ICU-AW rehabilitation robot remote cooperation and adaptive control system based on 5G network

Abstract

The invention provides an ICU-AW rehabilitation robot remote cooperation and adaptive control system based on a 5G network, and belongs to the technical field of visual inspection, and the system comprises an operation simulation state evaluation module, an ICU environment monitoring module, an initial plan simulation execution module and an initial plan judgment and adjustment module. The patient state monitoring assembly collects initial muscle oxygen state data, and the rehabilitation robot simulates and verifies a muscle oxygen state evaluation value according to the initial muscle oxygen state data and uploads the muscle oxygen state evaluation value to the rehabilitation control platform; meanwhile, the ICU environment monitoring assembly obtains environment state data, evaluates an environment characteristic influence coefficient and then uploads the environment state data, the rehabilitation control platform matches an initial plan according to a muscle oxygen simulation state evaluation value, remotely cooperates with the rehabilitation robot to simulate execution, obtains execution data, evaluates an execution quality index according to the environment characteristic influence coefficient, and compares the execution quality index with an expected index; and judging whether self-adaptive adjustment is carried out or not, thereby realizing remote cooperation and self-adaptive control of the ICU-AW rehabilitation robot.

Description

Technical Field

[0001] The present invention relates to the technical field of visual detection, and particularly to a remote collaboration and adaptive control system for an ICU-AW rehabilitation robot based on a 5G network. Background Art

[0002] When critically ill patients are immobilized in bed / mechanically ventilated for more than 5-7 days, they are extremely prone to developing intensive care unit-acquired weakness (ICU-AW), with an incidence rate that can be as high as 85%, and a hospital mortality rate as high as 27%. Currently, the clinical intervention effect is poor and the long-term prognosis is unfavorable. The United States pays $16 billion for it every year. ICU-AW patients have multiple system weaknesses occurring simultaneously, including brain weakness, cardiopulmonary weakness, and trunk and limb weakness, resulting in cognitive communication disorders, activity intolerance, and impaired executive function. Therefore, rehabilitation requires overall coverage of the above problems, and none can be missing.

[0003] Existing rehabilitation assessment / intervention means cannot meet the clinical needs, mainly having three problems: lacking, being few, and being poor. "Lacking" means the absence of an assessment system and difficulty in early diagnosis. "Being few" means that ICU-AW patients are in a special state of illness and the ICU environment, such as sedation, intubation, etc., and there are few rehabilitation means suitable for special conditions. "Being poor" means that the coordination of the rehabilitation prescriptions for patients is poor. The existing patchwork prescriptions have miscellaneous contents and rely mostly on manpower, with low efficiency. Summary of the Invention

[0004] The present invention provides a remote collaboration and adaptive control system for an ICU-AW rehabilitation robot based on a 5G network, which solves the problems of the absence of an assessment system, the particularity of the environment, and the low efficiency relying mostly on manpower in the prior art.

[0005] To achieve the above-mentioned invention objectives, the technical solutions provided by the present invention are as follows:

[0006] A remote cooperation and adaptive control system for an ICU-AW rehabilitation robot based on a 5G network, comprising: a muscle oxygen simulation state evaluation module, which is used for a patient state monitoring component to perform initial state monitoring on an ICU-AW patient, collect the initial muscle oxygen state data of the patient, the rehabilitation robot simulates the muscle oxygen state of the patient according to the initial muscle oxygen state data, determines the muscle oxygen simulation state evaluation value of the rehabilitation robot, and verifies it with a predefined muscle oxygen simulation state evaluation threshold, so as to determine whether to optimize the muscle oxygen simulation state of the rehabilitation robot, and upload the muscle oxygen simulation state evaluation value of the rehabilitation robot to a rehabilitation control platform through a 5G network; an ICU environment monitoring module, which is used for an ICU environment monitoring component to monitor the ICU environment where the rehabilitation robot is located, obtain the ICU environment state data where the rehabilitation robot is located, evaluate the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, and upload it to the rehabilitation control platform through a 5G network; an initial plan simulation execution module, which is used for the rehabilitation control platform to match and obtain an initial rehabilitation control plan according to the muscle oxygen simulation state evaluation value of the rehabilitation robot, so that the rehabilitation control platform simulates and executes the initial rehabilitation control plan through remote cooperation with the rehabilitation robot, and obtains the execution data of the initial rehabilitation control plan; an initial plan determination and adjustment module, which is used for the rehabilitation control platform to evaluate the execution quality index of the initial rehabilitation control plan according to the execution data of the initial rehabilitation control plan and the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, compare it with a predefined execution quality expectation index, so as to determine whether to adaptively adjust the initial rehabilitation control plan, and finally complete the remote cooperation and adaptive control of the ICU-AW rehabilitation robot.

[0007] Optionally, the determination process of the myo-oxygen simulation state evaluation value of the rehabilitation robot is as follows: The rehabilitation robot simulates the myo-oxygen state of the patient according to the initial myo-oxygen state data to obtain the myo-oxygen simulation state data of the rehabilitation robot, specifically including the output power of the rehabilitation robot at each myo-oxygen simulation time point, the number of repetitions of the set actions of the rehabilitation robot within the myo-oxygen simulation cycle, the response delay duration of the rehabilitation robot within the myo-oxygen simulation cycle, and the average value of the sensor linearity error of the rehabilitation robot within the myo-oxygen simulation cycle; The output powers of the rehabilitation robot at each myo-oxygen simulation time point are averaged to obtain the average output power of the rehabilitation robot within the myo-oxygen simulation cycle; The electromagnetic radiation intensity values of the environment where the rehabilitation robot is located at each myo-oxygen simulation time point are obtained; The average value of the output power adaptation and the number of repetitions of the set action adaptation are extracted from the rehabilitation control information library; The average output power of the rehabilitation robot within the myo-oxygen simulation cycle, the number of repetitions of the set actions of the rehabilitation robot within the myo-oxygen simulation cycle, the response delay duration of the rehabilitation robot within the myo-oxygen simulation cycle, the average value of the sensor linearity error of the rehabilitation robot within the myo-oxygen simulation cycle, and the electromagnetic radiation intensity values of the environment where the rehabilitation robot is located at each myo-oxygen simulation time point are comprehensively processed to obtain the myo-oxygen simulation state evaluation value of the rehabilitation robot.

[0008] Optionally, to determine whether to optimize the myo-oxygen simulation state of the rehabilitation robot, specifically, the myo-oxygen simulation state evaluation value of the rehabilitation robot is verified with a predefined myo-oxygen simulation state evaluation threshold to obtain a verification result, and based on the verification result, it is determined whether to optimize the myo-oxygen simulation state of the rehabilitation robot; The verification result is the first verification result or the second verification result; The first verification result is specifically that the myo-oxygen simulation state evaluation value of the rehabilitation robot is greater than or equal to the myo-oxygen simulation state evaluation threshold; The second verification result is specifically that the myo-oxygen simulation state evaluation value of the rehabilitation robot is less than the myo-oxygen simulation state evaluation threshold; If the verification result shows the first verification result, there is no need to optimize the myo-oxygen simulation state of the rehabilitation robot, and if the verification result shows the second verification result, it is necessary to optimize the myo-oxygen simulation state of the rehabilitation robot.

[0009] Optionally, the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located is evaluated. The specific evaluation process is as follows: The environmental state data of the ICU where the rehabilitation robot is located specifically includes the real-time environmental temperature of the ICU where the rehabilitation robot is located during the evaluation period, the real-time environmental humidity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time noise decibel value of the ICU where the rehabilitation robot is located during the evaluation period, and the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located during the evaluation period. The environmental temperature adaptation value, environmental humidity adaptation value, and carbon dioxide concentration adaptation value are extracted from the rehabilitation control information library. The real-time environmental temperature of the ICU where the rehabilitation robot is located during the evaluation period, the real-time environmental humidity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time noise decibel value of the ICU where the rehabilitation robot is located during the evaluation period, and the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located during the evaluation period are comprehensively analyzed to obtain the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located.

[0010] Optionally, the specific analysis method of the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located is as follows:

[0011]

[0012] In the formula, HJ is the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, t is the time variable, t ∈ [t0, t1], t0 is the start time point of the evaluation period, t1 is the end time point of the evaluation period, β(t) is the real-time environmental temperature of the ICU where the rehabilitation robot is located at time t during the evaluation period, β′ is the environmental temperature adaptation value, γ(t) is the real-time environmental humidity of the ICU where the rehabilitation robot is located at time t during the evaluation period, γ′ is the environmental humidity adaptation value, δ(t) is the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located at time t during the evaluation period, θ(t) is the real-time noise decibel value of the ICU where the rehabilitation robot is located at time t during the evaluation period, τ(t) is the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located at time t during the evaluation period, τ′ is the carbon dioxide concentration adaptation value, YX is the myo-oxygen simulation state evaluation value of the rehabilitation robot, b1 is the environmental characteristic parameter corresponding to the real-time electromagnetic interference intensity predefined in the rehabilitation control information library, b2 is the environmental characteristic parameter corresponding to the real-time noise decibel value predefined in the rehabilitation control information library, and k1 is the environmental characteristic parameter corresponding to the myo-oxygen simulation state evaluation value predefined in the rehabilitation control information library.

[0013] Optionally, the initial rehabilitation control plan is obtained through matching. The specific matching process is as follows: The myo-oxygen simulation state evaluation value of the rehabilitation robot is matched with the initial rehabilitation control plans corresponding to the pre-defined myo-oxygen simulation state evaluation value intervals. The specific matching process is as follows: Extract the mapping set between the myo-oxygen simulation state evaluation value of the rehabilitation robot and the initial rehabilitation control plans from the rehabilitation control information database, determine the interval to which the myo-oxygen simulation state evaluation value of the rehabilitation robot belongs, and obtain the initial rehabilitation control plan corresponding to this interval, thereby obtaining the initial rehabilitation control plan through matching.

[0014] Optionally, the execution data of the initial rehabilitation control plan specifically includes the number of completed rehabilitation training sessions within the execution cycle of the initial rehabilitation control plan, the feedback response duration within the execution cycle of the initial rehabilitation control plan, the number of faults of the rehabilitation robot within the execution cycle of the initial rehabilitation control plan, and the final simulated myo-oxygen saturation of the rehabilitation robot within the execution cycle of the initial rehabilitation control plan; Obtain the initial simulated myo-oxygen saturation of the rehabilitation robot during the myo-oxygen simulation cycle; Obtain the pre-set number of rehabilitation training sessions within the execution cycle of the initial rehabilitation control plan from the rehabilitation control information database; Perform a ratio process on the number of completed rehabilitation training sessions within the execution cycle of the initial rehabilitation control plan and the pre-set number of rehabilitation training sessions within the execution cycle of the initial rehabilitation control plan to obtain the simulated completion rate of rehabilitation training within the execution cycle of the initial rehabilitation control plan.

[0015] Optionally, the execution quality index of the initial rehabilitation control plan is specifically analyzed as follows: Match the number of action repetitions of the rehabilitation robot during the myo-oxygen simulation cycle with the defined number of fault definition times corresponding to each action repetition number interval to obtain the number of fault definition times of the rehabilitation robot through such matching; Match the response delay duration of the rehabilitation robot during the myo-oxygen simulation cycle with the defined feedback response defined duration corresponding to each response delay duration interval to obtain the feedback response defined duration of the initial rehabilitation control plan; Comprehensively analyze the simulated completion rate of rehabilitation training within the execution cycle of the initial rehabilitation control plan, the feedback response duration within the execution cycle of the initial rehabilitation control plan, the feedback response defined duration of the initial rehabilitation control plan, the number of faults of the rehabilitation robot within the execution cycle of the initial rehabilitation control plan, the number of fault definition times of the rehabilitation robot, the final simulated myo-oxygen saturation of the rehabilitation robot within the execution cycle of the initial rehabilitation control plan, the initial simulated myo-oxygen saturation of the rehabilitation robot during the myo-oxygen simulation cycle, and the environmental characteristic influence coefficient of the ICU to which the rehabilitation robot belongs to obtain the execution quality index of the initial rehabilitation control plan.

[0016] Optionally, the determination of whether to adaptively adjust the initial rehabilitation control plan is as follows: If the execution quality index of the initial rehabilitation control plan is equal to the default execution quality index of the initial rehabilitation control plan, then execute the predefined optimization and adjustment plan for the initial rehabilitation control plan; if the execution quality index of the initial rehabilitation control plan is not equal to the default execution quality index of the initial rehabilitation control plan, compare the execution quality index of the initial rehabilitation control plan with the predefined execution quality reference index. If the execution quality index of the initial rehabilitation control plan is greater than the execution quality reference index, then there is no need to adaptively adjust the initial rehabilitation control plan. If the execution quality index of the initial rehabilitation control plan is less than or equal to the execution quality reference index, then it is necessary to adaptively adjust the initial rehabilitation control plan.

[0017] Optionally, the adaptive adjustment of the initial rehabilitation control plan is as follows: Calculate the difference between the execution quality index of the initial rehabilitation control plan and the execution quality reference index to obtain the execution quality deviation value of the initial rehabilitation control plan, and match it with the adaptive adjustment plans corresponding to the predefined execution quality deviation value intervals, so as to obtain the adaptive adjustment plan of the initial rehabilitation control plan, and finally adaptively adjust the initial rehabilitation control plan.

[0018] The technical solution provided by the present invention has at least the following beneficial effects compared with the prior art:

[0019] In the above solution, the initial muscle oxygen state data of the patient is collected by the patient status monitoring component, and the rehabilitation robot simulates and verifies the muscle oxygen state evaluation value based on this and uploads it to the rehabilitation control platform; at the same time, the ICU environment monitoring component obtains the environmental state data, evaluates the environmental characteristic influence coefficient and uploads it. The rehabilitation control platform matches the initial plan according to the muscle oxygen simulation state evaluation value, and the remote collaborative rehabilitation robot simulates and executes to obtain the execution data, comprehensively evaluates the execution quality index with the environmental characteristic influence coefficient, compares it with the expected index, and determines whether to adaptively adjust, so as to realize the remote collaboration and adaptive control of the ICU-AW rehabilitation robot.

[0020] By collecting the initial muscle oxygen state data of the patient, the rehabilitation robot simulates the muscle oxygen state of the patient according to the initial muscle oxygen state data, determines the muscle oxygen simulation state evaluation value of the rehabilitation robot, helps the rehabilitation robot accurately simulate the oxygenation of the patient's muscles, and according to the individual differences of the patient, the rehabilitation robot can simulate the best rehabilitation plan as accurately as possible for the patient, improving the adaptability and effectiveness of the rehabilitation robot.

[0021] By obtaining the environmental status data of the ICU where the rehabilitation robot is located and evaluating the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, environmental factors that may affect the performance of the rehabilitation robot can be identified in advance, and corresponding safeguard measures can be taken in a timely manner, so that the rehabilitation robot can work in a suitable environment, keep its performance stable, help reduce robot failures caused by environmental factors, and ensure the smooth progress of subsequent rehabilitation simulation training.

[0022] By evaluating the execution quality index of the initial rehabilitation control plan and comparing it with the predefined expected execution quality index, it can be determined whether to adaptively adjust the initial rehabilitation control plan. Problems that may exist in the rehabilitation simulation plan can be discovered in a timely manner, so as to make different degrees of adaptive adjustment plans, improve the simulation effect of the rehabilitation robot, make the initial rehabilitation control plan more suitable for the patient's condition, and improve the feasibility of the plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0024] Figure 1 It is a schematic diagram of the system modules related to the embodiments of the present invention;

[0025] Figure 2 It is a sensor linearity curve graph.

[0026] Reference numerals: 1, reference line; 2, sensor linearity curve. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0028] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings as understood by those of ordinary skill in the art to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Similarly, the similar terms such as "a", "an" or "the" do not denote a quantity limitation, but mean that there is at least one. The similar terms such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The similar terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0029] It should be noted that the "up", "down", "left", "right", "front", "back", etc. used in the present invention are only used to represent relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0030] Aiming at the problems of the lack of existing evaluation systems, the particularity of the environment, and the low efficiency relying on manpower, the present invention provides a 5G network-based ICU-AW rehabilitation robot remote collaboration and adaptive control system that can effectively improve the evaluation system, reduce the negative impacts brought by the special environment, and improve the control efficiency.

[0031] As Figure 1 As shown, an embodiment of the present invention provides a 5G network-based ICU-AW rehabilitation robot remote collaboration and adaptive control system, including: a muscle oxygen simulation state evaluation module, which is used for the patient state monitoring component to perform initial state monitoring on ICU-AW patients, collect the initial muscle oxygen state data of the patients, the rehabilitation robot simulates the muscle oxygen state of the patients according to the initial muscle oxygen state data, determines the muscle oxygen simulation state evaluation value of the rehabilitation robot, checks it against a predefined muscle oxygen simulation state evaluation threshold, thereby determining whether to optimize the muscle oxygen simulation state of the rehabilitation robot, and uploads the muscle oxygen simulation state evaluation value of the rehabilitation robot to the rehabilitation control platform through the 5G network.

[0032] The patient state monitoring component is a device system for comprehensively monitoring the physical state of ICU-AW (ICU-acquired weakness) patients. Its main function is to accurately collect the initial muscle oxygen state data of the patients, and may also monitor other physiological parameters of the patients, such as heart rate, blood pressure, respiratory rate, body temperature, etc., to provide basic data support for subsequent rehabilitation simulation plans. It can be a wearable real-time muscle oxygen content monitoring device, and uses wireless communication technology to send the data to the control system of the rehabilitation robot.

[0033] The ICU-AW rehabilitation robot is an intelligent device specifically designed to assist ICU-AW patients in their rehabilitation treatment. This robot combines technologies from multiple fields such as mechanical engineering, electronics, computer science, and rehabilitation medicine. It can provide precise and personalized rehabilitation training in the ICU environment for patients with debilities such as muscle weakness and limited joint mobility caused by factors like long-term bed rest, illness, or drug treatment. The mechanical structure design of the ICU-AW rehabilitation robot can simulate the natural movements of the human body and is equipped with sensors to monitor the myo-oxygen status of patients. The robot can strictly provide customized rehabilitation services for patients according to the simulation programs set by the rehabilitation control platform, such as training time, training frequency, and combinations of motion patterns.

[0034] The rehabilitation control platform is an integrated intelligent system that plays a core control and coordination role throughout the rehabilitation process. It is like the "brain center" of rehabilitation treatment, integrating data from various aspects, including patient status data, rehabilitation robot data, and ICU environment data, etc. Then, based on this data, it conducts analysis and decision-making, remotely collaboratively controls the work of the rehabilitation robot, and can adaptively adjust the rehabilitation plan to achieve the best rehabilitation effect.

[0035] The initial myo-oxygen status data of the patient can specifically be the myo-oxygen saturation at each collection time point of the patient, the oxygen uptake rate at each collection time point of the patient, and the muscle oxygen metabolism rate at each collection time point of the patient.

[0036] Each of the collection time points is specifically several collection time points obtained by dividing the collection cycle into time points. The division method can be 30 seconds. The determination of the collection cycle is obtained through comprehensive analysis by control managers based on factors such as the patient's status, equipment monitoring status, and specific collection requirements.

[0037] The rehabilitation robot simulates the myo-oxygen state of the patient according to the initial myo-oxygen state data. Specifically, the rehabilitation robot uses a data-driven machine learning model, such as a neural network model. The historical myo-oxygen state data of the patient is used as the training set of the model, and the neural network model is trained using the training set. The weights and biases of the network are adjusted through the backpropagation algorithm so that the model can learn the complex relationship between the input parameters and the myo-oxygen state. The initial myo-oxygen state data of the patient is used as the input of the neural network model, and the data is cleaned to remove possible noise and outliers. For example, instantaneous extremely high or low myo-oxygen saturation data points caused by poor sensor contact or external interference are identified and corrected or removed. Data normalization is performed to convert myo-oxygen data in different ranges into a standard interval for convenient subsequent model processing. For example, the value range of myo-oxygen saturation is converted from 0-100% to a numerical interval of 0-1, making the data comparable and consistent. The model calculates the simulated myo-oxygen state through the forward propagation algorithm based on the learned weights and biases. For example, in a neural network, the input data undergoes linear and non-linear transformations between layers, and finally parameters such as simulated myo-oxygen saturation and oxygen uptake rate are obtained at the output layer.

[0038] Among them, the rehabilitation robot simulates the myo-oxygen state of the patient according to the initial myo-oxygen state data and obtains the myo-oxygen simulation state data of the rehabilitation robot, which specifically includes the output power of the rehabilitation robot at each myo-oxygen simulation time point, the number of set action repetitions of the rehabilitation robot within the myo-oxygen simulation period, the response delay duration of the rehabilitation robot within the myo-oxygen simulation period, and the average value of the sensor linearity error of the rehabilitation robot within the myo-oxygen simulation period.

[0039] The myo-oxygen simulation period is specifically a period of time for the rehabilitation robot to simulate the patient's state according to the patient's data. Each myo-oxygen simulation time point is specifically several myo-oxygen simulation time points obtained by dividing the myo-oxygen simulation period. The division method can be 30 seconds. The determination of the myo-oxygen simulation period is obtained through comprehensive analysis by the control management personnel based on factors such as the simulation state of the rehabilitation robot, the simulation environment, and the actual simulation requirements.

[0040] The myo-oxygen simulation state data of the rehabilitation robot is specifically extracted from the myo-oxygen simulation report of the rehabilitation robot.

[0041] In a specific embodiment, the average value of the sensor linearity error can be obtained by applying a series of input physical quantities of different magnitudes to the sensor within the myo-oxygen simulation period. For example, different magnitudes of myo-oxygen saturation are applied to the myo-oxygen sensor, and at the same time, the output signal of the sensor, such as the current signal, is recorded. These input-output data are plotted into a curve to construct a sensor linearity curve graph, such as Figure 2As shown in the figure, the abscissa is the muscle oxygen saturation, with the unit of percentage, and the ordinate is the current signal, with the unit of ampere. Ideally, the sensor linearity curve should be a straight line. Therefore, a reference line 1 is located in the sensor linearity curve graph. In actual testing, the difference between the sensor linearity curve 2 and the reference line 1 can be calculated and averaged to obtain the average value of the sensor linearity error of the rehabilitation robot during the muscle oxygen simulation period.

[0042] The output powers of the rehabilitation robot at each muscle oxygen simulation time point are averaged to obtain the average output power of the rehabilitation robot during the muscle oxygen simulation period.

[0043] Obtain the electromagnetic radiation intensity values of the environment where the rehabilitation robot is located at each muscle oxygen simulation time point, where the electromagnetic radiation intensity values can be detected by an electromagnetic radiation detector.

[0044] It should be noted that the environment where the rehabilitation robot is located in this embodiment may be the same as or different from the ICU environment where the following rehabilitation robot is located.

[0045] Extract the average value of output power adaptation and the set number of action repetitions adaptation from the rehabilitation control information database.

[0046] Among them, the specific determination process of the muscle oxygen simulation state evaluation value of the rehabilitation robot is as follows:

[0047] The average output power of the rehabilitation robot during the muscle oxygen simulation period, the set number of action repetitions of the rehabilitation robot during the muscle oxygen simulation period, the response delay duration of the rehabilitation robot during the muscle oxygen simulation period, the average value of the sensor linearity error of the rehabilitation robot during the muscle oxygen simulation period, and the electromagnetic radiation intensity values of the environment where the rehabilitation robot is located at each muscle oxygen simulation time point are comprehensively processed to obtain the muscle oxygen simulation state evaluation value of the rehabilitation robot. The specific method is as follows:

[0048]

[0049] In the formula, YX is the muscle oxygen simulation state evaluation value of the rehabilitation robot, GL is the average output power of the rehabilitation robot during the muscle oxygen simulation period, GL ′ is the average value of output power adaptation, DZ is the set number of action repetitions of the rehabilitation robot during the muscle oxygen simulation period, DZ ′ is the set number of action repetitions adaptation, WY is the response delay duration of the rehabilitation robot during the muscle oxygen simulation period, WC is the average value of the sensor linearity error of the rehabilitation robot during the muscle oxygen simulation period, DF nIt is the electromagnetic radiation intensity value of the environment where the rehabilitation robot is located at the nth myo-oxygen simulation time point. Here, n is the number of each myo-oxygen simulation time point, n = 1, 2, 3,..., N, where N is the total number of myo-oxygen simulation time points. max represents the maximum value, a2 is the myo-oxygen simulation impact factor corresponding to the predefined response delay duration in the rehabilitation control information library, a3 is the myo-oxygen simulation impact factor corresponding to the predefined average value of the sensor linearity error in the rehabilitation control information library, a4 is the myo-oxygen simulation impact factor corresponding to the predefined maximum electromagnetic radiation intensity value in the rehabilitation control information library, and e is the natural constant.

[0050] In this embodiment, the myo-oxygen simulation state evaluation value of the rehabilitation robot is used to measure the accuracy and effectiveness of the rehabilitation robot in simulating the muscle oxygenation state of the patient during the simulation of the patient's limb movement. The higher the myo-oxygen simulation state evaluation value, the more accurately the rehabilitation robot grasps the myo-oxygen state of the patient, and the greater the help for subsequent operations.

[0051] It should be explained that the average output power refers to the average value of the power output by the rehabilitation robot during the myo-oxygen simulation cycle. For the rehabilitation robot, the output power reflects the amount of energy output per unit time during the process of simulating the patient's movement; the average output power adaptation value refers to the reference average value corresponding to the preset output power; the set number of action repetitions refers to the number of times the rehabilitation robot repeats the operation according to the preset simulation actions during the myo-oxygen simulation cycle. For example, in a myo-oxygen simulation cycle of upper limb rehabilitation simulation training, the preset action of the rehabilitation robot is to simulate the flexion and extension movement of the arm. From the state of straightening the arm to bending the arm so that the palm approaches the shoulder and then back to the straight state is one complete action. If the robot is set to complete such arm flexion and extension actions 30 times in this simulation cycle, then this 30 times is the set number of action repetitions; the set number of action repetition adaptation refers to the reference value corresponding to the predefined set number of action repetitions; the response delay duration refers to the time interval from when the rehabilitation robot receives the simulation instruction sent by the rehabilitation control platform to when the rehabilitation robot actually makes the simulation action during the myo-oxygen simulation cycle; the average sensor linearity error refers to the average value of the deviation between the actual output signal and the ideal linear output signal when the myo-oxygen sensor carried by the rehabilitation robot measures the myo-oxygen related physical quantities during the myo-oxygen simulation cycle; the electromagnetic radiation intensity value refers to the intensity of the electromagnetic radiation existing in the space environment where the rehabilitation robot is located during the myo-oxygen simulation cycle. This electromagnetic radiation includes the electromagnetic radiation generated by the operation of the rehabilitation robot's own electronic devices and the electromagnetic radiation emitted by other surrounding medical devices, electrical devices, etc.

[0052] Among them, the myo-oxygen simulation impact factor corresponding to the response delay duration, the myo-oxygen simulation impact factor corresponding to the average value of the sensor linearity error, and the myo-oxygen simulation impact factor corresponding to the maximum electromagnetic radiation intensity value are all obtained in advance in the rehabilitation control information library. The mapping relationship therein can be a one-to-one or many-to-one relationship. For example, the response delay duration forms a mapping set with the myo-oxygen simulation impact factor corresponding to the preset response delay duration in the rehabilitation control information library, and the real-time response delay duration is brought into the mapping set to obtain the myo-oxygen simulation impact factor corresponding to the response delay duration; the average value of the sensor linearity error forms a mapping set with the myo-oxygen simulation impact factor corresponding to the preset average value of the sensor linearity error in the rehabilitation control information library, and the real-time average value of the sensor linearity error is brought into the mapping set to obtain the myo-oxygen simulation impact factor corresponding to the average value of the sensor linearity error; the maximum electromagnetic radiation intensity value forms a mapping set with the myo-oxygen simulation impact factor corresponding to the preset maximum electromagnetic radiation intensity value in the rehabilitation control information library, and the real-time maximum electromagnetic radiation intensity value is brought into the mapping set to obtain the myo-oxygen simulation impact factor corresponding to the maximum electromagnetic radiation intensity value. In this embodiment, the value ranges of the myo-oxygen simulation impact factor corresponding to the response delay duration, the myo-oxygen simulation impact factor corresponding to the average value of the sensor linearity error, and the myo-oxygen simulation impact factor corresponding to the maximum electromagnetic radiation intensity value are all (0, 1).

[0053] It should be elaborated that during the process of the rehabilitation robot simulating the myo-oxygen state of the patient, if the output power of the rehabilitation robot is too high and deviates greatly from the adapted output power during this process, it may cause the rehabilitation robot to repeatedly execute the set actions, resulting in a significant increase in the number of times the set actions are executed, and making it also deviate from the adapted number of executions, thus greatly reducing the myo-oxygen simulation state of the rehabilitation robot. On the contrary, if the output power is low and much smaller than the adapted output power, there may be a situation where the output power is insufficient to support the execution of the set actions, resulting in the number of times the set actions are executed being much lower than the adapted number of executions, which will also reduce the myo-oxygen simulation state of the rehabilitation robot; if the average value of the sensor linearity error is large, the number of times of the actions set based on these inaccurate data may not be suitable for simulating the actual situation of the patient. For example, if the error of the myo-oxygen sensor is large, it may misjudge the oxygen metabolism ability of the patient's muscles, thus setting too high or too low number of times of actions, affecting the simulation effect of the rehabilitation robot and reducing the simulation state; if the electromagnetic radiation intensity value is too high, it may cause an increase in the response delay duration of the rehabilitation robot, and electromagnetic interference may affect the communication system and control system of the robot, slowing down the data transmission and processing speed, thus increasing the response delay duration and reducing the myo-oxygen simulation state of the rehabilitation robot.

[0054] Among them, determining whether to optimize the myo-oxygen simulation state of the rehabilitation robot specifically involves verifying the myo-oxygen simulation state evaluation value of the rehabilitation robot against a predefined myo-oxygen simulation state evaluation threshold to obtain a verification result, and determining whether to optimize the myo-oxygen simulation state of the rehabilitation robot based on the verification result.

[0055] The verification result is the first verification result or the second verification result.

[0056] Specifically, the first verification result is that the myo-oxygen simulation state evaluation value of the rehabilitation robot is greater than or equal to the predefined myo-oxygen simulation state evaluation threshold in the rehabilitation control information database.

[0057] Specifically, the second verification result is that the myo-oxygen simulation state evaluation value of the rehabilitation robot is less than the myo-oxygen simulation state evaluation threshold.

[0058] If the verification result shows the first verification result, there is no need to optimize the myo-oxygen simulation state of the rehabilitation robot. If the verification result shows the second verification result, it is necessary to optimize the myo-oxygen simulation state of the rehabilitation robot.

[0059] Specifically, optimizing the myo-oxygen simulation state of the rehabilitation robot involves using professional calibration equipment, such as a high-precision oxygen content standard gas source, a simulated tissue sample with known myo-oxygen saturation, etc., to calibrate the myo-oxygen sensor carried by the rehabilitation robot to improve the initial accuracy of myo-oxygen data collection; installing electromagnetic shielding materials, such as nickel-plated copper mesh, electromagnetic shielding foil, etc., on key parts such as the electronic equipment cabin and sensor housing of the rehabilitation robot to block the intrusion of external electromagnetic radiation and reduce the negative impact brought by electromagnetic radiation; adopting a redundant communication link design so that when the main link fails, the rehabilitation robot network system can automatically switch to the standby link within milliseconds to avoid myo-oxygen simulation interruption caused by network interruption; introducing a data verification and retransmission mechanism to perform integrity verification on each packet of simulated myo-oxygen data at the receiving end, and immediately request retransmission from the sending end once an error or missing data packet is found to ensure that the data received by the control platform is accurate.

[0060] The ICU environment monitoring module is used for the ICU environment monitoring component to monitor the ICU environment where the rehabilitation robot is located, obtain the ICU environment state data of the rehabilitation robot, evaluate the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, and upload it to the rehabilitation control platform through the 5G network.

[0061] The ICU environmental monitoring component is a device system specifically used to monitor the environmental conditions in the intensive care unit (ICU). Its main function is to obtain various physical and chemical parameters in the ICU environment in real time, providing data support for evaluating the impact of the environment on the rehabilitation robot and the initial rehabilitation control plan. The ICU environmental monitoring component receives various analog signals from various sensors, converts them into digital signals, and then sends the data to the rehabilitation control platform through wireless communication.

[0062] The various sensors can specifically be temperature sensors, humidity sensors, electromagnetic interference detection sensors, noise sensors, gas sensors, etc.

[0063] Among them, the specific evaluation process of the impact coefficient of the environmental characteristics of the ICU where the rehabilitation robot is located is as follows:

[0064] The environmental status data of the ICU where the rehabilitation robot is located specifically includes the real-time environmental temperature of the ICU where the rehabilitation robot is located during the evaluation period, the real-time environmental humidity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time noise decibel value of the ICU where the rehabilitation robot is located during the evaluation period, and the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located during the evaluation period.

[0065] The evaluation period is specifically a period of time used for the ICU environmental monitoring component to monitor the environment of the ICU where the rehabilitation robot is located. The determination of the evaluation period is obtained through comprehensive analysis by control managers based on factors such as the environmental status, the monitoring status of the component, and the actual monitoring requirements.

[0066] The ICU environmental status data can specifically be extracted from the monitoring report of the ICU environmental monitoring component.

[0067] The environmental temperature adaptation value, environmental humidity adaptation value, and carbon dioxide concentration adaptation value are extracted from the rehabilitation control information database.

[0068] The real-time environmental temperature of the ICU where the rehabilitation robot is located during the evaluation period, the real-time environmental humidity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time noise decibel value of the ICU where the rehabilitation robot is located during the evaluation period, and the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located during the evaluation period are comprehensively analyzed to obtain the impact coefficient of the environmental characteristics of the ICU where the rehabilitation robot is located. The specific method is as follows:

[0069]

[0070] Where HJ is the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, t is the time variable, t ∈ [t0, t1], t0 is the start time point of the evaluation period, t1 is the end time point of the evaluation period, β(t) is the real-time environmental temperature of the ICU where the rehabilitation robot is located at the moment t of the evaluation period, β′ is the environmental temperature adaptation value, γ(t) is the real-time environmental humidity of the ICU where the rehabilitation robot is located at the moment t of the evaluation period, γ′ is the environmental humidity adaptation value, δ(t) is the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located at the moment t of the evaluation period, θ(t) is the real-time noise decibel value of the ICU where the rehabilitation robot is located at the moment t of the evaluation period, τ(t) is the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located at the moment t of the evaluation period, τ′ is the carbon dioxide concentration adaptation value, YX is the myo-oxygen simulation state evaluation value of the rehabilitation robot, b ′ is the environmental characteristic parameter corresponding to the real-time electromagnetic interference intensity predefined in the rehabilitation control information library, b2 is the environmental characteristic parameter corresponding to the real-time noise decibel value predefined in the rehabilitation control information library, and k1 is the environmental characteristic parameter corresponding to the myo-oxygen simulation state evaluation value predefined in the rehabilitation control information library.

[0071] In this embodiment, the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located is used to measure the degree of influence of various factors in the ICU environment, such as temperature, humidity, electromagnetic interference, air quality, noise, etc. on the performance of the rehabilitation robot and the process of the rehabilitation robot simulating and executing the initial rehabilitation control plan.

[0072] It should be explained that the real-time environmental temperature refers to the instantaneous temperature value of the space around the location of the rehabilitation robot within the evaluation period for evaluating the ICU environment; the environmental temperature adaptation value refers to the reference value corresponding to the pre-set environmental temperature; the real-time environmental humidity refers to the instantaneous humidity value of the space around the location of the rehabilitation robot within the evaluation period for evaluating the ICU environment; the environmental humidity adaptation value refers to the reference value corresponding to the pre-set environmental humidity; the real-time electromagnetic interference intensity refers to the instantaneous electromagnetic interference intensity value of the space around the location of the rehabilitation robot within the evaluation period for evaluating the ICU environment; the real-time noise decibel value refers to the decibel value of the instantaneous noise of the space around the location of the rehabilitation robot within the evaluation period for evaluating the ICU environment; the real-time carbon dioxide concentration refers to the instantaneous carbon dioxide concentration value of the space around the location of the rehabilitation robot within the evaluation period for evaluating the ICU environment; the carbon dioxide concentration adaptation value refers to the reference value corresponding to the pre-set carbon dioxide concentration.

[0073] Among them, the environmental characteristic parameters corresponding to the real-time electromagnetic interference intensity, the environmental characteristic parameters corresponding to the real-time noise decibel value, and the environmental characteristic parameters corresponding to the myo-oxygen simulation state evaluation value are all extracted from the rehabilitation control information library. The mapping relationship therein can be a one-to-one or many-to-one relationship. For example, the real-time electromagnetic interference intensity and the environmental characteristic parameters corresponding to the preset real-time electromagnetic interference intensity in the rehabilitation control information library form a mapping set, and the real-time electromagnetic interference intensity is brought into the mapping set to obtain the environmental characteristic parameters corresponding to the real-time electromagnetic interference intensity; the real-time noise decibel value and the environmental characteristic parameters corresponding to the preset real-time noise decibel value in the rehabilitation control information library form a mapping set, and the real-time noise decibel value is brought into the mapping set to obtain the environmental characteristic parameters corresponding to the real-time noise decibel value; the myo-oxygen simulation state evaluation value and the environmental characteristic parameters corresponding to the preset myo-oxygen simulation state evaluation value in the rehabilitation control information library form a mapping set, and the real-time myo-oxygen simulation state evaluation value is brought into the mapping set to obtain the environmental characteristic parameters corresponding to the myo-oxygen simulation state evaluation value. In this embodiment, the value ranges of the environmental characteristic parameters corresponding to the real-time electromagnetic interference intensity, the environmental characteristic parameters corresponding to the real-time noise decibel value, and the environmental characteristic parameters corresponding to the myo-oxygen simulation state evaluation value are all (0, 1).

[0074] In this embodiment, temperature has an impact on the performance of electronic devices, and electronic devices are one of the main sources of electromagnetic interference. When the environmental temperature is high and deviates from the adapted environmental temperature, for example, the performance of the electronic components inside various medical instruments and rehabilitation robots in the ICU may change, resulting in the drift of the electromagnetic emission frequency of these electronic devices, increasing the complexity of the electromagnetic interference intensity. At the same time, high temperature may also reduce the anti-interference ability of electronic devices, making them more vulnerable to external electromagnetic interference. Similarly, high humidity may also cause changes in the electromagnetic radiation frequency of electronic devices, or make the devices more likely to absorb external electromagnetic interference, thereby affecting the distribution of the electromagnetic interference intensity, so as to increase the environmental characteristic influence coefficient, bringing a greater negative impact on the execution performance of the rehabilitation robot; when there is strong electromagnetic interference, it may cause the electronic devices to operate abnormally, thus generating additional noise, resulting in an increase in the noise decibel value, also increasing the environmental characteristic influence coefficient; in addition, too high or too low carbon dioxide concentration in the ICU environment will bring negative impacts on the execution performance of the rehabilitation robot. High-concentration carbon dioxide may corrode the electronic components inside the rehabilitation robot. In an environment with high-concentration carbon dioxide, if the environmental humidity is also high, carbon dioxide will react with water vapor in the air to form carbonic acid. After being in a high-carbon dioxide concentration environment for a long time, the aging speed of electronic components will accelerate. When the carbon dioxide concentration is low, in order to maintain the normal carbon dioxide concentration, the ventilation system in the ICU may work excessively, which may cause changes in the air flow speed and pressure around the ventilation equipment, thereby affecting the stability of the rehabilitation robot, thus bringing a negative impact on the execution quality of the rehabilitation robot.

[0075] In this embodiment, the larger the myo-oxygen simulation state evaluation value of the rehabilitation robot, the more accurate the simulation of the patient's myo-oxygen state by the rehabilitation robot and the better it can meet the actual needs. A high myo-oxygen simulation state evaluation value indicates that the system operation of the rehabilitation robot is more stable and accurate, which means that the electromagnetic signal interference generated during the data collection, transmission, and processing of the robot is smaller. The stable operation state also reduces the sensitivity of the rehabilitation robot itself to external electromagnetic interference because it can better resist interference and maintain the accuracy of myo-oxygen simulation. When the rehabilitation robot works in a better myo-oxygen simulation state, its own electronic components can also operate in a more suitable temperature environment, reducing the impact of environmental temperature fluctuations on the robot. And when the myo-oxygen simulation state evaluation value is large, the rehabilitation robot can perform state simulation with a more reasonable motion mode and rhythm, which can avoid additional mechanical noise caused by unsmooth movement or frequent adjustment of actions.

[0076] The initial plan simulation execution module is used for the rehabilitation control platform to match and obtain the initial rehabilitation control plan according to the myo-oxygen simulation state evaluation value of the rehabilitation robot. Thus, the rehabilitation control platform simulates and executes the initial rehabilitation control plan through remote cooperation with the rehabilitation robot and obtains the execution data of the initial rehabilitation control plan.

[0077] Among them, the process of matching and obtaining the initial rehabilitation control plan is specifically as follows:

[0078] Match the myo-oxygen simulation state evaluation value of the rehabilitation robot with the initial rehabilitation control plans corresponding to the pre-defined myo-oxygen simulation state evaluation value intervals in the rehabilitation control information library. The specific matching process is as follows: Extract the mapping set between the myo-oxygen simulation state evaluation value of the rehabilitation robot and the initial rehabilitation control plan from the rehabilitation control information library, determine the interval to which the myo-oxygen simulation state evaluation value of the rehabilitation robot belongs, and obtain the initial rehabilitation control plan corresponding to this interval, thereby matching and obtaining the initial rehabilitation control plan.

[0079] The initial rehabilitation control plan specifically adjusts the assistance or resistance based on the muscle oxygen saturation simulated by the rehabilitation robot. When the muscle oxygen simulation status evaluation value shows that the muscle oxygen saturation is at a relatively high level, such as 75%-85%, it indicates that the patient's muscle oxygenation is good and the current muscle bearing capacity is strong. The initial rehabilitation control plan can set the rehabilitation robot to moderately increase the assistance when performing limb simulation assisted movement. On the contrary, if the muscle oxygen saturation is relatively low, such as 60%-70%, it means that the muscle may be in a relatively hypoxic state. At this time, the assistance or resistance of the robot's simulated movement should be reduced to avoid exceeding the patient's acceptance level. If the oxygen uptake rate is within the normal range, such as 20%-30%, it indicates that the efficiency of the patient's muscle in using oxygen is normal, and the rehabilitation robot can proceed according to the standard rehabilitation simulation training intensity. If the muscle oxygen metabolism rate shows that the speed of the muscle consuming oxygen is moderate, such as 2-3 milliliters per minute per 100 grams of muscle, the duration of a single simulation training of the rehabilitation robot can be set to 30-40 minutes. If the various simulated muscle oxygen indexes of the rehabilitation robot fluctuate smoothly and return to normal within the muscle oxygen simulation cycle, it is determined that the simulation training interval can be set to 1-2 days.

[0080] The initial plan judgment and adjustment module is used for the rehabilitation control platform to evaluate the execution quality index of the initial rehabilitation control plan based on the execution data of the initial rehabilitation control plan and the environmental characteristic influence coefficient of the ICU where the rehabilitation robot is located, and compare it with the predefined execution quality expected index, so as to determine whether to adaptively adjust the initial rehabilitation control plan, and finally complete the remote collaboration and adaptive control of the ICU-AW rehabilitation robot.

[0081] Among them, the execution data of the initial rehabilitation control plan specifically includes the number of completed rehabilitation trainings of the initial rehabilitation control plan within the execution cycle, the feedback response duration of the initial rehabilitation control plan within the execution cycle, the number of failures of the rehabilitation robot of the initial rehabilitation control plan within the execution cycle, and the final simulated muscle oxygen saturation of the rehabilitation robot of the initial rehabilitation control plan within the execution cycle.

[0082] The execution cycle is specifically a period of time for the rehabilitation robot to simulate the execution of the initial rehabilitation control plan. The execution cycle is specifically a period of time from the start of the first operation of the rehabilitation robot to execute the initial rehabilitation control plan to the completion of all operations by the rehabilitation robot.

[0083] The execution data of the initial rehabilitation control plan can specifically be extracted from the control report of the rehabilitation control platform.

[0084] Obtain the initial simulated muscle oxygen saturation of the rehabilitation robot within the muscle oxygen simulation cycle, where the initial simulated muscle oxygen saturation can specifically be extracted from the simulation report of the rehabilitation robot.

[0085] Match the number of repetitions of the set actions of the rehabilitation robot within the myo-oxygen simulation cycle with the fault definition times corresponding to the predefined action repetition number intervals in the rehabilitation control information database. The specific matching process is as follows: Extract from the rehabilitation control information database the mapping set between the number of repetitions of the set actions of the rehabilitation robot within the myo-oxygen simulation cycle and the fault definition times, determine the specific interval of the number of repetitions of the set actions of the rehabilitation robot within the myo-oxygen simulation cycle, and assign the fault definition times corresponding to this interval to the rehabilitation robot corresponding to the number of repetitions of the set actions, so as to match and obtain the fault definition times of the rehabilitation robot.

[0086] Match the response delay duration of the rehabilitation robot within the myo-oxygen simulation cycle with the feedback response definition duration corresponding to the predefined response delay duration intervals in the rehabilitation control information database. The specific matching process is as follows: Extract from the rehabilitation control information database the mapping set between the response delay duration of the rehabilitation robot within the myo-oxygen simulation cycle and the feedback response definition duration, determine the specific interval of the response delay duration of the rehabilitation robot within the myo-oxygen simulation cycle, and obtain the feedback response definition duration of this interval, so as to match and obtain the feedback response definition duration of the initial rehabilitation control plan.

[0087] Obtain the preset number of rehabilitation training times of the initial rehabilitation control plan within the execution cycle from the rehabilitation control information database.

[0088] Perform a ratio process on the number of completed rehabilitation training times of the initial rehabilitation control plan within the execution cycle and the preset number of rehabilitation training times of the initial rehabilitation control plan within the execution cycle to obtain the simulated completion rate of the rehabilitation training of the initial rehabilitation control plan within the execution cycle.

[0089] The number of completed rehabilitation training times specifically refers to the actual number of successfully completing the entire set of rehabilitation training actions or processes within the execution cycle according to the requirements of the initial rehabilitation control plan. For example, if the initial rehabilitation control plan stipulates that upper limb strength and joint range of motion rehabilitation simulations are carried out once a day, and the rehabilitation robot has completed 5 days of simulation training within a week, then the number of completed rehabilitation training times within this week is 5 times.

[0090] Among them, the specific analysis process of the execution quality index of the initial rehabilitation control plan is as follows:

[0091] Comprehensively analyze the rehabilitation training simulation completion rate of the initial rehabilitation control plan within the execution cycle, the feedback response duration of the initial rehabilitation control plan within the execution cycle, the feedback response defined duration of the initial rehabilitation control plan, the number of failures of the rehabilitation robot in the initial rehabilitation control plan within the execution cycle, the defined number of failures of the rehabilitation robot, the final simulated muscle oxygen saturation of the rehabilitation robot in the initial rehabilitation control plan within the execution cycle, the initial simulated muscle oxygen saturation of the rehabilitation robot within the muscle oxygen simulation cycle, and the environmental characteristic influence coefficient of the ICU to which the rehabilitation robot belongs to obtain the execution quality index of the initial rehabilitation control plan. The specific method is as follows:

[0092]

[0093] In the formula, ZX is the execution quality index of the initial rehabilitation control plan, ZX ′ is the default execution quality index of the initial rehabilitation control plan predefined in the rehabilitation control information library. SmO is the final simulated muscle oxygen saturation of the rehabilitation robot in the initial rehabilitation control plan within the execution cycle, SmO ′ is the initial simulated muscle oxygen saturation of the rehabilitation robot within the muscle oxygen simulation cycle. CL is the rehabilitation training simulation completion rate of the initial rehabilitation control plan within the execution cycle. FK is the feedback response duration of the initial rehabilitation control plan within the execution cycle, FK ′ is the feedback response defined duration of the initial rehabilitation control plan. GZ is the number of failures of the rehabilitation robot in the initial rehabilitation control plan within the execution cycle, GZ ′ is the defined number of failures of the rehabilitation robot. HJ is the environmental characteristic influence coefficient of the ICU to which the rehabilitation robot belongs. d1 is the execution quality influence factor corresponding to the predefined rehabilitation training simulation completion rate in the rehabilitation control information library. d2 is the execution quality influence factor corresponding to the predefined environmental characteristic influence coefficient in the rehabilitation control information library. e is the natural constant.

[0094] It should be elaborated that the default execution quality index of the initial rehabilitation control plan refers to the minimum permitted value corresponding to the execution quality index of the initial rehabilitation control plan set in advance. Setting the minimum permitted value of the execution quality index is to ensure that the rehabilitation simulation training meets certain basic requirements, so as to ensure that the rehabilitation training better meets the needs of patients. When the actual execution quality index is equal to the minimum permitted value, this indicates that there are relatively large problems in the simulation execution of the initial rehabilitation control plan and timely adjustment is required to make the initial rehabilitation control plan more in line with the actual situation of patients and improve the use effect of the rehabilitation robot.

[0095] The environmental characteristic influence coefficient of the rehabilitation robot belonging to the ICU is obtained by comprehensively analyzing the real-time environmental temperature of the ICU where the rehabilitation robot is located during the evaluation period, the real-time environmental humidity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time electromagnetic interference intensity of the ICU where the rehabilitation robot is located during the evaluation period, the real-time noise decibel value of the ICU where the rehabilitation robot is located during the evaluation period, and the real-time carbon dioxide concentration of the ICU where the rehabilitation robot is located during the evaluation period.

[0096] In this embodiment, the execution quality index of the initial rehabilitation control plan is a quantitative or evaluable standard used to measure whether the initial rehabilitation control plan meets the expected goals and has good execution effects during the actual execution process. The larger the execution quality index of the initial rehabilitation control plan, the higher the degree of compliance of this plan with the patient during the execution process and the greater the effect on the patient.

[0097] It should be explained that the final simulated muscle oxygen saturation of the rehabilitation robot refers to the final value of the simulated muscle oxygen saturation of the rehabilitation robot at the end of a complete cycle of rehabilitation simulation training according to the initial rehabilitation control plan; the initial simulated muscle oxygen saturation refers to the simulated muscle oxygen saturation of the rehabilitation robot before simulating the execution of the initial rehabilitation control plan; the rehabilitation training simulation completion rate refers to the ratio of the actual amount of rehabilitation training simulated by the rehabilitation robot to the planned amount of simulated rehabilitation training set in the initial plan during the execution cycle of the initial rehabilitation control plan; the feedback response duration refers to the time interval from the occurrence of a situation that requires feedback during the rehabilitation training process, such as abnormal working state of the rehabilitation robot, unmet training effects, etc., to the effective feedback made by the rehabilitation control center based on these situations, such as adjusting the parameters of the rehabilitation robot, changing the training plan, etc.; the feedback response defined duration refers to the maximum value corresponding to the pre-set feedback response duration; the number of failures of the rehabilitation robot refers to the cumulative number of times that the rehabilitation robot fails to work properly during the process of rehabilitation simulation training according to the initial rehabilitation control plan. These failure situations may include various types such as mechanical component damage, electronic component failure, software system crash, sensor failure, etc. Each occurrence of such a situation is counted as one failure; the defined number of failures of the rehabilitation robot refers to the maximum value corresponding to the pre-set number of failures of the rehabilitation robot.

[0098] Among them, the execution quality impact factor corresponding to the rehabilitation training simulation completion rate and the execution quality impact factor corresponding to the environmental feature impact coefficient are both extracted from the rehabilitation control information database, and the mapping relationship therein can be a one-to-one or many-to-one relationship. For example, the rehabilitation training simulation completion rate and the execution quality impact factor corresponding to the preset rehabilitation training simulation completion rate in the rehabilitation control information database form a mapping set, and the real-time rehabilitation training simulation completion rate is brought into the mapping set to obtain the execution quality impact factor corresponding to the rehabilitation training simulation completion rate; the environmental feature impact coefficient and the execution quality impact factor corresponding to the preset environmental feature impact coefficient in the rehabilitation control information database form a mapping set, and the real-time environmental feature impact coefficient is brought into the mapping set to obtain the execution quality impact factor corresponding to the environmental feature impact coefficient. In this embodiment, the value ranges of the execution quality impact factor corresponding to the rehabilitation training simulation completion rate and the execution quality impact factor corresponding to the environmental feature impact coefficient are both (0, 1).

[0099] In this embodiment, the feedback response duration has a direct impact on the rehabilitation training simulation completion rate. If the feedback response duration is short, when problems occur during the rehabilitation training, such as deviations in the simulation parameters of the rehabilitation robot, etc., it can be quickly adjusted and solved, and the rehabilitation robot can continue to complete the simulation training, thereby increasing the rehabilitation training simulation completion rate and improving the execution quality of the initial rehabilitation control plan. On the contrary, if the feedback response duration is long, the rehabilitation training simulation completion rate may decrease because the problems cannot be solved in time, and the rehabilitation robot may be forced to interrupt the simulation training, resulting in a decrease in the rehabilitation training simulation completion rate; similarly, the number of failures of the rehabilitation robot will also directly affect the rehabilitation training simulation completion rate. Each time the rehabilitation robot fails, it will cause the interruption of the rehabilitation simulation training, thereby reducing the rehabilitation training simulation completion rate and greatly reducing the execution quality index of the initial rehabilitation control plan; in addition, when the muscle oxygen saturation does not increase but decreases after the initial rehabilitation control plan is executed, it indicates that the current initial rehabilitation control plan may make the simulation state of the rehabilitation robot in a state of excessive oxygen consumption. Setting the execution quality to the minimum value of quality permission can timely adjust the intensity of the rehabilitation model training, prevent large deviations in the initial rehabilitation control plan, and this situation indicates that the current initial rehabilitation control plan may not be suitable for the patient's physical condition, and the parameters of the initial rehabilitation control plan need to be adjusted.

[0100] In this embodiment, the smaller the environmental characteristic influence coefficient of the rehabilitation robot belonging to the ICU, the smaller the interference of the ICU environment on the rehabilitation robot. For example, lower electromagnetic interference can ensure the stable operation of the motor control system of the rehabilitation robot, enabling the rehabilitation robot to work accurately according to the set assistance or resistance parameters during simulation training. The rehabilitation robot has good adaptability to the ICU environment, enabling it to continuously and stably simulate the execution of the initial rehabilitation control plan, effectively improving the execution quality of the initial rehabilitation control plan. At the same time, the smaller the interference of environmental factors on the rehabilitation robot, the more accurate the rehabilitation robot is during the simulation training of movement coordination. The sensors and control system of the rehabilitation robot can accurately perceive its own simulated actions, improving the applicability of the initial rehabilitation control plan.

[0101] Among them, the specific determination process for determining whether to adaptively adjust the initial rehabilitation control plan is as follows:

[0102] If the execution quality index of the initial rehabilitation control plan is equal to the default execution quality index of the initial rehabilitation control plan, then execute the predefined optimization adjustment plan for the initial rehabilitation control plan;

[0103] If the execution quality index of the initial rehabilitation control plan is not equal to the default execution quality index of the initial rehabilitation control plan, compare the execution quality index of the initial rehabilitation control plan with the predefined execution quality reference index in the rehabilitation control information library. If the execution quality index of the initial rehabilitation control plan is greater than the execution quality reference index, there is no need to adaptively adjust the initial rehabilitation control plan. If the execution quality index of the initial rehabilitation control plan is less than or equal to the execution quality reference index, then the initial rehabilitation control plan needs to be adaptively adjusted.

[0104] It should be noted that the default execution quality index of the initial rehabilitation control plan is expressed as the minimum permitted value corresponding to the execution quality index of the initial rehabilitation control plan, and is less than the execution quality reference index.

[0105] In this embodiment, when the final simulated muscle oxygen saturation of the rehabilitation robot during the execution cycle of the initial rehabilitation control plan is greater than the initial simulated muscle oxygen saturation of the rehabilitation robot during the muscle oxygen simulation cycle, the execution quality index of the initial rehabilitation control plan is greater than the default execution quality index of the initial rehabilitation control plan; when the final simulated muscle oxygen saturation of the rehabilitation robot during the execution cycle of the initial rehabilitation control plan is less than or equal to the initial simulated muscle oxygen saturation of the rehabilitation robot during the muscle oxygen simulation cycle, the execution quality index of the initial rehabilitation control plan is equal to the default execution quality index of the initial rehabilitation control plan, that is, there is no situation where the execution quality index of the initial rehabilitation control plan is less than the default execution quality index of the initial rehabilitation control plan.

[0106] Among them, the adaptive adjustment of the initial rehabilitation control plan is carried out as follows:

[0107] The difference between the execution quality index of the initial rehabilitation control plan and the execution quality reference index is processed to obtain the execution quality deviation value of the initial rehabilitation control plan, and it is matched with the adaptive adjustment plan corresponding to each predefined execution quality deviation value interval in the rehabilitation control information library. The specific matching process is as follows: Extract the mapping set between the execution quality deviation value of the initial rehabilitation control plan and the adaptive adjustment plan from the rehabilitation control information library, determine the specific interval of the execution quality deviation value of the initial rehabilitation control plan, and assign the adaptive adjustment plan corresponding to this interval to the initial rehabilitation control plan corresponding to the execution quality deviation value, so as to match the adaptive adjustment plan of the initial rehabilitation control plan, and finally carry out the adaptive adjustment of the initial rehabilitation control plan.

[0108] The optimized adjustment plan can specifically be that when the execution quality index is equal to the execution quality default index, it may mean that the current simulation training intensity is too high or too low and needs to be optimized. For the case where the simulation training intensity is too high, the simulated assistance or resistance of the rehabilitation robot can be reduced, and the number of repetitions of each group of simulation training can be reduced. If the training intensity is too low, the simulated assistance or resistance can be increased to improve the difficulty and intensity of the simulation training; it may be necessary to adjust the duration of a single simulation training or the frequency of simulation training. If the completion rate of the simulation training is low during the simulation training of the rehabilitation robot, it may be necessary to shorten the time of a single simulation training. On the contrary, if the time of the simulation training is too short and the effect of the initial rehabilitation control plan is not obvious, the time of a single simulation training can be appropriately extended, and at the same time, the simulation training intensity can be adjusted; calibrate the various parameters of the rehabilitation robot to ensure its accuracy and stability, which includes the calibration of sensors such as myo-oxygen sensors and force sensors, so that it can more accurately sense the physiological data and motion data during the simulation process and provide a more reliable basis for the rehabilitation simulation training.

[0109] The adaptive adjustment plan can specifically be that when the execution quality index is less than or equal to the execution quality reference index, intensity adjustments of different degrees are made according to the magnitude of the deviation value. If the deviation value is small and within the mild deviation range, the simulated assistance or resistance of the rehabilitation robot can be slightly adjusted. For example, it can be increased or decreased by 5%-10%, and the simulated training time can be adjusted. For example, the single simulated training time can be increased or decreased by 5 minutes. If the deviation value is within the moderate deviation range, a larger adjustment of the simulated training intensity may be required. For example, the simulated assistance or resistance is changed by 15%-30%, and at the same time, the training action mode is adjusted to add some targeted simulated training actions to make up for the deficiencies in the effect of the initial rehabilitation control plan. For the severe deviation range, it may be necessary to suspend the current training simulation plan, re-evaluate the patient's physical condition and rehabilitation needs, formulate a new initial rehabilitation control plan, and gradually resume the simulated training starting from a lower training simulation intensity; and considering the influence coefficient of the ICU environment characteristics, if the environmental factors have a greater impact on the rehabilitation training, such as electromagnetic interference causing an increase in the number of failures of the rehabilitation robot and affecting the performance of the rehabilitation robot, the adaptive adjustment plan can include taking corresponding environmental improvement measures. For example, an electromagnetic shielding device can be added to reduce electromagnetic interference.

[0110] It should be elaborated that the optimization adjustment plan and the adaptive adjustment plan can specifically be to adjust the initial rehabilitation control plan by monitoring the myo-oxygen simulation state data of the rehabilitation robot, the action execution accuracy data of the rehabilitation robot, and the ICU environment data during the adjustment process. All of the above data can be obtained by extraction in the rehabilitation control platform.

[0111] The following points need to be explained:

[0112] (1) The drawings of the embodiments of the present invention only relate to the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0113] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layer or region is enlarged or reduced, that is, these drawings are not drawn according to the actual scale. It can be understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element can be "directly" on or under the other element or there can be an intermediate element.

[0114] (3) Without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0115] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network, characterized by: include: The muscle oxygen simulation state evaluation module is used for the patient state monitoring component to perform initial state monitoring on ICU-AW patients, collect the initial muscle oxygen state data of the patients, and the rehabilitation robot simulates the muscle oxygen state of the patients according to the initial muscle oxygen state data, determines the muscle oxygen simulation state evaluation value of the rehabilitation robot, verifies it with the predefined muscle oxygen simulation state evaluation threshold, so as to determine whether to optimize the muscle oxygen simulation state of the rehabilitation robot, and uploads the muscle oxygen simulation state evaluation value of the rehabilitation robot to the rehabilitation control platform through the 5G network; ICU environment monitoring module, which is used by the ICU environment monitoring component to monitor the ICU environment of the rehabilitation robot, obtain the ICU environment status data of the rehabilitation robot, evaluate the environmental characteristic influence coefficient of the ICU of the rehabilitation robot, and upload it to the rehabilitation control platform through the 5G network; The initial plan simulation execution module is used for the rehabilitation control platform to match the rehabilitation control initial plan according to the muscle oxygen simulation state evaluation value of the rehabilitation robot, so that the rehabilitation control platform can simulate and execute the rehabilitation control initial plan through the remote collaborative rehabilitation robot and obtain the execution data of the rehabilitation control initial plan; The initial plan judgment and adjustment module is used for the rehabilitation control platform to evaluate the execution quality indicators of the initial rehabilitation control plan based on the execution data of the initial rehabilitation control plan and the environmental characteristics influence coefficient of the ICU to which the rehabilitation robot belongs, and compare them with the predefined expected execution quality indicators to determine whether to make adaptive adjustments to the initial rehabilitation control plan, and finally complete the remote collaboration and adaptive control of the ICU-AW rehabilitation robot.

2. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 1 is characterized in that: The specific determination process of determining the muscle oxygen simulation state evaluation value of the rehabilitation robot is as follows: The rehabilitation robot simulates the patient's muscle oxygen state according to the initial muscle oxygen state data, and obtains the muscle oxygen simulation state data of the rehabilitation robot, specifically including the output power of the rehabilitation robot at each muscle oxygen simulation time point, the number of set action repetitions of the rehabilitation robot within the muscle oxygen simulation cycle, the response delay duration of the rehabilitation robot within the muscle oxygen simulation cycle, and the average value of the sensor linearity error of the rehabilitation robot within the muscle oxygen simulation cycle; The output power of the rehabilitation robot at each muscle oxygen simulation time point is averaged to obtain the average output power of the rehabilitation robot during the muscle oxygen simulation period; Obtain the electromagnetic radiation intensity value of the environment of the rehabilitation robot at each muscle oxygen simulation time point; Extract the output power adaptation average value and the set action repetition adaptation times from the rehabilitation control information library; The average output power of the rehabilitation robot during the muscle oxygen simulation cycle, the number of set action repetitions of the rehabilitation robot during the muscle oxygen simulation cycle, the response delay duration of the rehabilitation robot during the muscle oxygen simulation cycle, the average sensor linearity error of the rehabilitation robot during the muscle oxygen simulation cycle, and the electromagnetic radiation intensity value of the environment to which the rehabilitation robot belongs at each muscle oxygen simulation time point are comprehensively processed to obtain the muscle oxygen simulation state evaluation value of the rehabilitation robot.

3. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 1 is characterized in that: The determining whether to optimize the muscle oxygen simulation state of the rehabilitation robot is specifically to verify the muscle oxygen simulation state evaluation value of the rehabilitation robot with a predefined muscle oxygen simulation state evaluation threshold to obtain a verification result, and determine whether to optimize the muscle oxygen simulation state of the rehabilitation robot based on the verification result; The verification result is the first verification result or the second verification result; The first verification result is specifically that the muscle oxygen simulation state evaluation value of the rehabilitation robot is greater than or equal to the muscle oxygen simulation state evaluation threshold; The second test result is specifically that the muscle oxygen simulation state evaluation value of the rehabilitation robot is less than the muscle oxygen simulation state evaluation threshold; If the verification result shows the first verification result, there is no need to optimize the muscle oxygen simulation state of the rehabilitation robot. If the verification result shows the second verification result, the muscle oxygen simulation state of the rehabilitation robot needs to be optimized.

4. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 1 is characterized in that: The specific evaluation process of evaluating the environmental characteristic influence coefficient of the ICU to which the rehabilitation robot belongs is as follows: The ICU environment status data of the rehabilitation robot specifically includes the real-time ambient temperature of the ICU environment of the rehabilitation robot during the evaluation period, the real-time ambient humidity of the ICU environment of the rehabilitation robot during the evaluation period, the real-time electromagnetic interference intensity of the ICU environment of the rehabilitation robot during the evaluation period, the real-time noise decibel value of the ICU environment of the rehabilitation robot during the evaluation period, and the real-time carbon dioxide concentration of the ICU environment of the rehabilitation robot during the evaluation period; Extracting the ambient temperature adaptation value, the ambient humidity adaptation value and the carbon dioxide concentration adaptation value from the rehabilitation control information database; The real-time ambient temperature of the ICU environment to which the rehabilitation robot belongs during the evaluation period, the real-time ambient humidity of the ICU environment to which the rehabilitation robot belongs during the evaluation period, the real-time electromagnetic interference intensity of the ICU environment to which the rehabilitation robot belongs during the evaluation period, the real-time noise decibel value of the ICU environment to which the rehabilitation robot belongs during the evaluation period, and the real-time carbon dioxide concentration of the ICU environment to which the rehabilitation robot belongs during the evaluation period are comprehensively analyzed to obtain the environmental characteristic influence coefficient of the ICU to which the rehabilitation robot belongs.

5. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 4 is characterized in that: The environmental characteristics influence coefficient of the ICU to which the rehabilitation robot belongs is analyzed in the following specific method: Where HJ is the environmental characteristic influence coefficient of the ICU to which the rehabilitation robot belongs, t is the time variable, t∈[t0, t1], t0 is the start time of the evaluation period, t1 is the end time of the evaluation period, β(t) is the real-time ambient temperature of the ICU environment to which the rehabilitation robot belongs at time t in the evaluation period, β′ is the ambient temperature adaptation value, γ(t) is the real-time ambient humidity of the ICU environment to which the rehabilitation robot belongs at time t in the evaluation period, γ′ is the ambient humidity adaptation value, δ(t) is the real-time electromagnetic interference intensity of the ICU environment to which the rehabilitation robot belongs at time t in the evaluation period, θ(t ) is the real-time noise decibel value of the ICU environment to which the rehabilitation robot belongs at the evaluation period t, τ(t) is the real-time carbon dioxide concentration of the ICU environment to which the rehabilitation robot belongs at the evaluation period t, τ′ is the carbon dioxide concentration adaptation value, YX is the muscle oxygen simulation state evaluation value of the rehabilitation robot, b1 is the environmental characteristic parameter corresponding to the real-time electromagnetic interference intensity predefined in the rehabilitation control information library, b2 is the environmental characteristic parameter corresponding to the real-time noise decibel value predefined in the rehabilitation control information library, k1 is the environmental characteristic parameter corresponding to the muscle oxygen simulation state evaluation value predefined in the rehabilitation control information library.

6. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 1 is characterized in that: The matching obtains the initial rehabilitation control plan, and the specific matching process is: The muscle oxygen simulation state evaluation value of the rehabilitation robot is matched with the initial rehabilitation control plan corresponding to each predefined muscle oxygen simulation state evaluation value interval. The specific matching process is: extracting the mapping set between the muscle oxygen simulation state evaluation value of the rehabilitation robot and the initial rehabilitation control plan from the rehabilitation control information library, determining the interval to which the muscle oxygen simulation state evaluation value of the rehabilitation robot belongs, and obtaining the initial rehabilitation control plan corresponding to the interval, so as to obtain the initial rehabilitation control plan by matching.

7. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 1 is characterized in that: The execution data of the initial rehabilitation control plan specifically includes the number of rehabilitation training completed during the execution cycle of the initial rehabilitation control plan, the feedback response duration during the execution cycle of the initial rehabilitation control plan, the number of rehabilitation robot failures during the execution cycle of the initial rehabilitation control plan, and the final simulated muscle oxygen saturation of the rehabilitation robot during the execution cycle of the initial rehabilitation control plan; Obtaining the initial simulated muscle oxygen saturation of the rehabilitation robot during the muscle oxygen simulation cycle; Obtaining the preset number of rehabilitation trainings of the rehabilitation control initial plan within the execution cycle from the rehabilitation control information database; The number of completed rehabilitation trainings of the initial rehabilitation control plan within the execution cycle is ratioed with the preset number of rehabilitation trainings of the initial rehabilitation control plan within the execution cycle to obtain the rehabilitation training simulation completion rate of the initial rehabilitation control plan within the execution cycle.

8. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 7 is characterized in that: The execution quality index of the initial rehabilitation control plan is analyzed in detail as follows: The number of action repetitions of the rehabilitation robot in the muscle oxygen simulation cycle is matched with the number of fault definition corresponding to each predefined interval of action repetitions, thereby obtaining the number of fault definition of the rehabilitation robot; Matching the response delay duration of the rehabilitation robot within the muscle oxygen simulation cycle with the feedback response limit duration corresponding to each predefined response delay duration interval, thereby obtaining the feedback response limit duration of the initial rehabilitation control plan; A comprehensive analysis was conducted on the completion rate of rehabilitation training simulation within the execution cycle of the initial rehabilitation control plan, the feedback response time of the initial rehabilitation control plan within the execution cycle, the feedback response definition time of the initial rehabilitation control plan, the number of rehabilitation robot failures within the execution cycle of the initial rehabilitation control plan, the number of rehabilitation robot failure definitions, the final simulated muscle oxygen saturation of the rehabilitation robot within the execution cycle of the initial rehabilitation control plan, the initial simulated muscle oxygen saturation of the rehabilitation robot within the muscle oxygen simulation cycle, and the influence coefficient of environmental characteristics of the ICU to which the rehabilitation robot belongs, to obtain the execution quality indicators of the initial rehabilitation control plan.

9. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 1 is characterized in that: The specific process of determining whether to adaptively adjust the initial rehabilitation control plan is as follows: If the execution quality index of the initial rehabilitation control plan is equal to the default execution quality index of the initial rehabilitation control plan, a predefined optimization adjustment plan is executed for the initial rehabilitation control plan; If the execution quality index of the initial rehabilitation control plan is not equal to the default execution quality index of the initial rehabilitation control plan, compare the execution quality index of the initial rehabilitation control plan with the predefined execution quality reference index. If the execution quality index of the initial rehabilitation control plan is greater than the execution quality reference index, there is no need to make adaptive adjustments to the initial rehabilitation control plan. If the execution quality index of the initial rehabilitation control plan is less than or equal to the execution quality reference index, it is necessary to make adaptive adjustments to the initial rehabilitation control plan.

10. The ICU-AW rehabilitation robot remote collaboration and adaptive control system based on 5G network according to claim 9 is characterized in that: The adaptive adjustment of the initial rehabilitation control plan is specifically carried out as follows: The execution quality index of the initial rehabilitation control plan is differenced with the execution quality reference index to obtain the execution quality deviation value of the initial rehabilitation control plan, which is matched with the adaptive adjustment plan corresponding to each predefined execution quality deviation value interval, so as to obtain the adaptive adjustment plan of the initial rehabilitation control plan, and finally the initial rehabilitation control plan is adaptively adjusted.

Images