Gait reference angle generation method and device, robot, equipment and medium

Abstract

The invention provides a gait reference angle generation method and device, a robot, equipment and a medium, and relates to the technical field of robot assistion.The method comprises the steps that target gait auxiliary physiological parameters and target walking speed of a target user are obtained; inputting the target gait auxiliary physiological parameters and the target walking speed into a hip joint gait reference trajectory generation model to obtain a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; acquiring a real-time gait phase generated when the target user wears the exoskeleton robot, and according to the real-time gait phase and a hip joint angle fitting polynomial, generating corresponding hip joint gait reference angles of the target user at different moments; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle value corresponding to each moment in the hip joint gait reference trajectory. According to the invention, the precision and individuation degree of rehabilitation training are improved.

Description

Technical Field

[0001] The present invention relates to the field of robot-assisted technology, and in particular to a gait reference angle generation method, device, robot, equipment and medium. Background Art

[0002] In the current field of medical rehabilitation science, the recovery of walking ability for stroke patients is a research focus that has attracted much attention. Due to the damage to the motor center caused by stroke, patients often face many challenges such as weakened strength on the affected side, abnormal muscle tension, decreased tactile sensitivity, and impaired body balance. These conditions not only limit their freedom of walking, but also bring many difficulties to daily life. Although traditional walking recovery exercise methods have achieved certain results, they generally have the problems of low efficiency and lack of customized solutions for individual specific conditions, making it difficult to fully meet the unique and changing rehabilitation needs of each patient.

[0003] With the continuous development of robotics technology, exoskeleton robots are increasingly used in the field of medical rehabilitation. Exoskeleton robots can simulate normal gait patterns and provide patients with accurate training trajectories and motion support, thereby helping them gradually recover their walking ability. However, existing exoskeleton robots mainly provide gait rehabilitation for patients based on fixed rehabilitation training patterns, and are unable to provide more accurate gait reference angles based on the patient's own conditions, resulting in low efficiency in gait rehabilitation training.

[0004] Therefore, there is an urgent need for a gait reference angle generation method, device, robot, equipment and medium to solve the above problems. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention provides a gait reference angle generation method, device, robot, equipment and medium.

[0006] The present invention provides a gait reference angle generation method, comprising: Obtain target gait assistance physiological parameters and target walking speed of the target user; Inputting the target gait auxiliary physiological parameter and the target walking speed into a hip joint gait reference trajectory generation model, and obtaining a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds; The real-time gait phase generated by the target user when wearing the exoskeleton robot is obtained, and a hip joint gait reference angle corresponding to the target user at different times is generated according to a fitting polynomial based on the real-time gait phase and the hip joint angle; wherein the hip joint angle fitting polynomial is constructed based on the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0007] According to a gait reference angle generation method provided by the present invention, the hip joint gait reference trajectory generation model is obtained by training through the following steps: The sample gait auxiliary physiological parameters are constructed according to the subject's gender information, age information, calf length, thigh length, height data, weight data, anterior superior iliac spine width and knee diameter; Obtaining hip joint angle sample trajectories corresponding to the sample gait auxiliary physiological parameters at different sample walking speeds; Using the hip joint angle sample trajectory as a hip joint gait reference trajectory label, marking the sample gait auxiliary physiological parameters and the sample walking speed, and obtaining a training sample set; Based on the training sample set, the multilayer perceptron is trained to obtain the hip joint gait reference trajectory generation model.

[0008] According to a gait reference angle generation method provided by the present invention, the multilayer perceptron includes an input layer, a hidden layer and an output layer; wherein the input layer includes 9 neurons, the hidden layer is provided with 10 neurons, and the output layer is composed of 101 neurons.

[0009] According to a gait reference angle generation method provided by the present invention, the step of obtaining the real-time gait phase generated by the target user when wearing an exoskeleton robot comprises: Based on a plurality of adaptive frequency oscillators in the exoskeleton robot, obtaining an oscillator frequency corresponding to each of the adaptive frequency oscillators at a current moment; Sorting the frequencies of the oscillators at the current moment, and determining the adaptive frequency oscillator with the smallest oscillator frequency as the target adaptive frequency oscillator according to the sorting result; The real-time gait phase is acquired according to the gait phase corresponding to the target adaptive frequency oscillator.

[0010] According to a gait reference angle generation method provided by the present invention, the real-time gait phase is obtained according to the gait phase corresponding to the target adaptive frequency oscillator, including: The phase change rate of the target adaptive frequency oscillator is calculated according to the oscillator frequency corresponding to the target adaptive frequency oscillator, the reconstruction error and the current oscillator phase; wherein the reconstruction error is constructed based on the oscillator amplitude corresponding to the target adaptive frequency oscillator; Based on the phase change rate, the phase of the target adaptive frequency oscillator is updated to obtain the real-time gait phase.

[0011] According to a gait reference angle generation method provided by the present invention, after fitting a polynomial according to the real-time gait phase and the hip joint angle to generate the hip joint gait reference angle corresponding to the target user at different times, the method further includes: Acquire the actual movement angle of the hip joint of the target user collected in real time by the exoskeleton robot; According to the actual movement angle of the hip joint and the gait reference angle of the hip joint, an auxiliary force adjustment parameter corresponding to the exoskeleton robot is generated.

[0012] The present invention also provides a gait reference angle generating device, comprising: A basic data acquisition module is used to obtain the target gait auxiliary physiological parameters and target walking speed of the target user; A hip joint gait reference trajectory generation module is used to input the target gait auxiliary physiological parameters and the target walking speed into a hip joint gait reference trajectory generation model to obtain a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds; The hip joint gait reference angle generation module is used to obtain the real-time gait phase generated by the target user when wearing the exoskeleton robot, and to generate the hip joint gait reference angle corresponding to the target user at different times according to the real-time gait phase and the hip joint angle fitting polynomial; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0013] The present invention also provides a robot, which is a hip joint exoskeleton robot and includes the above-mentioned gait reference angle generating device.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-described methods for generating a gait reference angle is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for generating a gait reference angle as described above is implemented.

[0016] The gait reference angle generation method, device, robot, equipment and medium provided by the present invention generate personalized hip joint gait reference trajectories through human physiological parameters and expected walking speed, using multimodal fusion technology and a multi-layer perceptron network model. These hip joint gait reference trajectories take into account the individual differences and rehabilitation needs of patients, thereby providing hip joint gait reference angles that are more in line with the user's own, thereby improving the accuracy and personalization of rehabilitation training. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of the flow of the gait reference angle generation method provided by the present invention; Figure 2 A schematic diagram of the structure of a multilayer perceptron provided by the present invention; Figure 3 An overall schematic diagram of a hip joint gait reference angle generation process provided by the present invention; Figure 4 A schematic diagram of the structure of a gait reference angle generating device provided by the present invention; Figure 5 A schematic diagram of the structure of the hip joint exoskeleton robot provided by the present invention; Figure 6 This is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Exoskeleton robot-assisted rehabilitation programs are gradually becoming a widely accepted trend in gait recovery training for stroke patients after surgery. The lower limb mobility of stroke patients after surgery shows significant individual differences. However, the current gait training path generation methods are insufficient in personalized customization, which not only limits the effectiveness of gait rehabilitation training, but also may bring additional risks, such as potential secondary injuries, thus affecting the overall rehabilitation process and effect.

[0021] In response to the above problems, the present invention proposes a method for generating a gait reference lower limit of the lower limb hip joint based on multimodal fusion of physiological parameters and expected walking speed. This method collects multiple physiological parameters of the patient, such as thigh length, calf length, height, weight, iliac width and knee diameter, and combines the patient's expected walking speed, uses multimodal fusion technology and a multi-layer perceptron network model to generate personalized hip joint angle training trajectories. These trajectories take into account the individual differences and rehabilitation needs of the patients, and improve the accuracy and personalization of gait rehabilitation training.

[0022] Figure 1 A schematic diagram of the process of the gait reference angle generation method provided by the present invention is shown in FIG. Figure 1 As shown, the present invention provides a gait reference angle generation method, comprising: Step 101: Obtain target gait auxiliary physiological parameters and target walking speed of a target user.

[0023] In the present invention, a series of target gait-assisted physiological parameters and target walking speed of the target user (i.e., the user who needs gait rehabilitation training) are firstly obtained, which not only helps to personalize the settings of the exoskeleton robot, but also can significantly improve the safety and efficiency of the training.

[0024] Specifically, the target gait assistance physiological parameters are composed of the following physiological parameters: Gender information: Gender differences may affect bone structure and muscle strength, which has a direct impact on the fit and comfort of the exoskeleton. Gender information can be obtained by directly asking the user for their gender and recorded as male or female.

[0025] Age information: As people age, their flexibility, strength, and bone density change, which helps adjust training intensity and assistance. Age information can be obtained by directly asking the user for their date of birth and calculating their age based on the current date.

[0026] Calf length: The user lies on his back with his legs straight and together. Then use a soft tape measure to measure the actual length of the calf from the inner lower edge of the knee joint to the inner ankle tip.

[0027] Thigh length: The user lies on his back with his legs straight and together. Then use a soft tape measure to measure the actual length of the thigh from the anterior superior iliac spine to the medial upper edge of the knee joint.

[0028] Height data: The user stands barefoot on the measuring machine, keeps the head straight, eyes looking straight ahead, arms hanging naturally, heels together and toes separated by about 60 degrees. Then, use the height measuring machine with a horizontal ruler to close to the top of the user's head and accurately read the height measurement value.

[0029] Weight data: The user wears light clothing and stands barefoot on a calibrated electronic scale to obtain the weight measurement displayed on the electronic scale.

[0030] Anterior superior iliac spine width: The user is in a standing position with feet shoulder-width apart. Then, the maximum horizontal distance between the anterior superior iliac spines on both sides is measured with a tape measure to obtain the anterior superior iliac spine width.

[0031] Knee diameter: With the user in a sitting or supine position, with legs straight and together, use a tape measure to measure the circumference of the knee at its most prominent part (i.e. above the patella), and estimate the knee diameter based on the circumference value.

[0032] The target walking speed refers to the expected walking speed set by the user based on his or her own situation and rehabilitation progress during the current stage of gait rehabilitation training. This parameter directly reflects the user's rehabilitation goals and current physical condition. By adjusting the exoskeleton's drive system to match the user's target walking speed, it can not only enhance the pertinence and effectiveness of the training, but also increase the user's participation and enthusiasm. In the early stages of gait rehabilitation training, the target walking speed may be low. As rehabilitation progresses, the speed is gradually increased to promote the user's gradual return to a normal gait.

[0033] Step 102, input the target gait auxiliary physiological parameters and the target walking speed into the hip joint gait reference trajectory generation model to obtain the hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multi-layer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds.

[0034] In the present invention, the hip joint gait reference trajectory generation model is a trained machine learning model, which receives a series of input parameters (such as gait auxiliary physiological parameters and walking speed) and outputs a hip joint gait reference trajectory that matches these parameters.

[0035] Specifically, the target gait assistance physiological parameters include gender, age, calf length, thigh length, height, weight, anterior superior iliac spine width and knee diameter, which together describe the user's physical characteristics and physiological state. The target walking speed is the walking speed that the user expects to achieve in the current rehabilitation stage, reflecting the user's rehabilitation goals and physical state.

[0036] During the model training phase, a large amount of sample data is collected, including gait auxiliary physiological parameters, walking speed, and corresponding hip joint angle sample trajectories of different individuals. These hip joint angle sample trajectories can be obtained through experimental measurement, motion capture system or simulation software, or directly obtained through relevant sample databases.

[0037] In the present invention, a multilayer perceptron (MLP) is used for model training. The multilayer perceptron is a feedforward neural network consisting of an input layer, a hidden layer, and an output layer. During the training process, sample data (gait auxiliary physiological parameters and walking speed as input, hip joint angle sample trajectory as output) are input into the MLP, and the network weights are adjusted through the back propagation algorithm to make the model's predicted output as close to the actual output as possible.

[0038] The trained multi-layer perceptron model can receive new target gait assistance physiological parameters and target walking speed as input and output a personalized hip joint gait reference trajectory. This hip joint gait reference trajectory can be used as the basis for exoskeleton robot control, gait analysis or rehabilitation training plan.

[0039] Step 103, obtaining the real-time gait phase generated by the target user when wearing the exoskeleton robot, and generating the hip joint gait reference angle corresponding to the target user at different times by fitting a polynomial based on the real-time gait phase and the hip joint angle; wherein the hip joint angle fitting polynomial is constructed based on the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0040] In the present invention, the exoskeleton robot system is mainly composed of two parts: the host computer and the hip joint exoskeleton robot. The core components of the hip joint exoskeleton robot include the main control module, the actuator and the angle sensor. The detailed working principle is as follows: First, the angle sensor equipped with the DC brushless motor of the hip exoskeleton robot will capture and record the actual movement angles of the user's hip joints on both sides in real time. These actual movement angle information is then efficiently transmitted to the host computer through the RS485 serial communication protocol for further analysis and processing. Subsequently, the host computer receives these angle sequences and calculates the real-time reference angle, that is, the hip joint gait reference angle. Then, the host computer will determine the required output torque value based on the deviation between the actual angle and the reference angle, combined with the preset precise control strategy. Finally, this precisely calculated torque value will be transmitted back to the main control module of the hip exoskeleton robot through the RS485 serial communication protocol again. After receiving the torque value, the main control module will immediately drive the actuator to make corresponding adjustments to achieve the gait assistance function, thereby helping users to conduct more accurate and effective rehabilitation training.

[0041] In the present invention, when the user wears the exoskeleton robot, the gait phase information of the target user is obtained in real time, and the gait phase information is used to generate the hip joint gait reference angle of the target user at different times through a polynomial fitting method. The hip joint gait reference trajectory obtained by the hip joint gait reference trajectory generation model is used to extract the hip joint angle values ​​corresponding to each moment. Then, a polynomial fitting model is constructed using these hip joint angle values ​​to obtain a model that can accurately describe the change of the hip joint angle over time. This model will be used to generate the hip joint gait reference angle of the target user at different times.

[0042] After the hip joint angle fitting polynomial model is established, the real-time gait phase information can be used as input to calculate the hip joint gait reference angle at the corresponding moment through the model. These reference angles can be used for real-time control of the exoskeleton robot to ensure that the robot can follow the user's natural gait and provide necessary support and assistance.

[0043] In the present invention, the hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model only contains 101 discrete angle values, and the gait phase during gait training can vary in a continuous range of 0 to 2π. Therefore, the present invention needs to accurately map any gait phase acquired in real time to the corresponding hip joint angle. Specifically, the present invention adopts a polynomial fitting method, and in order to avoid the problems of overfitting and underfitting, a quadratic polynomial is selected for fitting. The quadratic polynomial can avoid the overfitting phenomenon that may be caused by too high a degree, and can also prevent the problem that too low a degree may not fully capture the nonlinear characteristics of the data. The formula of the hip joint angle fitting polynomial is as follows: ; in, is the hip joint gait reference angle, is the real-time gait phase, , and They represent the constant term, linear term and quadratic term of the hip joint angle fitting polynomial respectively. When the gait phase at a certain moment is obtained during gait training, the hip joint gait reference trajectory and hip joint angle fitting polynomial output by the hip joint gait reference trajectory generation model can be used to calculate the hip joint gait reference angle at the current moment.

[0044] The gait reference angle generation method provided by the present invention generates personalized hip joint gait reference trajectories through human physiological parameters and expected walking speed, using multimodal fusion technology and a multi-layer perceptron network model. These hip joint gait reference trajectories take into account the individual differences and rehabilitation needs of patients, thereby providing hip joint gait reference angles that are more in line with the user's own, thereby improving the accuracy and personalization of rehabilitation training.

[0045] Based on the above embodiment, the hip joint gait reference trajectory generation model is obtained by training through the following steps: The sample gait auxiliary physiological parameters are constructed according to the subject's gender information, age information, calf length, thigh length, height data, weight data, anterior superior iliac spine width and knee diameter; Obtaining hip joint angle sample trajectories corresponding to the sample gait auxiliary physiological parameters at different sample walking speeds; Using the hip joint angle sample trajectory as a hip joint gait reference trajectory label, marking the sample gait auxiliary physiological parameters and the sample walking speed, and obtaining a training sample set; Based on the training sample set, the multilayer perceptron is trained to obtain the hip joint gait reference trajectory generation model.

[0046] Based on the above embodiment, the multilayer perceptron includes an input layer, a hidden layer and an output layer; wherein the input layer includes 9 neurons, the hidden layer is provided with 10 neurons, and the output layer is composed of 101 neurons.

[0047] In the present invention, a public data set of 50 healthy subjects was first collected, with an age span of 6 to 72 years old, covering a wide age range. In order to build a personalized gait assistance model, the present invention selected 8 key physiological parameters as input features, namely gender, age, calf length, thigh length, height, weight, anterior superior iliac spine width and knee diameter (i.e. knee circumference). These physiological parameters, as sample gait assistance physiological parameters, were used as input features of a multi-layer perceptron (MLP) model to build a personalized gait assistance model, i.e., a hip joint gait reference trajectory generation model.

[0048] In order to train the multi-layer perceptron model, it is also necessary to collect sample trajectories of the subject's hip joint angle at different walking speeds, which can be achieved through a motion capture system or similar equipment to accurately record the changes in the subject's hip joint angle during walking.

[0049] Furthermore, the subjects are required to walk at different walking speeds (for example, slow, normal speed and fast) and collect the corresponding hip joint angle trajectories at each speed. These hip joint angle trajectories will be used as the output targets of the model, namely, the hip joint gait reference trajectory labels. Then, the present invention matches and labels the collected hip joint angle sample trajectories with the corresponding sample gait auxiliary physiological parameters and sample walking speeds. Each training sample contains 8 physiological parameters (input features) and 1 walking speed (input feature), 101 hip joint angle values ​​(output targets, corresponding to 101 sampling time points after the gait cycle is divided into 100 equal parts), thereby obtaining a training sample set containing multiple training samples for subsequent multi-layer perceptron model training.

[0050] Figure 2 The schematic diagram of the structure of the multilayer perceptron provided by the present invention can be referred to Figure 2 As shown, in the present invention, the multilayer perceptron includes 9 neurons, which correspond to 8 physiological parameters (i.e., 8 parameters such as gender, age, calf length, thigh length, height, weight, anterior superior iliac spine width, and knee circumference) input and 1 walking speed input. These neurons receive input features and pass them to the hidden layer. The hidden layer is provided with 10 neurons for processing the information transmitted by the input layer and transforming it through a nonlinear activation function (such as ReLU). The neurons in the hidden layer are connected through weights and biases and trained through a back propagation algorithm. The output layer is composed of 101 neurons, each of which corresponds to the hip joint angle value at a sampling time point in the gait cycle. The neurons in the output layer output predicted values ​​through a linear activation function, which will be compared with the actual hip joint angle sample trajectory to calculate the loss and perform model optimization.

[0051] The present invention trains the multilayer perceptron model through a training sample set to obtain a generative model that can generate personalized hip joint gait reference trajectory. This model can generate a hip joint angle trajectory that conforms to the individual's gait characteristics based on the individual's physiological parameters and walking speed, providing an accurate reference for subsequent gait training and assistance.

[0052] Based on the above embodiment, the step of obtaining the real-time gait phase of the target user when wearing the exoskeleton robot includes: Based on a plurality of adaptive frequency oscillators in the exoskeleton robot, obtaining an oscillator frequency corresponding to each of the adaptive frequency oscillators at a current moment; Sorting the frequencies of the oscillators at the current moment, and determining the adaptive frequency oscillator with the smallest oscillator frequency as the target adaptive frequency oscillator according to the sorting result; The real-time gait phase is acquired according to the gait phase corresponding to the target adaptive frequency oscillator.

[0053] An exoskeleton robot is a mechanical device that assists human walking. It is closely integrated with the human body through sensors and actuators to help improve or restore walking ability. In the design of exoskeleton robots, multiple adaptive frequency oscillators are usually integrated, which can dynamically adjust their parameters to accurately match the frequency of any periodic input signal generated by the human body when walking.

[0054] Since walking is a complex quasi-periodic motion, the motion frequencies of different parts may be different. In order to determine the frequency that best represents the current gait characteristics, the present invention sorts the frequencies of all current adaptive frequency oscillators, among which the smallest frequency is considered to best reflect the basic gait rhythm of the walker (i.e., the most stable frequency component in the walking cycle). Therefore, in the present invention, the adaptive frequency oscillator with the smallest frequency is determined as the target adaptive frequency oscillator.

[0055] Furthermore, after determining the target adaptive frequency oscillator, its phase information can be used to track and analyze the user's real-time gait phase at the current moment. Gait phase refers to the division of different phases in the walking cycle, such as stance phase, swing phase, etc. Through the phase change of the target adaptive frequency oscillator, the transition of the walker from one step to the next and the subtle movement changes within each step can be accurately captured.

[0056] The present invention obtains real-time gait phase based on multiple adaptive frequency oscillators, which not only improves the accuracy of gait analysis, but also provides strong technical support for personalized gait rehabilitation training.

[0057] On the basis of the above embodiment, the step of acquiring the real-time gait phase according to the gait phase corresponding to the target adaptive frequency oscillator includes: The phase change rate of the target adaptive frequency oscillator is calculated according to the oscillator frequency corresponding to the target adaptive frequency oscillator, the reconstruction error and the current oscillator phase; wherein the reconstruction error is constructed based on the oscillator amplitude corresponding to the target adaptive frequency oscillator; Based on the phase change rate, the phase of the target adaptive frequency oscillator is updated to obtain the real-time gait phase.

[0058] In the present invention, an adaptive frequency oscillator is used to extract the gait phase from the angle trajectory of the hip joint. Specifically, the reconstruction error is calculated , which represents the original signal (i.e., the angle trajectory of the hip joint) and the reconstructed signal The difference between the two. Reconstructed signal is the amplitude of all adaptive frequency oscillators and Phase The specific formula is: ; in, M is the number of adaptive oscillators. Reconstruction error The original signal With the reconstruction signal The difference is: .

[0059] Furthermore, by using the parameters of the target adaptive frequency oscillator (frequency , Amplitude and the current phase ) and reconstruction error , calculate the phase change rate of the oscillator . This process involves the following formula: ; ; ; in, K is the coupling strength, which determines the influence of the reconstruction error on the dynamic characteristics of the oscillator. The coupling strength is set to 20 in the present invention; is the learning rate, which controls the speed of adjusting the amplitude of the oscillator. The present invention sets the learning rate to 1; Indicates The frequency change rate of an adaptive frequency oscillator, Indicates The amplitude change rate of an adaptive frequency oscillator, Represents a time variable.

[0060] The phase change rate is calculated After that, the phase of the target adaptive frequency oscillator is updated to obtain the real-time gait phase. This process is achieved by integrating the phase change rate. By continuously updating the current phase of the target adaptive frequency oscillator, the user's gait phase can be tracked and analyzed in real time. This process not only provides accurate time information within the gait cycle, but can also be used to evaluate the stability, symmetry and other key features of the gait, thereby providing valuable feedback for gait rehabilitation training.

[0061] On the basis of the above embodiment, after fitting a polynomial according to the real-time gait phase and the hip joint angle to generate the hip joint gait reference angle corresponding to the target user at different times, the method further includes: Acquire the actual movement angle of the hip joint of the target user collected in real time by the exoskeleton robot; According to the actual movement angle of the hip joint and the gait reference angle of the hip joint, an auxiliary force adjustment parameter corresponding to the exoskeleton robot is generated.

[0062] In the present invention, based on the comparison between the actual movement angle of the hip joint collected in real time and the reference angle of the hip joint gait, the control system of the exoskeleton robot will calculate how to adjust the auxiliary force it provides. If the user's actual movement angle deviates from the reference angle, the control system may increase or decrease the auxiliary force in a certain direction to help the user get closer to the ideal gait pattern. In the present invention, the auxiliary force adjustment parameters include the size, direction and action time of the auxiliary force, which together determine how the exoskeleton robot dynamically responds to the user's movement to provide the best auxiliary effect, so as to achieve more efficient and personalized gait assistance.

[0063] Figure 3 The overall schematic diagram of the hip joint gait reference angle generation process provided by the present invention can be referred to Figure 3 As shown, in the present invention, a model for generating different reference trajectories of the hip joint based on multimodal fusion of human physiological parameters and expected walking speed is used to generate a personalized hip joint angle reference trajectory. Before the user starts robot-assisted gait rehabilitation training, the physiological parameters and expected walking speed that are highly related to their gait trajectory need to be input into the system, and then a personalized hip joint gait reference trajectory will be generated. Then, according to this hip joint gait reference trajectory, the corresponding hip joint angle fitting polynomial and the extracted gait phase are constructed to calculate the reference angle, i.e., the hip joint gait reference angle.

[0064] After gait rehabilitation training begins, the main controller of the exoskeleton robot will monitor the actual angle parameters in real time, and drive the motor to provide corresponding assistance based on the actual angle parameters and the hip joint gait reference angle to ensure that the actual motion trajectory is as close to the reference trajectory as possible, thereby performing personalized gait rehabilitation training, improving rehabilitation training efficiency and ensuring rehabilitation training safety.

[0065] The gait reference angle generating device provided by the present invention is described below. The gait reference angle generating device described below and the gait reference angle generating method described above can be referenced to each other.

[0066] Figure 4 The schematic diagram of the structure of the gait reference angle generating device provided by the present invention is as follows: Figure 4 As shown, the present invention provides a gait reference angle generation device, including a basic data acquisition module 401, a hip joint gait reference trajectory generation module 402 and a hip joint gait reference angle generation module 403, wherein the basic data acquisition module 401 is used to obtain the target gait auxiliary physiological parameters and the target walking speed of the target user; the hip joint gait reference trajectory generation module 402 is used to input the target gait auxiliary physiological parameters and the target walking speed into the hip joint gait reference trajectory generation model to obtain the hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speed; the hip joint gait reference angle generation module 403 is used to obtain the real-time gait phase generated by the target user when wearing an exoskeleton robot, and generate the hip joint gait reference angle corresponding to the target user at different times according to the real-time gait phase and the hip joint angle fitting polynomial; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0067] The gait reference angle generating device provided by the present invention generates personalized hip joint gait reference trajectories through human physiological parameters and expected walking speed, using multimodal fusion technology and a multi-layer perceptron network model. These hip joint gait reference trajectories take into account the individual differences and rehabilitation needs of patients, thereby providing a hip joint gait reference angle that is more in line with the user's own, thereby improving the accuracy and personalization of rehabilitation training.

[0068] The device provided in the embodiment of the present invention is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for the specific process and detailed content, which will not be repeated here.

[0069] The present invention also provides a robot, which is a hip joint exoskeleton robot and includes the gait reference angle generating device described in the above embodiment.

[0070] Figure 5 The schematic diagram of the hip joint exoskeleton robot provided by the present invention can be referred to Figure 5 As shown, the hip joint exoskeleton robot provided by the present invention includes a gait reference angle generating device 501, which is connected to a host computer 502 and is used to generate hip joint gait reference angles corresponding to the target user at different times, so that the host computer 502 compares the received hip joint gait reference angle with the corresponding actual motion angle, thereby calculating the output torque value currently required by the hip joint exoskeleton robot.

[0071] The robot provided by the present invention generates personalized hip joint gait reference trajectories through human physiological parameters and expected walking speed, using multimodal fusion technology and a multi-layer perceptron network model. These hip joint gait reference trajectories take into account the individual differences and rehabilitation needs of patients, thereby providing hip joint gait reference angles that are more in line with the user's own, thereby improving the accuracy and personalization of rehabilitation training.

[0072] Figure 6 A schematic diagram of the structure of an electronic device provided by the present invention, such as Figure 6 As shown, the electronic device may include: a processor (Processor) 601, a communication interface (Communications Interface) 602, a memory (Memory) 603 and a communication bus 604, wherein the processor 601, the communication interface 602, and the memory 603 communicate with each other through the communication bus 604. The processor 601 can call the logic instructions in the memory 603 to execute the gait reference angle generation method, which includes: obtaining the target gait auxiliary physiological parameters and target walking speed of the target user; inputting the target gait auxiliary physiological parameters and the target walking speed into the hip joint gait reference trajectory generation model to obtain the hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speed; obtaining the real-time gait phase generated by the target user when wearing the exoskeleton robot, and generating the hip joint gait reference angle corresponding to the target user at different times according to the real-time gait phase and the hip joint angle fitting polynomial; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0073] In addition, the logic instructions in the above-mentioned memory 603 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0074] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the gait reference angle generation method provided by the above methods, the method including: obtaining the target gait auxiliary physiological parameters and target walking speed of the target user; inputting the target gait auxiliary physiological parameters and the target walking speed into a hip joint gait reference trajectory generation model to obtain a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds; obtaining the real-time gait phase generated by the target user when wearing an exoskeleton robot, and generating the hip joint gait reference angle corresponding to the target user at different times according to the real-time gait phase and the hip joint angle fitting polynomial; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0075] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the gait reference angle generation method provided in the above-mentioned embodiments, the method comprising: obtaining target gait auxiliary physiological parameters and target walking speed of the target user; inputting the target gait auxiliary physiological parameters and the target walking speed into a hip joint gait reference trajectory generation model to obtain a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds; obtaining the real-time gait phase generated by the target user when wearing an exoskeleton robot, and generating the hip joint gait reference angle corresponding to the target user at different times according to the real-time gait phase and the hip joint angle fitting polynomial; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

[0076] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0077] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a gait reference angle, characterized in that: include: Obtain target gait assistance physiological parameters and target walking speed of the target user; Inputting the target gait auxiliary physiological parameter and the target walking speed into a hip joint gait reference trajectory generation model, and obtaining a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds; The real-time gait phase generated by the target user when wearing the exoskeleton robot is obtained, and a hip joint gait reference angle corresponding to the target user at different times is generated according to a fitting polynomial based on the real-time gait phase and the hip joint angle; wherein the hip joint angle fitting polynomial is constructed based on the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

2. The gait reference angle generation method according to claim 1, characterized in that: The hip joint gait reference trajectory generation model is obtained by training through the following steps: The sample gait auxiliary physiological parameters are constructed according to the subject's gender information, age information, calf length, thigh length, height data, weight data, anterior superior iliac spine width and knee diameter; Obtaining hip joint angle sample trajectories corresponding to the sample gait auxiliary physiological parameters at different sample walking speeds; Using the hip joint angle sample trajectory as a hip joint gait reference trajectory label, marking the sample gait auxiliary physiological parameters and the sample walking speed, and obtaining a training sample set; Based on the training sample set, the multilayer perceptron is trained to obtain the hip joint gait reference trajectory generation model.

3. The gait reference angle generation method according to claim 2, characterized in that: The multilayer perceptron includes an input layer, a hidden layer and an output layer; wherein the input layer includes 9 neurons, the hidden layer is provided with 10 neurons, and the output layer is composed of 101 neurons.

4. The gait reference angle generation method according to claim 1, characterized in that: The step of obtaining the real-time gait phase of the target user when wearing the exoskeleton robot comprises: Based on a plurality of adaptive frequency oscillators in the exoskeleton robot, obtaining an oscillator frequency corresponding to each of the adaptive frequency oscillators at a current moment; Sorting the frequencies of the oscillators at the current moment, and determining the adaptive frequency oscillator with the smallest oscillator frequency as the target adaptive frequency oscillator according to the sorting result; The real-time gait phase is acquired according to the gait phase corresponding to the target adaptive frequency oscillator.

5. The gait reference angle generation method according to claim 4, characterized in that: The acquiring the real-time gait phase according to the gait phase corresponding to the target adaptive frequency oscillator includes: The phase change rate of the target adaptive frequency oscillator is calculated according to the oscillator frequency corresponding to the target adaptive frequency oscillator, the reconstruction error and the current oscillator phase; wherein the reconstruction error is constructed based on the oscillator amplitude corresponding to the target adaptive frequency oscillator; Based on the phase change rate, the phase of the target adaptive frequency oscillator is updated to obtain the real-time gait phase.

6. The gait reference angle generation method according to claim 1, characterized in that: After fitting a polynomial according to the real-time gait phase and the hip joint angle to generate the hip joint gait reference angles corresponding to the target user at different times, the method further includes: Acquire the actual movement angle of the hip joint of the target user collected in real time by the exoskeleton robot; According to the actual movement angle of the hip joint and the gait reference angle of the hip joint, an auxiliary force adjustment parameter corresponding to the exoskeleton robot is generated.

7. A gait reference angle generating device, characterized in that: include: A basic data acquisition module is used to obtain the target gait auxiliary physiological parameters and target walking speed of the target user; A hip joint gait reference trajectory generation module is used to input the target gait auxiliary physiological parameters and the target walking speed into a hip joint gait reference trajectory generation model to obtain a hip joint gait reference trajectory output by the hip joint gait reference trajectory generation model; wherein the hip joint gait reference trajectory generation model is obtained by training a multilayer perceptron based on sample gait auxiliary physiological parameters and sample walking speeds; The hip joint gait reference angle generation module is used to obtain the real-time gait phase generated by the target user when wearing the exoskeleton robot, and to generate the hip joint gait reference angle corresponding to the target user at different times according to the real-time gait phase and the hip joint angle fitting polynomial; wherein the hip joint angle fitting polynomial is constructed according to the hip joint angle values ​​corresponding to each time in the hip joint gait reference trajectory.

8. A robot, characterized in that: The robot is a hip joint exoskeleton robot, comprising the gait reference angle generating device as described in claim 7.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the gait reference angle generation method according to any one of claims 1 to 6 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for generating a gait reference angle according to any one of claims 1 to 6 is implemented.

Images