Joint torque prediction method and device based on myoelectricity and muscle-bone physiological information

Abstract

The invention discloses a lower limb exoskeleton rehabilitation robot joint torque prediction method and device based on an electromyographic signal, and the method comprises the steps: firstly collecting the electromyographic, joint angle, angular velocity and joint torque data of a user, and carrying out the modeling of four muscles related to a knee joint through a neuromusculoskeletal model; and identifying physiological parameters used in the model by using a sparrow optimization algorithm, and outputting an optimal solution to obtain an estimated torque. Then, the real torque and the estimated torque are endowed with the weights of # imgabs0 # respectively, and a fusion torque is obtained; and training a neural network model by using the collected data and the fusion moment, and meanwhile, enabling the neuromusculoskeletal model to participate in the training process of the neural network model by adopting a method of bringing the neuromusculoskeletal model into a loss function, so that the constraint of the model is increased. And finally, storing the trained model for subsequent calling. The method has the advantages of high precision and strong generalization, and model training can converge more quickly under the condition of less data volume.

Description

Technical Field

[0001] The present invention relates to a joint torque prediction method and device based on electromyography and musculoskeletal physiological information, which is applied to a lower limb rehabilitation exoskeleton robot and is suitable for lower limb joint torque prediction in active control strategies. The advance characteristics of electromyography signals are used to achieve reliable motion intention prediction. Background Art

[0003] Rehabilitation strategies vary with different rehabilitation stages. In order to make users feel more comfortable during training, it is necessary to detect and identify the user's movement intention, such as predicting the joint angle or joint torque, and then apply it to the control strategy to make it more smooth and comfortable. At present, the characteristic that the electromyographic signal is 30-200ms ahead of the muscle movement is often used to identify the movement intention. There are two main methods for joint torque prediction. The first method is to use the neuromusculoskeletal model to model a single muscle, explain the conversion relationship between the electromyographic signal and the muscle force, and then obtain the joint torque prediction value of the corresponding joint. Another method is to use a neural network model to extract features from the data, such as using a convolutional neural network to extract spatial features, or using a long short-term memory neural network to extract temporal features, and then predict the joint torque.

[0004] However, the neuromusculoskeletal model needs to model each related muscle separately, with numerous physiological parameters, poor individual adaptability, and relatively low accuracy. The training of the neural network model requires a large amount of data support, and it is also difficult to obtain high-quality data. At the same time, due to its black box nature, it is difficult to explain its mechanism of action, and it is difficult to ensure that the output of the neural network model can meet the needs and safety requirements. Therefore, a lightweight model with fewer parameters is needed that can simultaneously guarantee the real-time requirements, accuracy requirements, universality requirements, and safety requirements for predicting joint torque during exercise. Summary of the invention

[0005] The present invention aims to overcome the above-mentioned shortcomings of the prior art and provide a method and device for predicting joint torque based on electromyography and musculoskeletal physiological information. The present invention designs a neural network model that integrates muscle physiological information and incorporates the neuromusculoskeletal model into the training process of the neural network model, that is, muscle physiological information is introduced into the neural network model, and the advantages of the neuromusculoskeletal model and the neural network model are combined to construct a new lower limb joint torque prediction model with muscle physiological information constraints. Compared with the method that only uses the neural network model, this method has improved accuracy.

[0006] To achieve the above effects, the first aspect of the present invention relates to a joint torque prediction method based on electromyography and musculoskeletal physiological information, comprising the following steps:

[0007] S1: Build a data acquisition platform. Use Raspberry Pi, MATLAB, myoelectric sensors, and exoskeleton robots to build the entire data acquisition system to achieve offline data acquisition. According to muscle synergy, determine to collect the myoelectric signals of the rectus femoris, vastus medialis, medial head of the gastrocnemius, and lateral head of the gastrocnemius, as well as the joint angle signals and joint angular velocity signals of the hip joint, knee joint, and ankle joint.

[0008] S2: Establishment of neuromusculoskeletal model. On the pycharm platform, the four muscles were modeled using the collected electromyographic signals and joint angle signals, and the torque of each muscle was obtained and the torque of the knee joint as it changes with movement was calculated. Since the number of muscles used was small, in order to make the output of the model closer to the real torque, an output coefficient was added to the output to obtain the joint torque of the knee joint.

[0009] S3: Identify the parameters of the neuromusculoskeletal model. The physiological parameters required by the neuromusculoskeletal model, such as the optimal muscle force F m0 , maximum feather angle φ0, optimal muscle fiber length l m0 , tendon length l t As well as the muscle contraction velocity f(v), there are individual differences. All parameters are taken as parameters to be optimized, and a reasonable range of each parameter is given. The Sparrow Optimization Algorithm is used to optimize them, and then they are passed back to the neuromusculoskeletal model to improve the estimation accuracy of the neuromusculoskeletal model.

[0010] S4: Build a neural network model. Take the collected electromyographic signals, joint angle signals and joint angular velocity signals as input features, and use the attention mechanism to give different features a weight matrix to indicate the importance of different features. Use two convolutional layers and pooling layers to extract the spatial relationship between features and perform dimensionality reduction on the data. Add a bidirectional long short-term memory layer to extract long-term temporal dependencies. Add a dropout layer in between to reduce overfitting. Finally, add a fully connected layer to output the predicted joint torque.

[0011] S5: Construct loss function. During training, based on the above neuromusculoskeletal model, the joint torque during movement is estimated using electromyographic signals and joint angles, and combined with the true torque, different weights are assigned to the estimated torque and the true torque, respectively, to obtain a fused torque estimate as a label for offline training of the neural network. The mean absolute error between the fused torque and the predicted torque output by the neural network model is calculated as the loss function for training, so that the model continuously fits and approaches the true torque.

[0012] S6: Design evaluation indicators. The mean absolute error between the predicted moment and the true moment of the neural network model is used as the evaluation indicator of the model.

[0013] S7: Model training and saving. The model is trained using data and saved for subsequent use.

[0014] The second aspect of the present invention relates to a joint torque prediction device based on electromyography and musculoskeletal physiological information, comprising a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the joint torque prediction method based on electromyography and musculoskeletal physiological information of the present invention.

[0015] The third aspect of the present invention relates to a computer-readable storage medium on which a program is stored, including data collection and processing, model structure parameters, parameters of the trained model, and a model training method. When the program is executed by a processor, the joint torque prediction method based on electromyography and musculoskeletal physiological information of the present invention is implemented.

[0016] Compared with the existing scheme, the present invention has the following beneficial effects: the neuromusculoskeletal model only needs 4 muscles in the estimation process, which greatly reduces the number of parameters in the model application process. Introducing the muscle physiological information of the neuromusculoskeletal model into the training process of the neural network model can enhance the interpretability of the neural network model, add a physical constraint to the neural network model, and ensure that the output of the model meets the expected effect, thereby ensuring the safety of the exoskeleton. This method also makes up for the problem that the collected torque signal is unstable and has large fluctuations. By using the fused torque signal of the two, smoother torque label data can be obtained, which enhances the generalization performance of the neural network model and has higher prediction accuracy under the conditions of shorter training time and less training data. At the same time, this method avoids the problem of numerous parameters and low accuracy when the neuromusculoskeletal model is used directly, making the neural network model more suitable for prediction tasks with high real-time requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a display diagram of the joint torque prediction model experimental platform of the present invention.

[0018] Figure 2 This is a position diagram of the myoelectric signal collection patch of the present invention.

[0019] Figure 3 This is the effect diagram of the wavelet packet decomposition of the electromyographic signal of the present invention.

[0020] Figure 4 It is a diagram of the joint angle and joint angular velocity processing results of the present invention.

[0021] Figure 5 It is a diagram of the joint torque processing results of the present invention.

[0022] Figure 6 Schematic diagram of the neuromusculoskeletal model of the present invention.

[0023] Figure 7 It is a flow chart of the sparrow optimization algorithm of the present invention.

[0024] Figure 8 It is a structural diagram of the neural network model of the present invention.

[0025] Fig. 9 It is a structural diagram of the technical solution of the present invention.

[0026] Fig.10 It is the loss function curve of the neural network model training effect of the present invention.

[0027] Fig.11 It is a loss function curve of the training effect of the neural network model of the fusion neuromusculoskeletal model of the present invention.

[0028] Fig.12 This is a diagram showing the final knee joint torque prediction effect of the neural network model of the fusion neuromusculoskeletal model of the present invention.

[0029] Fig.13 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0030] The following examples of the invention, combined with the accompanying drawings, specifically explain the experimental plan, data processing, model construction and training, etc., to further clarify the application field, design ideas and technical solutions of the invention.

[0031] Example 1

[0032] This embodiment relates to a joint torque prediction method based on electromyography and musculoskeletal physiological information, comprising the following steps:

[0033] S1: According to Figure 1 As shown in the figure, design an experimental plan on the experimental platform of the built joint torque prediction model to collect offline data. Use Simulink to build a data collection program on the PC to achieve simultaneous collection of joint angles, joint torques, electromyographic signals, and IMU signals. The PC and exoskeleton communicate through the Raspberry Pi, and the electromyographic sensor and IMU sensor are directly connected to the PC through the USB interface. When collecting data, you need to use Figure 1 The exoskeleton is worn by the user, and the exoskeleton's built-in torque sensor and angle sensor are used to collect the joint angle and joint torque signals of the knee and hip joints during the movement, and the signals are saved in Simulink. The electromyographic signals are collected using a six-channel electromyographic sensor, such as Figure 2Each muscle corresponds to a channel. The yellow and green electrodes are attached to the corresponding muscles. The red patch is attached to the knee as a reference electrode. In Simulink, it is collected synchronously with the joint angle and other signals. For the angle data of the ankle joint, two IMU sensors are used to collect the pitch angle respectively, and then processed. There are 7 male subjects in the experiment, with a height of 178±7cm and a weight of 78±14kg. The acquisition frequency of the electromyographic signal, joint angle and joint torque signal is 1000Hz. The data were collected at walking speeds of 0.5km / h, 1.0km / h and 1.5km / h respectively. Two groups were collected at the same speed, and each group of data was collected for 30s.

[0034] S2: Preprocess the collected data. Use wavelet packet decomposition to analyze the energy distribution of the electromyographic signal. According to the decomposition results, filter the electromyographic signal to remove noise, such as 50Hz power frequency interference, high-frequency noise, etc. In the 10-level wavelet packet decomposition, the high-frequency decomposition results of the first level show that the electromyographic signal has higher energy before each movement begins. At this time, the frequency range is about 0-500Hz. The low-frequency decomposition results of the seventh level better retain the overall trend of the electromyographic signal and have a certain filtering effect. The results are as follows: Figure 3 As shown. After experiments, it was finally determined that the bandpass filter cutoff frequency was 300-450Hz, and the low-pass filter cutoff frequency was 2Hz. After filtering, the original electromyographic signal, high-frequency signal and low-frequency signal are retained respectively. The joint angle and joint torque data are processed by sliding average window to make the data smoother, and the window length is set to 250 and 400 time steps respectively. The angular velocity signal of each joint is calculated by the central difference method, and the processed joint angle, joint angular velocity and joint torque signals are saved. The processing effect is shown as follows: Figure 4 , 5 As shown. The original electromyographic signal, high-frequency electromyographic signal, low-frequency electromyographic signal, joint angle signal and joint angular velocity signal are used as input features, and the joint torque is used as a label. Finally, the sliding window method is used to process the data set. All feature data of a window corresponds to the label data of its last time step. As a sample, the window length is 50 time steps (should not exceed the time length of the electromyographic signal leading the muscle movement), and the sliding step length is 1. The data set for neural network model training is produced for subsequent model training.

[0035] S3: According to Figure 6The neuromusculoskeletal model in the paper models the four muscles of rectus femoris, vastus medialis, medial gastrocnemius and lateral gastrocnemius respectively, and inputs the processed low-frequency electromyographic signals and joint angle data into the neuromusculoskeletal model. The electromyographic signals are converted from nerve signals to muscle activation through muscle activation dynamics, muscle contraction dynamics convert muscle activation into muscle tension, and joint dynamics convert muscle tension into joint torque, thus obtaining the estimated torque output by the neuromusculoskeletal model. The modeling method is as follows:

[0036] For the obtained electromyographic signal e, calculate its neural activation:

[0037] u(t)=ae(tT)-bu(t-1)-cu(t-2) (1)

[0038] Among them, u(t-1) and u(t-2) represent the neural activation at the moment t-1 and t-2 respectively, e(tT) is the surface electromyographic signal with a time delay of T, and the constraints are:

[0039] b=d1+d2,c=d1d2,abc=1

[0040] |d1|<1, |d2|<1 (2)

[0041] Using the u(t) obtained above, calculate the muscle activation a(t) as follows:

[0042]

[0043] Where A is the nonlinear shape factor, ranging from A∈[-3,0].

[0044] Then the muscle contraction force F is obtained according to the following calculation formula m (t):

[0045] F m (t)=f(v)f(l)a(t)F m0 (4)

[0046] In the formula, F m (t) is the time-varying muscle fiber contraction force, f(v) is the normalized muscle contraction velocity-force relationship, f(l) is the normalized muscle fiber length-force relationship, F m0 is the maximum isometric muscle fiber contraction force. Since the muscle fiber contraction force is less affected by the muscle contraction speed under slow walking conditions, f(v) can be regarded as a constant in the modeling process.

[0047] The muscle contraction force is obtained and then converted into muscle tendon force. The relationship between muscle force and muscle tendon force is as follows:

[0048] F mt=F m cosφ (5)

[0049] Here φ represents the feathering angle.

[0050] Finally, the joint torque of the knee joint is calculated according to the following formula:

[0051]

[0052] Among them, T J (θ,t) is the joint torque of joint J, r i (θ) is the moment arm of the i-th muscle of joint J when the joint angle is θ, F mt(i) (θ,t) refers to the muscle tendon force of the i-th muscle at angle θ and time t.

[0053] In addition, the time-varying muscle-tendon length l required in the calculation process mt The estimation can be done according to the following method:

[0054]

[0055]

[0056] Where θ1, θ2, θ3, θ4 are the joint angles of muscle-related joints, a1, a2, a3, ..., a n is the parameter to be identified.

[0057] S4: There are four muscles involved in the neuromusculoskeletal model, each of which contains the maximum isometric muscle force F m0 , optimal feather angle φ0, optimal muscle fiber length l m0 , tendon length l t , normalized muscle force-velocity relationship f(v), and individual-specific physiological parameters d1, d2, and A, a total of 23 parameters need to be identified. The sparrow optimization algorithm is used for parameter identification. Table 1 lists the reasonable range of each parameter. The sparrow optimization algorithm first randomly generates multiple populations as possible solutions based on the reasonable range of each parameter in Table 1, and calculates the fitness of each solution separately. Each population is sorted according to the fitness, and all populations are updated to approach the population with the highest fitness. It is continuously iterated until the stopping condition is met to generate the optimal solution, and then the optimal solution is passed back to the neuromusculoskeletal model to obtain the estimated torque of the knee joint that is closest to the actual torque. The flow chart of the sparrow optimization algorithm is shown below. Figure 7 As shown. Among them, the fitness function uses the mean absolute error between the estimated torque and the true torque of the neuromusculoskeletal model, and the fitness function is:

[0058]

[0059] Where y represents the true moment, represents the estimated torque output by the neuromusculoskeletal model, and n represents the number of data points.

[0060] Table 1 Parameter range setting list

[0061]

[0062]

[0063] The conditions for determining whether optimization should be stopped are:

[0064] (1) If the fitness improvement is less than 0.01 for seven consecutive rounds, or the difference between the maximum and minimum fitness in the same round is less than 0.01, and the fitness function value is less than 1.45, the optimization is stopped. Otherwise, the population is regenerated for parameter optimization.

[0065] (2) The maximum number of iterations is 30 rounds. If the condition of less than 1.45 is not met in 30 rounds, the population with the smallest fitness function value in all rounds is taken.

[0066] S5: Build Figure 8 The neural network model shown in Figure 1 is shown in Figure 1. First, a feature selection attention mechanism is used to assign different weights to different features. Then, two layers of two-dimensional convolution layers with different numbers of convolution kernels and a convolution kernel size of 3×3 are used to extract the spatial relationship between different features. The number of convolution kernels in the first convolution layer is 32, and the number of convolution kernels in the second convolution layer is 16. The sliding step size is (1,1), and the activation function is ReLU. An average pooling layer is added between the two convolution layers to reduce the dimension of the data. The pooling window size is 3×3 and the step size is (1,1). At the same time, a dropout layer is added with a dropout rate of 0.25 to prevent overfitting. A layer normalization operation is added before the activation function of the second convolution layer to stabilize the training process and prevent gradient disappearance or gradient explosion. Then, a bidirectional long short-term memory neural network is used to extract the long-range dependency of features in time. The number of hidden layer units is 32. Finally, a fully connected layer is used to achieve feature integration and improve the overall performance of the model. The number of neurons in the fully connected layer is 64, and the output layer uses a linear activation function. The model uses the Adam optimizer with a learning rate of 0.001.

[0067] S6: Construct a loss function. Incorporating the neuromusculoskeletal model into the training process of the neural network model, through the neuromusculoskeletal model, the physiological information of the muscle and the transformation process from electromyographic signals to joint torque are introduced into the training process of the neural network model, which can add constraints to the system and enhance its interpretability. By constructing a suitable loss function, the fusion of the neuromusculoskeletal model and the neural network model can be achieved. The method adopted by the present invention is to convert the output estimated torque (TNMS ) and true moment (T true ) is assigned a weight and added to obtain a fusion torque (T label ), and the estimated moments are given in specific implementation and the true torque The weight of the neural network model is defined as the prediction moment (T Pred ) and true torque (T true ) is the loss function, and the neural network model is trained by calculating and minimizing the loss function. That is,

[0068] Fusion torque:

[0069] T label =αT NMS +βT true (7)

[0070] Loss function L:

[0071]

[0072] Where y represents the true moment, represents the fusion moment, and n represents the number of data points in the training set.

[0073] S7: Define the model evaluation index. To evaluate the final effect of the model, define the absolute average error between the model's predicted output and the true torque as the model evaluation index:

[0074]

[0075] Where y represents the true moment, It represents the prediction moment of the model output, and n represents the number of data points in the test set.

[0076] S8: Train the model and save it. The data is divided into a training set and a test set in a ratio of 8:2 for training the fusion model. After all the data are trained, the model is saved for subsequent model calls. The overall design idea is as follows Fig. 9 Framework diagram of the technical solution presented.

[0077] Finally, the sparrow optimization algorithm was used for parameter identification. The average error between the estimated torque output by the neuromusculoskeletal model and the actual torque was 1.604NM. Compared with the simulated annealing algorithm, the average error was 2.805NM. The sparrow optimization algorithm has a significant improvement. Only the neural network model was trained. The training error of the predicted joint torque output was about 1.457NM, and the model gradually converged after more than 80 rounds of training. Fig.10As shown in the figure. The output of the neural network model that integrates the neuromusculoskeletal model has a predicted joint torque training error of about 0.617 N·M. In the model evaluation index, it is only 0.939 NM. In addition, the model gradually converges after only 40 rounds of training. Fig.11 The final knee joint torque prediction effect is shown in Fig.12 shown.

[0078] Example 2

[0079] The present embodiment relates to a joint torque prediction device based on electromyography and musculoskeletal physiological information, including a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the joint torque prediction method based on electromyography and musculoskeletal physiological information of Example 1.

[0080] Example 3

[0081] The third aspect of the present invention relates to a computer-readable storage medium on which a program is stored, including data collection and processing, modeling of a neuromusculoskeletal model, neural network model structure and parameters, parameters of the trained model, and a model training method. When the program is executed by a processor, the joint torque prediction method based on electromyography and musculoskeletal physiological information of the present invention is implemented.

[0082] In summary, the present invention uses electromyography to predict joint torque during exercise, combined with the physiological prior knowledge of the neuromusculoskeletal model, which can increase the constraints of the neural network model output, more reliably predict the joint torque when wearing an exoskeleton, and better ensure the safety of the user. At the same time, fewer muscles make it more convenient to use and more suitable for strong real-time tasks, providing reliable prediction values ​​for motion control, which is crucial to improving the flexibility and comfort of the exoskeleton rehabilitation robot's motion control process.

[0083] The above embodiment describes an implementation of the present invention. For those skilled in the art, various adjustments, changes and improvements may be made to the present invention. All adjustments, changes and improvements made within the scope of the core principle of the present invention are considered to be within the scope of the present invention.

Claims

1. A joint torque prediction method based on electromyography and musculoskeletal physiological information includes the following steps: S1: Build a data acquisition platform; use Raspberry Pi, MATLAB, electromyographic sensors and exoskeleton robots to build the entire data acquisition system to achieve offline data acquisition; according to muscle synergy, determine to collect electromyographic signals of the rectus femoris, vastus medialis, medial head of gastrocnemius and lateral head of gastrocnemius, as well as joint angle signals and joint angular velocity signals of the hip joint, knee joint and ankle joint; S2: Establishment of neuromusculoskeletal model; on the pycharm platform, four muscles were modeled using the collected electromyographic signals and joint angle signals, the torque of each muscle was obtained and the torque of the knee joint as it changes with movement was calculated; due to the small number of muscles used, in order to make the output of the model closer to the real torque, an output coefficient was added to the output to obtain the joint torque of the knee joint; S3: Identify the parameters of the neuromusculoskeletal model; the physiological parameters required by the neuromusculoskeletal model, including the optimal muscle force F m0 , maximum feather angle φ0, optimal muscle fiber length l m0 , tendon length l t As well as the muscle contraction speed f(v), there are individual differences. All parameters are taken as parameters to be optimized, and a reasonable range of each parameter is given. The Sparrow Optimization Algorithm is used to optimize them, and then they are passed back to the neuromusculoskeletal model to improve the estimation accuracy of the neuromusculoskeletal model; S4: Build a neural network model; The collected electromyographic signals, joint angle signals and joint angular velocity signals are used as input features, and the attention mechanism is used to give different features a weight matrix to indicate the importance of different features; Use two convolutional layers and a pooling layer to extract the spatial relationship between features and perform dimensionality reduction on the data. Add a bidirectional long short-term memory layer to extract long-term temporal dependencies. Add a dropout layer in between to reduce overfitting. Finally, add a fully connected layer to output the predicted joint torque. S5: construct a loss function; during training, based on the above neuromusculoskeletal model, use electromyographic signals and joint angles to estimate the joint torque during movement, and combine with the true torque, assign different weights to the estimated torque and the true torque, and obtain a fused torque estimate as a label for offline training of the neural network; The mean absolute error between the fusion torque and the predicted torque output by the neural network model is calculated as the loss function for training, so that the model can continuously fit and approach the real torque; S6: Design evaluation indicators; use the mean absolute error between the predicted moment and the true moment of the neural network model as the evaluation indicator of the model; S7: Model training and saving: Train the model using data and save it for subsequent calls.

2. The joint torque prediction method based on myoelectricity and musculoskeletal physiological information as claimed in claim 1, characterized in that: The modeling method in step S2 is as follows: For the obtained electromyographic signal e, calculate its neural activation: u(t)=ae(tT)-bu(t-1)-cu(t-2) (1) Among them, u(t-1) and u(t-2) represent the neural activation at moments t-1 and t-2 respectively, e(tT) is the surface electromyographic signal with a time delay of T, and the constraints are: b=d1+d2,c=d1d2,abc=1 |d1|<1, |d2|<1 (2) Using the u(t) obtained above, calculate the muscle activation a(t) as follows: Where A is the nonlinear shape factor, ranging from A∈[-3,0]; Then the muscle contraction force F is obtained according to the following calculation formula m (t): F m (t)=f(v)f(l)a(t)F m0 (4) In the formula, F m (t) is the time-varying muscle fiber contraction force, f(v) is the normalized muscle contraction velocity-force relationship, f(l) is the normalized muscle fiber length-force relationship, F m0 is the maximum isometric muscle fiber contraction force; since the muscle fiber contraction force is less affected by the muscle contraction speed under slow walking conditions, f(v) is regarded as a constant in the modeling process; The muscle contraction force is obtained and then converted into muscle tendon force. The relationship between muscle force and muscle tendon force is as follows: F t =F m cosφ (5) Here φ represents the feathering angle; Finally, the joint torque of the knee joint is calculated according to the following formula: Among them, T J (θ,t) is the joint torque of joint J, r i (θ) is the moment arm of the i-th muscle of joint J when the joint angle is θ, F mt (i)(θ,t) refers to the muscle-tendon force of the i-th muscle at angle θ and time t; In addition, the time-varying muscle-tendon length l required in the calculation process mt The estimation of is based on formula (7): Where θ1, θ2, θ3, θ4 are the joint angles of muscle-related joints, a1, a2, a3, ..., a n is the parameter to be identified.

3. The joint torque prediction method based on electromyography and musculoskeletal physiological information as claimed in claim 1, characterized in that: Step S3 specifically includes: There are four muscles involved in the neuromusculoskeletal model, each of which contains the maximum isometric muscle force F m0 , optimal feather angle φ0, optimal muscle fiber length l m0 , tendon length l t , normalized muscle force-velocity relationship f(v), and individual-specific physiological parameters d1, d2, A, a total of 23 parameters need to be identified; the sparrow optimization algorithm is used for parameter identification. The sparrow optimization algorithm first randomly generates multiple populations as possible solutions according to the reasonable range of each parameter, and calculates the fitness of each solution respectively, sorts each population according to the fitness, and all populations are updated to the population with the highest fitness. It is continuously iterated until the stop condition is met to generate the optimal solution, and then the optimal solution is passed back to the neuromusculoskeletal model to obtain the estimated torque of the knee joint that is closest to the true torque; the fitness function uses the mean absolute error between the estimated torque of the neuromusculoskeletal model and the true torque, and the fitness function is: Where y represents the true moment, represents the estimated torque output by the neuromusculoskeletal model, and n represents the number of data points; The conditions for determining whether optimization should be stopped are: (1) If the fitness improvement is less than 0.01 for seven consecutive rounds, or the difference between the maximum and minimum fitness in the same round is less than 0.01, and the fitness function value is less than 1.45, the optimization is stopped. Otherwise, the population is regenerated for parameter optimization. (2) The maximum number of iterations is 30 rounds. If the condition of less than 1.45 is not met in 30 rounds, the population with the smallest fitness function value in all rounds is taken.

4. The joint torque prediction method based on myoelectricity and musculoskeletal physiological information as claimed in claim 1, characterized in that: Step S4 specifically includes: First, a feature selection attention mechanism is used to assign different weights to different features. Then, two layers of two-dimensional convolutional layers with different numbers of convolutional kernels and a convolutional kernel size of 3×3 are used to extract the spatial relationship between different features. The number of convolutional kernels in the first convolutional layer is 32, and the number of convolutional kernels in the second convolutional layer is 16. The sliding step size is (1,1), and the activation function is ReLU. An average pooling layer is added between the two convolutional layers to reduce the dimension of the data. The pooling window size is 3×3 and the step size is (1,1). At the same time, a drop layer is added with a drop rate of 0.25 to prevent overfitting. A layer normalization operation is added before the activation function of the second convolutional layer to stabilize the training process and prevent gradient disappearance or gradient explosion. Then, a bidirectional long short-term memory neural network is used to extract the long-range dependency of features in time. The number of hidden layer units is 32. Finally, a fully connected layer is used to achieve feature integration and improve the overall performance of the model. The number of neurons in the fully connected layer is 64, and the output layer uses a linear activation function. The model uses the Adam optimizer with a learning rate of 0.

001.

5. The joint torque prediction method based on electromyography and musculoskeletal physiological information as claimed in claim 1, characterized in that: The specific method of constructing the loss function in step S5 is: the output estimated torque (T NMS ) and true moment (T true ) is assigned a weight and added to obtain a fusion torque (T label ), and the estimated moments are given in specific implementation and the true torque The weight of the neural network model; define the prediction moment (T Pred ) and the true moment (T true ) is the loss function. The neural network model is trained by calculating and minimizing the loss function; that is, Fusion torque: T label =αT NMS +βT true (7) Loss function L: Where y represents the true moment, represents the fusion moment, and n represents the number of data points in the training set.

6. The joint torque prediction method based on myoelectricity and musculoskeletal physiological information as claimed in claim 1, characterized in that: The model evaluation index of step S6 is specifically: Where y represents the true moment, It represents the prediction moment of the model output, and n represents the number of data points in the test set.

7. A joint torque prediction device based on electromyography and musculoskeletal physiological information, characterized in that: It comprises a memory and one or more processors, wherein the memory stores executable codes, and when the one or more processors execute the executable codes, they are used to implement the joint torque prediction method based on electromyography and musculoskeletal physiological information as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: A program is stored thereon, and when the program is executed by a processor, the joint torque prediction method based on myoelectricity and musculoskeletal physiological information described in any one of claims 1 to 6 is implemented.

Images