Robot platform stability keeping method, system and equipment and storage medium

Abstract

The invention relates to the technical field of robot control, provides a robot platform stability keeping method, system and equipment and a storage medium, and solves the problem of low attitude stability caused by poor anti-interference capability of a high-altitude operation robot in a complex dynamic environment. The method comprises the steps that acceleration, angular velocity and displacement data of a robot platform are collected, and attitude deviation characteristics are extracted through filtering fusion processing; the characteristics are input into a joint mechanical model, and real-time damping parameters are generated in combination with hydraulic damping characteristics and joint motion constraints; coupling operation is conducted on the attitude deviation characteristics and the damping parameters through a self-adaptive sliding mode control algorithm, and a compensation control instruction of the hydraulic drive joint is generated; an instruction acts on the deformable supporting mechanism, and the hydraulic damping force and the joint motion trail are continuously adjusted so that the gravity center projection of the platform can be kept within the stable boundary range of the non-continuous supporting face. According to the invention, the anti-interference capability and the attitude stability of the aerial work robot in a complex dynamic environment are improved.

Description

Technical Field

[0001] This application relates to the field of robot control technology, and in particular to a method, system, device and storage medium for maintaining the stability of a robot platform. Background Technology

[0002] In high-altitude working environments, robot platforms often face the dual challenges of discontinuous support surfaces and wind disturbances. They need to dynamically adjust their posture in real time to maintain stability and avoid imbalance or even overturning due to sudden changes in the support surface or strong wind interference. This places extremely high demands on the response speed, environmental adaptability, and anti-interference capabilities of the control system.

[0003] Existing solutions employ data fusion technology based on inertial measurement units and position sensors, combined with proportional-integral-derivative control algorithms, to achieve dynamic balance of the platform's attitude by adjusting the output torque of the joint motors. This solution triggers control commands through a preset stability threshold to correct the platform's tilt or offset.

[0004] However, this scheme relies on a control model with fixed parameters, which makes it difficult to adapt to the dynamic changes in the support surface and wind disturbances in the high-altitude environment, resulting in adjustment lag or overshoot. At the same time, relying solely on motor torque adjustment lacks synergistic optimization of hydraulic damping characteristics, resulting in limited stability maintenance under strong disturbances. Summary of the Invention

[0005] This application provides a method, system, device, and storage medium for maintaining the stability of a robot platform, in order to solve the problem of low attitude stability of high-altitude operation robots in the prior art due to their poor anti-interference ability in complex dynamic environments.

[0006] Firstly, this application provides a method for maintaining the stability of a robot platform, including:

[0007] Acquire real-time motion data from the robot platform;

[0008] The real-time motion data is filtered and fused to generate attitude deviation features;

[0009] The posture deviation features are input into a pre-trained joint mechanics model, which generates real-time damping parameters based on the posture deviation features, hydraulic damping characteristics, and joint motion constraints.

[0010] Based on the real-time damping parameters and combined with the adaptive sliding mode control algorithm, the posture deviation characteristics are calculated to generate compensation control commands for the hydraulically driven joints. These compensation control commands are applied to the deformable support mechanism of the robot platform.

[0011] Optionally, the step of calculating the attitude deviation characteristics based on the real-time damping parameters and in conjunction with an adaptive sliding mode control algorithm to generate compensation control commands for the hydraulically driven joint includes:

[0012] Based on the real-time damping parameters, the decay rate of the boundary layer thickness corresponding to the control command in the adaptive sliding mode control algorithm is dynamically adjusted.

[0013] The tilt angle change rate and the center of gravity offset value in the attitude deviation characteristics are directionally weighted to generate a deviation correction vector that matches the deformation direction of the deformable support mechanism.

[0014] Phase lag compensation is performed on the deviation correction vector based on the adjusted attenuation rate to generate compensation control commands.

[0015] Optionally, the step of performing phase lag compensation on the deviation correction vector based on the adjusted attenuation rate to generate a compensation control command includes:

[0016] Based on the correlation between the adjusted attenuation rate and the disturbance frequency, determine the time delay required for phase lag compensation;

[0017] The component in the deviation correction vector that is in the same direction as the abrupt change direction of the discontinuous support surface is segmented and translated according to the time delay amount to generate the correction component.

[0018] Based on the correction component, the extension and retraction acceleration of the hydraulically driven joint is nonlinearly scaled;

[0019] Compensation control commands are generated based on the nonlinearly scaled scaling acceleration and the correction component.

[0020] Optionally, the nonlinear scaling of the extension and retraction acceleration of the hydraulically driven joint based on the correction component includes:

[0021] Based on the abrupt change amplitude of the discontinuous support surface, determine the scaling range corresponding to the amplitude change rate of the correction component;

[0022] Based on the proportion of the component in the correction component that is in the same direction as the abrupt change direction of the discontinuous support surface, the dynamic adjustment coefficient of the amplitude change rate within the scaling range is calculated.

[0023] The product of the dynamic adjustment coefficient and the adjusted decay rate is used as the scaling factor of the stretching acceleration.

[0024] The stretching acceleration is nonlinearly scaled according to the scaling factor.

[0025] Optionally, the joint mechanics model generates real-time damping parameters based on the posture deviation characteristics, hydraulic damping characteristics, and joint motion constraints, including:

[0026] The inverse correlation between the hydraulic damping characteristics and the joint extension / retraction rate is established through the joint mechanics model building module.

[0027] The target deformation angle of each hydraulically driven joint is calculated using the calculation module of the joint mechanics model based on the tilt angle change rate in the posture deviation characteristics.

[0028] Based on the target deformation angle of each hydraulically driven joint, and combined with the target angle threshold in the joint motion constraint, the allowable deformation range of each hydraulically driven joint is determined.

[0029] Within the allowable deformation range, the center of gravity offset value in the attitude deviation feature is mapped to the pressure adjustment coefficient of the hydraulic circuit according to the reverse correlation to generate real-time damping parameters.

[0030] Optionally, within the allowable deformation range, mapping the center of gravity offset value in the attitude deviation feature to a pressure adjustment coefficient of the hydraulic circuit based on the reverse correlation to generate real-time damping parameters includes:

[0031] The adjustable range of the pressure adjustment coefficient is determined based on the proportional relationship between the abrupt change amplitude of the discontinuous support surface and the allowable deformation range.

[0032] Based on the reverse correlation, the center of gravity acceleration of the center of gravity offset value is converted into the pressure growth gradient of the hydraulic circuit;

[0033] Based on the distribution pattern of the pressure growth gradient within the adjustable range, and combined with the pressure fluctuation of the hydraulic circuit, real-time damping parameters are generated.

[0034] Optionally, the real-time motion data includes acceleration, angular velocity, and visual positioning data;

[0035] The step of filtering and fusing the real-time motion data to generate attitude deviation features includes:

[0036] Based on the abrupt displacement characteristics of the discontinuous support surface, the acceleration and the angular velocity are filtered respectively.

[0037] The displacement offset in the visual positioning data is matched with the filtered acceleration to determine the trajectory. The first deviation component in the vertical direction and the second deviation component in the horizontal direction corresponding to the robot platform are extracted from the matching result.

[0038] Based on the combined vector direction of the first deviation component and the second deviation component, and combined with the filtered angular velocity, attitude deviation features are generated.

[0039] Secondly, this application provides a robot platform stability maintenance system, comprising:

[0040] The acquisition module is used to acquire real-time motion data of the robot platform;

[0041] The processing module is used to filter and fuse the real-time motion data to generate attitude deviation features;

[0042] The input module is used to input the posture deviation features into a pre-trained joint mechanics model, which generates real-time damping parameters based on the posture deviation features, hydraulic damping characteristics, and joint motion constraints.

[0043] The generation module is used to calculate the attitude deviation characteristics based on the real-time damping parameters and in combination with the adaptive sliding mode control algorithm, and generate compensation control commands for the hydraulically driven joints. The compensation control commands are applied to the deformable support mechanism of the robot platform.

[0044] Thirdly, this application provides a computing device including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform a robot platform stability maintenance method as described in any of the first aspects.

[0045] Fourthly, this application provides a computer storage medium storing computer program instructions thereon, which, when executed by a processor, implement a robot platform stability maintenance method as described in any one of the first aspects.

[0046] The technical solution provided in this application has the following beneficial effects:

[0047] This application first comprehensively perceives the instantaneous motion state of the robot platform in a high-altitude dynamic environment through multi-sensor collaborative acquisition, thereby providing accurate input data for subsequent control. Based on this, sensor noise interference is effectively eliminated, and feature quantities reflecting the actual attitude changes of the platform are accurately extracted, thus providing a reliable basis for the platform's stability control. Subsequently, damping adjustment parameters matching the current attitude deviation are output by combining mechanical characteristics and environmental constraints, serving as a physically adaptable adjustment benchmark in subsequent control algorithms. Finally, the optimal control quantity is calculated in real time through intelligent algorithms, precisely guiding the actuators to perform corresponding actions, thereby ensuring that the platform maintains a stable state in complex high-altitude dynamic environments.

[0048] Furthermore, this application dynamically adjusts the boundary layer thickness attenuation rate of the control algorithm by real-time damping parameters, and simultaneously generates a deviation correction vector by directionally weighting the tilt angle change rate and the center of gravity offset value. Based on the adjusted attenuation rate, phase lag compensation is performed on this vector to finally generate a compensation control command. This method achieves dynamic adaptation of control parameters to environmental disturbances, effectively solves the problem of hydraulic system response lag, and ensures the synchronization between control commands and actual needs through precise phase compensation, thereby improving the stability control accuracy of the aerial work robot in complex dynamic environments.

[0049] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 A flowchart illustrating a method for maintaining the stability of a robot platform, as provided in this application embodiment;

[0052] Figure 2 This is a schematic diagram of a robot platform stability maintenance system provided in an embodiment of this application;

[0053] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0054] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0055] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0056] In the field of stability control for aerial work robots, existing technologies mainly adopt a scheme based on fixed-parameter PID control combined with motor torque adjustment. The core defect of this approach is that the control model lacks adaptability to the dynamic characteristics of discontinuous support surfaces and wind disturbances. On the one hand, the preset stability threshold cannot match the sudden changes in amplitude and direction of the support surface in real time, resulting in adjustment lag or overshoot. On the other hand, relying solely on motor torque adjustment while ignoring the synergistic optimization of the damping characteristics of the hydraulic system makes it difficult to achieve accurate phase compensation under strong wind disturbances, causing platform oscillation or response delay. These problems essentially stem from the structural contradiction between the static control architecture and the requirements of the dynamic environment.

[0057] To address the aforementioned limitations, this application proposes a method for maintaining the stability of a robot platform. Its innovation lies in constructing posture deviation characteristics using real-time motion data, quantifying hydraulic damping characteristics into real-time parameters for environmental adaptation using a joint mechanics model, and finally generating control commands with phase compensation capabilities via an adaptive sliding mode algorithm. This method achieves breakthrough dynamic adaptation at three levels: online matching of support surface abrupt changes with filtering parameters, real-time mapping of wind disturbance intensity with damping parameters, and collaborative optimization of hydraulic delay effects and control boundaries. This fundamentally solves the stability defects caused by insufficient identification of environmental dynamics and actuator response mismatch in existing technologies.

[0058] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] Figure 1 A flowchart of a method for maintaining the stability of a robot platform provided in this application embodiment is shown below. Figure 1 As shown, the method includes:

[0060] Step 101: Obtain real-time motion data from the robot platform.

[0061] In step 101, the robot platform is the robot platform of the overhead contact line self-propelled maintenance equipment vehicle group;

[0062] The real-time motion data mainly includes the following: the acceleration and angular velocity collected in real time by the robot platform of the self-propelled contact network maintenance equipment vehicle through a six-axis inertial sensor, and the visual positioning data relative to the contact network support structure obtained by the platform through a visual sensor.

[0063] Furthermore, the aforementioned real-time motion data was obtained in a high-altitude dynamic environment where the robot platform was on a discontinuous support surface and subject to wind interference. The discontinuous support surface refers to the characteristic of the support surface that the robot platform contacts during high-altitude operations, which has discontinuous, abrupt, or intermittent contact features, such as gaps in high-altitude scaffolding or joints in temporary platforms, resulting in a lack of continuity in the support surface.

[0064] Wind disturbance refers to the dynamic interference force generated by airflow movement in the high-altitude environment on the robot platform, which can cause the platform to produce unexpected displacement or tilting; "high-altitude dynamic environment" refers to the working environment at a height of 5 meters or more above the ground, at which height the impact of wind disturbance and support surface discontinuity on platform stability is enhanced.

[0065] Step 102: Filter and fuse the real-time motion data to generate attitude deviation features.

[0066] In step 102, the attitude deviation feature represents the feature quantity obtained after processing and used to reflect the difference between the actual attitude and the desired attitude of the platform.

[0067] In this embodiment, acceleration and angular velocity are dynamically balanced and filtered to eliminate high-frequency noise interference caused by wind disturbance. Simultaneously, the platform displacement information in the visual positioning data is matched with the filtered acceleration data to extract the first deviation component in the vertical direction and the second deviation component in the horizontal direction. Finally, combined with the filtered angular velocity data, attitude deviation features containing the platform tilt rate of change and center of gravity offset are generated through vector synthesis.

[0068] For example, when the robot platform of the self-propelled overhead contact line maintenance equipment is performing overhead contact line maintenance work at a height of 5 meters, the six-axis inertial sensor detects changes in the longitudinal acceleration and angular velocity around the horizontal axis of the platform. At the same time, the vision sensor identifies that the platform has generated lateral displacement relative to the overhead contact line support structure. These data are collected in real time and transmitted to the processing unit. The longitudinal acceleration change is the motion characteristic of the platform in the track direction, the angular velocity change around the horizontal axis reflects the possible tilt of the platform, and the lateral displacement indicates the change in the relative position of the platform and the support structure.

[0069] In the aforementioned high-altitude operation scenario, the processing unit first matches and analyzes the filtered longitudinal acceleration data with the lateral displacement data from visual positioning to extract the vertical gravity projection deviation. This deviation indicates that the platform's center of gravity has shifted from the predetermined position. Simultaneously, it can also extract the horizontal inertial offset data, which further indicates that the platform is at risk of lateral slippage. Finally, it combines the angular velocity data for comprehensive analysis to obtain the attitude deviation characteristics. These characteristics clearly show that the platform is tilting outward from the track at a certain rate, and its center of gravity offset is showing a continuous increasing trend.

[0070] Step 103: Input the posture deviation features into the pre-trained joint mechanics model. The joint mechanics model generates real-time damping parameters based on the posture deviation features, hydraulic damping characteristics, and joint motion constraints.

[0071] In step 103, the hydraulic damping characteristic represents the relationship between the resistance and speed generated by the hydraulic system during motion, and the joint motion constraint represents the range of motion and mechanical characteristics of each joint under the constraints of the mechanical structure.

[0072] The real-time damping parameter is a hydraulic adjustment that is dynamically adapted to the current environment and calculated through the joint mechanics model. Its value is determined by the tilt angle change rate in the attitude deviation characteristics, the damping characteristic curve of the hydraulic system, and the joint motion constraint range. It is used to guide the hydraulic system to generate the precise damping force required to suppress the attitude deviation of the platform.

[0073] In this embodiment, the attitude deviation characteristics are input into a pre-established joint mechanics model, which calculates the target deformation angle of each joint based on the current tilt angle change rate of the platform. At the same time, the optimal real-time damping parameters are determined by combining the damping characteristic curves of the hydraulic system at different speeds and the mechanical motion constraint range of each joint through mechanical balance calculation. These parameters reflect the amount of hydraulic system adjustment required to maintain platform stability under the current environmental conditions.

[0074] For example, in the case of the platform tilting outwards from the track, the mechanical model is first used to calculate the amount of extension and retraction of the four corner support joints of the platform that needs to be adjusted to restore the balance. Then, based on the characteristic curve of the hydraulic system, it is determined that the damping force of the outer support joints needs to be increased to effectively suppress the tilting trend. At the same time, considering the constraints such as the maximum allowable extension and retraction of each support joint, real-time damping parameters are generated. These damping parameters can clearly indicate that the two outer support joints need to increase the damping force, while the two inner support joints maintain the current damping force.

[0075] Step 104: Based on the real-time damping parameters and combined with the adaptive sliding mode control algorithm, the attitude deviation characteristics are calculated to generate compensation control commands for the hydraulically driven joints. The compensation control commands are applied to the deformable support mechanism of the robot platform.

[0076] In step 104, the hydraulically driven joint is the power execution component of the robot platform. It drives the deformation and movement of the deformable support mechanism through the hydraulic circuit, thereby adjusting the posture and position of the entire robot platform. The compensation control command represents the control signal used to correct the posture deviation of the platform. The deformable support mechanism represents the mechanical structure that can adjust the posture of the platform through deformation.

[0077] In this embodiment, an adaptive sliding mode control algorithm is first used to adjust the thickness change rate of the control boundary layer based on real-time damping parameters to match the intensity of the current environmental disturbance. Then, the attitude deviation characteristics are weighted by direction sensitivity to generate a deviation correction vector corresponding to the platform deformation direction. Finally, the effect of hydraulic system response delay is eliminated by a phase compensation algorithm to generate precise hydraulic drive joint control commands. These commands are applied to the deformable support mechanism of the platform, and by adjusting the hydraulic damping force and extension position of each support point, the center of gravity of the platform returns to the stable range.

[0078] For example, in a 5-meter high-altitude operation scenario, the control algorithm adjusts the control strategy based on real-time damping parameters, generating instructions to increase the pressure of the outer support hydraulic cylinder in response to the platform's external tilt. Simultaneously, considering the response delay characteristics of the hydraulic system, control signals can be issued at preset intervals, such as 0.3 seconds. After the hydraulic system executes these instructions, the support force on the outer side of the platform gradually increases, effectively suppressing the tilting trend. After a certain period of adjustment, the platform's center of gravity returns to the center of the support surface, thus restoring stability.

[0079] The robot platform stability maintenance method provided in this application accurately identifies platform attitude changes through multi-sensor data fusion, then calculates optimal adjustment parameters based on a mechanical model, and generates precise control commands using an intelligent control algorithm. This achieves rapid and stable control of the catenary self-propelled maintenance equipment unit in complex high-altitude environments. This method effectively overcomes the stability challenges posed by discontinuous support surfaces and wind disturbances, ensuring the safety and reliability of high-altitude operations.

[0080] To further improve the control accuracy of the robot platform in complex dynamic environments, in some embodiments, step 104: based on the real-time damping parameters and combined with an adaptive sliding mode control algorithm, the attitude deviation characteristics are calculated to generate compensation control commands for the hydraulically driven joints, including:

[0081] Step 201: Based on the real-time damping parameters, dynamically adjust the decay rate of the boundary layer thickness corresponding to the control command in the adaptive sliding mode control algorithm.

[0082] In step 201, the decay rate of the boundary layer thickness refers to the parameter adjustment rate used in the adaptive sliding mode control algorithm to smooth changes in control commands. Its magnitude directly affects the balance between the system's anti-interference capability and response speed.

[0083] In this embodiment, the intensity of the current environmental disturbance is determined based on the magnitude of the real-time damping parameter. When the damping parameter is large, it corresponds to a strong wind disturbance or a sudden change in the support surface. At this time, the attenuation rate of the boundary layer thickness is accelerated to improve the system response speed; conversely, the attenuation rate is slowed down to ensure control stability. For example, this adjustment process can be achieved through a preset damping parameter-attenuation rate mapping relationship.

[0084] Step 202: The tilt angle change rate and the center of gravity offset value in the attitude deviation characteristics are directionally weighted to generate a deviation correction vector that matches the deformation direction of the deformable support mechanism.

[0085] In step 202, the tilt angle change rate refers to the amount of change in the tilt angle of the robot platform per unit time, which can be obtained by extracting the posture deviation features. Specifically, it can be calculated by the angular velocity after filtering and fusion and coordinate transformation, and is used to reflect the degree of platform posture imbalance. The center of gravity offset value represents the deviation distance of the platform's center of gravity from the ideal support position, which is obtained by vector synthesis of the first deviation component and the second deviation component, and is used to reflect the severity of platform imbalance.

[0086] The deviation correction vector is a spatial vector composed of the direction of the tilt rate of change and the center of gravity offset value. Its direction indicates the orientation of the deformable support mechanism that needs to be deformed and adjusted, and its amplitude represents the required adjustment force. This deviation correction vector is used to generate precise compensation control commands.

[0087] In this embodiment, the direction of the tilt rate of change is first analyzed to determine the main tilt orientation of the platform. Then, the required adjustment intensity in each orientation is calculated by combining the center of gravity offset value. Finally, a deviation correction vector that perfectly matches the deformation requirements of the support mechanism is generated by vector synthesis. This vector not only contains the spatial direction information that needs to be adjusted, but also the specific adjustment amount in each direction.

[0088] Step 203: Perform phase lag compensation on the deviation correction vector based on the adjusted attenuation rate to generate a compensation control command.

[0089] In step 203, the adjusted decay rate reflects the optimal control rhythm under the current environment.

[0090] In this embodiment, the expected delay time of the hydraulic system is calculated based on the adjusted decay rate. Then, the control quantity in the deviation correction vector is shifted forward by this time amount. At the same time, the amplitude of the control quantity is dynamically scaled according to the magnitude of the decay rate. The resulting compensation control command can accurately match the actual response characteristics of the hydraulic system.

[0091] Here is a specific example:

[0092] When the robot platform of the overhead contact line self-propelled maintenance equipment is performing maintenance work at a height of 5 meters, when the system detects that the platform is tilting outward from the track at a rate of 0.5 degrees per second and the center of gravity shift reaches 15 centimeters, it first determines that the outer support joint needs to increase the damping force by 30% based on the real-time damping parameters, and then uses an adaptive sliding mode control algorithm to adjust the attenuation rate of the boundary layer thickness to decrease by 2 millimeters per second. The attenuation rate of the boundary layer thickness can be obtained by querying the damping parameters and the preset mapping relationship table.

[0093] Next, the tilt change rate of 0.5 degrees per second and the center of gravity offset of 15 centimeters per second are directionally weighted in a ratio of 1:2. During the weighting process, based on the platform structural characteristics, the proportional coefficient is determined as the deformation sensitivity parameter of the support mechanism, and a deviation correction vector with an amplitude of 35 centimeters and pointing towards the inside of the track is generated.

[0094] Finally, based on the adjusted attenuation rate, it was calculated that the hydraulic system has a response delay of 0.3 seconds. This delay time is determined by the ratio of the hydraulic pipeline length to the oil flow rate. Based on this, the deviation correction vector is sent 0.3 seconds in advance, and the control quantity amplitude is amplified by 1.2 times to compensate for the attenuation effect, generating a compensation control command.

[0095] In the embodiments of this application, the method achieves rapid and precise control of the aerial work robot in complex environments by dynamically adjusting control parameters, accurately calculating correction requirements, and intelligently compensating for system delays, thereby improving the platform's stability under discontinuous support surfaces and wind disturbance conditions.

[0096] To further improve the control accuracy of the robot platform in complex dynamic environments, in some embodiments, step 203: performing phase lag compensation on the deviation correction vector based on the adjusted attenuation rate to generate compensation control commands includes:

[0097] Step 301: Determine the time delay required for phase lag compensation based on the correlation between the adjusted attenuation rate and the disturbance frequency.

[0098] In step 301, the disturbance frequency can refer to the disturbance frequency caused by wind, which can be obtained by analyzing the periodic fluctuation components in the inertial sensor data; the time delay refers to the length of time that the control command needs to be issued in advance to compensate for the lag in the response of the hydraulic system. Its magnitude is determined by the decay rate and the disturbance frequency. The faster the decay rate or the higher the disturbance frequency, the smaller the required time delay.

[0099] In this embodiment, the system response requirements reflected by the adjusted attenuation rate are first analyzed. Then, based on the system response requirements and combined with the real-time monitored wind disturbance frequency components, the optimal compensation time is determined by a preset delay calculation model to ensure that the control command and the hydraulic execution timing are accurately matched.

[0100] Step 302: The component in the deviation correction vector that is in the same direction as the abrupt change direction of the discontinuous support surface is segmented and translated according to the time delay amount to generate the correction component.

[0101] In step 302, the direction of mutation can be obtained by analyzing the trend of displacement offset. The trend of displacement offset is the displacement of the robot platform relative to the discontinuous support surface. The correction component is a control vector with time-shift adjustment and time-lead characteristics.

[0102] In this embodiment, the component in the deviation correction vector that is consistent with the direction of abrupt change in the support surface is first identified. Then, the component is shifted forward according to the calculated time delay. While keeping other directional components unchanged, a correction component that can both compensate for hydraulic delay and maintain spatial coordination is generated.

[0103] Step 303: Based on the correction component, nonlinearly scale the extension and retraction acceleration of the hydraulically driven joint.

[0104] In step 303, the telescopic acceleration can be determined by the rate of change of the amplitude of the control quantity in the correction component. Specifically, it can be obtained by calculating the change of this component per unit time. This telescopic acceleration is used to reflect the speed change required by the hydraulically driven joint to match the platform's attitude adjustment requirements.

[0105] In this embodiment, the speed adjustment requirement of the hydraulically driven joint is first dynamically scaled according to the amplitude change trend of the control quantity in each direction in the correction component. Then, the adjustment gradient is increased in the direction of rapid change and decreased in the direction of gradual change, thereby achieving precise and efficient speed control.

[0106] Step 304: Generate compensation control commands based on the nonlinearly scaled stretching acceleration and the correction component.

[0107] In this embodiment, the spatial information of the nonlinearly scaled velocity adjustment gradients in each direction and the correction components is fused to generate multidimensional control commands that have both time accuracy and amplitude rationality, which can ensure that each hydraulically driven joint works together to complete the stable adjustment of the platform.

[0108] Here is a specific example:

[0109] During the maintenance operation of the self-propelled contact network maintenance equipment robot platform at a height of 5 meters, when the system detects that the platform is tilting outward from the track at a rate of 0.5 degrees per second and the center of gravity shifts by 15 centimeters, the time delay is first determined by the formula T=K / (v×f) based on the adjusted boundary layer thickness attenuation rate of 2 millimeters per second and the detected disturbance frequency of 0.4 Hz. Here, T is the time delay, K is the system's inherent constant, v is the attenuation rate, and f is the wind frequency. For example, when K is 0.24, v is 2 millimeters per second, and f is 0.4 Hz, the calculated time delay is 0.3 seconds.

[0110] Subsequently, the component in the deviation correction vector that is in the same direction as the sudden change of the outer support surface of the track is processed 0.3 seconds in advance, while keeping the other directional components unchanged, to generate a correction component that includes time compensation; then, based on the change trend of the amplitude of the outer support control quantity in this component from 0 to 35 cm degrees in 0.5 seconds, the expansion acceleration is calculated to be 70 cm degrees / second, and the gradient value is nonlinearly amplified by 1.3 times according to the dynamic response requirements of the platform.

[0111] The final compensation control command is used to control the outer support hydraulic cylinder to act 0.3 seconds in advance at an adjustment speed of 91 cm / s, so that the tilted platform is corrected to a safe range within 3.5 seconds and the center of gravity offset is controlled within 2 cm. The 3.5-second stabilization time is a value calculated using parameters such as the platform mass of 10 tons and the maximum output of the hydraulic system of 50 kN.

[0112] In the embodiments of this application, the method achieves effective compensation for the response delay of the hydraulic system by accurately calculating the compensation time, intelligently adjusting the timing of the control quantity, and dynamically optimizing the adjustment range, thereby ensuring the rapid, accurate, and stable control of the aerial work robot under wind disturbance and discontinuous support surface conditions.

[0113] To further improve the control accuracy of the robot platform in environments with discontinuous support surfaces, in some embodiments, step 303: nonlinearly scaling the extension and retraction acceleration of the hydraulically driven joint based on the correction component, includes:

[0114] Step 401: Determine the scaling range corresponding to the amplitude change rate of the correction component based on the abrupt change amplitude of the discontinuous support surface.

[0115] In step 401, the amplitude change rate is calculated from the instantaneous amplitude derivative of the correction component, reflecting the rate of change of the robot platform's center of gravity offset; the scaling range refers to the reasonable range of control quantity adjustment determined according to the degree of abrupt change in the support surface, and its upper and lower limit thresholds are positively correlated with the abrupt change in the support surface amplitude. The larger the abrupt change amplitude, the wider the allowable adjustment range.

[0116] In this embodiment, based on the specific dimensional characteristics of the current support surface mutation and in combination with the platform structural stability requirements, the minimum and maximum allowable adjustment thresholds for the amplitude change rate of the correction component are determined to provide a reference range for subsequent scaling.

[0117] For example, when the self-propelled maintenance equipment robot platform of the overhead contact line is operating at a height of 5 meters, if a sudden 8-centimeter subsidence of the right-side support surface of the track is detected, the minimum stability adjustment threshold corresponding to this dimensional change is first determined to be 0.6 and the maximum safety adjustment threshold is 1.8 based on the platform's structural stability analysis model. The minimum threshold of 0.6 is calculated using the formula: Minimum threshold = 0.5 + 8 × 0.0125, where 0.5 is the base value, 8 is the change size (unit: cm), and 0.0125 is a preset coefficient. The maximum threshold of 1.8 is determined using the formula: Maximum threshold = 1.5 + Change size 8 cm × Coefficient 0.0375, where 1.5 is the base value, 8 is the change size (unit: cm), and 0.0375 is a preset coefficient. Both base values ​​and coefficients can be obtained from a preset database. Simultaneously, considering the current wind force level of 3, a reduction factor of 0.9 is applied to the maximum threshold, ultimately determining the allowable adjustment threshold range for the rate of change of the correction component amplitude to be 0.6 to 1.62.

[0118] Step 402: Based on the proportion of the component in the correction component that is in the same direction as the abrupt change direction of the discontinuous support surface, calculate the dynamic adjustment coefficient of the amplitude change rate within the scaling range.

[0119] In step 402, the mutation direction is used to describe the spatial trend of the mutation, while the mutation amplitude is used to describe the degree of mutation. Therefore, the two are used together to quantify the disturbance characteristics of the dynamic environment at high altitudes. The component proportion refers to the proportion of the control quantity in the correction component that is consistent with the mutation direction of the support surface, which can reflect the demand intensity of the current main adjustment direction. The dynamic adjustment coefficient is the adjustment intensity parameter calculated based on this proportion within the threshold range.

[0120] In this embodiment of the application, the proportion of the modulus of the vector in the same direction as the abrupt change of the support surface in the correction component is calculated to the total modulus, and then this proportion is mapped to a preset threshold range. During the mapping process, a dynamic adjustment coefficient reflecting the current environmental requirements can be obtained by interpolation calculation.

[0121] Step 403: The product of the dynamic adjustment coefficient and the adjusted decay rate is used as the scaling factor of the stretching acceleration.

[0122] In step 403, the scaling factor is the final comprehensive parameter used to adjust the scaling acceleration. It is determined by the dynamic adjustment coefficient and the decay rate, thus taking into account both environmental requirements and system response characteristics.

[0123] In this embodiment of the application, multiplying the calculated dynamic adjustment coefficient by the adjusted attenuation rate yields a scaling factor that considers both the abrupt change requirements of the support surface and the system's response capability.

[0124] Step 404: Perform nonlinear scaling on the stretching acceleration according to the scaling factor.

[0125] In this embodiment, a scaling factor is used to adjust the original stretching acceleration in segments. For example, the scaling ratio used on the gradient component corresponding to the abrupt change direction is greater than the scaling ratio on other directions, thereby achieving precise directional control.

[0126] In this embodiment, the method achieves rapid, precise and stable control of the aerial work robot under discontinuous support surface conditions by dynamically determining the adjustment range, accurately calculating the adjustment intensity and directionally optimizing the control gradient, thereby effectively overcoming the risk of imbalance caused by abrupt changes in the support surface.

[0127] To further improve the accuracy of multi-degree-of-freedom joint cooperative control of the work robot platform, in some embodiments, step 103: the joint mechanics model generates real-time damping parameters based on the posture deviation characteristics, hydraulic damping characteristics, and joint motion constraints, including:

[0128] Step 501: Establish the inverse correlation between the hydraulic damping characteristics and the joint extension and contraction rate through the joint mechanics model building module.

[0129] In step 501, the module is established by fitting the parameterized mathematical model obtained through experimental data; the inverse correlation refers to the negative correlation between the damping force of the hydraulic system and the joint movement speed, specifically, the joint extension and retraction speed decreases accordingly when the damping force increases. This relationship can be determined by the fluid dynamic characteristics of the hydraulic system.

[0130] In this embodiment of the application, a quantitative relationship curve between damping force and joint speed is established by analyzing the pressure loss characteristics of hydraulic oil at different flow rates. This curve reflects the inherent energy dissipation characteristics of the hydraulic system and provides a theoretical basis for subsequent parameter mapping.

[0131] Step 502: Calculate the target deformation angle of each hydraulically driven joint using the calculation module of the joint mechanics model, based on the tilt angle change rate in the posture deviation characteristics.

[0132] In step 502, the calculation module is a solver based on the robot kinematics model; the target deformation angle refers to the angle that each hydraulically driven joint needs to reach to compensate for the platform's posture deviation, and this angle can be calculated by the relationship between the tilt rate of change and the joint kinematics.

[0133] In this embodiment, based on the overall tilt angle change trend of the platform in the attitude deviation characteristics, and combined with the proportional relationship between each joint and the platform's center of gravity, the angle that each hydraulically driven joint needs to be adjusted is calculated. These angles constitute the control target for joint coordinated motion. For example, when the robot platform encounters lateral wind disturbance during operation at a height of 5 meters, the tilt angle change rate is detected as 5° / s based on the attitude deviation characteristics. Based on the motion constraints of the robot's hip joint in the joint mechanics model, the target deformation angle of the robot's hip joint is calculated to be 15°, and that of the knee joint is 10°. This target deformation angle is used to subsequently generate matching real-time damping parameters to stabilize the platform's attitude.

[0134] Step 503: Based on the target deformation angle of each hydraulically driven joint and in combination with the target angle threshold in the joint motion constraint, determine the allowable deformation range of each hydraulically driven joint.

[0135] In step 503, the target angle threshold is an important parameter to ensure mechanical safety; the allowable deformation range refers to the angle range in which each hydraulically driven joint can work safely under the constraints of the mechanical structure, and the upper and lower limits of this range can be determined by the physical structural parameters of the joint.

[0136] In this embodiment of the application, the calculated target deformation angle is compared with the maximum motion angle allowed by the design of each joint, and then the angle that exceeds the safe range is limited to ensure that all joints work within the range allowed by the mechanical structure.

[0137] Step 504: Within the allowable deformation range, the center of gravity offset value in the attitude deviation feature is mapped to the pressure adjustment coefficient of the hydraulic circuit according to the reverse correlation to generate real-time damping parameters. The hydraulic circuit is generated by the joint mechanics model based on the hydraulic damping characteristics.

[0138] In step 504, the pressure adjustment coefficient is a dimensionless parameter that reflects the magnitude of the damping force required by the hydraulic system, and its value is positively correlated with the degree of center of gravity offset and joint movement requirements.

[0139] In this embodiment of the application, within the determined allowable deformation range, the center of gravity offset value is converted into the corresponding hydraulic system pressure demand according to the reverse correlation relationship. At the same time, the real-time motion state of each joint is considered to generate damping parameters that can effectively stabilize the platform without causing hydraulic system overload.

[0140] In the embodiments of this application, the method achieves rapid, accurate and stable operation of the aerial work robot platform under multi-degree-of-freedom collaborative control by establishing an accurate hydraulic-mechanical relationship model, reasonably calculating joint motion requirements, strictly ensuring the mechanical safety range, and intelligently generating damping parameters, effectively overcoming the posture control problem in complex environments.

[0141] To further improve the accuracy of damping parameter generation for the work robot platform, in some embodiments, step 504: within the allowable deformation range, mapping the center of gravity offset value in the attitude deviation feature to the pressure adjustment coefficient of the hydraulic circuit according to the reverse correlation to generate real-time damping parameters includes:

[0142] Step 601: Determine the adjustable range of the pressure adjustment coefficient based on the proportional relationship between the abrupt change amplitude of the discontinuous support surface and the allowable deformation range.

[0143] In step 601, the proportional relationship reflects the dynamic balance between environmental disturbance and mechanical capability. For example, when the single-sided support surface of the robot platform suddenly sinks by 200mm, and the maximum deformation range allowed by the joint mechanics model is ±150mm, and the proportional relationship is 200:150=4:3, the upper limit of the adjustable range of the pressure adjustment coefficient needs to be compressed to 75% of the original design to avoid joint over-deformation.

[0144] Furthermore, the aforementioned adjustable range refers to the effective range within which the pressure adjustment coefficient is allowed to change, and its upper and lower thresholds can be determined by the ratio of the degree of abrupt change in the support surface to the allowable deformation capacity of the joint. For example, the greater the abrupt change or the smaller the deformation capacity, the narrower the corresponding range.

[0145] In this embodiment, the ratio of the abrupt change size of the support surface to the maximum allowable deformation angle of the joint is first calculated, and then the ratio is mapped to a preset pressure adjustment range reference value to obtain an operable range that meets stability requirements without exceeding mechanical limitations.

[0146] Step 602: Based on the reverse correlation, convert the center of gravity acceleration of the center of gravity offset value into the pressure growth gradient of the hydraulic circuit.

[0147] In step 602, the center of gravity acceleration is a trend indicator obtained by differentiating the center of gravity offset value over time; the pressure growth gradient represents the rate of change of hydraulic circuit pressure over time, and its value is positively correlated with the rate of increase of the center of gravity offset.

[0148] In this embodiment, the instantaneous rate of change of the center of gravity shift is converted into the equivalent rate of change of pressure demand in the hydraulic circuit based on the reverse correlation, which can ensure that the damping force adjustment process matches the platform imbalance development speed.

[0149] Step 603: Based on the distribution pattern of the pressure growth gradient within the adjustable range, and combined with the pressure fluctuation of the hydraulic circuit, generate real-time damping parameters.

[0150] In step 603, the distribution law describes the reasonable distribution of the pressure gradient within the adjustable range; the current wind disturbance intensity refers to the magnitude of the lateral thrust generated by the instantaneous wind speed on the robot platform in the high-altitude environment, and this disturbance intensity can be measured in real time by the fluctuation amplitude of the lateral acceleration in the six-axis inertial sensor; the pressure fluctuation refers to the periodic fluctuation component of the hydraulic circuit pressure caused by wind disturbance, and its amplitude is positively correlated with the wind disturbance intensity, which is usually obtained in real time by looking up the mapping relationship table between wind level and pressure fluctuation in historical data.

[0151] In this embodiment, the distribution of the pressure growth gradient within the adjustable range is first analyzed, then the pressure fluctuation component corresponding to the current wind force level is superimposed, and finally a comprehensive damping parameter that can effectively suppress abrupt changes in the support surface and offset wind interference is generated by a weighted fusion method.

[0152] For example, firstly, the pressure growth gradient is determined to be 0.5 MPa per second based on the growth rate of the center of gravity offset value of 10 mm per second; then, the distribution position of this pressure growth gradient within the adjustable range [0.3, 0.8] MPa / s is calculated to be 40%, specifically calculated as: (0.5-0.3) / (0.8-0.3)=40%; at the same time, based on the additional fluctuation amount corresponding to the current wind force level 2 being ±0.2 MPa, a sinusoidal fluctuation synchronized with the wind disturbance is superimposed at the 40% distribution position; finally, a dynamic damping parameter with an amplitude of 0.5±0.12 MPa is generated.

[0153] In this embodiment, the method achieves real-time adaptation of damping parameters to complex high-altitude working conditions by dynamically determining the adjustment range, accurately converting the changing trend, and intelligently integrating environmental disturbances, thereby ensuring the reliable stability of the robot platform under discontinuous support surfaces and wind disturbances.

[0154] In cases where the real-time operational data includes acceleration, angular velocity, and visual positioning data, to further improve the accuracy of posture perception on the robot platform, in some embodiments, step 102: the filtering and fusion processing of the real-time motion data to generate posture deviation features includes:

[0155] Step 701: Based on the displacement abrupt change characteristics of the discontinuous support surface, the acceleration and the angular velocity are filtered respectively.

[0156] In step 701, the displacement mutation characteristic refers to the sudden change in platform position caused by discontinuous support surfaces, which is manifested as step fluctuations in acceleration and angular velocity data; acceleration represents the linear acceleration data of the platform in three-dimensional space measured by inertial sensors, reflecting the rate of change of the platform's motion state; angular velocity represents the angular velocity data of the platform rotating around the three-dimensional coordinate axes measured by inertial sensors, reflecting the speed of changes in the platform's attitude.

[0157] In this embodiment of the application, an adaptive filtering algorithm can be used to process the original acceleration and angular velocity signals to effectively separate the effective signals generated by the actual platform motion from the high-frequency noise caused by environmental interference, thus providing clean input data for subsequent analysis.

[0158] Step 702: Perform trajectory matching between the displacement offset in the visual positioning data and the filtered acceleration, and extract the first deviation component corresponding to the robot platform in the vertical direction and the second deviation component corresponding to the robot platform in the horizontal direction from the matching result.

[0159] In step 702, the visual positioning data represents the displacement and orientation information of the platform relative to the surrounding environment obtained by the visual sensor, which is used to assist in locating the spatial position of the platform; the first deviation component reflects the degree of imbalance in the vertical direction of the platform, and the second deviation component characterizes the motion deviation in the horizontal direction.

[0160] In this embodiment, the overall displacement change trend of the platform detected by the vision system is first analyzed in a spatiotemporal correlation with the filtered acceleration data. Then, the positional deviation of the platform in the direction of gravity and the inertial motion components on the horizontal plane are extracted by coordinate transformation and vector decomposition methods, thereby comprehensively describing the imbalance state of the contact network self-wheeled maintenance equipment robot platform.

[0161] Step 703: Generate attitude deviation features based on the combined vector direction of the first deviation component and the second deviation component, combined with the filtered angular velocity.

[0162] In step 703, the composite vector direction is the overall imbalance trend that combines the deviations in the vertical and horizontal directions.

[0163] In this embodiment, the first deviation component and the second deviation component are vector-synthesized to obtain the main direction of the overall offset trend of the platform. Then, combined with the rotational motion characteristics reflected by the filtered angular velocity, a complete attitude deviation description containing both displacement and rotation information is generated, providing a comprehensive basis for subsequent control.

[0164] The specific implementation process is as follows: First, the vector synthesis result of the first deviation component and the second deviation component is calculated to obtain the overall offset trend direction of the platform. This synthesized vector direction is obtained by vector addition. Specifically, the first deviation component is multiplied by the vertical unit vector and the second deviation component is multiplied by the horizontal unit vector to obtain the synthesized vector direction. Then, the filtered angular velocity is converted into the rotation axis direction vector according to the right-hand rule. Finally, the synthesized vector direction and the rotation axis direction vector are cross-producted to obtain the attitude deviation feature. The magnitude of this feature represents the degree of attitude deviation, and the direction represents the deviation correction direction.

[0165] A specific implementation is as follows: When the high-altitude operation robot is located at a height of 5 meters and the wind force level is 4, the first deviation component measured by the sensor is 0.3 m / s. 2 The direction is vertically downward, and the second deviation component is 0.2 m / s. 2 The direction is horizontal eastward, where m / s 2 Using the unit of gravitational acceleration, the calculated direction of the composite vector is arctan(0.2 / 0.3) = 33.7°, where 33.7° represents east-southeast. The measured angular velocity is 1.5 rad / s, and the vector rotates around the north axis. After cross product operation, the magnitude of the attitude deviation feature is 0.45 grad / s, and the direction is east-southeast at 57.3°. The 0.45 is calculated from 0.3 × 1.5, and the 57.3° is obtained by subtracting 33.7° from 90°.

[0166] In this embodiment, the method achieves accurate perception of the posture of the high-altitude operation robot platform through intelligent filtering to eliminate environmental interference, precise matching of multi-source data, and comprehensive fusion of motion features, thereby providing reliable state input for stability control.

[0167] Figure 2 This is a schematic diagram of a robot platform stability maintenance system provided in an embodiment of this application, as shown below. Figure 2 As shown, the system includes:

[0168] The acquisition module 21 is used to acquire real-time motion data of the robot platform.

[0169] The processing module 22 is used to perform filtering and fusion processing on the real-time motion data to generate attitude deviation features.

[0170] Input module 23 is used to input the posture deviation features into a pre-trained joint mechanics model, which generates real-time damping parameters based on the posture deviation features, hydraulic damping characteristics, and joint motion constraints.

[0171] The generation module 24 is used to calculate the attitude deviation characteristics based on the real-time damping parameters and combined with the adaptive sliding mode control algorithm, and generate compensation control commands for the hydraulically driven joints. The compensation control commands are applied to the deformable support mechanism of the robot platform.

[0172] Figure 2 The aforementioned robot platform stability maintenance system can perform... Figure 1 The implementation principle and technical effects of the robot platform stability maintenance method described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the robot platform stability maintenance system in the above embodiments perform operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0173] In one possible design, Figure 2 The robot platform stability maintenance system of the illustrated embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0174] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0175] The processing component 32 is used for the above Figure 1 The embodiment describes a method for maintaining the stability of a robot platform.

[0176] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-described method.

[0177] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0178] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0179] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0180] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0181] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0182] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates a method for maintaining the stability of a robot platform.

[0183] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0184] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0185] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for maintaining the stability of a robot platform, characterized in that, include: Acquire real-time motion data from the robot platform; The real-time motion data is filtered and fused to generate attitude deviation features; The posture deviation features are input into a pre-trained joint mechanics model, which generates real-time damping parameters based on the posture deviation features, hydraulic damping characteristics, and joint motion constraints. Based on the real-time damping parameters and combined with the adaptive sliding mode control algorithm, the posture deviation characteristics are calculated to generate compensation control commands for the hydraulically driven joints. These compensation control commands are applied to the deformable support mechanism of the robot platform.

2. The method for maintaining the stability of a robot platform according to claim 1, characterized in that, Based on the real-time damping parameters and combined with the adaptive sliding mode control algorithm, the attitude deviation characteristics are calculated to generate compensation control commands for the hydraulically driven joint, including: Based on the real-time damping parameters, the decay rate of the boundary layer thickness corresponding to the control command in the adaptive sliding mode control algorithm is dynamically adjusted. The tilt angle change rate and the center of gravity offset value in the attitude deviation characteristics are directionally weighted to generate a deviation correction vector that matches the deformation direction of the deformable support mechanism. Phase lag compensation is performed on the deviation correction vector based on the adjusted attenuation rate to generate compensation control commands.

3. The method for maintaining the stability of a robot platform according to claim 2, characterized in that, The step of performing phase lag compensation on the deviation correction vector based on the adjusted attenuation rate to generate a compensation control command includes: Based on the correlation between the adjusted attenuation rate and the disturbance frequency, determine the time delay required for phase lag compensation; The component in the deviation correction vector that is in the same direction as the abrupt change direction of the discontinuous support surface is segmented and translated according to the time delay amount to generate the correction component. Based on the correction component, the extension and retraction acceleration of the hydraulically driven joint is nonlinearly scaled; Compensation control commands are generated based on the nonlinearly scaled scaling acceleration and the correction component.

4. The method for maintaining the stability of a robot platform according to claim 3, characterized in that, The nonlinear scaling of the extension and retraction acceleration of the hydraulically driven joint based on the correction component includes: Based on the abrupt change amplitude of the discontinuous support surface, determine the scaling range corresponding to the amplitude change rate of the correction component; Based on the proportion of the component in the correction component that is in the same direction as the abrupt change direction of the discontinuous support surface, the dynamic adjustment coefficient of the amplitude change rate within the scaling range is calculated. The product of the dynamic adjustment coefficient and the adjusted decay rate is used as the scaling factor of the stretching acceleration. The stretching acceleration is nonlinearly scaled according to the scaling factor.

5. The method for maintaining the stability of a robot platform according to claim 4, characterized in that, The joint mechanics model generates real-time damping parameters based on the posture deviation characteristics, hydraulic damping characteristics, and joint motion constraints, including: The inverse correlation between the hydraulic damping characteristics and the joint extension / retraction rate is established through the joint mechanics model building module. The target deformation angle of each hydraulically driven joint is calculated using the calculation module of the joint mechanics model based on the tilt angle change rate in the posture deviation characteristics. Based on the target deformation angle of each hydraulically driven joint, and combined with the target angle threshold in the joint motion constraint, the allowable deformation range of each hydraulically driven joint is determined. Within the allowable deformation range, the center of gravity offset value in the attitude deviation feature is mapped to the pressure adjustment coefficient of the hydraulic circuit according to the reverse correlation to generate real-time damping parameters.

6. The method for maintaining the stability of a robot platform according to claim 5, characterized in that, Within the allowable deformation range, mapping the center of gravity offset value in the attitude deviation feature to the pressure adjustment coefficient of the hydraulic circuit according to the reverse correlation to generate real-time damping parameters includes: The adjustable range of the pressure adjustment coefficient is determined based on the proportional relationship between the abrupt change amplitude of the discontinuous support surface and the allowable deformation range. Based on the reverse correlation, the center of gravity acceleration of the center of gravity offset value is converted into the pressure growth gradient of the hydraulic circuit; Based on the distribution pattern of the pressure growth gradient within the adjustable range, and combined with the pressure fluctuation of the hydraulic circuit, real-time damping parameters are generated.

7. The method for maintaining the stability of a robot platform according to claim 6, characterized in that, The real-time motion data includes acceleration, angular velocity, and visual positioning data; The real-time motion data is filtered and fused to generate attitude deviation features, including: Based on the abrupt displacement characteristics of the discontinuous support surface, the acceleration and the angular velocity are filtered respectively. The displacement offset in the visual positioning data is matched with the filtered acceleration to determine the trajectory. The first deviation component in the vertical direction and the second deviation component in the horizontal direction corresponding to the robot platform are extracted from the matching result. Based on the combined vector direction of the first deviation component and the second deviation component, and combined with the filtered angular velocity, attitude deviation features are generated.

8. A robot platform stability maintenance system, characterized in that, include: The acquisition module is used to acquire real-time motion data of the robot platform; The processing module is used to filter and fuse the real-time motion data to generate attitude deviation features; The input module is used to input the posture deviation features into a pre-trained joint mechanics model, which generates real-time damping parameters based on the posture deviation features, hydraulic damping characteristics, and joint motion constraints. The generation module is used to calculate the attitude deviation characteristics based on the real-time damping parameters and in combination with the adaptive sliding mode control algorithm, and generate compensation control commands for the hydraulically driven joints. The compensation control commands are applied to the deformable support mechanism of the robot platform.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a robot platform stability maintenance method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements a method for maintaining the stability of a robot platform as described in any one of claims 1 to 7.

Images