton is mainly used for assisting walking, physical energy consumption is reduced, and the timely power-assisted model is used for providing proper power for the wearer at proper time while ensuring that the exoskeleton and the wearer coordinate compliant motion.

In some embodiments, each trajectory model is configured with a model tag that corresponds to a limb health level, and determining a target trajectory model that corresponds to a target limb health level includes: selecting a model label corresponding to the target limb health grade as a target model label, and selecting a track model with the model label as the target model label as a target track model.

For example, model tag I D1 for the passive control model is 1, where the model tag for the "complete paralysis level" is I D. The model label I D of the adaptive model is 2, wherein the model labels of the "certain sports health grade" and the "auxiliary sports health grade" are I D. Model label I D of the timely power-assisted model is 3, wherein model labels of the "near health grade" and the "normal grade" are I D. When the target model label is 1, the exoskeleton selects the passive control model as the target track model, and so on, which are not described herein.

S53: and planning the joint track of the exoskeleton according to the target track model.

In this step, the joint track may be a joint angle curve or a joint moment curve, etc.

According to the embodiment, the corresponding target track model is flexibly selected to plan the joint track which accords with the limb health condition of the user by combining the target limb health grade of the user, so that the embodiment can be compatible with various lower limb rehabilitation scenes, and the application range of the exoskeleton provided by the embodiment is favorably improved.

In some embodiments, the exoskeleton comprises a plurality of joint locations, and when the target trajectory model is a timely assistance model, planning the joint trajectory of the exoskeleton according to the target trajectory model comprises the steps of:

s531: gait data is acquired, and a gait phase is determined from the gait data.

S532: the power type of the target joint location is determined based on the gait phase, the target joint location being one of a plurality of joint locations.

S533: and planning the joint track of the target joint part according to the power type of the target joint part.

In S531, the gait data includes joint angle data acquired by respective IMU sensors located at different joint portions of the exoskeleton and/or plantar pressure data acquired by respective pressure sensors provided on the intelligent shoe of the exoskeleton.

Gait phase is used to describe the state of the feet when the person walks, as previously described, different researchers may give a division of the unsynchronized state phase, in some embodiments the gait phase may be divided as follows: gait phase ① of "right foot forward bipedal support", gait phase ② of "left foot swing", gait phase ③ of "left foot forward bipedal support", and gait phase ④ of "right foot swing". The corresponding gait phases are sequentially carried out under one gait cycle, and under normal conditions, the gait phases under the same gait cycle form a standard gait sequence. As previously described, under normal conditions, one acts in accordance with the following standard gait sequence, the first standard gait sequence, as follows: gait phase ① → gait phase ② → gait phase ③ → gait phase ④. Second standard gait timing: gait phase ② → gait phase ③ → gait phase ④ → gait phase ①. Third standard gait timing: gait phase ③ → gait phase ④ → gait phase ① → gait phase ②. Fourth standard gait timing: gait phase ④ → gait phase ① → gait phase ② → gait phase ③.

In some embodiments, the gait phase may also be divided into: gait phase ① of "left foot on rear right foot on front ground contact", gait phase ② of "right foot support left foot swing early", gait phase ③ of "right foot support left foot swing middle", gait phase ④ of "right foot support left foot swing late", gait phase ⑤ of "right foot on rear left foot on front double foot support", and gait phase ⑥ of "left foot on rear right foot on front ground contact".

In some embodiments, determining the gait phase from the gait data comprises: and determining gait phases according to the joint angle data acquired by each IMU sensor.

In some embodiments, determining gait phases from joint angle data acquired by the respective IMU sensors includes: filtering abnormal joint angle data acquired by each IMU sensor to obtain filtered joint angle data, performing equal length processing on each filtered joint angle data to obtain equal length joint angle data, performing normalization processing on each equal length joint angle data to obtain normalized joint angle data, performing feature processing on the normalized joint angle data to obtain gait feature values, and determining gait phases according to the gait feature values.

In some embodiments, determining the gait phase from the gait feature value comprises: and inputting the gait characteristic value into a phase analysis model to obtain the gait phase, wherein the phase analysis model comprises a neural network model, a fuzzy algorithm model or a fuzzy neural network model.

In some embodiments, determining the gait phase from the gait data comprises: gait phases are determined from plantar pressure data.

In some embodiments, the plantar pressure data includes a pressure signal of the right heel and a pressure signal of the right toe, a pressure signal of the left heel and a pressure signal of the left toe, and determining the gait phase from the plantar pressure data includes: and determining a first level change of the pressure signals of the right heel and the right toe and a second level change of the pressure signals of the left heel and the left toe, and determining the gait phase according to the first level change and the second level change.

In S532, the exoskeleton includes a plurality of joint locations including a hip joint location, a knee joint location, and an ankle joint location, and the hip joint location is a target joint location when it is desired to plan a joint trajectory of the hip joint location. Similarly, when it is desired to plan the joint trajectory of the knee joint portion, the knee joint portion is the target joint portion. When the joint track of the ankle joint part needs to be planned, the ankle joint part is a target joint part.

The power type is a type that is required to provide energy to the joint part for completing the motion of the corresponding gait phase, and when the power type is a positive power type, the joint part is required to perform positive power to complete the motion of the corresponding gait phase. When the power type is negative, the joint part needs to be subjected to negative power to complete the movement of the corresponding gait phase.

It will be appreciated that when the power type is a positive power type, the corresponding joint portion of the exoskeleton will need to perform positive work on the corresponding joint portion of the user in order to assist the user in walking, reducing the walking load of the user. When the power type is a negative power type, the corresponding joint part of the exoskeleton does not need to do positive work on the corresponding joint part of the user, and the corresponding joint part of the user can be driven to walk by utilizing the energy accumulated by the muscle force of the corresponding joint part of the user through subsequent automatic release, so that the intervention of the exoskeleton on the corresponding joint part can be reduced, the energy is saved, and the control difficulty is reduced.

Referring to fig. 6, taking the right foot as an example, when the gait phase is "the left foot is touching the ground at the rear and the right foot is touching the ground at the front", the power type H1 of the hip joint part of the right foot is a positive power type, and the power type K1 of the knee joint part is a negative power type. When the gait phase is "early swing of right foot support left foot", the power type K2 of the knee joint portion of the right foot is a positive power type. When the gait phase is "middle swing of right foot support left foot", the power type A1 of the knee joint portion and the ankle joint portion of the right foot is a negative power type. When the gait phase is "late swing of right foot support left foot", the power type H2 of the hip joint part of the right foot is a negative power type, and the power type A2 of the knee joint part and the ankle joint is a positive power type. When the gait phase is "right foot posterior left foot anterior bipedal support", the power type H3 of the hip joint portion of the right foot is a positive power type, and the power type K3 of the knee joint portion is a negative power type. When the gait phase is "left foot behind right foot in front of the front ground contact", the power type K4 of the knee joint portion of the right foot is a negative power type.

In S533, the exoskeleton can plan the joint track of the target joint part according to the power type of the target joint part, so that the exoskeleton can reduce the control difficulty, and is beneficial to improving the planning efficiency of the joint track.

In some embodiments, planning the joint trajectory of the target joint location according to the power type of the target joint location comprises the steps of: when the power type of the target joint part is positive, the target joint part is controlled to work according to a preset standard joint track, and when the power type of the target joint part is negative, the target joint part is controlled to stop working.

For example, as described above, when the gait phase is "the left foot is on the front ground with the right foot behind", the power type H1 of the hip joint portion is a positive power type, so the exoskeleton controls the joint driving unit of the hip joint portion to operate, so that the hip joint portion of the exoskeleton operates according to the preset standard joint track, thus assisting the hip joint portion of the user, reducing the walking load of the user, and reducing the fatigue of muscles. Meanwhile, since the power type K1 of the knee joint part is a negative power type, that is, the muscles of the knee joint part are accumulating energy to counter the forward movement of gravity, the exoskeleton does not need to control the joint driving unit of the knee joint part to work at the moment, that is, the knee joint part of the exoskeleton does not need to drive the knee joint part of the user to act, so that the knee joint part of the exoskeleton is controlled to stop working, the control difficulty and calculation force can be reduced, and the planning efficiency of the joint track is improved.

When the joint track is a joint moment curve, referring to fig. 7, fig. 7 shows moment curves of various joint parts of the subject in a normal gait cycle, and a power-assisted instruction table can be constructed according to the moment curves and a required power-assisted proportion system C (0= < C < = 1), and the vertical axis of fig. 7 also shows that the joint moment is periodically standardized on the horizontal axis and weight standardized on the vertical axis. In fig. 7, when assistance is required between 50% and 65% of the gait cycle, the exoskeleton performs negative work on the hip joint part of the right foot and positive work on the knee joint part and the ankle joint according to the power type of each target joint part, that is, the hip joint part of the exoskeleton is controlled to stop working, the knee joint part and the ankle joint of the exoskeleton are controlled to work according to the preset standard joint track, an upward dotted line 71 exists at 50% to 65% of the gait cycle, and the dotted line 71 represents that positive work is performed on the knee joint part and the ankle joint of the exoskeleton to improve the joint moment.

In some embodiments, the exoskeleton is configured with a trajectory database that stores a plurality of standard joint trajectories, and when the power type of the target joint portion is a positive power type, controlling the target joint portion to operate according to the preset standard joint trajectories includes the following steps:

s5331: and accessing a track database to select the standard joint track of the target joint part as the target standard joint track.

S5332: and controlling the target joint part to work according to the target standard joint track.

In S5331, each joint portion in the exoskeleton corresponds to a corresponding standard joint track, for example, referring to fig. 8, the hip joint portion corresponds to a first standard joint track 81, the knee joint portion corresponds to a second standard joint track 82, and the ankle joint portion corresponds to a third standard joint track 83, where each standard joint track is used to describe a change of a joint angle of the corresponding joint portion in a gait cycle, and the exoskeleton can control the corresponding joint portion to act according to the standard joint track according to the corresponding standard joint track.

When the target joint location is a hip joint location, the exoskeleton selects the first standard joint trajectory 81 as the target standard joint trajectory. When the target joint location is a knee joint location, the exoskeleton selects the second standard joint trajectory 82 as the target standard joint trajectory. When the target joint portion is an ankle joint portion, the exoskeleton selects the third standard joint trajectory 83 as the target standard joint trajectory.

In some embodiments, the present embodiment may generate a standard joint track of each joint portion under each user feature in advance, and store the standard joint track of each joint portion under the user feature in a track database, so that the exoskeleton is configured with the track database, which stores a plurality of standard joint tracks, each of which corresponds to a joint portion of the corresponding user feature.

Referring to fig. 9, user characteristics are constrained by at least gender and/or age and/or height, such as age may be divided into the following age groups: [18,28],[29,39],[40,50],[51,61],[62,72],[73,83]. Height may be divided into the following height segments: 150,155,160,165,170,175,180,185,190, wherein an age group may correspond to multiple height groups, e.g., age groups [18,28] may be associated with any of 150,155,160,165,170,175,180,185,190 height groups, respectively.

Typically, there are individual differences in joint trajectories, which are associated with user information, including gender, age, and height. The track database provided in this embodiment can perform experiments according to gender, age and height, so as to use the joint tracks collected by the gait normal person under each user feature as standard joint tracks. For the tested population of each age range, a person with a height of 150cm to 190cm is selected, walks at a standard gait and a constant speed on a horizontal road surface, and simultaneously records the track curve of the tested person in the sagittal plane.

In some embodiments, generating a standard joint trajectory for each joint location under each user characteristic comprises: acquiring joint angle data of a plurality of gait cycles of an experimental joint part under each user characteristic, and generating a standard joint track of the experimental joint part according to the joint angle data of the plurality of gait cycles, wherein the joint angle data comprises a plurality of joint angles acquired in one gait cycle, the acquisition points of the joint angle data in one gait cycle are the same, and the experimental joint part is one joint part of a hip joint part, a knee joint part and an ankle joint part.

In some embodiments, generating a standard joint trajectory for the experimental joint region from the joint angle data for the plurality of gait cycles comprises: according to the joint angle data of each gait cycle, the kth joint angle of the experimental joint part is obtained, the angle average value of each kth joint angle is obtained, and according to the angle average value of each joint angle, the standard joint track of the experimental joint part is generated.

For example, when the joint track is collected in this embodiment, the time corresponding to the joint track immediately before the next right foot heel strike is taken as a complete gait cycle T. For the same subject, a plurality of gait cycle data are continuously collected and recorded as N, and for the joint track of the subject, the present embodiment takes the angular average value of the joint angles at the same position in the N gait cycles. Wherein the joint angle data for the ith gait cycle includes M joint angles, denoted as [1,2,3,4 … … k, k+1 … … M ] in each gait cycle.

The joint angle at the kth point is: avg (k) = [ y1 (k) +y2 (k) + for use, +yi (k) … +yn (k) ]/N, i=1, 2.

According to the method, the standard joint track is generated by calculating the angle average value of each kth joint angle, so that the reliability and the accuracy of the standard joint track are improved, and the exoskeleton can drive corresponding joint parts to act more reliably and accurately.

As previously described, each standard joint trajectory corresponds to a joint location of a corresponding user characteristic, such as standard joint trajectory A1 for a hip joint location of a male, age group [18,28], height group 160, standard joint trajectory A2 for a knee joint location of a male, age group [18,28], height group 160, and standard joint trajectory A3 for a hip joint location of a female, age group [18,28], height group 165.

In some embodiments, accessing the trajectory database to select a standard joint trajectory of the target joint location as the target standard joint trajectory comprises: obtaining user information, judging whether a user characteristic matched with the user information exists in a track database, if so, selecting a standard joint track corresponding to the user information as a target standard joint track, if not, determining two user characteristics adjacent to the user information as reference user characteristics, and fitting to generate the target standard joint track according to the standard joint track of the two reference user characteristics.

The user information includes a user age, the user feature having an age less than the user age being one reference user feature, the user feature having an age greater than the user age being another reference user feature. Similarly, the user information includes a user height, the user feature having a height less than the user height is one reference user feature, and the user feature having a height greater than the user height is another reference user feature.

In some embodiments, generating the target standard joint trajectory by fitting from the standard joint trajectories of the two reference user features includes: and fitting to generate a target standard joint track according to the interpolation algorithm and the standard joint tracks of the two reference user characteristics.

In some embodiments, for a patient with lower extremities having no locomotion capability, the target limb health level is a complete paralysis level, the exoskeleton may select the passive control model as the target trajectory model, and therefore, when the target trajectory model is the passive control model, planning the joint trajectory of the exoskeleton according to the target trajectory model comprises the steps of:

S534: and obtaining an adjustment coefficient and a standard joint track of a target joint part, wherein the target joint part is one of a plurality of joint parts.

S535: and adjusting the standard joint track of the target joint part according to the adjustment coefficient and the standard joint track of the target joint part.

S536: and controlling the target joint part to work according to the adjusted standard joint track.

In S534, the adjustment coefficient is a coefficient for adjusting the amplitude of the standard joint trajectory. In some embodiments, obtaining the adjustment coefficient includes: the coefficient corresponding to the target limb health grade is selected as the adjustment coefficient, and the adjustment coefficient is selected according to the target limb health grade, so that artificial subjective interference can be eliminated, the adjustment coefficient can be selected quantitatively, and the track planning effect can be improved.

In S535, in some embodiments, adjusting the standard joint trajectory of the target joint portion according to the adjustment coefficient and the standard joint trajectory of the target joint portion comprises: and multiplying the adjustment coefficient by the standard joint track of the target joint part to obtain an adjusted standard joint track. When the adjustment coefficient is smaller than 1, the exoskeleton can reduce the amplitude of the standard joint track, so that the secondary strain of the exoskeleton on the muscles of the human body can be reduced. When the adjustment coefficient is greater than 1, the exoskeleton can increase the amplitude of the standard joint trajectory.

In S536, the exoskeleton adjusts the standard joint track of the target joint part according to the adjustment coefficient and the standard joint track of the target joint part, so that the exoskeleton can meet the joint action requirement when the patient is completely paralyzed, avoid secondary damage to the patient, and can help the patient to recover the activities of the lower limbs.

In some embodiments, for patients recovering to a certain stage, the muscles of the patient have a certain muscle strength and may exercise autonomously to a certain extent. If the fully passive control model is continuously adopted to plan the joint path, and the autonomous movement capability of the subject is not considered, the exoskeleton and the patient can not move in coordination, and the better rehabilitation effect can not be achieved, so that when the target limb health grade is an auxiliary movement health grade or a certain movement health grade, the exoskeleton can select the self-adaptive model as the target track model.

When the target track model is an adaptive model, planning the joint track of the exoskeleton according to the target track model comprises the following steps:

s537: the method comprises the steps of obtaining expected interaction force and man-machine interaction force of a target joint part, wherein the target joint part is one of a plurality of joint parts.

S538: inputting the expected interaction force and the man-machine interaction force into a preset admittance control model so that the admittance control model outputs the angle adjustment quantity.

S539: and adjusting the standard joint track of the target joint part according to the angle adjustment quantity to obtain the self-adaptive joint track.

S540: and controlling the target joint part to work according to the self-adaptive joint track.

In S537, the desired interaction force is customized by the designer based on engineering experience. The man-machine interaction force is the force detected when the target joint part interacts with the user.

Referring to fig. 10, the exoskeleton applies admittance control strategy to standard joint trajectoriesAnd performing online adjustment to generate a self-adaptive joint track. Assuming that we expect the interaction vertical F to be equal to 0, if the patient is in a complete paralysis state, i.e. the lower limb of the person cannot move, the human-computer interaction force feedback value is equal to 0, the corrected angle command q ^amt is also equal to 0, and then the control is converted into passive control. When the man-machine interaction force is not equal to 0, the exoskeleton can adjust the standard joint track on line in real time so as to incorporate the movement capacity of the patient to plan the joint track.

In S538, inputting the desired interaction force and the man-machine interaction force into the preset admittance control model includes: subtracting the man-machine interaction force from the expected interaction force to obtain an interaction force difference value, and inputting the interaction force difference value into a preset admittance control model.

In S539, the exoskeleton calculates an adaptive joint trajectory according to the following equation, as follows: Wherein, omega _n is the response angular frequency of human-machine interaction force and adjustment joint angle, and ζ is the attenuation coefficient.

In S540, the exoskeleton controls the target joint portion to operate according to the adaptive joint trajectory. According to the embodiment, the admittance control model is adopted, so that the influence of the force of a user can be fully considered, and the safety and the comfort are improved. In addition, the relation between man-machine interaction force and adjustment angle is set to be a second-order system, and parameters of an admittance control model are directly adjusted, so that the debugging is convenient. Meanwhile, the angular frequency and the attenuation coefficient can be properly adjusted according to the comfort of people, and compared with the current common parameter setting method, the parameter setting is quick.

In some embodiments, obtaining the target limb health level comprises the steps of:

S41: and acquiring limb detection data, wherein the limb detection data is data of detecting that the limb of the user performs actions in a gait cycle after the user wears the exoskeleton.

S42: and calculating a limb action value according to the limb detection data and a preset standard joint track, wherein the limb action value is used for representing the flexibility degree of the limb.

S43: and determining the target limb health grade according to the limb action value and a preset health grade table.

In S41, after the user wears the exoskeleton, the IMU sensor of the exoskeleton can detect a joint angle of a limb of the user, and/or the muscle force sensor can detect a muscle force signal of the limb of the user, and the exoskeleton generates limb detection data according to the joint angle and/or the muscle force signal. Wherein the limb can be an upper limb or a lower limb.

In some embodiments, the limb test data comprises a class of limb test values over a gait cycle, wherein the limb test values may be represented by parameters such as joint angle, muscle force signals, joint moment or muscle force moment, and the like, and thus the limb test data comprises joint angle or muscle force signals or joint moment or muscle force moment over a gait cycle.

In some embodiments, the limb detection data includes at least two types of limb detection values over the same gait cycle. As previously mentioned, joint angle, muscle force signals, joint moment or muscle force moment are all limb measurements belonging to different classes.

Each type of limb detection value can be obtained by sampling at least two limb sensors distributed at different joint positions of the exoskeleton, wherein the limb sensors are sensors for sampling the limb detection values. When the limb sensor is an IMU sensor, the limb detection value is a joint angle or joint moment, and when the limb sensor is a muscle force sensor, the limb detection value is a muscle force signal or muscle force moment.

As described above, the joint angle may be sampled by a first IMU sensor or a second IMU sensor or a third IMU sensor disposed at different joint locations of the exoskeleton, and the muscle force signal may be sampled by any one of a first muscle force sensor to a sixth muscle force sensor disposed at different joint locations of the exoskeleton.

When each type of limb detection value is composed of at least two limb sensors distributed at different joint parts of the exoskeleton, gait cycles of a plurality of limb detection values belonging to the same type are the same, namely, the limb detection values with the same sampling points are contained in the same gait cycle. For example, during the kth gait cycle, the first IMU sensor samples N first joint angles, the second IMU sensor samples N second joint angles, and the third IMU sensor samples N third joint angles. Similarly, in the k+1th gait cycle, the first IMU sensor samples N first joint angles, the second IMU sensor samples N second joint angles, and the third IMU sensor samples N third joint angles.

In S42, each joint location in the exoskeleton corresponds to a respective standard joint trajectory.

In some embodiments, when the limb detection data includes a type of limb detection value, and the limb detection value is obtained by sampling a limb sensor disposed at a corresponding joint portion of the exoskeleton, the exoskeleton accesses the track database to select a standard joint track corresponding to the limb detection value, and calculates a limb motion value according to the limb detection value and the standard joint track corresponding to the limb detection value.

For example, the limb detection value is a joint angle, wherein the joint angle G1 is obtained by sampling an IMU sensor B1 disposed at the hip joint position, and the exoskeleton calculates the limb motion value according to a standard joint track C1 and the joint angle G1 corresponding to the IMU sensor B1.

In some embodiments, when the limb detection data includes a type of limb detection value, and the limb detection value is obtained by sampling at least two limb sensors arranged at corresponding joint positions of the exoskeleton, the exoskeleton accesses the track database to determine a standard joint track corresponding to each limb sensor, and calculates a limb motion value according to the limb detection values sampled by the plurality of target limb sensors and the standard joint track corresponding to the target limb sensors, wherein the target limb sensor is one of the at least two limb sensors.

For example, the limb detection value is a joint angle, wherein the joint angle G1 is obtained by sampling an IMU sensor B1 disposed at the hip joint portion, the joint angle G2 is obtained by sampling an IMU sensor B2 disposed at the knee joint portion, and the exoskeleton calculates the limb motion value according to a standard joint track C1 and the joint angle G1 corresponding to the IMU sensor B1 and a standard joint track C2 and the joint angle G2 corresponding to the IMU sensor B2.

In some embodiments, when the limb test data includes at least two types of limb test values, each type of limb test value is sampled by a limb sensor disposed at a corresponding joint portion of the exoskeleton.

For example, the limb detection values include joint angles and muscle force signals, wherein the joint angles G1 are obtained by sampling by IMU sensors B1 arranged at the hip joint positions, the muscle force signals J1 are obtained by sampling by muscle force sensors E1 arranged at the hip joint positions, and the exoskeleton calculates the limb motion values Z1 according to the standard joint trajectories C1 and the joint angles G1 corresponding to the IMU sensors B1. The exoskeleton calculates a limb motion value Z2 according to a standard joint track C3 corresponding to the muscle force sensor E1 and a muscle force signal J1.

In some embodiments, when the limb test data includes at least two types of limb test values, each type of limb test value is sampled by a plurality of limb sensors disposed at corresponding joint locations of the exoskeleton.

For example, the limb detection values include joint angles and muscle force signals, wherein the joint angles G1 are sampled by IMU sensors B1 arranged at the hip joint positions, the joint angles G2 are sampled by IMU sensors B2 arranged at the knee joint positions, the muscle force signals J1 are sampled by muscle force sensors E1 arranged at the hip joint positions, and the exoskeleton calculates the limb motion value Z3 according to standard joint trajectories C1 and joint angles G1 corresponding to the IMU sensors B1 and standard joint trajectories C2 and joint angles G2 corresponding to the IMU sensors B2. The exoskeleton calculates a limb motion value Z2 according to a standard joint track C3 corresponding to the muscle force sensor E1 and a muscle force signal J1.

For another example, the limb detection values include a joint angle G1 and a muscle force signal, wherein the joint angle G1 is sampled by an IMU sensor B1 disposed at the hip joint portion, the joint angle G2 is sampled by an IMU sensor B2 disposed at the knee joint portion, the muscle force signal J1 is sampled by a muscle force sensor E1 disposed at the hip joint portion, and the muscle force signal J2 is sampled by a muscle force sensor E2 disposed at the knee joint portion. The exoskeleton calculates a limb action value Z3 according to a standard joint track C1 and a joint angle G1 corresponding to the IMU sensor B1 and a standard joint track C2 and a joint angle G2 corresponding to the IMU sensor B2. The exoskeleton calculates a limb motion value Z4 according to the standard joint track C3 and the muscle force signal J1 corresponding to the muscle force sensor E1 and the standard joint track C4 and the muscle force signal J2 corresponding to the muscle force sensor E2.

In S43, the health grade table is an evaluation table for evaluating the limb health condition of the user, wherein the health grade table includes at least two limb health grades, the target limb health grade is one of the at least two limb health grades, and the limb health grade is a health grade for representing the limb of the user, wherein the limb may be an upper limb or a lower limb.

The limb health grade comprises a normal grade, an adjacent health grade, an auxiliary exercise health grade, a certain exercise health grade and a complete paralysis grade, and the target limb health grade is one of the health grades. Wherein the normal level is a level at which the limb can normally act autonomously. The near health grade is a grade where the limb can act autonomously with less difficulty. The exercise health assistance level is a level that can assist the autonomous action of the limb with little assistance applied to the exoskeleton. A certain sports health grade is a grade that applies most of the assistance to the exoskeleton to assist in the autonomous movement of the limb. The complete paralysis grade is the grade that the limbs can not act at all and the whole process cooperation of the exoskeleton is required to act autonomously. After the user wears the exoskeleton, the exoskeleton acquires limb detection data, and a target limb health grade is determined according to the limb detection data.

In some embodiments, when the limb detection data includes a class of limb detection values, the exoskeleton traverses the health class table to select a limb health class that satisfies the limb movement value as the target limb health class.

In some embodiments, when the limb detection data includes at least two types of limb detection values, the exoskeleton traverses the health class table to select as the target limb health class a limb health class that satisfies at least two types of limb motion values simultaneously.

The embodiment can eliminate the factor interference of human subjectivity and reliably and accurately evaluate the limb health grade of the user. In addition, the embodiment can automatically evaluate the limb health grade of the user without excessive participation of rehabilitation doctors, thereby being beneficial to saving medical resources.

In some embodiments, the limb detection data comprises at least two types of limb detection values within the same gait cycle, and calculating the limb movement value according to the limb detection data and the preset standard joint trajectory comprises the following steps:

s421: and acquiring a target limb detection value sampled by the exoskeleton at each sampling moment, wherein the target limb detection value is one limb detection value in at least two types of limb detection values.

S422: and calculating a limb action value corresponding to the target limb detection value according to the target limb detection value and the standard joint track corresponding to the target limb detection value.

In S421, for example, the limb detection value may be any of a joint angle and a muscle force signal, and when the target limb detection value is the joint angle, the IMU sensor is capable of sampling the joint angle at each sampling time in one gait cycle, for example, the t _w represents the w sampling time,The ith joint angle of the ith I MU sensor is represented, the gait cycle is recorded as [ t ₁,t₂,t₃,..,t_j...,t_m ], and the 1 st IMU sensor samples at a sampling time t ₁ to obtain a joint angle/>Sampling at a sampling time t ₂ to obtain a joint angle/>By analogy, during the gait cycle [ t ₁,t₂,t₃,..,t_j...,t_m ], the 1 st IMU sensor is able to sample the joint angle/>Similarly, the 2 nd IMU sensor can sample the joint angles/>And so on.

When the target limb detection value is a muscle force signal, the muscle force sensor can sample the muscle force signal at each sampling time in one gait cycle, for example, the t _w represents the w sampling time,Representing the ith muscle force sensor and the ith muscle force signal, the gait cycle is recorded as [ t ₁,t₂,t₃,..,t_q...,t_n ],/>For the q-th sampling time, the 1 st muscle sensor samples at a sampling time t ₁ to obtain a muscle force signal F ₁ ¹, and samples at a sampling time t ₂ to obtain a muscle force signal/>By analogy, during the gait cycle [ t ₁,t₂,t₃,..,t_q...,t_n ], the 1 st muscle sensor can sample the following muscle force signalsSimilarly available, the 2 nd muscle sensor can sample the following muscle force signalsAnd so on.

In S422, the limb sensors for sampling the limb detection values all correspond to the standard joint trajectories, and as described above, since the 1 st IMU sensor is disposed on the hip joint region, the 1 st IMU sensor corresponds to the standard joint trajectoriesBecause the 2 nd I MU sensor is arranged on the knee joint part, the 2 nd IMU sensor corresponds to the standard joint track/>Because the 3 rd IMU sensor is arranged on the ankle joint part, the 3 rd IMU sensor corresponds to the standard joint track/>

Similarly, the 6 muscle force sensors can also respectively correspond to the following standard joint tracks, as follows:

In some embodiments, the exoskeleton is based on the following when the target limb detection value is the joint angle, with each type of limb detection value sampled by one limb sensor arranged at the corresponding joint position of the exoskeleton />And calculating a limb movement value M theta. When the target limb detection value is a muscle force signal, the exoskeleton is used for controlling the external force according to/>/>The limb movement value MF is calculated. Subsequently, the exoskeleton determines a target limb health level according to the limb motion value mθ and/or the limb motion value MF and the health level table.

In some embodiments, the exoskeleton is based on the following when the target limb detection values are joint angles, respectively, on the premise that each type of limb detection value is sampled by at least two limb sensors arranged at the corresponding joint portions of the exoskeleton/>And/>And calculating a limb movement value M theta.

When the target limb detection values are respectively muscle strength signals, the exoskeleton is used for controlling the body positionAnd/> />And/> />The limb movement value MF is calculated.

The embodiment can calculate the limb action value by using the limb detection values of the same type acquired by the limb sensors of the same type, so that the limb action value can be obtained more reliably and accurately, and the target limb health grade can be determined more reliably and accurately later.

As previously described, the standard joint trajectory includes reference detection values corresponding to respective sampling moments in the gait cycle, and the reference detection values may be joint angles or muscle force signals.

In some embodiments, calculating the limb motion value corresponding to the target limb detection value from the target limb detection value and the standard joint trajectory corresponding to the target limb detection value comprises the steps of:

s4221: and calculating the difference value between the target limb detection value and the reference detection value at the same sampling time.

S4222: and integrating the plurality of differences according to the gait cycle and a preset sampling cycle to obtain each type of limb action value, wherein the sampling cycle is the time interval between two adjacent sampling moments.

In S4221, for example, as described above, the target limb detection value is the joint angle, and the 1 st IMU sensor samples the following joint angles together in one gait cycleThe 1 st I MU sensor corresponds to the standard joint track as the standard joint track/>Exoskeleton follows the formula: /(I)Calculate the difference, wherein,/>For the difference of the 1 st I MU sensor at the j sampling moment,/>For the joint angle sampled by the 1 st IMU sensor at the j-th sampling moment,/>The joint angle of the standard joint track at the j-th sampling moment is obtained.

In S4222, for example, for the gait cycle [ T ₁,t₂,t₃,..,t_j...,t_m ], the sampling period t=t ₁-t₂＝t₃-t₂, and the exoskeleton can integrate a plurality of differences according to the gait cycle and the sampling period, so as to be able to perform each type of limb motion value.

According to the embodiment, the difference value between the real-time sampled target limb detection value and the reference detection value of the standard joint track is accumulated, so that the deviation of the limb of the user relative to the limb under the normal condition can be reliably obtained, and the target limb health grade can be determined.

In some embodiments, each type of limb detection value may be obtained by sampling at least two limb sensors disposed at different joint portions of the exoskeleton, and integrating the plurality of differences according to the gait cycle and a preset sampling period to obtain each type of limb motion value includes the steps of:

s44: the number of limb sensors belonging to the same class is determined.

S45: and integrating the difference values of all the limb sensors according to the gait cycle, the sampling cycle and the number of the limb sensors to obtain each type of limb motion values.

In S44, for example, the first to third IMU sensors all belong to the same type of sensor, and the number is 3. Similarly, the first muscle force sensor and the sixth muscle force sensor all belong to the same type of sensor, and the number is 6.

In S45, in some embodiments, when the limb detection value is the joint angle, the exoskeleton calculates the limb movement value according to equation one as follows:

wherein t=kt, k belongs to [0,1,2,3, … …, a/T-1], a is a gait cycle, M is the number of IMU sensors, For the difference value of the j sampling time of the i-th IMU sensor,/>For the joint angle of the ith I MU sensor at the jth sampling moment,/>The joint angle of the standard joint track at the j-th sampling moment is obtained.

Discretizing the first equation gives the third equation, as follows:

In some embodiments, when the limb detection value is a muscle force signal, the exoskeleton calculates the limb movement value according to equation four as follows:

wherein t=kt, k belongs to [0,1,2,3, … …, a/T-1], a is a gait cycle, N is the number of muscular sensors, For the difference of the (q) th sampling moment of the (i) th muscular force sensor,/>For the difference of the (q) th sampling moment of the (i) th muscular force sensor,/>For the muscle force signal of the ith muscle force sensor at the qth sampling moment,/>The muscle force signal of the standard joint track at the q-th sampling moment is obtained.

Discretizing the equation four can obtain the equation six as follows:

As can be seen from the above description, the present embodiment can be used to calculate each type of limb motion values by fusing the limb detection values sampled by at least two limb sensors disposed at different joint positions of the exoskeleton, which belong to the same type, so that the target limb health level of the limb can be reliably evaluated from the dimensions of the multiple joint positions.

In some embodiments, integrating the differences of all limb sensors to obtain each type of limb motion value according to the gait cycle, the sampling cycle and the number of limb sensors comprises the steps of:

s451: and obtaining weight factors corresponding to the difference values, wherein limb sensors at different joint positions correspond to different weight factors.

S452: and weighting the difference value according to the difference value and a weight factor corresponding to the difference value to obtain a weighted difference value.

S453: and integrating the weighted differences of all the limb sensors according to the gait cycle and the sampling cycle to obtain limb motion values of each type.

In S451, for example, the first IMU sensor disposed on the hip joint corresponds to the weight factor α ₁, the second IMU sensor disposed on the knee joint corresponds to the weight factor α ₂, and the third IMU sensor disposed on the ankle joint corresponds to the weight factor α ₃. Because the contributions of the limb detection values sampled by the limb sensors arranged at different joint positions to the limb action values are different, the exoskeleton can give corresponding weight factors according to the difference values corresponding to the limb detection values sampled by the different limb sensors, and the obtained limb action values can comprehensively reflect the contributions of the limb detection values sampled by the limb sensors at different joint positions, so that the limb action values can be obtained more reliably to evaluate the health degree of the limb reliably.

In S452, in some embodiments, when the limb-detection value is joint angle, the exoskeleton calculates a weighted difference according to equation seven, as follows:

Wherein, For the weighted difference of the ith IMU sensor at the jth time, α _i is the weighting factor of the ith IMU sensor.

In some embodiments, when the limb detection value is a muscle force signal, the exoskeleton calculates a weighted difference according to equation eight as follows:

Wherein, For the weighted difference of the ith force sensor at the q-th moment, β _i is the weight factor of the ith force sensor.

In S453, in some embodiments, when the limb detection value is the joint angle, the exoskeleton calculates the limb movement value according to equation nine as follows:

In some embodiments, when the limb detection value is a muscle force signal, the exoskeleton calculates the limb movement value according to equation ten, as follows:

In some embodiments, the health rating table comprises at least two limb health ratings, the limb health ratings being collectively constrained by at least two categories of rehabilitation assessment conditions, each category of rehabilitation assessment conditions corresponding to each category of limb movement values. When the limb detection values include joint angle and muscle strength signals, please refer to table 1:

TABLE 1

As can be seen from table 1, the muscle strength signal corresponds to the first type of rehabilitation evaluation condition, the joint angle corresponds to the second type of rehabilitation evaluation condition, and each limb health grade is constrained by the first type of rehabilitation evaluation condition and the second type of rehabilitation evaluation condition.

In some embodiments, the limb detection data includes at least two classes of limb detection values within the same gait cycle, each class of limb detection values being sampled by at least two limb sensors disposed at different joint locations of the exoskeleton, the exoskeleton-based limb health determination method further comprising the steps of:

S46: a health zone corresponding to each limb sensor is determined, the health zone including a standard joint trajectory corresponding to the target limb detection value.

S47: according to the plurality of health bands, rehabilitation evaluation conditions corresponding to each type of limb action values are generated.

In S46, for example, a first IMU sensor disposed on the hip joint corresponds to a first health belt, a second IMU sensor disposed on the knee joint corresponds to a second health belt, and a third IMU sensor disposed on the ankle joint corresponds to a third health belt. Similarly, the first muscle force sensor arranged on the hip joint also corresponds to the health belt of the hip joint, and the 6 muscle force sensors correspond to the 6 health belts, which are not described herein.

Referring to fig. 11, fig. 11 shows an ith health belt for an ith IMU sensor. As shown in fig. 11, the horizontal axis is a gait cycle of 100% and the vertical axis is a joint angle, wherein the dashed line 111 is the standard joint trajectory of the ith IMU sensorLine 112 is defined by the standard joint trajectory/>The resulting trace average/>A corresponding straight line. The solid line 113 and the solid line 114 are solid lines formed centering on the broken line 111, and the bandwidth ± λ _i, respectively, wherein:

Where β _i is the normal float range of the joint angle, which can be determined by the designer based on engineering experience. Defined as the joint angle at which movement begins, gamma _i is negative. /(I)Defined as the joint angle when fully relaxed.

Referring to fig. 12, fig. 12 shows an ith health belt for an ith muscular force sensor. As shown in fig. 12, the horizontal axis represents a gait cycle of 100% and the vertical axis represents a muscle force signal, wherein the broken line 121 represents a standard joint trajectory of the ith muscle force sensorLine 122 is defined by the standard joint trajectory/>The resulting trace average/>A corresponding straight line. The solid line 123 and the solid line 124 are solid lines formed centering on the dashed line 121, and the bandwidth ±δ _i is the solid line formed, respectively, in which:

/>

b₄＝0

wherein phi _i is the normal floating range of the muscle force signal, which can be determined by the designer according to engineering experience. Defined as the muscle force signal at the beginning of exercise. /(I)Defined as the muscle force signal at full relaxation.

In S47, each health belt includes a plurality of reference evaluation values, each of which is fixed at the evaluation position of the health belt, and adjacent two reference evaluation values on the same health belt may constitute a rehabilitation evaluation condition, such asAnd/>Are all used as reference evaluation values,/>Evaluation position and/>, of health zone at 1 st I MU sensorThe evaluation locations of the health zones at the 2 nd IMU sensor are the same, and similarly available,/>Evaluation position and/>, of health zone at 1 st I MU sensorThe evaluation position of the health belt at the 2 nd IMU sensor is also the same, wherein the reference evaluation value/>And reference evaluation value/>Can form a rehabilitation evaluation condition, a reference evaluation value/>And reference evaluation value/>A rehabilitation evaluation condition can be constructed.

In some embodiments, generating rehabilitation evaluation conditions corresponding to each type of limb movement value from a plurality of health bands comprises the steps of:

s471: and determining a reference evaluation value of each evaluation position in each similar health belt.

S472: the respective reference evaluation values belonging to the same evaluation position are aggregated.

S473: the reference evaluation value of the evaluation position is updated based on the respective reference evaluation values belonging to the same evaluation position.

In S471, for example, the health belt corresponding to the first IMU sensor includes the reference evaluation values of the following evaluation positions: The health zone corresponding to the second I MU sensor comprises the reference evaluation values of the following evaluation positions: the health zone corresponding to the third I MU sensor comprises the reference evaluation values of the following evaluation positions: Wherein, reference evaluation value/> />Reference evaluation values belonging to the same evaluation position, reference evaluation value/>/>Reference evaluation values belonging to the same evaluation position, reference evaluation value/>/>Reference evaluation values belonging to the same evaluation position, reference evaluation value/> />All belonging to the same evaluation position.

In S472, for example, the reference evaluation values of the exoskeleton aggregate belonging to the same evaluation position are respectively recorded as

In S473, in some embodiments, updating the reference evaluation value of the evaluation position according to the respective reference evaluation values belonging to the same evaluation position includes the steps of: a reference average value of each reference evaluation value belonging to the same evaluation position is obtained, and the reference average value is used as a final reference evaluation value of the evaluation position, for example, as follows:

/>

in some embodiments, determining the target limb health level according to the limb motion value and the preset health level table comprises the steps of:

s431: and obtaining multiple types of limb action values.

S432: and determining the target limb health grade according to the multiple types of limb action values, wherein the multiple types of limb action values simultaneously meet at least two types of rehabilitation evaluation conditions of the target limb health grade.

In S431, for example, acquiring the multiple types of limb movement values includes acquiring a limb movement value mθ and a limb movement value MF.

In S432, the exoskeleton traverses the health grade table to traverse the limb health grade for which the multiple types of limb motion values all satisfy at least two types of rehabilitation evaluation conditions simultaneously as the target limb health grade. For example, the exoskeleton obtains a limb motion value mθ1 and a limb motion value MF1, and the limb motion value mθ1∈ [ a ₀,a₁ ] is due to the limb motion value mθ1∈ [ a ₀,a₁ ]And the first type of rehabilitation evaluation condition of the near health level satisfies the limb motion value MF1 and the second type of rehabilitation evaluation condition satisfies the limb motion value mθ1, the near health level may be selected as the target limb health level.

According to the method and the device, the multiple types of rehabilitation evaluation conditions are fused to be compatible in evaluating multiple types of limb action values, namely, the multiple types of rehabilitation evaluation conditions are set to serve as constraint conditions to select the target limb health grade, so that the limb health grade of the limb of the user can be reliably determined.

It should be noted that, in the foregoing embodiments, there is not necessarily a certain sequence between the steps, and those skilled in the art will understand that, according to the description of the embodiments of the present invention, the steps may be performed in different orders in different embodiments, that is, may be performed in parallel, may be performed interchangeably, or the like.

Referring to fig. 13, fig. 13 is a schematic circuit diagram of an electronic device according to an embodiment of the present invention, where the electronic device may be an exoskeleton or any other suitable type of device or electronic product. As shown in fig. 13, the electronic device 130 includes one or more processors 131 and a memory 132. In fig. 13, a processor 131 is taken as an example.

The processor 131 and the memory 132 may be connected by a bus or otherwise, for example in fig. 13. The memory 132 is used as a non-volatile computer readable storage medium for storing non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions/modules corresponding to the exoskeleton-based trajectory planning method in the embodiments of the present invention. The processor 131 performs the functions of the exoskeleton-based trajectory planning method provided by the above method embodiments by running non-volatile software programs, instructions and modules stored in the memory 132. The memory 132 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 132 may optionally include memory remotely located relative to processor 131, such remote memory being connectable to processor 131 through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof. The program instructions/modules are stored in the memory 132, which when executed by the one or more processors 131, perform the exoskeleton-based track planning method of any of the method embodiments described above.

Embodiments of the present invention also provide a storage medium storing computer-executable instructions for execution by one or more processors, such as the one processor 131 of fig. 13, to cause the one or more processors to perform the exoskeleton-based track planning method of any of the method embodiments described above.

Embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by an electronic device, cause the electronic device to perform any of the exoskeleton-based track planning methods described in any of the above.

The above-described embodiments of the apparatus or device are merely illustrative, in which the unit modules illustrated as separate components may or may not be physically separate, and the components shown as unit modules may or may not be physical units, may be located in one place, or may be distributed over multiple network module units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. From the above description of embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus a general purpose hardware platform, or may be implemented by hardware. Based on such understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the related art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; the technical features of the above embodiments or in the different embodiments may also be combined within the idea of the invention, the steps may be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims (10)

1. An exoskeleton-based trajectory planning method, comprising:

Obtaining a target limb health grade;

Determining a target track model corresponding to the target limb health grade;

and planning the joint track of the exoskeleton according to the target track model.

2. The method of claim 1, wherein the exoskeleton comprises a plurality of joint locations, and wherein when the target trajectory model is a timely assistance model, planning the joint trajectory of the exoskeleton according to the target trajectory model comprises:

acquiring gait data, and determining a gait phase according to the gait data;

Determining a power type of a target joint part according to the gait phase, wherein the target joint part is one of a plurality of joint parts;

And planning the joint track of the target joint part according to the power type of the target joint part.

3. The method of claim 2, wherein the planning the joint trajectory of the target joint region according to the power type of the target joint region comprises:

When the power type of the target joint part is positive, controlling the target joint part to work according to a preset standard joint track;

And when the power type of the target joint part is a negative power type, controlling the target joint part to stop working.

4. A method according to claim 3, wherein the exoskeleton is configured with a trajectory database storing a plurality of standard joint trajectories, and wherein controlling the target joint location to operate according to a preset standard joint trajectory comprises:

Accessing the track database to select a standard joint track of the target joint part as a target standard joint track;

And controlling the target joint part to work according to the target standard joint track.

5. The method of claim 4, wherein each of the standard joint trajectories corresponds to a joint location of a respective user characteristic, the accessing the trajectory database to select the standard joint trajectory of the target joint location as the target standard joint trajectory comprising:

Acquiring user information;

judging whether the track database has user characteristics matched with the user information or not;

if the user information exists, selecting a standard joint track corresponding to the user information as a target standard joint track;

If the user information does not exist, two user features adjacent to the user information are determined to serve as reference user features, and a target standard joint track is generated through fitting according to the standard joint tracks of the two reference user features.

6. The method of claim 1, wherein the exoskeleton comprises a plurality of joint locations, and wherein when the target trajectory model is a passive control model, the planning the joint trajectory of the exoskeleton according to the target trajectory model comprises:

Obtaining an adjustment coefficient and a standard joint track of a target joint part, wherein the target joint part is one of a plurality of joint parts;

According to the adjustment coefficient and the standard joint track of the target joint part, adjusting the standard joint track of the target joint part;

and controlling the target joint part to work according to the adjusted standard joint track.

7. The method of claim 1, wherein the exoskeleton comprises a plurality of joint locations, and wherein when the target trajectory model is an adaptive model, the planning the joint trajectory of the exoskeleton according to the target trajectory model comprises:

acquiring expected interaction force and man-machine interaction force of a target joint part, wherein the target joint part is one of a plurality of joint parts;

inputting the expected interaction force and the man-machine interaction force into a preset admittance control model so that the admittance control model outputs an angle adjustment amount;

Adjusting the standard joint track of the target joint part according to the angle adjustment quantity to obtain a self-adaptive joint track;

and controlling the target joint part to work according to the self-adaptive joint track.

8. The method of any one of claims 1 to 7, wherein each trajectory model is configured with a model tag, the model tag corresponding to the limb health level, the determining a target trajectory model corresponding to the target limb health level comprising:

Selecting a model label corresponding to the target limb health grade as a target model label;

And selecting a track model with the model label as the target track model.

9. A storage medium having stored thereon computer executable instructions for causing an electronic device to perform the exoskeleton-based trajectory planning method of any one of claims 1 to 8.

10. An exoskeleton, comprising:

at least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the exoskeleton-based trajectory planning method of any one of claims 1 to 8.