Exoskeleton joint regulation and control method based on edge calculation

Abstract

The invention discloses an exoskeleton joint regulation and control method based on edge calculation, and relates to the technical field of exoskeleton program control. The method comprises the following steps: performing man-machine exoskeleton metal steam temperature component evaluation; performing judgment, and performing heterojunction temperature coupling man-machine exoskeleton wrist joint influence component analysis; according to the temperature component evaluation result, first regulation and control over ship engine compartment maintenance exoskeleton joints are carried out; performing man-machine exoskeleton wrist joint multi-dimensional coupling evaluation; and performing man-machine exoskeleton wrist joint multi-dimensional coupling regulation and control according to the man-machine exoskeleton wrist joint multi-dimensional coupling evaluation. According to the invention, through heterojunction state monitoring, temperature coupling influence analysis, hierarchical regulation and control execution, long-term damage accumulation monitoring and multi-dimensional coupling safety protection, the effect of improving the joint regulation and control environment adaptability of ship engine compartment man-machine exoskeleton maintenance operation is achieved; the problem that in the prior art, joint regulation and control environment adaptability for ship engine compartment man-machine exoskeleton overhauling operation is insufficient is solved.

Description

Technical Field

[0001] This invention relates to the field of exoskeleton program control technology, and in particular to an exoskeleton joint control method based on edge computing. Background Technology

[0002] As humanoid exoskeletons become increasingly prevalent in various fields, technical challenges in different application scenarios are gradually emerging. In the application of exoskeletons for ship engine room maintenance, the condition monitoring of critical components such as the wrist joint typically relies on contact or offline measurement methods to assess their thickness or performance degradation. This approach fails to adequately consider the nonlinear coupling amplification effect of thermoelectric characteristics and contact resistance changes under high-temperature environments. Control strategies are mainly based on a single heterojunction thickness threshold or a simple temperature threshold, triggering limited operations and lacking tiered proactive control based on thickness gradient degradation or quantified temperature coupling risk levels.

[0003] For example, the invention patent with publication number CN114986478B discloses an arm exoskeleton, an upper body exoskeleton, and a teleoperation system. The arm exoskeleton includes a shoulder exoskeleton, an elbow exoskeleton, and a wrist exoskeleton; the shoulder, elbow, and wrist exoskeletons are connected sequentially. During the process of the arm exoskeleton actively moving according to control commands or passively moving in accordance with the user's arm movements, the shoulder motor group simulates movements guided by the human shoulder joint, the elbow motor group simulates movements guided by the human elbow joint, and the wrist joint group simulates movements guided by the human wrist joint.

[0004] For example, the invention patent with publication number CN114986478B discloses an arm exoskeleton, an upper body exoskeleton, and a teleoperation system. The arm exoskeleton includes a shoulder exoskeleton, an elbow exoskeleton, and a wrist exoskeleton; the shoulder, elbow, and wrist exoskeletons are connected sequentially. During the process of the arm exoskeleton actively moving according to control commands or passively moving in accordance with the user's arm movements, the shoulder motor group simulates movements guided by the human shoulder joint, the elbow motor group simulates movements guided by the human elbow joint, and the wrist joint group simulates movements guided by the human wrist joint.

[0005] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0006] In the existing technology, under the high temperature and high humidity maintenance environment of the ship's engine room, the wrist joint of the human-machine exoskeleton faces a combination of problems such as metal vapor deposition, vibration frequency locking and electrochemical corrosion during operation. The volatilized metal vapor deposits on the surface of the joint strain gauge to form a heterojunction semiconductor layer, inducing the Seebeck effect parasitic potential. This effect exhibits nonlinear temperature drift characteristics at high temperatures. Traditional linear temperature compensation algorithms, due to their failure to model the enhancement effect of deposition layer thickness on the temperature sensitivity coefficient, significantly increase detection errors. Simultaneously, the wrist joint of the exoskeleton couples with the fundamental frequency vibration of the engine during operation, accelerating the formation of conduction pathways in the deposition layer microcracks. Vibration energy promotes the expansion of microcracks, and oil mist droplets bridge within the cracks to form conductive paths, causing the parasitic potential to resonate and abruptly change with the dominant vibration frequency. This results in torque fluctuations in the PID controller of the exoskeleton wrist joint under these conditions. Furthermore, the parasitic potential catalyzes the electrochemical decomposition of polyurea-based grease. Traditional life models, neglecting this electrochemical corrosion coupling mechanism and relying solely on the number of mechanical cycles to estimate life, lead to the sudden failure of the harmonic reducer without obvious signs of mechanical wear. There is a problem of insufficient adaptability to the joint control environment of exoskeleton operations for ship engine room maintenance. Summary of the Invention

[0007] This application provides an edge computing-based exoskeleton joint control method, which solves the problem of insufficient adaptability of joint control environment in the prior art for human-machine exoskeleton operations in ship engine room maintenance, and achieves the effect of improving the adaptability of joint control environment in human-machine exoskeleton operations in ship engine room maintenance.

[0008] This application provides an edge computing-based exoskeleton joint control method, comprising the following steps: evaluating the metal vapor temperature component of the exoskeleton; determining and analyzing the influence component of the heterojunction temperature coupling on the exoskeleton wrist joint; performing a first control of the exoskeleton joint for ship engine room maintenance based on the temperature component evaluation results; performing a long-term damage dynamic accumulation analysis control based on the long-term damage dynamic accumulation analysis results, wherein the first control of the exoskeleton joint for ship engine room maintenance includes long-term damage dynamic accumulation analysis control; performing a multi-dimensional coupling evaluation of the exoskeleton wrist joint; and performing multi-dimensional coupling control of the exoskeleton wrist joint based on the multi-dimensional coupling evaluation.

[0009] Furthermore, the metal vapor temperature component of the exoskeleton is assessed, specifically including: inducing the breakdown of the heterojunction of the exoskeleton wrist joint using a pulsed laser emitted from the wrist joint pulsed laser; collecting the heterojunction excitation spectrum of the exoskeleton wrist joint using a fiber optic spectral sensor to obtain the characteristic spectral lines of the heterojunction; extracting features from the characteristic spectral lines of the heterojunction and comparing them with predefined heterojunction characteristic spectral lines in the exoskeleton wrist joint control database to obtain the actual thickness of the heterojunction; if the actual thickness of the heterojunction is less than a first threshold, then the influence component of the temperature coupling of the heterojunction on the exoskeleton wrist joint is analyzed; if the actual thickness of the heterojunction is equal to or greater than the first threshold and less than a second threshold, then a preventive nitrogen curtain is activated; if the actual thickness of the heterojunction is equal to or greater than the second threshold, then a warning notification is issued to relevant personnel.

[0010] Further, the analysis of the influence components of the heterojunction temperature-coupled human-machine exoskeleton wrist joint is determined. Specifically, this includes: acquiring the ambient temperature of the human-machine exoskeleton wrist joint using a temperature sensor; if the ambient temperature is greater than the inflection point of the heterojunction temperature influence, then the analysis of the influence components of the heterojunction temperature-coupled human-machine exoskeleton wrist joint is performed; otherwise, no analysis is performed. The specific process of this analysis is as follows: obtaining the basic Seebeck coefficient and the heterojunction temperature sensitivity coefficient of the human-machine exoskeleton wrist joint by measuring a standard heterojunction sample in the laboratory; directly extracting the initial contact resistance from the human-machine exoskeleton wrist joint control database; measuring the current contact resistance on the front end cap of the human-machine exoskeleton wrist joint using the four-probe method; and then performing a minimum value function analysis with the unit resistance ratio to obtain the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component. The following methods were used to determine the effects of different deposition thicknesses on temperature sensitivity: a Seebeck coefficient was measured using accelerated aging tests to assess the temperature sensitivity of the heterojunction deposition thickness at the inflection point where the ambient temperature of the exoskeleton wrist joint exceeds the heterojunction temperature. A temperature sensitivity enhancement coefficient was obtained by comparing the actual heterojunction thickness with a first threshold value, then performing a squared percentage analysis, and finally coupling this result with the temperature sensitivity enhancement coefficient to obtain a deposition thickness enhancement temperature sensitivity coefficient. Finally, the results of the analysis of the differences between the base Seebeck coefficient, the heterojunction coverage coefficient, the exoskeleton wrist joint heterojunction temperature sensitivity coefficient, the deposition thickness enhancement temperature sensitivity coefficient, and the inflection point of the effect of ambient temperature on the exoskeleton wrist joint and the heterojunction temperature were coupled to obtain the heterojunction temperature-coupled influence component on the exoskeleton wrist joint.

[0011] Furthermore, based on the temperature component assessment results, the first adjustment of the exoskeleton joint for ship engine room maintenance is implemented, specifically including: conducting long-term damage dynamic cumulative analysis; if the influence component of the heterojunction temperature coupling on the wrist joint of the exoskeleton is less than the first threshold of the heterojunction temperature coupling influence component, no additional adjustment is performed; if the influence component of the heterojunction temperature coupling on the wrist joint of the exoskeleton is equal to or greater than the first threshold of the heterojunction temperature coupling influence component and less than the second threshold of the heterojunction temperature coupling influence component, then the first adjustment of the exoskeleton joint for ship engine room maintenance is implemented; the first adjustment of the exoskeleton joint for ship engine room maintenance initiates a laser precision stripping operation, and immediately after the laser precision stripping operation, a plasma spray gun is activated; if the influence component of the heterojunction temperature coupling on the wrist joint of the exoskeleton is equal to or greater than the second threshold of the heterojunction temperature coupling influence component, then the second adjustment of the exoskeleton joint for ship engine room maintenance is implemented; the second adjustment of the exoskeleton joint for ship engine room maintenance initiates a phase change material rapid cooling operation, and immediately after the phase change material rapid cooling operation, an electromagnetic mechanical locking operation is activated, and a warning notification is issued to relevant personnel.

[0012] Furthermore, a long-term dynamic cumulative damage analysis was conducted, specifically including: obtaining the basic wear coefficient by directly testing the harmonic reducer under the wrist joint of the human-machine exoskeleton using a rotary wear tester in a laboratory accelerated aging test device; directly extracting the activation energy, gas constant, and regulation correction factor of the polyurea-based lubricating grease for the human-machine exoskeleton wrist joint harmonic reducer from the human-machine exoskeleton wrist joint regulation database; directly extracting the potential enhancement coefficient from the electrochemical decomposition experimental data of the lubricating grease; directly measuring the parasitic potential by using differential electrodes on the anterior cap of the wrist joint; and applying the polyurea-based lubricating grease to the human-machine exoskeleton wrist joint harmonic reducer. The activation energy of the base grease was coupled with the gas constant and the ambient temperature of the wrist joint of the human-machine exoskeleton for proportional analysis. This result was used as an index for index analysis, and then coupled with the base wear coefficient to obtain the dynamic temperature component of the human-machine exoskeleton wrist joint for long-term damage. The potential enhancement coefficient and parasitic potential were coupled for analysis to obtain the dynamic potential component of the human-machine exoskeleton wrist joint for long-term damage. The dynamic temperature component, the dynamic potential component, and the regulation and correction factor of the human-machine exoskeleton wrist joint for long-term damage were analyzed together to obtain the dynamic cumulative index of the human-machine exoskeleton wrist joint for long-term damage.

[0013] Furthermore, the long-term injury dynamic accumulation analysis also includes: if the comparative analysis results of the heterojunction temperature coupling influence component of the human-machine exoskeleton wrist joint and the first threshold or the second threshold of the heterojunction temperature coupling influence component change, then the current long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is recorded as the base value of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint. The corresponding adjustment correction factor is adjusted according to the current comparative analysis results, the cumulative time duration is cleared, and the specific constraint model of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is updated.

[0014] Furthermore, based on the results of long-term injury dynamic accumulation analysis, long-term injury dynamic accumulation analysis and regulation are carried out, specifically including: if the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is less than the long-term injury dynamic accumulation threshold of the human-machine exoskeleton wrist joint, a phosphate anti-wear film is injected through a micro liposuction pump and a multi-dimensional coupling assessment of the human-machine exoskeleton wrist joint is performed; if the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is equal to or greater than the long-term injury dynamic accumulation threshold of the human-machine exoskeleton wrist joint, an electromagnetic mechanical locking operation is immediately initiated and a warning notification is issued to relevant personnel.

[0015] Furthermore, a multi-dimensional coupling evaluation of the wrist joint of the human-machine exoskeleton was conducted. The specific process was as follows: The oil mist concentration conductivity coupling coefficient was obtained under different oil mist concentrations and deposition layer thicknesses through an oil mist concentration conductivity calibration experiment; the oil mist concentration in the wrist joint environment was collected and analyzed using a laser forward scattering sensor under the wrist joint's side protective shield; the oil mist concentration conductivity coupling coefficient and the oil mist concentration in the wrist joint environment were coupled and analyzed; the real-time vibration main frequency was collected using a triaxial accelerometer on the input shaft flange of the harmonic reducer; based on the real-time vibration main frequency, the average fundamental frequency of the human-machine exoskeleton operation was extracted from historical data; a vibration locking correction factor was obtained based on the real-time vibration main frequency and the average fundamental frequency of the human-machine exoskeleton operation; and the parasitic potential coefficient of the heterojunction deposition layer was obtained by coupling and analyzing the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component, the oil mist concentration conductivity coupling coefficient, the oil mist concentration in the wrist joint environment, and the vibration locking correction factor.

[0016] Furthermore, based on the multidimensional coupling assessment of the human-machine exoskeleton wrist joint, multidimensional coupling regulation of the human-machine exoskeleton wrist joint is carried out. Specifically, this includes: if the parasitic potential coefficient of the heterojunction deposition layer is less than the first threshold of the parasitic potential of the heterojunction deposition layer, dynamic adjustment of baseline noise filtering is performed; if the parasitic potential coefficient of the heterojunction deposition layer is equal to or greater than the first threshold of the parasitic potential of the heterojunction deposition layer and less than the second threshold of the parasitic potential of the heterojunction deposition layer, a preventive nitrogen curtain is activated; if the parasitic potential coefficient of the heterojunction deposition layer is equal to or greater than the second threshold of the parasitic potential of the heterojunction deposition layer, an early warning notification is issued to relevant personnel.

[0017] Furthermore, dynamic adjustment of baseline noise filtering is performed. The specific process is as follows: the first threshold of the parasitic potential of the heterojunction deposition layer is divided by the parasitic potential coefficient of the heterojunction deposition layer to obtain the parasitic potential difference coefficient of the heterojunction deposition layer; if the parasitic potential difference coefficient of the heterojunction deposition layer is less than the first threshold of the parasitic potential difference of the heterojunction deposition layer, then the basic moving average filtering is enabled for the encoder signal baseline of the wrist joint of the human-machine exoskeleton; if the parasitic potential difference coefficient of the heterojunction deposition layer is equal to or greater than the first threshold of the parasitic potential difference of the heterojunction deposition layer, then the dynamic window moving average and recursive filter low-pass filtering are switched on and started.

[0018] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0019] 1. In the extreme environment of high temperature and high oil mist in ship engine rooms, heterojunctions formed by metal vapor deposition may induce joint malfunction. Therefore, a micro-LIBS probe is used to emit pulsed lasers to penetrate the deposition layer, and characteristic spectral lines are extracted using a fiber optic spectral sensor to quantify the actual thickness of the heterojunction in real time. A four-probe method is used to dynamically measure changes in contact resistance to accurately determine the deposition coverage thickness. Then, when the deposition coverage thickness is compared with a threshold, a preventative nitrogen curtain is immediately activated to block metal vapor adsorption onto the substrate in the early stages of deposition. This approach physically inhibits heterojunction formation and avoids the accumulation of parasitic potential caused by the Seebeck effect. Compared to traditional periodic maintenance, this method moves the heterojunction control point forward, significantly reducing the risk of sudden joint lock-up.

[0020] 2. To address the nonlinear Seebeck effect induced by the thickening of the exoskeleton wrist joint deposit layer and high temperature synergistically caused by the extreme environment of high temperature and high oil mist in the ship's engine room, a multi-level dynamic control is proposed: Based on the infrared thermal imager to locate the highest temperature point, the sidewall pulsed laser precisely peels off the deposit layer, and simultaneously starts the plasma spray gun to spray a hexagonal boron nitride coating to block secondary deposition and strengthen the substrate protection; when the high temperature phase change causes the motor current to run away, the heat pipe array directs the heat to the PCM phase change module to absorb the latent heat, and the circulation pump removes the waste heat; at the same time, the electromagnetic mechanical lock drives the tungsten carbide pawl to engage in the output shaft groove, and the piezoelectric ceramic brake locks simultaneously, forcibly switching to passive damping mode; this mechanism directly eliminates the physical basis of the Seebeck effect or cuts off the energy transfer path through hardware coordinated response, ensuring that the joint torque fluctuation is always below the human safety threshold.

[0021] 3. The failure of grease electrolysis in harmonic reducers is the main cause of joint mechanical damage. Therefore, a long-term damage dynamic accumulation assessment and analysis is constructed. The thermochemical wear component is calculated and the electrochemical decomposition rate is quantified based on the parasitic potential enhancement coefficient. When the long-term damage dynamic accumulation reaches the warning threshold, a micro grease pump injects zinc dialkyl dithiophosphate into the meshing tooth surface, which reacts with the metal surface to form a phosphate anti-wear film, blocking the electrolytic chain reaction. When the limit is exceeded, the electromagnetic ratchet mechanism immediately locks the output shaft, and the LoRa wireless module sends a replacement alarm. This mechanism links the damage accumulation with the real-time control status, realizing progressive maintenance from delayed failure to active repair.

[0022] 4. Oil mist penetration in the ship's engine room and engine vibration amplify the interference of parasitic potential on electronic systems. Therefore, oil mist concentration is monitored in real time using a laser forward scattering sensor. Combined with the oil mist conductivity coupling coefficient obtained from calibration experiments, the current flux jump caused by oil droplet bridging microcracks is quantified. Then, the vibration main frequency is captured based on a triaxial accelerometer. When it approaches the engine's fundamental frequency, the vibration frequency locking correction factor is automatically activated, reflecting the amplification effect of resonance on parasitic potential fluctuations. Furthermore, the difference coefficient of parasitic potential in the heterojunction deposition layer is used to dynamically adjust the encoder signal processing chain: a moving average filter is used when the risk is low, and an adaptive notch filter is switched to and the signal tolerance threshold is relaxed when the risk is medium to high. This ensures that the control system maintains stable position feedback under strong interference and avoids malfunctions caused by encoder frame loss. Attached Figure Description

[0023] Figure 1 A flowchart of the exoskeleton joint control method based on edge computing provided in this application embodiment. Detailed Implementation

[0024] This application provides an edge computing-based exoskeleton joint control method. The technical solution in this application addresses the aforementioned problem of insufficient environmental adaptability of joint control in exoskeleton operations for ship engine room maintenance. The overall approach is as follows: This method uses edge computing to monitor the metal vapor deposition thickness, ambient temperature, and vibration state of the exoskeleton wrist joint in real time. When heterojunction deposition exceeds a threshold, a nitrogen curtain is activated to block deposition. Based on the temperature-heterojunction coupling effect, hierarchical control is implemented: in medium-risk situations, laser stripping of the deposition layer and spraying of a protective coating are performed; in high-risk situations, phase change cooling and mechanical locking are initiated. The influence of oil mist concentration and vibration frequency on parasitic potential is analyzed simultaneously, and signal filtering strategies are dynamically adjusted. When long-term damage accumulation exceeds limits, anti-wear additives are injected or the joint is locked, achieving active safety protection for the exoskeleton joints in the harsh environment of the ship engine room.

[0025] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0026] like Figure 1 The diagram shows a flowchart of an exoskeleton joint control method based on edge computing provided in this application. The method includes the following steps: evaluating the metal vapor temperature component of the exoskeleton; determining and analyzing the influence component of the heterojunction temperature coupling on the exoskeleton wrist joint; performing a first control of the exoskeleton joint for ship engine room maintenance based on the temperature component evaluation results; performing long-term damage dynamic accumulation analysis control based on the long-term damage dynamic accumulation analysis results, wherein the first control of the exoskeleton joint for ship engine room maintenance includes long-term damage dynamic accumulation analysis control; performing a multi-dimensional coupling evaluation of the exoskeleton wrist joint; and performing multi-dimensional coupling control of the exoskeleton wrist joint based on the multi-dimensional coupling evaluation.

[0027] In this embodiment, the grease in the harmonic reducer undergoes saponification and failure within 48 hours under sulfide catalysis, leading to a surge in the coefficient of friction. Furthermore, the carbon fiber joint support experiences a decrease in interlaminar shear strength in a humid and hot environment, increasing the risk of structural fracture. More seriously, when a sudden load signal, such as a sudden change in bolt disassembly reaction force, couples with parasitic potentials, it can easily trigger false protection, causing joint lock-up and potentially resulting in wrist injuries for maintenance personnel. Therefore, timely assessment and control are necessary.

[0028] Furthermore, the metal vapor temperature component of the exoskeleton is assessed, specifically including: inducing the breakdown of the heterojunction of the exoskeleton wrist joint using a pulsed laser emitted from the wrist joint pulsed laser; collecting the heterojunction excitation spectrum of the exoskeleton wrist joint using a fiber optic spectral sensor to obtain the characteristic spectral lines of the heterojunction; extracting features from the characteristic spectral lines of the heterojunction and comparing them with predefined heterojunction characteristic spectral lines in the exoskeleton wrist joint control database to obtain the actual thickness of the heterojunction; if the actual thickness of the heterojunction is less than a first threshold, then the influence component of the temperature coupling of the heterojunction on the exoskeleton wrist joint is analyzed; if the actual thickness of the heterojunction is equal to or greater than the first threshold and less than a second threshold, then a preventive nitrogen curtain is activated; if the actual thickness of the heterojunction is equal to or greater than the second threshold, then a warning notification is issued to relevant personnel.

[0029] In this embodiment, the front end cap of the wrist joint of the human-machine exoskeleton is the area with the most severe metal vapor deposition, and all detection devices are avoided in the wrist joint's movement space.

[0030] The fiber optic spectral sensor uses a miniature LIBS probe, which is installed in the countersunk hole at the front end of the wrist joint of the human-machine exoskeleton.

[0031] Feature extraction was performed on the heterojunction characteristic spectral lines of the human-machine exoskeleton wrist joint, and the results were compared and analyzed with the predefined heterojunction characteristic spectral lines in the human-machine exoskeleton wrist joint control database to detect the actual thickness of the heterojunction.

[0032] If the actual thickness of the heterojunction is equal to or greater than the first threshold of the actual thickness of the heterojunction and less than the second threshold of the actual thickness of the heterojunction, and the heterojunction has not yet formed, a preventive nitrogen curtain is activated to block the contact of metal vapor in the early stage of deposition, thereby significantly reducing the probability of heterojunction formation.

[0033] Further, the analysis of the influence components of the heterojunction temperature-coupled human-machine exoskeleton wrist joint is determined. Specifically, this includes: acquiring the ambient temperature of the human-machine exoskeleton wrist joint using a temperature sensor; if the ambient temperature is greater than the inflection point of the heterojunction temperature influence, then the analysis of the influence components of the heterojunction temperature-coupled human-machine exoskeleton wrist joint is performed; otherwise, no analysis is performed. The specific process of this analysis is as follows: obtaining the basic Seebeck coefficient and the heterojunction temperature sensitivity coefficient of the human-machine exoskeleton wrist joint by measuring a standard heterojunction sample in the laboratory; directly extracting the initial contact resistance from the human-machine exoskeleton wrist joint control database; measuring the current contact resistance on the front end cap of the human-machine exoskeleton wrist joint using the four-probe method; and then performing a minimum value function analysis with the unit resistance ratio to obtain the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component. The following methods were used to determine the effects of different deposition thicknesses on temperature sensitivity: a Seebeck coefficient was measured using accelerated aging tests to assess the temperature sensitivity of the heterojunction deposition thickness at the inflection point where the ambient temperature of the exoskeleton wrist joint exceeds the heterojunction temperature. A temperature sensitivity enhancement coefficient was obtained by comparing the actual heterojunction thickness with a first threshold value, then performing a squared percentage analysis, and finally coupling this result with the temperature sensitivity enhancement coefficient to obtain a deposition thickness enhancement temperature sensitivity coefficient. Finally, the results of the analysis of the differences between the base Seebeck coefficient, the heterojunction coverage coefficient, the exoskeleton wrist joint heterojunction temperature sensitivity coefficient, the deposition thickness enhancement temperature sensitivity coefficient, and the inflection point of the effect of ambient temperature on the exoskeleton wrist joint and the heterojunction temperature were coupled to obtain the heterojunction temperature-coupled influence component on the exoskeleton wrist joint.

[0034] In this embodiment, the constraint formula for the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component is as follows: a(T)=SBKB*λ+WYJ*K*(T-GYZ), where a(T) represents the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component.

[0035] SBKB represents the basic Seebeck coefficient, which is used to represent the Seebeck coefficient of the heterojunction below the inflection point affected by the heterojunction temperature when there is no deposition on the heterojunction. The basic Seebeck coefficient is specifically obtained by measuring on laboratory standard samples. For example: using a standard sample of FeS / ZnS heterojunction with a thickness of 1 μm prepared by vapor deposition, in a constant temperature environment of 100 °C, applying different temperature differences, measuring the generated thermoelectric potential, obtaining the Seebeck coefficient, repeating the experiment and taking the average value, getting 100 μV / K, with a standard deviation of ±0.5 μV / K, then the corresponding basic Seebeck coefficient is taken as 100.

[0036] T represents the environmental temperature of the wrist joint of the human-machine exoskeleton, and GYZ represents the inflection point affected by the heterojunction temperature. The inflection point affected by the heterojunction temperature is used to describe from this heterojunction temperature, the heterojunction. The inflection point affected by the heterojunction temperature is calibrated by taking the average value of experimental tests and is obtained as 100 °C.

[0037] Among them, λ represents the heterojunction coverage degree coefficient. RC0 represents the initial contact resistance, which is used to describe the reference value of the contact resistance measured in the factory log of the wrist joint of the human-machine exoskeleton. For example: after the initial installation of the exoskeleton wrist joint or maintenance and replacement of the seal, in an oil-free mist environment of 25 °C, the resistance value measured at a specific detection point on the front end cover of the joint using the four-probe method. This value is stored as a reference in the regulation database of the wrist joint of the human-machine exoskeleton; RC1 represents the current contact resistance, which is a real-time measurement value. For example: during the actual operation process, through the four-probe sensor installed on the front end cover of the wrist joint, which is the same position as the measurement of the initial contact resistance, the current contact resistance is measured. A constant current is applied during the measurement, and then the voltage drop is measured to calculate the resistance.

[0038] When there is no heterojunction deposition, RC1 is close to RC0, and λ≈1; as the deposition thickens, the deposition layer may form a conductive path, and the formation of the heterojunction causes the contact resistance to decrease, making RC1 < RC0, so λ decreases; the lower the value of λ, the higher the heterojunction coverage, because the coverage of the heterojunction changes the overall properties of the material, and the basic Seebeck coefficient needs to be reduced proportionally.

[0039] WYJ represents the heterojunction temperature sensitivity coefficient of the wrist joint of the human-machine exoskeleton; it is used to represent the growth rate of the Seebeck coefficient with the increase in temperature when the temperature exceeds the inflection point affected by the heterojunction temperature. For example: conducting a temperature change experiment on a standard sample. The sample is gradually heated from 100 °C to 150 °C, and the Seebeck coefficient is measured every 5 °C increase. By linear fitting, the slope of the change of the Seebeck coefficient with temperature is obtained. If it is measured as 1.3 μV / K 2That is, for every 1K increase in temperature, the Seebeck coefficient increases by 1.3μV / K. The second threshold of the actual thickness of the heterojunction represents the immediate alarm threshold. Based on prior knowledge from experts, the temperature range for the sample temperature change experiment here is from the first threshold of the actual thickness of the heterojunction to the second threshold of the actual thickness of the heterojunction.

[0040] Wherein, K represents the deposition thickness enhancement temperature sensitivity coefficient, JZH1 represents the first threshold of the actual thickness of the heterojunction, and δ represents the actual thickness of the heterojunction; the first threshold and the second threshold of the actual thickness of the heterojunction are directly extracted from the human-machine exoskeleton wrist joint control database. The first threshold of the actual thickness of the heterojunction represents the reference value of the deposition thickness, which is obtained based on expert prior knowledge.

[0041] ZQX1 represents the temperature sensitivity enhancement coefficient of heterojunction deposition thickness, obtained through fitting experimental data. It reflects the degree to which deposition thickness enhances temperature sensitivity. For example, in accelerated aging tests in the laboratory, the Seebeck coefficient temperature sensitivity under different deposition layers is measured. This is the slope of the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component when the ambient temperature of the human-machine exoskeleton wrist joint is greater than the inflection point of the heterojunction temperature influence, used to describe the enhancing effect of heterojunction deposition thickness on sensitivity.

[0042] Furthermore, based on the temperature component assessment results, the first adjustment of the exoskeleton joint during ship engine room maintenance is implemented. Specifically, if the heterojunction temperature coupling effect component of the exoskeleton wrist joint is less than the first threshold of the heterojunction temperature coupling effect component, no additional adjustment is performed. If the heterojunction temperature coupling effect component of the exoskeleton wrist joint is equal to or greater than the first threshold of the heterojunction temperature coupling effect component and less than the second threshold of the heterojunction temperature coupling effect component, then the thickening of the deposition layer induces a nonlinear Seebeck effect, causing fluctuations in joint torque. The corresponding laser precision stripping operation procedure is as follows: the infrared thermal imager locates the point of maximum ambient temperature at the exoskeleton wrist joint, and the laser head on the side wall is used to... The pulse is emitted, and the depth of action is measured in real time to avoid damaging the substrate. After laser cleaning, the plasma spray gun is immediately started to spray hexagonal boron nitride and perform heat treatment to solidify it. If the heterojunction temperature coupling effect component of the human exoskeleton wrist joint is equal to or greater than the second threshold of the heterojunction temperature coupling effect component, the high-temperature phase change will cause the Seebeck coefficient to increase sharply, causing the motor current loop to run out of control. The rapid cooling operation process of the phase change material is as follows: the heat pipe array directs the heat to the PCM module, the PCM module absorbs the latent heat of the phase change, and the circulation pump conducts the residual heat to the external heat sink. The electromagnetic mechanical locking operation process is as follows: the coil is energized to drive the tungsten carbide pawl to engage in the output shaft groove and the standby piezoelectric ceramic brake locks synchronously, and the joint is switched to passive damping mode.

[0043] In this embodiment, the laser-precision stripping operation is used to remove the formed FeS / ZnS heterojunction semiconductor layer, directly eliminating the physical basis of the Seebeck effect.

[0044] Plasma spray guns are used to prevent secondary deposition of metal vapors.

[0045] Rapid cooling of phase change materials can be used to rapidly reduce temperature and avoid a jump in carrier mobility.

[0046] Electromagnetic mechanical locking is used when the parasitic potential is greater than 20mV when the component of the heterojunction temperature coupling effect on the wrist joint of the human exoskeleton is equal to or greater than the second threshold of the heterojunction temperature coupling effect component. This will cause the motor torque fluctuation to rise significantly, far exceeding the human safety threshold. Mechanical locking is the only reliable protection.

[0047] Furthermore, a long-term dynamic cumulative damage analysis was conducted, specifically including: obtaining the basic wear coefficient by directly testing the harmonic reducer under the wrist joint of the human-machine exoskeleton using a rotary wear tester in a laboratory accelerated aging test device; directly extracting the activation energy, gas constant, and regulation correction factor of the polyurea-based lubricating grease for the human-machine exoskeleton wrist joint harmonic reducer from the human-machine exoskeleton wrist joint regulation database; directly extracting the potential enhancement coefficient from the electrochemical decomposition experimental data of the lubricating grease; directly measuring the parasitic potential by using differential electrodes on the anterior cap of the wrist joint; and applying the polyurea-based lubricating grease to the human-machine exoskeleton wrist joint harmonic reducer. The activation energy of the base grease was coupled with the gas constant and the ambient temperature of the wrist joint of the human-machine exoskeleton for proportional analysis. This result was used as an index for index analysis, and then coupled with the base wear coefficient to obtain the dynamic temperature component of the human-machine exoskeleton wrist joint for long-term damage. The potential enhancement coefficient and parasitic potential were coupled for analysis to obtain the dynamic potential component of the human-machine exoskeleton wrist joint for long-term damage. The dynamic temperature component, the dynamic potential component, and the regulation and correction factor of the human-machine exoskeleton wrist joint for long-term damage were analyzed together to obtain the dynamic cumulative index of the human-machine exoskeleton wrist joint for long-term damage.

[0048] In this embodiment, the specific constraint model for the long-term injury dynamic cumulative index of the wrist joint of the human-machine exoskeleton is as follows: Where Da represents the dynamic cumulative index of long-term injury of the wrist joint of the human-machine exoskeleton, which is used to quantify the dynamic cumulative level of long-term injury of the wrist joint of the human-machine exoskeleton from the start of the detection to the present. Note that t here represents the cumulative timing duration, and 0 represents the start time of detection. As will be known below, under certain conditions, the cumulative timing duration will be reset to zero and accumulated again.

[0049] k0 represents the basic wear coefficient, which describes the basic wear rate of the wrist joint of the human-machine exoskeleton under normal temperature (25°C) and interference-free working conditions. Specifically, it is the basic wear rate of the steel-copper friction pair of the harmonic reducer under the wrist joint of the human-machine exoskeleton. For example, it can be obtained by direct experimentation using a rotary wear tester in a laboratory accelerated aging test equipment.

[0050] e represents the natural constant.

[0051] Ea represents the activation energy of polyurea-based grease in the wrist joint harmonic reducer of the human-machine exoskeleton. It is the energy value required for the grease molecular bond breaking. It is obtained in advance through differential scanning calorimetry peak analysis using a DSC analyzer and is directly extracted from the human-machine exoskeleton wrist joint control database.

[0052] R represents the gas constant, used to describe the constant value of the ideal gas law. It is a physical constant and is directly extracted from the human-machine exoskeleton wrist joint control database.

[0053] T represents the ambient temperature of the wrist joint of the human-machine exoskeleton.

[0054] Vp represents the parasitic potential, used to describe the Seebeck effect potential difference induced by heterojunction deposition. It is directly measured by a differential electrode, which is co-located with the LIBS probe and mounted on the front end cap of the wrist joint.

[0055] DQX represents the potential enhancement coefficient, which is directly extracted from the electrochemical decomposition experimental data of grease. For example, by using step voltage testing, the weight loss rate of grease under different step voltages is obtained, and the fitting coefficient is obtained, which is the potential enhancement coefficient. It is applicable to all polyurea-based / lithium-based marine greases and is used to quantify the different grease loss rates under different step voltages.

[0056] η represents the regulation correction factor, which is directly extracted from the human-machine exoskeleton wrist joint regulation database. It is used to quantify the inhibitory effect of regulation measures on damage accumulation. Based on the dynamic switching of real-time regulation state: through bench comparison test, the relative damage accumulation inhibition accumulation rate comparison data is obtained, and the corresponding regulation correction factor is obtained. The first-level regulation corresponds to laser precision peeling operation and plasma spray gun spraying, and the second-level regulation corresponds to phase change material rapid cooling operation and electromagnetic mechanical locking operation.

[0057] Furthermore, the long-term injury dynamic accumulation analysis also includes: if the comparative analysis results of the heterojunction temperature coupling influence component of the human-machine exoskeleton wrist joint and the first threshold or the second threshold of the heterojunction temperature coupling influence component change, then the current long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is recorded as the base value of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint. The corresponding regulation correction factor is adjusted according to the current comparative analysis results, and the specific constraint model of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is updated.

[0058] In this embodiment, for example, if the influence component of the heterojunction temperature coupling on the wrist joint of the human-machine exoskeleton is less than the first threshold of the heterojunction temperature coupling influence component, no additional adjustment is performed; at a certain moment, if the influence component of the heterojunction temperature coupling on the wrist joint of the human-machine exoskeleton is equal to or greater than the second threshold of the heterojunction temperature coupling influence component, then the second adjustment of the ship engine room maintenance exoskeleton joint is performed, and the current long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is recorded as the base value of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint. DaBASE represents the base value of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint. The specific constraint model of the long-term injury dynamic accumulation index of the human-machine exoskeleton wrist joint is updated as follows: The accumulated time is reset to zero, and t is accumulated again until the next change in the adjustment factor.

[0059] Furthermore, based on the results of long-term damage dynamic accumulation analysis, long-term damage dynamic accumulation analysis and control are carried out. Specifically, this includes: if the long-term damage dynamic accumulation index of the exoskeleton wrist joint is less than the long-term damage dynamic accumulation threshold, then dialkyl dithiophosphate zinc is injected into the meshing tooth surface of the harmonic reducer through a micro grease pump and radial oil passage. The dialkyl dithiophosphate zinc reacts with the metal surface to form a phosphate anti-wear film, which is used to block the grease electrolytic chain reaction caused by parasitic potential, thereby reducing the wear rate. The grease pump is integrated into the back cavity of the wrist joint, and the oil circuit system uses hard alloy oil pipes. If the long-term damage dynamic accumulation index of the exoskeleton wrist joint is equal to or greater than the long-term damage dynamic accumulation threshold, then the output shaft is physically locked through an electromagnetic ratchet mechanism, and a maintenance alarm is sent to the external control console through a LoRa wireless module to suggest replacing the reducer module. The accumulated damage exceeds the material fatigue limit, and the risk of fracture is high, so the machine must be shut down.

[0060] Furthermore, a multi-dimensional coupling evaluation of the wrist joint of the human-machine exoskeleton was conducted. The specific process was as follows: The oil mist concentration conductivity coupling coefficient was obtained under different oil mist concentrations and deposition layer thicknesses through an oil mist concentration conductivity calibration experiment; the oil mist concentration in the wrist joint environment was collected and analyzed using a laser forward scattering sensor under the wrist joint's side protective shield; the oil mist concentration conductivity coupling coefficient and the oil mist concentration in the wrist joint environment were coupled and analyzed; the real-time vibration main frequency was collected using a triaxial accelerometer on the input shaft flange of the harmonic reducer; based on the real-time vibration main frequency, the average fundamental frequency of the human-machine exoskeleton operation was extracted from historical data; a vibration locking correction factor was obtained based on the real-time vibration main frequency and the average fundamental frequency of the human-machine exoskeleton operation; and the parasitic potential coefficient of the heterojunction deposition layer was obtained by coupling and analyzing the heterojunction temperature-coupled human-machine exoskeleton wrist joint influence component, the oil mist concentration conductivity coupling coefficient, the oil mist concentration in the wrist joint environment, and the vibration locking correction factor.

[0061] In this embodiment, the parasitic potential coefficient of the heterojunction deposition layer is constrained as follows: VpG=a(T)*(1+YDSX*Co)*φ(fv), where VpG represents the parasitic potential coefficient of the heterojunction deposition layer, which is used to quantify the risk level of the surge in parasitic potential of the heterojunction deposition layer of the human-machine exoskeleton wrist joint under the coupling effect of the ship engine room exoskeleton wrist joint, and a(T) represents the influence component of the heterojunction temperature coupling of the human-machine exoskeleton wrist joint.

[0062] Co represents the oil mist concentration in the wrist joint environment of the human-machine exoskeleton, which is obtained by collecting and analyzing data from a laser forward scattering sensor under the protective shield on the side of the wrist joint.

[0063] YDSX represents the oil mist concentration-conductivity coupling coefficient, used to describe the probability density of oil droplets forming conductive bridges in the microcracks of the deposition layer. For example, per 1 mg / m³ 3 The increase in oil mist concentration enhances the parasitic potential. The oil droplet bridging of microcracks in the deposition layer greatly reduces the contact resistance, resulting in a several-fold increase in the Seebeck current flux. Through oil mist concentration conductivity calibration experiments, different potential enhancement rates were obtained under different oil mist concentrations and deposition layer thicknesses. The corresponding average fitting coefficient was obtained, which is the oil mist concentration conductivity coupling coefficient.

[0064] φ(fv) represents the vibration frequency locking correction factor, used to quantify the level at which resonance in actual operation causes the expansion of microcracks in the deposition layer, leading to an increase in the amplitude of parasitic potential fluctuations and thus an increase in the frame drop rate of the encoder controlling the wrist joint of the exoskeleton. If |fv-ZDPD*n|≤0.5, n=1,2,...,∞, then φ(fv)=ZCZ, where ZCZ represents the test resonance factor; otherwise, φ(fv)=1. ZDPD represents the average fundamental frequency of the exoskeleton operation extracted from historical data, which is generally the vibration frequency of the corresponding ship engine.

[0065] The resonance effect is obtained by testing the vibration table. For example, when the exoskeleton vibrates at the average fundamental frequency of operation, the average ratio of the rise of the three aspects, namely the vibration acceleration amplitude, the deposition crack propagation rate, and the parasitic potential fluctuation amplitude, is taken as the test resonance factor.

[0066] fv represents the real-time vibration frequency, which is acquired by a triaxial MEMS accelerometer on the input shaft flange of the harmonic reducer.

[0067] Furthermore, based on the multidimensional coupling assessment of the human-machine exoskeleton wrist joint, multidimensional coupling regulation of the human-machine exoskeleton wrist joint is carried out. Specifically, this includes: if the parasitic potential coefficient of the heterojunction deposition layer is less than the first threshold of the parasitic potential of the heterojunction deposition layer, dynamic adjustment of baseline noise filtering is performed; if the parasitic potential coefficient of the heterojunction deposition layer is equal to or greater than the first threshold of the parasitic potential of the heterojunction deposition layer and less than the second threshold of the parasitic potential of the heterojunction deposition layer, a preventive nitrogen curtain is activated; if the parasitic potential coefficient of the heterojunction deposition layer is equal to or greater than the second threshold of the parasitic potential of the heterojunction deposition layer, an early warning notification is issued to relevant personnel.

[0068] In this embodiment, the drastically fluctuating parasitic potential is coupled to the encoder signal line, sensor power supply line, control bus, etc. in the weak current control system of the human exoskeleton wrist joint, causing encoder signal frame loss; parasitic noise is superimposed on the encoder pulse signal, causing speed feedback errors; causing ADC sampling errors; causing logic circuit malfunctions; strong interference may cause microcontroller reset, logic errors, or even trigger false protection.

[0069] Furthermore, dynamic adjustment of the baseline noise filtering is performed, specifically as follows:

[0070] Divide the first threshold of the parasitic potential of the heterojunction deposit by the parasitic potential coefficient of the heterojunction deposit to obtain the parasitic potential difference coefficient of the heterojunction deposit.

[0071] If the parasitic potential difference coefficient of the heterojunction deposition layer is less than the first threshold of the parasitic potential difference of the heterojunction deposition layer, then the basic moving average filter is enabled for the encoder signal baseline of the wrist joint of the human-machine exoskeleton.

[0072] If the parasitic potential difference coefficient of the heterojunction deposition layer is equal to or greater than the first threshold of the parasitic potential difference of the heterojunction deposition layer, then switch to activate dynamic window moving average and recursive filter low-pass filtering.

[0073] In this embodiment, the window size of the basic moving average filter is fixed. The minimum and maximum window sizes are set according to the processing power of the edge computing chip to achieve low computational overhead while ensuring the filtering of high-frequency noise spikes.

[0074] The constraint formula for dynamic window moving average is as follows:

[0075] Wherein, CKN represents the dynamic window size of the moving average filter, which is used to quantify the dynamic window size of the moving average filter as the parasitic potential difference coefficient of the heterojunction deposit layer increases, MCKN represents the minimum window size, ACKN represents the maximum window size, and JYX represents the parasitic potential difference coefficient of the heterojunction deposit layer. This indicates rounding up to the nearest integer.

[0076] The cutoff frequency constraint formula for low-pass filtering of a recursive filter is as follows:

[0077] Where fc represents the actual low-pass filter cutoff frequency of the recursive filter, JZM represents the minimum low-pass filter cutoff frequency of the recursive filter, which is directly extracted from the historical minimum cutoff frequency of the encoder signal line of the exoskeleton wrist joint, and JZA represents the maximum low-pass filter cutoff frequency of the recursive filter, which is also directly extracted from the historical minimum cutoff frequency of the encoder signal line of the exoskeleton wrist joint. This ensures that the cutoff frequency constraint of the recursive filter low-pass filter decreases as the parasitic potential difference coefficient of the heterojunction deposition layer increases.

[0078] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0082] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0083] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for controlling exoskeleton joints based on edge computing, characterized in that, Includes the following steps: Evaluation of the metal vapor temperature components of human-machine exoskeletons; To determine the influence component of the wrist joint of the heterojunction temperature-coupled human-machine exoskeleton; Based on the temperature component assessment results, the first adjustment of the exoskeleton joints during ship engine room maintenance is carried out. Based on the results of long-term dynamic cumulative analysis of damage, long-term dynamic cumulative analysis and control should be carried out to address the first control of the exoskeleton joints in ship engine room maintenance, including long-term dynamic cumulative analysis and control. Perform multidimensional coupling assessment of the wrist joint of the human-machine exoskeleton; Multidimensional coupling regulation of the wrist joint of a human-machine exoskeleton is performed based on multidimensional coupling assessment of the wrist joint.

2. The exoskeleton joint control method based on edge computing as described in claim 1, characterized in that, The assessment of the metal vapor temperature component of the human-machine exoskeleton specifically includes: The pulsed laser emitted by the pulsed laser of the exoskeleton wrist joint induces the breakdown of the heterojunction of the exoskeleton wrist joint. The heterojunction excitation spectrum of the exoskeleton wrist joint is collected by the fiber optic spectral sensor to obtain the characteristic spectral lines of the heterojunction. The characteristic spectral lines of the heterojunction are extracted and compared with the predefined heterojunction characteristic spectral lines in the exoskeleton wrist joint control database to obtain the actual thickness of the heterojunction. If the actual thickness of the heterojunction is less than the first threshold of the actual thickness of the heterojunction, then it is determined that the influence component of the wrist joint of the heterojunction temperature coupling human-machine exoskeleton should be analyzed. If the actual thickness of the heterojunction is equal to or greater than the first threshold of the actual thickness of the heterojunction and less than the second threshold of the actual thickness of the heterojunction, then the preventive nitrogen curtain is activated. If the actual thickness of the heterojunction is equal to or greater than the second threshold of the actual thickness of the heterojunction, an early warning notification will be issued to the relevant personnel.

3. The exoskeleton joint control method based on edge computing as described in claim 2, characterized in that, The judgment involves analyzing the influence component of the heterojunction temperature-coupled human-machine exoskeleton wrist joint, specifically including: The ambient temperature of the wrist joint of the human-machine exoskeleton is collected by a temperature sensor. If the ambient temperature of the wrist joint of the human-machine exoskeleton is greater than the inflection point of the heterojunction temperature influence, the heterojunction temperature coupling human-machine exoskeleton wrist joint influence component analysis is performed; otherwise, no analysis is performed. The analysis of the influence component of the heterojunction temperature coupling on the wrist joint of the human-machine exoskeleton is as follows: The basic Seebeck coefficient and the temperature sensitivity coefficient of the human-machine exoskeleton wrist joint heterostructure were obtained by measuring standard samples of heterostructures in the laboratory. The initial contact resistance was directly extracted from the human-machine exoskeleton wrist joint control database; The current contact resistance was measured at the anterior cap of the wrist joint of the human-machine exoskeleton using a four-probe method. By analyzing the ratio of initial contact resistance to current contact resistance, and then performing a minimum value function analysis with unit 1, the heterojunction coverage coefficient is obtained. The Seebeck coefficient temperature sensitivity of different thickness deposited layers was measured by laboratory accelerated aging test at the inflection point of the effect of the ambient temperature of the wrist joint of the human-machine exoskeleton being greater than the temperature of the heterojunction. The temperature sensitivity enhancement coefficient of the heterojunction deposition thickness was obtained. The difference between the first threshold of the actual thickness of the heterojunction and the actual thickness of the heterojunction is analyzed, and then the square of the ratio analysis is performed on the first threshold of the actual thickness of the heterojunction. Finally, the temperature sensitivity enhancement coefficient of the deposition thickness is obtained by coupling analysis with the temperature sensitivity enhancement coefficient of the heterojunction deposition thickness. The results of the analysis of the difference between the basic Seebeck coefficient, the heterojunction coverage coefficient, the temperature sensitivity coefficient of the heterojunction of the human-machine exoskeleton wrist joint, the temperature sensitivity coefficient of the deposition thickness enhancement, and the inflection point of the influence of the ambient temperature of the human-machine exoskeleton wrist joint and the temperature of the heterojunction were coupled to obtain the heterojunction temperature coupled influence component of the human-machine exoskeleton wrist joint.

4. The exoskeleton joint control method based on edge computing as described in claim 1, characterized in that, The first adjustment of the exoskeleton joints for ship engine room maintenance based on temperature component assessment results specifically includes: Perform long-term dynamic cumulative damage analysis; If the component of the heterojunction temperature coupling effect on the wrist joint of the human exoskeleton is less than the first threshold of the heterojunction temperature coupling effect, no additional control will be performed. If the influence component of the heterojunction temperature coupling human-machine exoskeleton wrist joint is equal to or greater than the first threshold of the heterojunction temperature coupling influence component and the influence component of the heterojunction temperature coupling human-machine exoskeleton wrist joint is less than the second threshold of the heterojunction temperature coupling influence component, then the first adjustment of the exoskeleton joint for ship engine room maintenance will be performed. The first control of the ship engine room maintenance exoskeleton joint first adjustment starts the laser precision peeling operation, and the plasma spray gun is immediately started after the laser precision peeling operation. If the component of the heterojunction temperature coupling effect on the wrist joint of the human exoskeleton is equal to or greater than the second threshold of the heterojunction temperature coupling effect, then the second adjustment of the exoskeleton joint for ship engine room maintenance will be performed. The second control of the ship's engine room maintenance exoskeleton joint initiates a phase change material rapid cooling operation. Immediately after the phase change material rapid cooling operation, an electromagnetic mechanical locking operation is initiated, and a warning notification is issued to relevant personnel.

5. The exoskeleton joint control method based on edge computing as described in claim 4, characterized in that, The aforementioned long-term dynamic cumulative damage analysis specifically includes: The basic wear coefficient was obtained by directly testing the harmonic reducer under the wrist joint of the human-machine exoskeleton using a rotary wear tester in a laboratory accelerated aging test equipment. The activation energy, gas constant, and regulation correction factor of the polyurea-based lubricating grease in the harmonic reducer of the human-machine exoskeleton wrist joint were directly extracted from the human-machine exoskeleton wrist joint regulation database. The potential enhancement coefficient was directly extracted from the experimental data of electrochemical decomposition of lubricating grease. Parasitic potential was directly measured by differential electrodes on the anterior cap of the wrist joint; The activation energy of polyurea-based grease in the harmonic reducer of the human-machine exoskeleton wrist joint is coupled with the gas constant and the ambient temperature of the human-machine exoskeleton wrist joint for proportional analysis. The results are used as an index for index analysis, and then coupled with the basic wear coefficient for analysis to obtain the dynamic temperature component of long-term damage of the human-machine exoskeleton wrist joint. By coupling the potential enhancement coefficient and the parasitic potential, the dynamic potential components of the wrist joint of the human-machine exoskeleton for long-term injury are obtained. By jointly analyzing the dynamic temperature component, the dynamic potential component, and the regulation correction factor of the long-term injury of the wrist joint of the human-machine exoskeleton, the dynamic cumulative index of long-term injury of the wrist joint of the human-machine exoskeleton is obtained.

6. The exoskeleton joint control method based on edge computing as described in claim 4, characterized in that, The long-term dynamic accumulation analysis of damage also includes: If the comparative analysis results of the heterojunction temperature coupling influence component of the human-machine exoskeleton wrist joint and the first threshold or the second threshold of the heterojunction temperature coupling influence component change, then the current dynamic cumulative index of long-term injury of the human-machine exoskeleton wrist joint is recorded as the base value of the dynamic cumulative index of long-term injury of the human-machine exoskeleton wrist joint. The corresponding adjustment correction factor is adjusted according to the current comparative analysis results, the cumulative time duration is cleared, and the specific constraint model of the dynamic cumulative index of long-term injury of the human-machine exoskeleton wrist joint is updated.

7. The exoskeleton joint control method based on edge computing as described in claim 1, characterized in that, The regulation based on the results of long-term damage dynamic accumulation analysis specifically includes: If the long-term dynamic cumulative index of wrist joint injury of human-machine exoskeleton is less than the long-term dynamic cumulative threshold of wrist joint injury of human-machine exoskeleton, then a phosphate anti-wear film is generated by injecting through a micro liposuction pump and a multi-dimensional coupling assessment of wrist joint injury of human-machine exoskeleton is performed. If the long-term dynamic cumulative index of wrist joint injury of human-machine exoskeleton is equal to or greater than the long-term dynamic cumulative threshold of wrist joint injury of human-machine exoskeleton, the electromagnetic mechanical locking operation will be immediately activated and a warning notification will be issued to relevant personnel.

8. The exoskeleton joint control method based on edge computing as described in claim 7, characterized in that, The specific process for conducting the multidimensional coupling assessment of the wrist joint of the human-machine exoskeleton is as follows: The oil mist concentration conductivity coupling coefficient was obtained through an oil mist concentration conductivity calibration experiment at different oil mist concentrations and deposition layer thicknesses. The concentration of oil mist in the wrist joint environment of the human-machine exoskeleton was obtained by collecting and analyzing data from a laser forward scattering sensor under the side shield of the wrist joint. The coupling coefficients of oil mist concentration and conductivity and the oil mist concentration in the wrist joint environment of the human-machine exoskeleton were coupled and analyzed. The real-time vibration main frequency is obtained by using a triaxial accelerometer on the input shaft flange of the harmonic reducer; The average fundamental frequency of human-machine exoskeleton operation was extracted from historical data; The vibration frequency locking correction factor is obtained based on the real-time vibration main frequency and the average fundamental frequency of the human-machine exoskeleton operation. The parasitic potential coefficient of the heterojunction deposition layer was obtained by coupling the heterojunction temperature with the wrist joint effect component of the human-machine exoskeleton, the oil mist concentration conductivity coupling coefficient, the oil mist concentration of the human-machine exoskeleton wrist joint environment, and the vibration frequency locking correction factor.

9. The exoskeleton joint control method based on edge computing as described in claim 1, characterized in that, The multidimensional coupling regulation of the wrist joint of the human-machine exoskeleton based on the multidimensional coupling assessment specifically includes: If the parasitic potential coefficient of the heterojunction deposition layer is less than the first threshold of the parasitic potential of the heterojunction deposition layer, then dynamic adjustment of the baseline noise filtering is performed. If the parasitic potential coefficient of the heterojunction deposit layer is equal to or greater than the first threshold of the parasitic potential of the heterojunction deposit layer and the parasitic potential coefficient of the heterojunction deposit layer is less than the second threshold of the parasitic potential of the heterojunction deposit layer, then the preventive nitrogen curtain is activated. If the parasitic potential coefficient of the heterojunction deposition layer is equal to or greater than the second threshold of the parasitic potential of the heterojunction deposition layer, an early warning notice will be issued to the relevant personnel.

10. The exoskeleton joint control method based on edge computing as described in claim 9, characterized in that, The dynamic adjustment of the baseline noise filtering is specifically performed as follows: Divide the first threshold of the parasitic potential of the heterojunction deposit by the parasitic potential coefficient of the heterojunction deposit to obtain the parasitic potential difference coefficient of the heterojunction deposit. If the parasitic potential difference coefficient of the heterojunction deposition layer is less than the first threshold of the parasitic potential difference of the heterojunction deposition layer, then the basic moving average filter is enabled for the encoder signal baseline of the wrist joint of the human-machine exoskeleton. If the parasitic potential difference coefficient of the heterojunction deposition layer is equal to or greater than the first threshold of the parasitic potential difference of the heterojunction deposition layer, then switch to activate dynamic window moving average and recursive filter low-pass filtering.

Images