Exoskeleton robot control method and system

Abstract

The application provides an exoskeleton robot control method and system, the method comprises the following steps: acquiring motion data of a target and a pre-trained target model; performing trajectory prediction based on the target model and the motion data to obtain a predicted ankle joint motion trajectory; performing trajectory processing and filtering on the predicted ankle joint motion trajectory to obtain a processed motion trajectory; performing first motion solving of the exoskeleton robot based on the processed motion trajectory to output a first control parameter; performing second motion solving of the exoskeleton robot based on the processed motion trajectory to output a second control parameter; and sending the first control parameter and the second control parameter to the exoskeleton robot to complete control of the exoskeleton robot. The application optimizes the response performance of the motor, ensures accurate trajectory tracking, and improves the assistance effect of the exoskeleton robot.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of exoskeleton robot control, and particularly relates to an exoskeleton robot control method and system. BACKGROUND

[0002] An exoskeleton robot generally consists of an electric power system and a sensor for measuring motor torque, which can assist patients with limb movement disorders in rehabilitation training and is mainly applied in the medical rehabilitation field. As a new rehabilitation auxiliary tool, the exoskeleton technology shows significant therapeutic potential. Unlike traditional physical therapy which relies on slow recovery through manual training, the exoskeleton can provide customized motion assistance for patients by simulating human gait and motion mechanism combined with intelligent control, thereby significantly improving the rehabilitation effect. The exoskeleton can not only help patients recover walking ability and enhance muscle strength, but also adjust the assistance force in real time according to the patient's needs to improve the treatment efficiency.

[0003] For the ankle exoskeleton robot in the prior art, there is a great limitation, that is, it cannot recognize the motion intention according to the motion state of the motion person, which leads to insufficient real-time response capability and poor human-computer interaction experience. SUMMARY

[0004] To solve the above technical problems, the present application provides an exoskeleton robot control method and system to solve the technical problems in the prior art.

[0005] In a first aspect, the present application provides the following technical solution, an exoskeleton robot control method, comprising:

[0006] Obtaining motion data of a target and a pre-trained target model, performing trajectory prediction based on the target model and the motion data to obtain a predicted ankle joint motion trajectory;

[0007] Performing trajectory processing and filtering on the predicted ankle joint motion trajectory to obtain a processed motion trajectory;

[0008] Performing first motion solving of the exoskeleton robot based on the processed motion trajectory to output a first control parameter;

[0009] Performing second motion solving of the exoskeleton robot based on the processed motion trajectory to output a second control parameter;

[0010] Sending the first control parameter and the second control parameter to the exoskeleton robot to complete the control of the exoskeleton robot.

[0011] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention first acquires the target's motion data and a pre-trained target model, and then performs trajectory prediction based on the target model and motion data to obtain a predicted ankle joint motion trajectory; then, it processes and filters the predicted ankle joint motion trajectory to obtain a processed motion trajectory; then, it solves the first motion of the exoskeleton robot based on the processed motion trajectory to output first control parameters; then, it solves the second motion of the exoskeleton robot based on the processed motion trajectory to output second control parameters; finally, it sends the first and second control parameters to the exoskeleton robot to complete the control of the exoskeleton robot. This invention can achieve the purpose of recognizing the target's motion intention by predicting the target's ankle joint motion trajectory, and then processes the trajectory to improve its smoothness and accuracy. Finally, it controls the target by outputting the first and second control parameters, thereby optimizing the motor's response performance, ensuring accurate trajectory tracking, and improving the assistive effect of the exoskeleton robot.

[0012] Preferably, the motion data includes the angular velocity, speed, and angle of the wearable exoskeleton robot in different motion modes.

[0013] Preferably, the step of predicting the ankle joint movement trajectory by using the target model and based on the motion data includes:

[0014] The motion data is converted into vector form to obtain motion vectors. The motion vectors are then input into the encoder in the target model. The encoder maps the motion vectors to a high-dimensional space to obtain query vectors, key vectors, and value vectors.

[0015] Perform a Fourier transform on the query vector, the key vector, and the value vector to obtain the transformed query vector. Transform the key vector and the transformed value vector ;

[0016] Sparse self-attention is calculated based on the transformed query vector, the transformed key vector, and the transformed value vector. :

[0017] ;

[0018] In the formula, This is the inverse Fourier transform. For zero-filling operation, To transform the query vector Transform the key vector Dimensions To transform the query vector Sparse matrices of the same dimension;

[0019] encoding the motion vector by the encoder to obtain extracted features, splicing the extracted features to obtain spliced features;

[0020] inputting the spliced features and the sparse self-attention input into a decoder of the target model and performing feature decoding by using a multi-head self-attention mechanism to obtain decoded features;

[0021] inputting the decoded features into a fully connected layer in the target model and mapping the decoded features to a prediction space by the fully connected layer to obtain a predicted ankle motion trajectory.

[0022] Preferably, the step of performing trajectory processing and filtering on the predicted ankle motion trajectory to obtain a processed motion trajectory comprises:

[0023] extracting a predicted ankle motion angle in the predicted ankle motion trajectory, and identifying an angle error between the predicted ankle motion angle and a real ankle motion angle after wearing the exoskeleton robot;

[0024] performing trajectory correction on the predicted ankle motion trajectory based on the angle error and by using a linear translation mapping to obtain a first adjusted trajectory;

[0025] decomposing the first adjusted trajectory into a plurality of sub-trajectories, obtaining historical motion trajectories under the same motion mode, calculating a similarity between the sub-trajectories and the historical motion trajectories, splicing and performing trajectory continuous adjustment on the sub-trajectories with a similarity greater than a similarity threshold to obtain a second adjusted trajectory;

[0026] performing filtering processing on the second adjusted trajectory by using a preset filter to obtain a processed motion trajectory.

[0027] Preferably, the step of performing first motion solving of the exoskeleton robot based on the processed motion trajectory to output a first control parameter comprises:

[0028] calculating initial lengths of two leg active stretchers in the exoskeleton robot:

[0029] ; ;

[0030] wherein, are initial lengths of the first leg active stretcher and the second leg active stretcher, , are coordinates of connection points of the first leg active stretcher and the second leg active stretcher and the lower leg connecting component, , are coordinates of the connection points of the first and second leg active stretchers and the foot connecting component, respectively;

[0031] calculating coordinates of dynamic connection points of the first and second leg active stretchers and the lower leg connecting component:

[0032]

[0033]

[0034] wherein, is the rotation center, is the ankle joint motion angle in the processing motion trajectory, are coordinates of the dynamic connection points of the first and second leg active stretchers and the lower leg connecting component, respectively;

[0035] calculating length change amounts of the first and second leg active stretchers based on the coordinates of the dynamic connection points of the first and second leg active stretchers and the lower leg connecting component and initial lengths of the two leg active stretchers:

[0036]

[0037] wherein, are length change amounts of the first and second leg active stretchers, respectively;

[0038] calculating rotation turns of the two side driving motors in the exoskeleton robot based on the length change amounts of the two leg active stretchers to obtain the first control parameter:

[0039]

[0040] wherein, , are rotation turns and lead of the one side driving motor, respectively, , are rotation turns and lead of the other side driving motor, respectively.

[0041] Preferably, the step of solving the second motion of the exoskeleton robot based on the processing motion trajectory to output the second control parameter comprises:

[0042] extracting predicted motion parameters in the processing motion trajectory, constructing a coefficient matrix based on the predicted motion parameters constructing a balance equation based on the coefficient matrix:

[0043]

[0044] wherein, ​​​​​​​The actual internal stress of the leg active stretch element;

[0045] The constraint condition is determined:

[0046]

[0047] In the formula, is an external force matrix, is a relaxation amount, is the maximum internal stress that the leg active stretch element can withstand;

[0048] The balance equation is modified into a force position equation through the constraint condition, and the force position equation is solved by optimization to output an internal stress trajectory of the leg active stretch element, so as to obtain a second control parameter.

[0049] In a second aspect, the present application provides the following technical solution, a kind of exoskeleton robot control system, the system includes:

[0050] Prediction submodule, for obtaining the motion data of target and pre-trained target model, trajectory prediction is carried out by the target model and based on the motion data, to obtain predicted ankle motion trajectory;

[0051] Processing submodule, for trajectory processing and filtering to the predicted ankle motion trajectory, to obtain processing motion trajectory;

[0052] First solving submodule, for the first motion solving of exoskeleton robot based on the processing motion trajectory, to output first control parameter;

[0053] Second solving submodule, for the second motion solving of exoskeleton robot based on the processing motion trajectory, to output second control parameter;

[0054] Control submodule, for the first control parameter and the second control parameter are sent to exoskeleton robot, to complete the control of the exoskeleton robot.

[0055] Preferably, the motion data includes the angular velocity, the motion speed and the angle of the exoskeleton robot in different motion modes.

[0056] In a third aspect, the present application provides the following technical solution, a kind of computer, including memory, processor and computer program stored on the memory and can be run on the processor, when the processor executes the computer program, realize the exoskeleton robot control method as described above.

[0057] ​​In a fourth aspect, the present application provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the exoskeleton robot control method described above. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort based on these drawings.

[0059] Figure 1 A flow chart of the exoskeleton robot control method provided by the first embodiment of the present application;

[0060] Figure 2 A structural block diagram of the exoskeleton robot control system provided by the second embodiment of the present application;

[0061] Figure 3 A hardware structure schematic diagram of the computer provided by another embodiment of the present application.

[0062] The embodiments of the present application will be further described below with reference to the drawings. DETAILED DESCRIPTION

[0063] The embodiments of the present application will be described in detail below, and examples of the embodiments are shown in the drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are intended to explain the embodiments of the present application, and cannot be understood as a limitation of the present application.

[0064] Embodiment one

[0065] In the first embodiment of the present application, as shown in Figure 1 An exoskeleton robot control method comprises:

[0066] S1, obtaining motion data of a target and a pre-trained target model, performing trajectory prediction based on the motion data through the target model to obtain a predicted ankle joint motion trajectory;

[0067] Specifically, the motion data in this embodiment includes the angular velocity, the motion speed and the angle of the exoskeleton robot in different motion modes, and the angular velocity, the motion speed and the angle in this embodiment refer to the angular velocity, the motion speed and the angle of the ankle joint. Meanwhile, the target model is based on the Ankleformer model, and the original Ankleformer model is improved by setting the decoder, the encoder and the multi-head self-attention mechanism to obtain the target model.

[0068] The step S1 comprises:

[0069] S11, converting the motion data into a vector form to obtain a motion vector, inputting the motion vector into an encoder in the target model, and mapping the motion vector to a high-dimensional space by the encoder to obtain a query vector, a key vector and a value vector;

[0070] Specifically, the encoder is a structure repeatedly stacked by a plurality of modules, each module comprising a self-attention layer, a feedforward network, a residual connection and a layer normalization operation. For the encoder, the purpose is to fully capture the individualized gait features of different subjects in multiple motion modes. Specifically, the original multi-dimensional time series signal is first processed by an embedding layer to map it from the original input space to a high-dimensional feature space, so that the subsequent model can more effectively learn the complex time sequence structure and nonlinear relationship. Then, through the layer-by-layer progressive feature extraction process, the dynamic change features in different time scales are gradually captured, so as to realize comprehensive modeling of local and global gait patterns.

[0071] S12, Fourier transforming the query vector, the key vector and the value vector to obtain a transformed query vector , a transformed key vector and a transformed value vector .

[0072] S13, calculating sparse self-attention based on the transformed query vector, the transformed key vector and the transformed value vector .

[0073] ;

[0074] In the formula, is an inverse Fourier transform, is a zero padding operation, is the dimension of the transformed query vector , the transformed key vector , and is a sparse matrix with the same dimension as the transformed query vector .

[0075] S14, encode and feature extract the motion vector by the encoder to obtain extracted features, and splice the extracted features to obtain spliced features;

[0076] Specifically, in the process of encoding and feature extraction, the sparse self-attention calculated above can be used for attention weight distribution, which can significantly improve the modeling ability of the model for long-term dependencies and enhance its perception of the global structure of time series, thereby better adapting to the cross-subject and cross-motion mode gait prediction task. The spliced features obtained here are high-level gait features.

[0077] S15, input the spliced features and the sparse self-attention into the decoder of the target model and use the multi-head self-attention mechanism to decode the features to obtain decoded features;

[0078] Specifically, the decoder here has a similar structure to the encoder, which is a multi-layer stacked structure, and uses a multi-layer stacked method to gradually generate the prediction results of each time step. However, the multi-head self-attention mechanism is used in the decoder, which can utilize both frequency domain information and time domain information when generating target output. The sparse self-attention enables the decoder to capture global trends, periodic changes and local details in the input sequence, while the multi-head self-attention mechanism further enhances the modeling ability of different features. The stacked frequency domain and time domain attention mechanisms enable the decoder to efficiently extract key features from the context information of the generated partial sequence and the input sequence, thereby generating more accurate and consistent output sequences.

[0079] S16, input the decoded features into the fully connected layer in the target model and map the decoded features to the prediction space through the fully connected layer to obtain a predicted ankle motion trajectory;

[0080] Specifically, the output of the decoder is mapped to the final prediction space through the fully connected layer to generate a continuous ankle angle trajectory sequence, i.e., a predicted ankle motion trajectory.

[0081] S2, trajectory processing and filtering of the predicted ankle motion trajectory to obtain a processed motion trajectory;

[0082] The step S2 includes:

[0083] S21, extract the predicted ankle motion angle in the predicted ankle motion trajectory, and identify the angle error between the predicted ankle motion angle and the real ankle motion angle after wearing the exoskeleton robot;

[0084] Specifically, in actual situations, there is a certain angle deviation between the predicted ankle joint angle and the actual ankle joint angle after wearing the exoskeleton, resulting in that the predicted ankle joint angle range is between-5° and 25°, while the actual ankle joint movement angle range is usually between-20° and 10°, and there is a certain deviation between the two. Therefore, the method of linear translation mapping is adopted to adjust the predicted trajectory as a whole, so that the range is consistent with the actual ankle joint movement angle.

[0085] S22, based on the angle error and using the linear translation mapping method, the predicted ankle joint movement trajectory is corrected to obtain a first adjusted trajectory.

[0086] S23, the first adjusted trajectory is decomposed into a plurality of sub-trajectories, the historical movement trajectory under the same movement mode is obtained, the similarity between the sub-trajectory and the historical movement trajectory is calculated, and the sub-trajectory with a similarity greater than a similarity threshold is spliced and trajectory continuous adjustment is performed to obtain a second adjusted trajectory.

[0087] Specifically, the similarity here is the cosine similarity. By calculating the similarity with the historical movement trajectory, the selected second adjusted trajectory is more consistent with the trajectory synchronization in the real situation, improves the consistency of the exoskeleton and human gait, effectively avoids the phenomenon of gait disharmony or unnaturalness caused by delay, further improves the precision and fluency of exoskeleton control, and at the same time, the trajectory continuous adjustment here is to avoid trajectory interruption and ensure trajectory continuity.

[0088] S24, a preset filter is used to filter the second adjusted trajectory to obtain a processed movement trajectory.

[0089] Specifically, the preset filter here is a 6th order Butterworth low-pass filter. Through the filter, high-frequency noise components can be effectively removed, while the main dynamic characteristics of the movement trajectory are retained, ensuring the smoothness and accuracy of the exoskeleton control signal.

[0090] S3, based on the processed movement trajectory, a first movement of the exoskeleton robot is solved to output a first control parameter.

[0091] Wherein, the step S3 includes:

[0092] S31, the initial length of the two leg active stretchers in the exoskeleton robot is calculated:

[0093] ; ;

[0094] In the formula, are the initial lengths of the first and second leg active stretchers, , respectively are coordinates of the connection points of the first and second leg active stretchable members and the shank connecting member, 、 respectively are coordinates of the connection points of the first and second leg active stretchable members and the foot connecting member;

[0095] Specifically, for the ankle exoskeleton robot, it can be abstracted as a bar model, and the stretching of the leg active stretchable member realizes the motion assistance and rehabilitation process, and the ankle exoskeleton robot here is a bionic tension whole ankle exoskeleton, and the leg active stretchable member is an active tension cable.

[0096] S32, calculate the dynamic connection point coordinates of the first and second leg active stretchable members and the shank connecting member:

[0097] ;

[0098] ;

[0099] In the formula, is the rotation center, is the ankle motion angle in the processing motion trajectory, respectively are the dynamic connection point coordinates of the first and second leg active stretchable members and the shank connecting member.

[0100] S33, based on the dynamic connection point coordinates of the first and second leg active stretchable members and the shank connecting member and the initial length of the two leg active stretchable members, calculate the length change of the two leg active stretchable members:

[0101] ; ;

[0102] In the formula, respectively are the length changes of the first and second leg active stretchable members.

[0103] S34, based on the length changes of the two leg active stretchable members, calculate the rotation turns of the two side driving motors in the exoskeleton robot, to obtain the first control parameter:

[0104] ; ;

[0105] In the formula, 、 respectively are the rotation turns and lead of one side driving motor, 、 respectively are the rotation turns and lead of the other side driving motor;

[0106] Specifically, after the number of rotation circles is obtained, the number of rotation circles can be sent to a motor drive board of the exoskeleton robot.

[0107] S4, performing second motion solving of the exoskeleton robot based on the processed motion trajectory, to output a second control parameter;

[0108] The step S4 comprises:

[0109] S41, extracting a predicted motion parameter in the processed motion trajectory, and constructing a coefficient matrix based on the predicted motion parameter constructing a balance equation based on the coefficient matrix:

[0110]

[0111] In the formula, is an actual internal stress of the leg active telescopic member;

[0112] Specifically, the coefficient matrix is a matrix of cable force density, which can be constructed by the predicted motion parameter. In the actual motion process, the ankle exoskeleton will be subjected to non-zero external force from human motion and environment. In order to truly reflect this situation, the data measured by the three-axis force sensor is used as the external force input in this step, and then the subsequent external force matrix is obtained.

[0113] S42, determining a constraint condition:

[0114]

[0115] In the formula, is an external force matrix, is a slack variable, is a maximum internal stress that the leg active telescopic member can withstand;

[0116] In the actual motion process, there is a certain deviation between the quasi-static state and the theoretical statics model. In order to prevent the optimization problem from being unsolvable or singular due to the rigid equality constraint, the slack variable is introduced into the constraint condition, so that the constraint condition has better adaptability. In order to avoid the leg active telescopic member from being broken, the maximum internal stress that the leg active telescopic member can withstand is set to constrain it.

[0117] S43, modifying the balance equation to a force-position equation by the constraint condition, and performing optimization solving on the force-position equation, to output an internal stress trajectory of the leg active telescopic member, so as to obtain the second control parameter;

[0118] ​​​Specifically, after the comprehensive constraint condition, the balance equation is modified as the force potential equation, and after solving the force potential equation, the target stress distribution is obtained, and the target internal force distribution is converted into the target force trajectory of the two driving ropes.

[0119] S5, the first control parameter and the second control parameter are sent to the exoskeleton robot to complete the control of the exoskeleton robot.

[0120] Specifically, after obtaining the first control parameter and the second control parameter, they are respectively input into the respective PID controller to generate the corresponding signal, and then the motor is controlled by the signal to move, so that the exoskeleton robot performs corresponding motion control to assist the ankle joint of the target to realize the processing motion trajectory.

[0121] The exoskeleton robot control method provided by the embodiment one of the application first acquires the motion data of the target and the pre-trained target model, performs trajectory prediction based on the target model and the motion data to obtain a predicted ankle joint motion trajectory, then performs trajectory processing and filtering on the predicted ankle joint motion trajectory to obtain a processed motion trajectory, then performs first motion solving of the exoskeleton robot based on the processed motion trajectory to output a first control parameter, then performs second motion solving of the exoskeleton robot based on the processed motion trajectory to output a second control parameter, and finally sends the first control parameter and the second control parameter to the exoskeleton robot to complete the control of the exoskeleton robot. The application can predict the ankle joint motion trajectory of the target to achieve the purpose of recognizing the motion intention of the target, then perform corresponding processing on the trajectory to improve the smoothness and accuracy of the trajectory, and then control by outputting the first control parameter and the second control parameter, so as to optimize the response performance of the motor, ensure accurate trajectory tracking, and improve the assistance effect of the exoskeleton robot.

[0122] Embodiment two

[0123] As shown in Figure 2 The embodiment two of the application provides an exoskeleton robot control system, which comprises:

[0124] A prediction submodule 1 is configured to acquire motion data of a target and a pre-trained target model, and perform trajectory prediction based on the target model and the motion data to obtain a predicted ankle joint motion trajectory.

[0125] A processing submodule 2 is configured to perform trajectory processing and filtering on the predicted ankle joint motion trajectory to obtain a processed motion trajectory.

[0126] A first solving submodule 3 is configured to perform first motion solving of an exoskeleton robot based on the processed motion trajectory to output a first control parameter.

[0127] a second solving sub-module 4, configured to perform second motion solving of the exoskeleton robot based on the processed motion trajectory, to output a second control parameter;

[0128] a control sub-module 5, configured to send the first control parameter and the second control parameter to the exoskeleton robot, to complete control of the exoskeleton robot.

[0129] The motion data includes angular velocity, motion velocity and angle of the exoskeleton robot in different motion modes.

[0130] The prediction sub-module 1 includes:

[0131] a conversion sub-module, configured to convert the motion data into a vector form to obtain a motion vector, and input the motion vector into an encoder in the target model, to map the motion vector to a high-dimensional space by the encoder, to obtain a query vector, a key vector and a value vector;

[0132] a transformation sub-module, configured to perform Fourier transform on the query vector, the key vector and the value vector, to obtain a transformed query vector , a transformed key vector and a transformed value vector ;

[0133] an attention sub-module, configured to calculate sparse self-attention based on the transformed query vector, the transformed key vector and the transformed value vector:

[0134] ;

[0135] wherein, is inverse Fourier transform, is zero padding operation, is dimension of the transformed query vector , the transformed key vector , and is a sparse matrix with the same dimension as the transformed query vector ;

[0136] a splicing sub-module, configured to encode and extract features of the motion vector by the encoder, to obtain extracted features, and splice the extracted features, to obtain spliced features;

[0137] a decoding sub-module, configured to input the spliced features and the sparse self-attention into a decoder of the target model, and perform feature decoding by using a multi-head self-attention mechanism, to obtain decoded features;

[0138] The mapping submodule is configured to input the decoded feature into a full connection layer in the target model and map the decoded feature to a prediction space through the full connection layer to obtain a predicted ankle motion trajectory.

[0139] The processing submodule 2 comprises:

[0140] The extraction submodule is configured to extract a predicted ankle motion angle in the predicted ankle motion trajectory, and identify an angle error between the predicted ankle motion angle and a real ankle motion angle after wearing the exoskeleton robot;

[0141] The correction submodule is configured to correct the predicted ankle motion trajectory based on the angle error and in a manner of linear translation mapping to obtain a first adjusted trajectory;

[0142] The adjustment submodule is configured to decompose the first adjusted trajectory into a plurality of sub-trajectories, acquire a historical motion trajectory under a same motion mode, calculate a similarity between the sub-trajectories and the historical motion trajectory, and splice and adjust the continuity of the sub-trajectories with a similarity greater than a similarity threshold to obtain a second adjusted trajectory;

[0143] The filtering submodule is configured to filter the second adjusted trajectory by using a preset filter to obtain a processed motion trajectory.

[0144] The first solving submodule 3 comprises:

[0145] The first calculation submodule is configured to calculate initial lengths of two leg active stretchers in the exoskeleton robot:

[0146]

[0147] In the formula, L1 and L2 respectively represent the initial lengths of the first and second leg active stretchers,

[0148] The second calculation submodule is configured to calculate dynamic connection point coordinates of the first and second leg active stretchers and the shank connecting components:

[0149]

[0150]

[0151] In the formula,​​​​​​​​​ is the rotation center, is the ankle joint motion angle in the processing motion trajectory, respectively are the dynamic connection point coordinates of the first and second leg active stretch components and the lower leg connecting component;

[0152] The third calculation submodule is configured to calculate the length change of the two leg active stretch components based on the dynamic connection point coordinates of the first and second leg active stretch components and the lower leg connecting component and the initial lengths of the two leg active stretch components:

[0153] ; ;

[0154] In the formula, respectively are the length changes of the first and second leg active stretch components;

[0155] The fourth calculation submodule is configured to calculate the rotation turns of the two side driving motors in the exoskeleton robot based on the length changes of the two leg active stretch components to obtain the first control parameter:

[0156] ; ;

[0157] In the formula, , respectively are the rotation turns and lead of one side driving motor, , respectively are the rotation turns and lead of the other side driving motor.

[0158] The second solving submodule 4 includes:

[0159] The fifth calculation submodule is configured to extract a predicted motion parameter in the processing motion trajectory, construct a coefficient matrix based on the predicted motion parameter , and construct a balance equation based on the coefficient matrix:

[0160] ;

[0161] In the formula, is the actual internal stress of the leg active stretch component;

[0162] The sixth calculation submodule is configured to determine a constraint condition:

[0163] ; ;

[0164] In the formula, is an external force matrix, is a relaxation amount, The maximum internal stress that the leg active stretch component can bear;

[0165] The seventh calculation sub-module is configured to modify the balance equation into a force potential equation by the constraint condition, and to solve the force potential equation to output an internal stress trajectory of the leg active stretch component, so as to obtain the second control parameter.

[0166] In some embodiments of the present application, the present application provides the following technical solutions, a computer, comprising a memory 102, a processor 101 and a computer program stored in the memory 102 and executable on the processor 101, when the processor 101 executes the computer program, the method for controlling the exoskeleton robot is realized.

[0167] Specifically, the processor 101 can include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present application.

[0168] The memory 102 can include mass storage for data or instructions. By way of example, and not limitation, the memory 102 can include a Hard Disk Drive (HDD), a floppy disk drive, a solid-state drive (SSD), a flash drive, a Compact Disk Drive (CD Drive), a DVD Drive, a tape drive, a USB drive, or any combination of two or more of these. The memory 102 can be removable and / or non-removable (or fixed) as appropriate. The memory 102 can be internal or external as appropriate. In certain embodiments, the memory 102 is a non-volatile memory. In certain embodiments, the memory 102 includes a Read-Only Memory (ROM) and a Random Access Memory (RAM). The ROM can be a mask-programmed ROM, a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), an Electrically Alterable ROM (EAROM), or a FLASH memory, or any combination of two or more of these, as appropriate. The RAM can be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), which can be a Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), an Extended Data Output Dynamic Random-Access Memory (EDODRAM), a Synchronous Dynamic Random-Access Memory (SDRAM), or the like, as appropriate.

[0169] The memory 102 can be used to store or buffer various data files required for processing and / or communication, and possible computer program instructions executed by the processor 101.

[0170] The processor 101 realizes the above-mentioned exoskeleton robot control method by reading and executing the computer program instructions stored in the memory 102.

[0171] In some embodiments, the computer can further include a communication interface 103 and a bus 100. As shown, the processor 101, the memory 102, and the communication interface 103 are connected through the bus 100 and complete communication with each other. Figure 3

[0172] The communication interface 103 is used to realize the communication between the modules, devices, units and / or equipment in the embodiments of the present application. The communication interface 103 can also realize data communication with other components, such as external devices, image / data acquisition devices, databases, external storage, image / data processing workstations, etc.

[0173] ​Bus 100 includes hardware, software, or both, to couple components of computer device to each other and to couple components of computer device to other devices. Bus 100 includes, but is not limited to, at least one of the following: a data bus, an address bus, a control bus, an expansion bus, a local bus, and the like. By way of example and not limitation, bus 100 can include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand (IB) interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or another suitable bus or a combination of two or more of these. Where appropriate, bus 100 can include one or more buses. Although the present embodiments describe and show a particular bus, the present embodiments contemplate any suitable bus or interconnect.

[0174] The computer can acquire the exoskeleton robot control system and execute the exoskeleton robot control method of the present application, so as to realize the exoskeleton robot control.

[0175] In some embodiments of the present application, in combination with the exoskeleton robot control method described above, the present embodiments provide the following technical solutions: a storage medium, the storage medium has a computer program stored thereon, and the computer program is executed by a processor to implement the exoskeleton robot control method described above.

[0176] Those skilled in the art will appreciate that the logic and / or steps represented in the flow diagrams, or otherwise described herein, can be embodied in executable instructions, for example, that can be executed by a general-purpose computer, special-purpose computer, or other computing device to perform the steps illustrated in the flow diagrams, or otherwise described herein. As will be understood, such steps can be performed by a combination of hardware and / or software. As will be appreciated, any reference to a computer process, program, or the like, contains one or more sequences of instructions for implementing logic functions in accordance with the embodiments of this disclosure. Computer implemented steps can also be understood as a computer implemented process or process operating on an electronic device acting on electrical signals representing a computer process. The one or more sequences of instructions can be embodied within a computer readable medium, which is to be interpreted broadly to encompass a non-transitory computer readable medium. A non-transitory computer readable medium includes any medium that is not "transitory", where such terms are used as electrical or other physically transitory signals.

[0177] More specific examples (a non-exhaustive list) of the computer readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and stored in a computer memory.

[0178] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies, known in the art, or combinations thereof, can be used: a discrete logic circuit having logic gates for implementing logic functions upon data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), and / or the like.

[0179] The technical features of the above-described embodiments can be combined in any manner, and in order to make the description concise, not all possible combinations of the technical features in the above-described embodiments are described, however, as long as the combinations of the technical features do not contradict each other, it should be considered that they are within the scope of the present specification.

[0180] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are within the protection scope of the present application. Therefore, the protection scope of the present application patent should be subject to the appended claims.

Claims

1. An exoskeleton robot control method characterized by comprising: The method comprises the following steps: acquiring motion data of a target and a pre-trained target model, performing trajectory prediction on the target model based on the motion data to obtain a predicted ankle joint motion trajectory; performing trajectory processing and filtering on the predicted ankle joint motion trajectory to obtain a processed motion trajectory; performing first motion solving of an exoskeleton robot based on the processed motion trajectory to output first control parameters; performing second motion solving of the exoskeleton robot based on the processed motion trajectory to output second control parameters; sending the first control parameters and the second control parameters to the exoskeleton robot to complete control of the exoskeleton robot; the step of performing first motion solving of the exoskeleton robot based on the processed motion trajectory to output first control parameters comprises the following steps: calculating initial lengths of two leg active stretchers in the exoskeleton robot; ； ； wherein L1, L2 are the initial lengths of the first and second leg active stretch elements, respectively, , P1, P2 are the coordinates of the connection points of the first and second leg active stretch elements to the lower leg connecting element, respectively, , P3, P4 are the coordinates of the connection points of the first and second leg active stretch elements to the foot connecting element, respectively. calculating dynamic connection point coordinates of the first leg active stretcher, the second leg active stretcher and the lower leg connecting component; ； ； In the formula, is the rotation center, is the angle of the ankle joint in the processing of the motion trajectory, are the dynamic connection point coordinates of the first and second leg main telescopic components and the lower leg connecting components, respectively. calculating length change amounts of the two leg active stretchers based on the dynamic connection point coordinates of the first leg active stretcher, the second leg active stretcher and the lower leg connecting component and the initial lengths of the two leg active stretchers; ； ； In the formula, respectively, the length change amount of the first and second leg active extension members calculating rotation numbers of two side driving motors in the exoskeleton robot based on the length change amounts of the two leg active stretchers to obtain the first control parameters; ； ； In the formula, , are the number of rotations and the lead of the drive motor on one side, respectively, , are the number of rotations and the lead of the drive motor on the other side, respectively.

2. The exoskeleton robot control method according to claim 1, wherein the motion data comprises motion angular velocities, motion speeds and angles of the exoskeleton robot in different motion modes.

3. The exoskeleton robot control method of claim 1, wherein, the step of performing trajectory prediction on the target model based on the motion data to obtain a predicted ankle joint motion trajectory comprises the following steps: converting the motion data into a vector form to obtain a motion vector, inputting the motion vector into an encoder in the target model, mapping the motion vector to a high-dimensional space through the encoder to obtain a query vector, a key vector and a value vector; Fourier transforming the query vector, the key vector and the value vector to obtain a transformed query vector , a transformed key vector , and a transformed value vector ; computing sparse self-attention based on the transformed query vectors, the transformed key vectors, and the transformed value vectors : ； wherein is an inverse Fourier transform, is a zero padding operation, is a transform query vector , a transform key vector dimension of is a sparse matrix of the same dimension as the transform query vector dimension of encoding and feature extracting the motion vector through the encoder to obtain extracted features, splicing the extracted features to obtain spliced features; inputting the spliced features and the sparse self-attention into a decoder of the target model and performing feature decoding by using a multi-head self-attention mechanism to obtain decoded features; inputting the decoded features into a fully connected layer in the target model and mapping the decoded features to a prediction space through the fully connected layer to obtain a predicted ankle joint motion trajectory.

4. The exoskeleton robot control method of claim 1, wherein the step of performing trajectory processing and filtering on the predicted ankle joint motion trajectory to obtain a processed motion trajectory comprises the following steps: extracting a predicted ankle joint motion angle in the predicted ankle joint motion trajectory, identifying an angle error between the predicted ankle joint motion angle and a real ankle joint motion angle after wearing the exoskeleton robot; performing trajectory correction on the predicted ankle joint motion trajectory based on the angle error and by using a linear translation mapping method to obtain a first adjusted trajectory; The first adjustment trajectory is decomposed into a plurality of sub-trajectories, historical motion trajectories under the same motion mode are obtained, similarity between the sub-trajectories and the historical motion trajectories is calculated, and the sub-trajectories with similarity greater than a similarity threshold are spliced and adjusted continuously to obtain a second adjustment trajectory. The second adjustment trajectory is filtered by using a preset filter to obtain a processed motion trajectory.

5. The exoskeleton robot control method of claim 1, wherein, The step of performing second motion solving of the exoskeleton robot based on the processed motion trajectory to output a second control parameter comprises: extracting a predicted motion parameter in the processed motion trajectory, constructing a coefficient matrix based on the predicted motion parameter constructing a balance equation based on the coefficient matrix ； In the formula, is the actual internal stress of the leg active stretch member; determining a constraint condition; ； ； wherein is the outer force matrix, is the slack, is the maximum internal stress that the leg active extension can withstand; The balance equation is modified into a force-position equation through the constraint condition, and the force-position equation is optimized to output an internal stress trajectory of a leg active telescopic member to obtain the second control parameter.

6. An exoskeleton robot control system employing the exoskeleton robot control method according to claim 1, characterized by The system comprises: a prediction submodule configured to obtain motion data of a target and a pre-trained target model, perform trajectory prediction based on the target model and the motion data to obtain a predicted ankle motion trajectory; a processing submodule configured to perform trajectory processing and filtering on the predicted ankle motion trajectory to obtain a processed motion trajectory; a first solving submodule configured to perform first motion solving of an exoskeleton robot based on the processed motion trajectory to output a first control parameter; a second solving submodule configured to perform second motion solving of the exoskeleton robot based on the processed motion trajectory to output a second control parameter; a control submodule configured to send the first control parameter and the second control parameter to the exoskeleton robot to complete control of the exoskeleton robot.

7. The exoskeleton robot control system of claim 6, wherein, The motion data comprises motion angular velocity, motion speed and angle of the exoskeleton robot under different motion modes.

8. A computer comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the exoskeleton robot control method in any one of claims 1 to 5.

9. A storage medium, characterized by The storage medium stores the computer program, and the computer program is executed by the processor to implement the exoskeleton robot control method in any one of claims 1 to 5.

Images