A rehabilitation decision-making method based on virtual reality rehabilitation training system
By introducing flexible ankle rehabilitation robot and cuckoo algorithm improved fuzzy neural network model in virtual reality rehabilitation training system, personalized rehabilitation training for ankle patients is achieved, and the training mode is single and control accuracy in the existing system is solved. The problem has improved the fun and effectiveness of rehabilitation training.
Patent Information
- Application Number
- CN202210933875.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-04
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2042-08-04
AI Technical Summary
The existing virtual reality rehabilitation training system cannot achieve precise control and adaptive training decisions for ankle patients, and lacks real-time adjustment capabilities, resulting in a single training mode and cannot meet personalized needs.
Design a robot-assisted ankle rehabilitation training system based on virtual reality, combining flexible ankle rehabilitation robot, control system hardware module, virtual reality communication module and rehabilitation decision module, and realize the fuzzy neural network model improved by the cuckoo algorithm to achieve the patient's Real-time analysis of ankle joint movement and dynamic adjustment of training scenarios provide personalized rehabilitation training plans.
Improve the fun and effectiveness of the training, realize the closed loop of virtual reality rehabilitation training-adaptive rehabilitation training decision-making-virtual reality rehabilitation training, ensure the accuracy and personalization of rehabilitation training, and overcome the single model of traditional rehabilitation training and The problem of low control accuracy.
Smart Images

Figure CN115363907B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of rehabilitation training systems, and in particular to a rehabilitation decision-making method based on a virtual reality rehabilitation training system. Background Art
[0002] Ankle injuries are extremely common in life. Timely and effective rehabilitation treatment is essential to maximize the recovery of ankle function. Traditional sports rehabilitation methods are mostly assisted by physical therapists. There are no accurate training records and training feedback, and it is impossible to complete high-intensity repetitive training. In addition, the training mode is relatively boring and single, which cannot mobilize the patient's initiative in rehabilitation training.
[0003] Rehabilitation medical robot technology is a motor neurological rehabilitation treatment technology developed in recent years. It can assist patients in motion control and record their motion data in real time, greatly enhancing the purposefulness and scientific nature of the rehabilitation training process. Virtual reality technology provides patients with a rehabilitation training environment by building a highly realistic virtual scene. It can not only change the inherent rehabilitation training model, but also combine games with diagnosis and treatment, get rid of the previous boring training state, and greatly improve the fun of training.
[0004] The combination of virtual reality technology and robotics technology has opened up new development directions in the field of rehabilitation therapy and has been widely used. The virtual reality enhanced bicycle kit developed by Amir B. Farjadian et al. integrates ankle rehabilitation into the bicycle training module, which helps to train the muscle strength of the ankle joint. With the powerful data acquisition system, the accuracy of ankle rehabilitation training has been greatly improved. However, the system is too complex, inconvenient to wear and cannot complete passive training. Michael Girone et al. developed the "Rutgers Ankle" ankle rehabilitation mechanism, which can assist patients in autonomous rehabilitation training. Through the generation of virtual scenes by PC, patients can perform rehabilitation training while playing games, which can increase the fun of the entire rehabilitation process of patients and bring effective rehabilitation effects. However, the rehabilitation robot is difficult to achieve precise control and has a high cost. The above two rehabilitation training methods represent some of the problems existing in the current virtual reality rehabilitation training and lack rehabilitation decision-making during the training process, and cannot achieve real-time adjustment of the actual training of patients.
[0005] Therefore, it is necessary to improve the existing technology so as to provide patients with different training modes and achieve precise control, and realize the closed loop of virtual reality rehabilitation training-adaptive rehabilitation training decision-virtual reality rehabilitation training, so as to provide more effective rehabilitation training for ankle patients. Summary of the invention
[0006] The present invention provides a robot-assisted ankle joint rehabilitation training system based on virtual reality and a rehabilitation decision method thereof. A suitable rehabilitation training mode is formulated according to the patient's joint movement ability, and the robot's assistance is realized in combination with a control method. The ankle joint abduction / adduction, external rotation / internal rotation, and inversion / eversion rehabilitation training are completed based on virtual reality scene tasks, thereby realizing a closed loop of virtual reality rehabilitation training-adaptive rehabilitation training decision-virtual reality rehabilitation training. The entire training process can not only improve the interest of the training but also ensure the effectiveness of the rehabilitation training.
[0007] A rehabilitation decision-making method based on a virtual reality rehabilitation training system, characterized in that:
[0008] The virtual reality rehabilitation training system includes: a flexible ankle joint rehabilitation robot, a control system hardware module, a host computer display control module, a virtual reality communication module, a virtual environment module, and a rehabilitation decision module;
[0009] The flexible ankle joint rehabilitation robot, the virtual reality communication module, and the virtual environment module are connected in sequence;
[0010] The control system hardware module is connected to the flexible ankle joint rehabilitation robot;
[0011] The flexible ankle joint rehabilitation robot is connected to the host computer display control module;
[0012] The rehabilitation decision module is connected to the virtual environment module;
[0013] The flexible ankle joint rehabilitation robot is bound to the patient's ankle joint to assist the patient's ankle joint in rotating at different degrees of freedom;
[0014] The control system hardware module is used to drive the actual movement of the flexible ankle joint rehabilitation robot and perform real-time adjustments according to the control instructions issued by the host computer;
[0015] The host computer display control module is used to display the joint motion angle information of the patient's ankle in three different degrees of freedom, and can achieve a real-time trajectory tracking effect;
[0016] In the upper computer display control module, the user can select a rehabilitation mode, namely, a passive mode and an active mode. The passive mode is the active control of the flexible ankle joint rehabilitation robot based on a virtual task, and the active mode is the user's active control driven by a virtual task. Combined with the patient's ankle joint movement ability and the scene task difficulty planning, the auxiliary control of the flexible ankle joint rehabilitation robot under different scene tasks is realized;
[0017] The virtual reality communication module is used to input the angle change information of the ankle joint into the virtual scene in real time, and its functions include uploading the motion signal of the patient's ankle to the host computer and establishing a server and a client to realize the communication between the host computer and the virtual scene;
[0018] The virtual environment module is used to provide scene feedback during patient rehabilitation training and realize virtual reality interaction;
[0019] The rehabilitation decision module analyzes the electromyographic signals and activity range of the patient during rehabilitation training through a fuzzy neural network rehabilitation training decision model improved based on the cuckoo algorithm, establishes a mapping model, and realizes real-time adjustment of the speed and range of the rehabilitation training scene action;
[0020] Preferably, the various patient motion state information collected by the flexible ankle joint rehabilitation robot is used as the basis for rehabilitation decision-making to reasonably design rehabilitation training tasks. The host computer display control module provides an interactive interface and realizes robot control, giving the flexible ankle joint rehabilitation robot a motion trajectory. The virtual reality communication module realizes communication between the flexible ankle joint rehabilitation robot and the host computer display control module, and between the host computer display control module and the virtual environment module, so as to ensure natural interaction between the patient and the virtual scene.
[0021] Preferably, the virtual environment module can provide patients with an interesting and immersive virtual training environment and different training modes, allowing patients to get rid of the traditional boring training methods;
[0022] The control system hardware module obtains relevant control instructions according to specific training requirements and calculations, and controls the 3-DOF ankle rehabilitation robot dynamic platform by controlling the pneumatic muscle driver, so as to realize the inversion / eversion, adduction / abduction movement, and dorsiflexion / plantar flexion movement of the ankle joint, so as to assist the patient in corresponding sports rehabilitation training;
[0023] The flexible ankle rehabilitation robot is equipped with an angle sensor on each of its three axes for recording the angle changes of the ankle joint in real time. The virtual reality real-time communication module can transmit the angle change information of the ankle joint to the virtual scene, and drive the objects in the rehabilitation game to perform corresponding movements according to the formulated corresponding movement strategies, thereby realizing virtual reality synchronization and improving the fun and immersion of the training.
[0024] Preferably, the virtual environment module includes a user information module and three training scenarios, namely, a park roaming scenario, an indoor storage scenario and an aircraft combat scenario. From daily life scenarios to challenging aerial scenarios, from fully passive robot-driven training to patient-controlled robot autonomous training, the whole process is both interesting and enables the rehabilitation training process to proceed step by step.
[0025] Preferably, the host computer displays the control module, and in the display module, the real-time motion trajectory of the patient's ankle can be seen. Here, the motion information of the flexible ankle joint rehabilitation robot is decomposed into three axial motions of X, Y, and Z. There are two curves on the curve graph interface of each axis. The red curve is the expected trajectory of the flexible ankle joint rehabilitation robot, and the blue curve is its actual motion trajectory. The real-time trajectory tracking effect can be seen. The control module can be used to set and input relevant motion parameters, such as motion mode, motion range, motion speed, etc., and perform various control functions, such as initialization, stop, etc.
[0026] Preferably, the virtual reality real-time communication module comprises two parts. The first part is the communication between the flexible ankle joint rehabilitation robot and the upper computer display control module. The data acquisition card can be used as a communication medium to upload the patient's ankle motion signal to the upper computer display control module, or the upper computer display control module converts the calculated control amount into an analog voltage signal and transmits it to the pneumatic proportional valve, thereby realizing the control of the flexible ankle joint rehabilitation robot.
[0027] The second part is the communication between the host computer display control module and the virtual environment module. The server and the client are established, and the movement signal of the patient's ankle is transmitted to the virtual environment module in real time through TCP network communication to perform synchronous control of the virtual avatar, thereby realizing human-computer interaction in virtual reality and enhancing the interest of patient rehabilitation training.
[0028] Preferably, the training mode is divided into passive training and active training. The passive mode is active control of the robot based on virtual tasks, and the active mode is active control of the user driven by virtual tasks. The upper computer display control module and the rehabilitation decision model are combined to analyze and evaluate the user's ankle joint movement ability and the difficulty of the scene task. After planning, the control system hardware module can set the desired trajectory to achieve auxiliary control of the robot under different scene tasks.
[0029] Preferably, the virtual environment module enables the patient to perform ankle rehabilitation training exercises while completing game tasks. During the exercise, the host computer display control module can collect electromyographic data and angle data, display the patient's current rehabilitation status through rehabilitation evaluation, and automatically update the evolution training scene under the adaptive rehabilitation decision model, so that the exercise load can be regulated in real time, realizing a closed loop of virtual reality rehabilitation training-rehabilitation evaluation-adaptive rehabilitation training decision-virtual reality rehabilitation training.
[0030] Preferably, when making virtual reality rehabilitation training decisions, it is necessary to efficiently and accurately establish the connection between the patient's motion state and the parameters in the virtual training scene. This goal is achieved by the virtual reality communication module, that is, establishing a nonlinear mapping relationship between input and human output data, and the flexible ankle joint rehabilitation robot provides assistance to the patient's ankle joint. A rehabilitation training system with real-time regulation of motion load is established through the rehabilitation decision model, which can enable patients to undergo personalized and adaptive rehabilitation training.
[0031] The beneficial effects of the present invention are as follows: the present invention can not only help patients complete basic rehabilitation training, but also improve the patient's initiative in rehabilitation training in the form of virtual games, and realize the closed loop of virtual reality rehabilitation training-rehabilitation assessment-adaptive rehabilitation training decision-virtual reality rehabilitation training. In training, the robot has the advantages of high control accuracy, good repeatability and diversified training modes, which well solves the defects of manual assisted rehabilitation. Virtual reality technology, with its unique advantages, can integrate games, diagnosis and treatment, and psychological guidance, improve the fun of training and eliminate the patient's resistance, thereby improving rehabilitation efficiency and accelerating the patient's rehabilitation process. The designed adaptive rehabilitation training decision model can well meet the needs of adjusting the control parameters of the virtual training scene in real time according to the patient's motor training status during virtual reality rehabilitation training, thereby changing the difficulty of rehabilitation training tasks and allowing patients to receive personalized and appropriate training. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 :Basic implementation flow chart of virtual scene;
[0033] Figure 2 : Working principle diagram of ankle rehabilitation robot system;
[0034] Figure 3 : Overall flow chart of a robot-assisted ankle joint rehabilitation training system based on virtual reality according to an embodiment of the present invention;
[0035] Figure 4 : Ankle joint rehabilitation training system software module block diagram;
[0036] Figure 5 :Structure diagram of CS-FNN virtual reality rehabilitation training decision model established based on cuckoo optimized fuzzy neural network algorithm;
[0037] Figure 6 :CS-FNN virtual reality rehabilitation training decision model structure and workflow diagram; DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.
[0039] By referring to the exemplary embodiments, the objects and functions of the present invention and methods for achieving these objects and functions will be explained. However, the present invention is not limited to the exemplary embodiments disclosed below; it can be implemented in different forms.
[0040] The specific implementation of the present invention is a robot-assisted ankle joint rehabilitation training system based on virtual reality and a rehabilitation decision method thereof, comprising the following steps:
[0041] Enter the rehabilitation training system interface and fill in the participant information;
[0042] Select training mode and scenario tasks based on the patient's actual ankle joint motion angle information;
[0043] In passive mode, the rehabilitation movement trajectory is formulated according to the scene task, and in active mode, the rehabilitation movement is carried out according to the scene task;
[0044] Training scores and rehabilitation evaluations are given at the end of each round of training;
[0045] After the training, adaptive rehabilitation decisions are made based on the patient's training situation and the training strategy is adjusted.
[0046] Combine the following Figures 1 to 6 The embodiment of the present invention is a rehabilitation decision-making method based on a virtual reality rehabilitation training system, which is as follows:
[0047] like Figure 1 As shown, the present invention provides ankle dorsiflexion / plantar flexion corresponding direction movement training and three-degree-of-freedom ankle joint training. From daily life scenes to challenging aerial scenes, from fully passive training driven by robots to patient-controlled robot autonomous training, the whole process is both interesting and can make the rehabilitation training process step by step. The basic implementation process of the three scenes is as follows Figure 1 shown.
[0048] like Figure 2The figure shows the working principle of the robot system. The designed virtual training scenes can be set to a completely passive training mode, that is, the movement trajectory of the ankle robot is designed according to the task path of the scene, and the patient is completely driven by the robot. The path designed in the virtual scene includes forward, backward, left turn, right turn and other actions, which basically cover the movement of the ankle joint in daily walking, making it easy for patients to switch from virtual behavior to real life.
[0049] like Figure 3 The figure shows the whole rehabilitation training system process. The system is mainly divided into flexible ankle rehabilitation robot, control system hardware module, host computer display control module, virtual reality communication module, virtual environment module and rehabilitation decision module.
[0050] The flexible ankle joint rehabilitation robot, the virtual reality communication module, and the virtual environment module are connected in sequence;
[0051] The control system hardware module is connected to the flexible ankle joint rehabilitation robot;
[0052] The flexible ankle joint rehabilitation robot is connected to the host computer display control module;
[0053] The rehabilitation decision module is connected to the virtual environment module;
[0054] The flexible ankle joint rehabilitation robot is bound to the patient's ankle joint to assist the patient's ankle joint in rotating at different degrees of freedom;
[0055] The control system hardware module is used to drive the actual movement of the flexible ankle joint rehabilitation robot and perform real-time adjustments according to the control instructions issued by the host computer;
[0056] The host computer display control module is used to display the joint motion angle information of the patient's ankle in three different degrees of freedom, and can achieve a real-time trajectory tracking effect;
[0057] In the upper computer display control module, the user can select a rehabilitation mode, namely, a passive mode and an active mode. The passive mode is the active control of the flexible ankle joint rehabilitation robot based on a virtual task, and the active mode is the user's active control driven by a virtual task. Combined with the patient's ankle joint movement ability and the scene task difficulty planning, the auxiliary control of the flexible ankle joint rehabilitation robot under different scene tasks is realized;
[0058] The virtual reality communication module is used to input the angle change information of the ankle joint into the virtual scene in real time, and its functions include uploading the motion signal of the patient's ankle to the host computer and establishing a server and a client to realize the communication between the host computer and the virtual scene;
[0059] The virtual environment module is used to provide scene feedback during patient rehabilitation training and realize virtual reality interaction;
[0060] The rehabilitation decision module analyzes the electromyographic signals and activity range of the patient during rehabilitation training through a fuzzy neural network rehabilitation training decision model improved based on the cuckoo algorithm, establishes a mapping model, and realizes real-time adjustment of the speed and range of the rehabilitation training scene action;
[0061] The various patient motion state information collected by the flexible ankle joint rehabilitation robot is used as the basis for rehabilitation decision-making to reasonably design rehabilitation training tasks. The host computer display control module provides an interactive interface and realizes robot control, giving the flexible ankle joint rehabilitation robot a motion trajectory. The virtual reality communication module realizes communication between the flexible ankle joint rehabilitation robot and the host computer display control module, and between the host computer display control module and the virtual environment module, so as to ensure natural interaction between the patient and the virtual scene.
[0062] The virtual environment module can provide patients with an interesting and immersive virtual training environment and different training modes, allowing patients to get rid of the traditional boring training methods;
[0063] The control system hardware module obtains relevant control instructions according to specific training requirements and calculations, and controls the 3-DOF ankle rehabilitation robot dynamic platform by controlling the pneumatic muscle driver, so as to realize the inversion / eversion, adduction / abduction movement, and dorsiflexion / plantar flexion movement of the ankle joint, so as to assist the patient in corresponding sports rehabilitation training;
[0064] The flexible ankle rehabilitation robot is equipped with an angle sensor on each of its three axes for recording the angle changes of the ankle joint in real time. The virtual reality real-time communication module can transmit the angle change information of the ankle joint to the virtual scene, and drive the objects in the rehabilitation game to perform corresponding movements according to the formulated corresponding movement strategies, thereby realizing virtual reality synchronization and improving the fun and immersion of the training.
[0065] The virtual environment module includes a user information module and three training scenarios, namely, a park roaming scenario, an indoor storage scenario and an aircraft combat scenario. From daily life scenarios to challenging aerial scenarios, from fully passive robot-driven training to patient-controlled robot autonomous training, the whole process is both interesting and can make the rehabilitation training process step by step.
[0066] The host computer displays the control module. In the display module, the real-time motion trajectory of the patient's ankle can be seen. Here, the motion information of the flexible ankle joint rehabilitation robot is decomposed into three axial motions of X, Y, and Z. There are two curves on the curve interface of each axis. The red curve is the expected trajectory of the flexible ankle joint rehabilitation robot, and the blue curve is its actual motion trajectory. The real-time trajectory tracking effect can be seen. The control module can be used to set and input relevant motion parameters, such as motion mode, motion range, motion speed, etc., and perform various control functions, such as initialization, stop, etc.
[0067] The virtual reality real-time communication module includes two parts. The first part is the communication between the flexible ankle joint rehabilitation robot and the upper computer display control module. The data acquisition card can be used as a communication medium to upload the patient's ankle motion signal to the upper computer display control module, or the upper computer display control module converts the calculated control amount into an analog voltage signal and transmits it to the pneumatic proportional valve, thereby realizing the control of the flexible ankle joint rehabilitation robot;
[0068] The second part is the communication between the host computer display control module and the virtual environment module. The server and the client are established, and the movement signal of the patient's ankle is transmitted to the virtual environment module in real time through TCP network communication to perform synchronous control of the virtual avatar, thereby realizing human-computer interaction in virtual reality and enhancing the interest of patient rehabilitation training.
[0069] The training modes are divided into passive training and active training. The passive mode is the active control of the robot based on the virtual task, and the active mode is the active control of the user driven by the virtual task. The upper computer display control module and the rehabilitation decision model are combined to analyze and evaluate the user's ankle joint movement ability and the difficulty of the scene task. After planning, the control system hardware module can set the expected trajectory to achieve auxiliary control of the robot under different scene tasks.
[0070] The virtual environment module enables the patient to perform ankle joint rehabilitation training exercises while completing game tasks. During the exercise, the upper computer display control module can collect electromyographic data and angle data, display the patient's current rehabilitation status through rehabilitation evaluation, and automatically update the evolution training scene under the adaptive rehabilitation decision model, so that the exercise load can be regulated in real time, realizing a closed loop of virtual reality rehabilitation training-rehabilitation evaluation-adaptive rehabilitation training decision-virtual reality rehabilitation training.
[0071] When making virtual reality rehabilitation training decisions, it is necessary to efficiently and accurately establish the connection between the patient's movement state and the parameters in the virtual training scene. This goal is achieved by the virtual reality communication module, that is, establishing a nonlinear mapping relationship between input and human output data, and the flexible ankle joint rehabilitation robot provides assistance to the patient's ankle joint. Through the rehabilitation decision model, a rehabilitation training system with real-time regulation of motion load is established, which can enable patients to conduct personalized and adaptive rehabilitation training.
[0072] like Figure 4 The figure shows the ankle joint rehabilitation training system software module, which may be composed of a virtual scene software module, a real-time data display and control module, a virtual reality communication module, etc.
[0073] like Figure 5 As shown in the figure, this paper applies the fuzzy neural network algorithm based on cuckoo optimization to virtual reality rehabilitation training decision-making, and establishes a CS-FNN virtual reality rehabilitation training decision-making model. In the cuckoo-optimized fuzzy neural network, the main steps are the same as those of the fuzzy neural network. When updating the weights, the gradient descent method is replaced by the cuckoo optimization algorithm. From improving the convergence of the fuzzy neural network and improving the accuracy of the fuzzy neural network, the fuzzy neural network based on cuckoo optimization is realized. The main processes are: cluster center initialization, grouping according to the cluster center principle, updating the cluster center, initializing the weights, finding the Gaussian membership width and center, finding the optimal bird's nest, Levy flight to update the bird's nest position, finding the optimal bird's nest, outputting the optimal solution when the conditions are met, and updating the weights. The workflow of the fuzzy neural network based on cuckoo optimization is as follows: Figure 5 shown.
[0074] like Figure 6 As shown in FIG. 1 , a rehabilitation decision method applied to the above-mentioned virtual reality rehabilitation training system is used for virtual reality rehabilitation training decision making by using a fuzzy neural network algorithm based on cuckoo optimization, and a CS-FNN virtual reality rehabilitation training decision model is established. The model structure and workflow are shown in FIG. Figure 6 As shown, the specific steps include:
[0075] Step 1: Calculate the ankle joint range of motion and use the ROM values of the three motion axes of the ankle joint to form the joint range of motion feature matrix A j ;
[0076] Step 2: Calculate the average electromyography (AEMG) of the four muscles of the lower limbs to form A m ;
[0077]
[0078] In the above formula, x mis the mth electromyographic signal value. j and A m Construct the patient's rehabilitation training motion state characteristic matrix A;
[0079] A=[A j ,A m ]
[0080] Step 3: Substitute the motion state feature matrix A as training data into the CS-FNN model for training, and adjust the parameters of the virtual reality training scene in real time, that is, the speed and range of motion required for rehabilitation training. The control parameter set C of the virtual reality rehabilitation training scene is as follows:
[0081] C=[V i ,R j ](i=1,2,j=1,2,)
[0082] In the above formula, V i is the scene action speed control coefficient, i value range is [1,2], the larger the i value, the faster the scene action control speed, representing slow and fast respectively, R j is the scene motion range control coefficient, the j value range is [1,2]. The larger the j value, the larger the scene motion range. The two values represent a small range and a large range respectively.
[0083] Step 4: The ankle joint activity feature matrix A during the patient's rehabilitation training j and the average value of the electromyographic signal A m Normalize them separately and combine them to form the input X of the CS-FNN rehabilitation training decision model and the virtual scene action speed control coefficient V i and range control coefficient R j The combination is used as the model output Y;
[0084] X=[A j ,A m ]
[0085] Y=[V i ,R j ]
[0086] Step 5: Calculate the fitness function of the i-th cuckoo nest by the following formula f. In CS-FNN, the fitness function f is the sum of the absolute values of the errors between the actual output value and the predicted output value.
[0087]
[0088] In the above formula, N represents the number of times, and δ(t) represents the movement state index of the patient during rehabilitation training, including the CS-FNN training objective function constructed by the angular acceleration and the change in joint activity, δ(t) is as follows:
[0089] δ(t)=min(μ·ΔA j +λ·ΔA m )
[0090] In the above formula, ΔA j Represents the change in joint range of motion, ΔA m Represents the change in angular acceleration;
[0091]
[0092]
[0093] in is the range of motion of the joints at the initial stage of rehabilitation training. is the angular acceleration of the initial state, both are normalized matrices, μ and λ represent the weight coefficients of the cost function for the two input eigenvalues;
[0094] CS-FNN is used to construct a difficulty relationship mapping from the ankle joint motion state of rehabilitation patients to virtual reality rehabilitation training scenes. CS is used to optimize the width and center of the Gaussian membership function in the FNN algorithm. The optimized fuzzy neural network model is used to control the range and speed of task movements in virtual rehabilitation training scenes, so that the difficulty requirements of rehabilitation training tasks can adapt to the patient's motor ability.
[0095] It should be understood that parts not elaborated in detail in this specification belong to the prior art.
[0096] It should be understood that the above description of the preferred embodiment is relatively detailed and cannot be regarded as limiting the scope of patent protection of the present invention. Under the enlightenment of the present invention, ordinary technicians in this field can also make substitutions or modifications without departing from the scope of protection of the claims of the present invention, which all fall within the scope of protection of the present invention. The scope of protection requested for the present invention shall be based on the attached claims.
Claims
1. A rehabilitation decision-making method based on a virtual reality rehabilitation training system, characterized in that: The virtual reality rehabilitation training system includes: a flexible ankle joint rehabilitation robot, a control system hardware module, a host computer display control module, a virtual reality communication module, a virtual environment module, and a rehabilitation decision module; The flexible ankle joint rehabilitation robot, the virtual reality communication module, and the virtual environment module are connected in sequence; The control system hardware module is connected to the flexible ankle joint rehabilitation robot; The flexible ankle joint rehabilitation robot is connected to the host computer display control module; The rehabilitation decision module is connected to the virtual environment module; The flexible ankle joint rehabilitation robot is bound to the patient's ankle joint to assist the patient's ankle joint in rotating at different degrees of freedom; The control system hardware module is used to drive the actual movement of the flexible ankle joint rehabilitation robot and perform real-time adjustments according to the control instructions issued by the host computer; The host computer display control module is used to display the joint motion angle information of the patient's ankle in three different degrees of freedom, and can achieve a real-time trajectory tracking effect; In the upper computer display control module, the user can select a rehabilitation mode, namely, a passive mode and an active mode. The passive mode is the active control of the flexible ankle joint rehabilitation robot based on a virtual task, and the active mode is the user's active control driven by a virtual task. Combined with the patient's ankle joint movement ability and the scene task difficulty planning, the auxiliary control of the flexible ankle joint rehabilitation robot under different scene tasks is realized; The virtual reality communication module is used to input the angle change information of the ankle joint into the virtual scene in real time, and its functions include uploading the motion signal of the patient's ankle to the host computer and establishing a server and a client to realize the communication between the host computer and the virtual scene; The virtual environment module is used to provide scene feedback during patient rehabilitation training and realize virtual reality interaction; The rehabilitation decision module analyzes the electromyographic signals and activity range of the patient during rehabilitation training through a fuzzy neural network rehabilitation training decision model improved based on the cuckoo algorithm, establishes a mapping model, and realizes real-time adjustment of the speed and range of the rehabilitation training scene action; The virtual environment module enables the patient to perform ankle joint rehabilitation training while completing the game tasks. During the exercise, the upper computer display control module can collect electromyographic data and angle data, display the patient's current rehabilitation status through rehabilitation evaluation, and automatically update the evolution training scene under the adaptive rehabilitation decision model, so that the exercise load can be regulated in real time, realizing the closed loop of virtual reality rehabilitation training-rehabilitation evaluation-adaptive rehabilitation training decision-virtual reality rehabilitation training; When making virtual reality rehabilitation training decisions, it is necessary to efficiently and accurately establish the connection between the patient's movement state and the parameters in the virtual training scene. This goal is achieved by the virtual reality communication module, that is, establishing a nonlinear mapping relationship between input and human output data, and the flexible ankle joint rehabilitation robot provides assistance to the patient's ankle joint. Through the rehabilitation decision model, a rehabilitation training system with real-time regulation of motion load is established, which can enable patients to conduct personalized and adaptive rehabilitation training.
2. The rehabilitation decision-making method based on the virtual reality rehabilitation training system according to claim 1, characterized in that: The various patient motion state information collected by the flexible ankle joint rehabilitation robot is used as the basis for rehabilitation decision-making to reasonably design rehabilitation training tasks. The host computer display control module provides an interactive interface and realizes robot control, giving the flexible ankle joint rehabilitation robot a motion trajectory. The virtual reality communication module realizes communication between the flexible ankle joint rehabilitation robot and the host computer display control module, and between the host computer display control module and the virtual environment module, so as to ensure natural interaction between the patient and the virtual scene.
3. The rehabilitation decision-making method based on the virtual reality rehabilitation training system according to claim 1, characterized in that: The virtual environment module can provide patients with an interesting and immersive virtual training environment and different training modes, allowing patients to get rid of the traditional boring training methods; The control system hardware module obtains relevant control instructions according to specific training requirements and calculations, and controls the 3-DOF ankle rehabilitation robot dynamic platform by controlling the pneumatic muscle driver, so as to realize the inversion / eversion, adduction / abduction movement, and dorsiflexion / plantar flexion movement of the ankle joint, so as to assist the patient in corresponding sports rehabilitation training; The flexible ankle rehabilitation robot is equipped with an angle sensor on each of its three axes for recording the angle changes of the ankle joint in real time. The virtual reality real-time communication module can transmit the angle change information of the ankle joint to the virtual scene, and drive the objects in the rehabilitation game to perform corresponding movements according to the formulated corresponding movement strategies, thereby realizing virtual reality synchronization and improving the fun and immersion of the training.
4. The rehabilitation decision-making method based on the virtual reality rehabilitation training system according to claim 1, characterized in that: The virtual environment module includes a user information module and three training scenarios: a park roaming scenario, an indoor storage scenario, and an aircraft combat scenario. From daily life scenarios to challenging aerial scenarios, from fully passive robot-driven training to patient-controlled robot autonomous training, the whole process is both interesting and can make the rehabilitation training process step by step.
5. The rehabilitation decision-making method based on the virtual reality rehabilitation training system according to claim 1, characterized in that: The host computer displays the control module. In the display module, the real-time motion trajectory of the patient's ankle can be seen. Here, the motion information of the flexible ankle rehabilitation robot is decomposed into three axial movements of X, Y, and Z. There are two curves on the curve graph interface of each axis. The red curve is the expected trajectory of the flexible ankle rehabilitation robot, and the blue curve is its actual motion trajectory. The real-time trajectory tracking effect can be seen; the control module can be used to set and input relevant motion parameters, such as motion mode, motion range, motion speed, and perform various control functions, such as initialization and stop.
6. The rehabilitation decision-making method based on the virtual reality rehabilitation training system according to claim 1, characterized in that: The virtual reality real-time communication module includes two parts. The first part is the communication between the flexible ankle joint rehabilitation robot and the upper computer display control module. The data acquisition card can be used as a communication medium to upload the patient's ankle motion signal to the upper computer display control module, or the upper computer display control module converts the calculated control amount into an analog voltage signal and transmits it to the pneumatic proportional valve, thereby realizing the control of the flexible ankle joint rehabilitation robot; The second part is the communication between the host computer display control module and the virtual environment module. The server and the client are established, and the movement signal of the patient's ankle is transmitted to the virtual environment module in real time through TCP network communication to perform synchronous control of the virtual avatar, thereby realizing human-computer interaction in virtual reality and enhancing the interest of patient rehabilitation training.
7. The rehabilitation decision-making method based on the virtual reality rehabilitation training system according to claim 3, characterized in that: The training modes are divided into passive training and active training; the passive mode is the active control of the robot based on virtual tasks, and the active mode is the active control of the user driven by virtual tasks; the upper computer display control module and the rehabilitation decision model are combined to analyze and evaluate the user's ankle joint movement ability and the difficulty of the scene task. After planning, the control system hardware module can set the expected trajectory to achieve auxiliary control of the robot under different scene tasks.
Citation Information
Cited By
Severe illness rehabilitation treatment decision-making method based on virtual reality
CN121439126A
Bilateral difference cooperation-based hemiplegic upper limb mirror image induction rehabilitation method and system
CN121588427A
