Suction force self-adaptive adjusting method and system based on suction cup base

Abstract

The invention relates to the technical field of automatic control, and discloses a suction force self-adaptive adjusting method based on a suction cup base, comprising the following steps: collecting current attitude disturbance information of to-be-adsorbed objects including a lamp, glass, a steel plate, a plastic box body and a wood material plate, the attitude disturbance information comprises a pitch angle, a roll angle, an angular velocity and a vertical acceleration; estimating the offset of the gravity center of the to-be-adsorbed object relative to the center of the sucker array based on the attitude disturbance information; according to the offset, target adsorption force of the object to be adsorbed is determined, and distribution is carried out according to a set load distribution weight; the invention further discloses a suction force self-adaptive adjusting system based on the suction cup base. The system comprises a posture sensing module; an offset calculation module; an adsorption force distribution module; and an adsorption control module. According to the invention, dynamic coordination and accurate tracking control of multi-sucker adsorption force are realized by fusing attitude sensing, adsorption modeling and nonlinear predictive control.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, specifically to a method and system for adaptive adjustment of suction force based on a suction cup base. Background Technology

[0002] In scenarios such as stage lighting, exhibition set design, high-altitude maintenance, and construction site work, objects to be adsorbed often need to be temporarily fixed to smooth surfaces using suction cups. Compared to mechanical locking structures, suction cup installation offers advantages such as convenient assembly and disassembly, no surface damage, and strong adaptability. However, as the size of objects to be adsorbed continues to increase, their mass distribution becomes more uneven, and their posture changes frequently during use, traditional static adsorption methods are increasingly unable to meet the dual requirements of adsorption stability and safety.

[0003] Existing technologies typically employ a fixed negative pressure source to uniformly drive multiple suction cups to improve the visibility of the adsorption status. Adsorption force alarm mechanisms are also explored, alerting users to intervene when the suction force falls below a threshold. These methods offer a certain level of safety in applications with typical static loads and relatively stable environments, demonstrating practical value, particularly in static lighting installations and temporary setup systems where they offer advantages in terms of cost and implementation.

[0004] However, these existing technologies still have some significant drawbacks. First, most systems employ preset suction control strategies, failing to adjust the suction cup load based on the center of gravity shift of the object being adsorbed, resulting in poor performance in complex dynamic conditions. Second, while they possess adsorption state detection capabilities, the control loop is not closed-loop, meaning the data is not utilized and cannot drive control actions. Third, conventional control methods are mostly based on simple linear models, making them slow and inflexible in adjusting to factors such as multi-suction cup collaboration, disturbance coupling, and nonlinear adsorption forces. More importantly, existing solutions generally lack predictive control logic, failing to optimize control inputs in advance based on future attitude change trends. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for adaptive adjustment of suction force based on a suction cup base, which solves the problem that the suction force of the suction cup cannot be dynamically adjusted according to changes in the posture of the object to be suctioned and the shift of the center of gravity.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for adaptive adjustment of suction force based on a suction cup base, comprising the following steps: Collect current attitude disturbance information of the object to be adsorbed. The object to be adsorbed includes lamps, glass, steel plates, plastic boxes and wooden boards. The attitude disturbance information includes pitch angle, roll angle, angular velocity and vertical acceleration. The offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array is estimated based on attitude perturbation information. Based on the offset, the target adsorption force of the object to be adsorbed is determined and allocated according to the set load allocation weight; the current adsorption force state of the object to be adsorbed is obtained, and the corresponding air pump control input is calculated in combination with the adsorption response model; Based on a nonlinear predictive control algorithm, the control input of each suction cup is optimized so that the actual adsorption force gradually approaches the target adsorption force.

[0007] Preferably, the attitude perturbation information specifically includes: The pitch angle, roll angle, and attitude change rate of the object to be adsorbed in three-dimensional space can reflect the change in the center of gravity of the object to be adsorbed as a result of external disturbances. The acceleration of the object to be adsorbed in the vertical direction. This acceleration data can reflect the force situation of the object under gravity and external dynamic load, especially when the object vibrates and shakes.

[0008] Preferably, the offset specifically includes: The horizontal offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array is determined by the geometry and weight distribution of the object. The vertical offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array is affected by the external forces and operational disturbances on the object to be adsorbed.

[0009] Preferably, the target adsorption force includes: The required suction force for each suction cup; Based on the center of gravity shift of the object to be adsorbed and the load distribution rules, the target adsorption force is dynamically changing and can be adjusted according to the real-time attitude changes of the object to be adsorbed.

[0010] Preferably, the load allocation weights include: The distance between the suction cup and the center of gravity of the object to be adsorbed is weighted; suction cups closer to the center of gravity will receive a higher weight value. The load capacity and stability of the suction cup are also considered; suction cups with a stronger load capacity will be assigned more adsorption tasks. The interaction effect between suction cups may require adjustments to the distribution strategy when multiple suction cups work together.

[0011] Preferably, the adsorption response model includes: The first-order kinetic relationship between the suction force of the suction cup and the control input; The adsorption force F i (t) depends on the control input u i The relationship changes with the change of (t), and the specific formula is: Among them, F i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time; Control input u i (t) can be adjusted based on real-time feedback to achieve dynamic adjustment of the suction cup's adsorption force.

[0012] Preferably, the control input includes: The air pump drive voltage for each suction cup; The suction force control signal for each suction cup is connected to the air pump through the control system, and the driving voltage is adjusted according to the suction force requirement. Where, α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time; F is the target adsorption force of suction cup i, calculated through the center of gravity offset; i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; the function f is the suction cup control model, which is a control input function related to the adsorption force and is adjusted according to the response characteristics of the suction cup.

[0013] Preferably, the nonlinear predictive control algorithm includes: The suction force of each suction cup is optimized and adjusted by setting a control target function; The specific objective function is: Where J is the length of the prediction period; F i (t) represents the suction force of suction cup i at time t; λ is the target adsorption force of suction cup i, calculated through center of gravity offset; λ is the penalty coefficient for control input, used to limit excessive control input; u i (t) represents the control input; i is the index of the suction cup number; The objective function is optimized to ensure that the adsorption force error of the system is minimized in the future time period and that the control input is not too large.

[0014] Preferably, the actual adsorption force includes: The instantaneous suction force of each suction cup is measured using real-time feedback sensor data; The average suction force of each suction cup is used as an evaluation index for long-term stability.

[0015] The present invention also provides a suction force adaptive adjustment system based on a suction cup base, comprising: The attitude sensing module is used to collect attitude disturbance information of the object to be adsorbed, including pitch angle, roll angle, angular velocity, and vertical acceleration. The offset calculation module estimates the offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array based on the attitude perturbation information; the adsorption force distribution module determines the target adsorption force of each suction cup based on the offset and the load distribution strategy. The adsorption control module is used to obtain real-time adsorption force feedback from the suction cup and adjust the control input of the suction cup through a nonlinear predictive control algorithm so that the actual adsorption force gradually approaches the target adsorption force.

[0016] This invention provides a method and system for adaptive adjustment of suction force based on a suction cup base. It has the following beneficial effects: 1. This invention achieves the technical effect of dynamically identifying changes in the load center of gravity of the object to be adsorbed and responding to the adjustment of the adsorption force in real time by combining a joint strategy of attitude disturbance perception and center of gravity offset estimation. Compared with the existing technology where suction cup control mostly adopts fixed adsorption force or manually set parameters, which cannot respond to changes in external disturbances in real time, this invention improves the system's adaptability to dynamic loads and solves the problem of local instability caused by uneven adsorption force.

[0017] 2. This invention employs an adsorption response model-based reverse control input method, enabling reverse drive calculations based on the actual adsorption force state. This transforms suction cup control from "experience-based setting" to a closed-loop path of "data feedback + dynamic correction." Compared to traditional suction cup control, which relies on unidirectional output commands and lacks feedback self-adjustment, this invention provides more precise control of the adsorption process, reduces negative pressure waste, and avoids the risk of falling due to insufficient adsorption.

[0018] 3. This invention introduces a nonlinear predictive control algorithm to optimize the air pump control input, effectively making the process of adsorption force tracking the target value smoother and more forward-looking. Previous control methods often reacted slowly and adjusted slowly because they did not consider future trends. Through the rolling optimization mechanism, the continuous control capability of the system under complex working conditions is improved, which is particularly suitable for application scenarios with frequent disturbances or rapid attitude switching.

[0019] 4. This invention constructs a complete closed-loop structure from center of gravity identification, adsorption force distribution, state feedback to control optimization. Each module has clear logic and supports each other. Compared with the traditional system with loose modules, isolated data, and weak collaborative control capabilities, this structure not only improves the integration of the control system, but also significantly improves the operational stability and long-term reliability. It is particularly suitable for industrial-grade object suspension platforms that require high adsorption stability. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system architecture of the present invention. Detailed Implementation

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see the appendix Figure 1 This invention provides a method for adaptive adjustment of suction force based on a suction cup base, comprising the following steps: S1. Collect the current attitude disturbance information of the object to be adsorbed. The object to be adsorbed includes lamps, glass, steel plates, plastic boxes and wooden boards. The attitude disturbance information includes pitch angle, roll angle, angular velocity and vertical acceleration. The attitude information of the object to be adsorbed has a crucial impact on the control accuracy of the entire adsorption conditioning process. Changes in attitude disturbance directly cause a dynamic shift in the center of gravity of the object to be adsorbed, and the change in the position of the center of gravity will in turn require dynamic adjustments to the load distribution of the suction cup array. Therefore, in the entire conditioning process, it is necessary to continuously, stably, and in real time collect the attitude state of the object to be adsorbed.

[0023] In general, objects to be adsorbed are easily disturbed by external factors in actual use environments, such as wind disturbances, construction vibrations, and personnel adjustments. These disturbances manifest as changes in attitude, including angular shifts and sudden acceleration changes. To enable the suction cup control system to respond quickly to these disturbances, it is necessary to know the specific dynamic state of the object to be adsorbed in advance.

[0024] In this invention, during the control system initialization phase, the central controller initiates the attitude acquisition task, acquiring dynamic information of the object to be adsorbed through the built-in attitude sensing module. This module can integrate commonly available inertial measurement unit (IMU) sensors, such as the MPU9250 and BNO055 nine-axis sensors, and combine them with high-frequency data sampling and filtering methods to achieve attitude tracking under high dynamic conditions.

[0025] In this embodiment, the collected attitude disturbance information includes pitch, roll, angular velocity, and vertical acceleration. By continuously detecting the angular velocity and acceleration along the XYZ axes, and using attitude fusion algorithms such as complementary filtering, Kalman filtering, or quaternion solving, the sensor can accurately estimate the current attitude state of the object to be adsorbed.

[0026] In one possible implementation, the pitch angle and roll angle represent the rotational offset of the object to be adsorbed around the horizontal and vertical axes, respectively, and are defined as follows: Pitch angle θ: represents the rotation angle of the object to be adsorbed around the Y-axis; Roll angle φ: represents the rotation angle of the object to be adsorbed around the X-axis; The above angle can be preliminarily estimated using the following approximate solution formula: Among them, a x is the output acceleration value of the accelerometer in the X-axis direction; arctan is the arctangent function; a y The accelerometer output acceleration value in the Y-axis direction; a z This represents the output acceleration value of the accelerometer in the Z-axis direction.

[0027] Vertical acceleration reflects the change in the net force acting on the object to be adsorbed in the Z direction (usually the vertical direction). This parameter fluctuates most dramatically when the object is subjected to impact or rapid movement, and is an important reference value for judging short-term dynamic disturbances.

[0028] As an alternative, the attitude sensing module uses a sampling frequency of 100Hz and applies a moving average filter to the sensor data with a window length of 5 points to reduce high-frequency noise interference.

[0029] In some embodiments, sensor data is preprocessed locally before entering the central controller to reduce communication latency. For communication protocols, I2C, SPI, UART, etc., can be used to transmit data to the central processing module. The processing module generates an attitude perturbation vector based on the sampled data. D=[θ,φ,ωx ,ω y ,ω z ,a z ]; Where D is the attitude perturbation vector; θ is the pitch angle; φ is the roll angle; a z ω represents the output acceleration value of the accelerometer in the Z-axis direction. x This represents the output angular velocity value of the angular velocity sensor in the X-axis direction; ω y This represents the output angular velocity value of the angular velocity sensor in the Y-axis direction; ω z This indicates the output angular velocity value of the angular velocity sensor in the Z-axis direction.

[0030] In a typical implementation, the controller completes a full attitude sampling and processing cycle every 50ms. This cycle is synchronized with the adsorption force control cycle to ensure that the control system has complete dynamic closed-loop control capability.

[0031] Furthermore, the attitude disturbance acquisition in this invention is not only used for determining the current control cycle, but also for building a predictive model. The controller stacks historical multi-cycle disturbance data to form a time-series feature, which serves as the input to a nonlinear predictive control algorithm to predict potential offset trends in advance, thereby achieving feedforward correction control.

[0032] In some implementations, to improve attitude estimation accuracy, magnetometer data can be incorporated into the solution process to construct a complete three-axis directional cosine matrix (DCM) or quaternion model, thereby achieving high-precision, low-drift spatial attitude recovery. This enhanced sensor fusion mechanism is particularly important in lighting environments requiring long-term stable adsorption.

[0033] S2. Estimate the offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array based on attitude perturbation information; After acquiring the attitude disturbance information of the object to be adsorbed, the system needs to further convert this disturbance information into a reference quantity that can be used to adjust the suction force. To achieve the core objective of adaptive suction control, the current center of gravity position of the object to be adsorbed needs to be reasonably estimated. Since the object to be adsorbed may tilt, rotate, or vibrate during use, its center of gravity relative to the center of the suction cup array will constantly change.

[0034] Typically, the geometric center of the suction cup array is set as the initial reference point, i.e., the theoretical center of gravity in a static state. When the object to be attracted experiences attitude disturbance, its center of gravity will deviate from this reference position. Therefore, it is necessary to estimate the current actual center of gravity offset relative to this center based on attitude disturbance information to guide the subsequent redistribution of the attraction force.

[0035] In one possible implementation, assuming the geometry of the object to be adsorbed is known, the mass distribution is uniform, and the disturbance is relatively small, the offset estimation can use a linear approximation model: Where K is the offset mapping coefficient matrix, which can be obtained through actual calibration or determined by simulation modeling; θ is the pitch angle; φ is the roll angle; a z Δx is the output acceleration value of the accelerometer in the Z-axis direction; Δx is the offset of the center of gravity of the object to be adsorbed in the X-axis direction; Δy is the offset of the center of gravity of the object to be adsorbed in the Y-axis direction; Δz is the offset of the center of gravity of the object to be adsorbed in the Z-axis direction.

[0036] Alternatively, the system can use a neural network regression model to fit the relationship between perturbation and center of gravity shift. The network input layer is the attitude perturbation vector D, and the output layer is the three-dimensional offset [Δx, Δy, Δz]. Training data can be obtained through calibration experiments on the object to be adsorbed under different tilt angles and attitudes.

[0037] In some embodiments, the system also incorporates structural parameters of the object to be adsorbed, including the outer shell dimensions L, W, and H of the object to be adsorbed and the height h of the support center. c The offset estimation model was further optimized based on factors such as the initial attitude θ0 and φ0 of the installation angle. Δx=f1(θ,ω y ,L,h c ),Δy=f2(φ,ω x ,W,h c ),Δz=f3(a z ,ω z ,H); Wherein, functions f1, f2, and f3 can be linear or nonlinear mappings, depending on the accuracy requirements of the application scenario; Δx is the offset of the center of gravity of the object to be adsorbed in the X-axis direction; Δy is the offset of the center of gravity of the object to be adsorbed in the Y-axis direction; Δz is the offset of the center of gravity of the object to be adsorbed in the Z-axis direction; θ is the pitch angle; φ is the roll angle; a z This represents the output acceleration value of the accelerometer in the Z-axis direction.

[0038] To improve timing stability, the controller uses a sliding window to smooth the estimated values. The window size can be set to 3-5 control cycles, and it is updated using an exponentially weighted average (EMA). This effectively suppresses fluctuations in the center of gravity estimation caused by transient disturbances.

[0039] In addition, to assess the confidence level of the estimated migration results, the system incorporates a variance estimation module to perform statistical analysis on the intensity of disturbances over several past periods. If severe disturbances are detected and attitude data fluctuates significantly, the system automatically increases the response coefficient of the migration model, enhancing its sensitivity to suction adjustments.

[0040] During the actual control cycle, the offset estimation module and the attitude acquisition module operate at the same frequency, usually set to update once every 50ms, to ensure the timeliness and dynamic continuity of the estimation results.

[0041] S3. Determine the target adsorption force of the object to be adsorbed based on the offset, and distribute it according to the set load distribution weight. After estimating the offset of the object's center of gravity relative to the center of the suction cup array, the system needs to reconfigure the target adsorption force of each suction cup unit based on this offset result. This process not only determines the force applied by each suction cup but also directly affects the stability and response speed of the entire system. Therefore, the target adsorption force distribution mechanism constitutes a key aspect of this invention.

[0042] Generally, if the center of gravity of the object to be adsorbed is located near the geometric center of the suction cup array, all suction cups can apply uniform adsorption force to achieve system balance. However, in cases of significant offset, using equal suction force distribution will lead to some suction cups being underloaded or overloaded, resulting in adsorption failure or localized slippage. Therefore, dynamic adjustments based on the center of gravity offset information are necessary.

[0043] Specifically, the target adsorption force is calculated based on the following principle: the suction cup closer to the center of gravity bears more load, and the farther away, the lower the load. To quantify this distribution, the control system introduces a distance decay function: Among them, w i The original load weight of the i-th suction cup is represented by r; r0 represents the geometric center coordinates of the suction cup array; ∈ is a small positive constant to prevent division by zero errors; r i Δr represents the spatial geometric position vector of the i-th suction cup; Δr is the centroid offset vector, representing the offset value of the current center of mass of the object to be adsorbed relative to the geometric center of the suction cup array.

[0044] As an option, the system will process all w i Perform normalization to ensure that it satisfies: Among them, w i This represents the original load weight of the i-th suction cup; i is the index of the suction cup number; N is the total number of suction cups.

[0045] After normalization, let the total weight of the object to be adsorbed be G, then the target adsorption force F of each suction cup is... i (t) can be allocated as follows: F i (t)=G·w i ; Among them, F i (t) represents the suction force of suction cup i at time t; w i This represents the original load weight of the i-th suction cup.

[0046] In some embodiments, the system may also introduce adjustment factors for the state of the suction cup body, such as surface roughness, vacuum response speed, aging degree, etc., to improve the overall control robustness by modifying the weighting factor.

[0047] In one possible implementation, the control system establishes a load scheduling list, updates the target adsorption force of each suction cup in each control cycle, compares it with the current actual adsorption force, and generates a difference as the input of the vacuum pump control command.

[0048] For example, in one type of control mechanism, the system is based on the target adsorption force F i The difference between the current adsorption state and the negative pressure value is adjusted in the connecting pipelines of each suction cup using PID closed-loop control, so that the negative pressure value steadily approaches the target value. In some implementations, a feedforward-feedback composite control strategy can also be used to adjust the negative pressure in advance by predicting disturbances.

[0049] To avoid frequent switching of the target value by the suction cup due to drastic changes in disturbance within a short period of time, the system can introduce a smoothing filter or a limiter for the rate of change of adsorption force to limit the rate of change of the target value and improve system stability.

[0050] Furthermore, in the matrix suction cup arrangement structure, the system can also introduce a "regional weighting mechanism," which sets different basic adsorption weights for different regions by column or row, forming a compensation mechanism for the offset direction. For example, if the object to be adsorbed is biased towards the upper left corner region, the system can assign a higher initial weight to the corresponding suction cup group in that region to enhance the region's load response capability.

[0051] As an extension mechanism, in actual deployment, the system can also dynamically adjust the adsorption force distribution model parameters by combining real-time feedback adsorption effect evaluation values ​​(such as actual changes in adsorption pressure, suction cup load sensor output, etc.) to form an adaptive adjustment mechanism.

[0052] S4. Obtain the current adsorption force state of the object to be adsorbed, and calculate the corresponding air pump control input based on the adsorption response model; After estimating the center of gravity offset based on the attitude disturbance information of the object to be adsorbed, and determining the target adsorption force distribution for each suction cup accordingly, the system needs to enter the adsorption state detection and control input calculation stage to achieve closed-loop regulation of adsorption control. The core task of this stage is to obtain the current adsorption force state of each suction cup, and, in conjunction with the adsorption response model, deduce the corresponding air pump control input, thereby driving the suction cup to generate an adsorption force that matches the target value.

[0053] Generally, the actual adsorption force of a suction cup is controlled by the operating state of the connected vacuum pump, especially the driving voltage and the rate of change of vacuum negative pressure. Because the suction cup's response exhibits a certain hysteresis and nonlinear dynamic process, precise control cannot be achieved solely through static distribution; therefore, an adsorption response model is needed to assist in modeling.

[0054] In this embodiment, each suction cup unit is equipped with an independent suction force sensor module for real-time acquisition of the current suction force value. This sensor employs a measurement unit based on piezoresistive, capacitive, or differential pressure principles, and its output signal is input to the central controller via an A / D conversion module.

[0055] To ensure the timeliness and accuracy of the measurement data, the sensor sampling frequency is set within the range of 10Hz to 50Hz, and the specific frequency can be dynamically adjusted according to the system control cycle. In some embodiments, a multi-point filtering algorithm (such as weighted moving average) is also introduced to smooth out abrupt interference within short periods.

[0056] Specifically, the adsorption response model of the suction cup can be approximated using a first-order linear differential model, the mathematical form of which is shown below: Among them, F i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time.

[0057] As an alternative, to avoid instability caused by sensor errors and model biases, the system introduces a feedforward term to correct the model input: Among them, u i (t) represents the control input; α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time; This represents the target's adsorption force at the current moment; The rate of change of the target adsorption force can be approximated by the target sequence difference.

[0058] In one possible implementation, each suction cup has an adsorption force control signal, which is connected to the air pump via a control system to adjust the drive voltage according to the adsorption force requirement. Where, α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time; F is the target adsorption force of suction cup i, calculated through the center of gravity offset; i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; the function f is the suction cup control model, which is a control input function related to the adsorption force and is adjusted according to the response characteristics of the suction cup.

[0059] To further enhance the robustness of the control input, the system is also equipped with an input saturation limiting module to ensure that the control voltage or air pump PWM signal does not exceed its physical limits.

[0060] S5. Based on a nonlinear predictive control algorithm, optimize the control input of each suction cup so that the actual adsorption force gradually approaches the target adsorption force. After acquiring the current adsorption force state of the suction cups and deriving the initial control input using the adsorption response model, the system further introduces a nonlinear predictive control algorithm to continuously optimize the control input of each suction cup in a dynamic environment, gradually bringing the actual adsorption force closer to the target adsorption force. The core of this control stage lies in constructing an optimization problem with time-domain prediction. By solving for the optimal input sequence, the system's adaptability to disturbance changes is improved, and the adsorption performance is ensured to converge rapidly within a finite time.

[0061] In general, traditional proportional-integral (PI) control strategies suffer from response lag and large steady-state errors when facing complex operating conditions such as multi-suction-cup cooperative control, model nonlinearity, and system coupling. Therefore, adopting a nonlinear predictive control strategy can ensure dynamic stability while introducing constraints to limit the control input and maintaining the system's continuous adjustment capability through rolling time-domain optimization.

[0062] The suction force of each suction cup is optimized and adjusted by setting a control target function; The specific objective function is: Where J is the length of the prediction period; F i (t) represents the suction force of suction cup i at time t; λ is the target adsorption force of suction cup i, calculated through center of gravity offset; λ is the penalty coefficient for control input, used to limit excessive control input; u i (t) represents the control input; i is the index of the suction cup number.

[0063] In one possible implementation, the prediction model is discretized, and the sampling period is Δt. The adsorption force update formula then becomes: F i (t+1)=F i (t)+Δt·(α i u i (t)-β i F i (t)); Among them, F i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time.

[0064] Specifically, this nonlinear predictive control algorithm forms a continuous control link with the aforementioned adsorption response modeling and target adsorption force dynamic allocation modules. The target adsorption force is jointly determined by the center of gravity offset and load weight, the initial control input is provided by the adsorption model estimation, and the predictive controller further updates the control strategy with the overall system optimization as the goal, while meeting the adsorption force accuracy and control amplitude limitations.

[0065] At the system architecture level, the predictive control algorithm runs in the central controller, periodically acquiring real-time suction force feedback and historical input values ​​for each suction cup to form a control closed loop. To ensure sufficient computing resources, the central processing unit uses an embedded high-performance processor and reserves FPGA or DSP resources to support optimized computing acceleration.

[0066] Furthermore, as a further extension of the nonlinear predictive control mechanism, the system can integrate a disturbance observer or an online model adaptive mechanism to respond when the suction cup response parameters drift slowly or environmental dynamic disturbances intensify.

[0067] The suction force adaptive adjustment system based on a suction cup base described below and the suction force adaptive adjustment method based on a suction cup base described above can be referred to in correspondence.

[0068] Please see the appendix Figure 2 The present invention also provides an adaptive suction force adjustment system based on a suction cup base, comprising: an attitude sensing module for collecting attitude disturbance information of the object to be adsorbed, including pitch angle, roll angle, angular velocity and vertical acceleration; The offset calculation module estimates the offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array based on attitude perturbation information; the adsorption force distribution module determines the target adsorption force of each suction cup based on the offset and the load distribution strategy. The adsorption control module is used to obtain real-time adsorption force feedback from the suction cup and adjust the control input of the suction cup through a nonlinear predictive control algorithm so that the actual adsorption force gradually approaches the target adsorption force.

[0069] The system in this embodiment can be used to execute the above method embodiments, and its principle and technical effect are similar, so they will not be described again here.

[0070] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for adaptive adjustment of suction force based on a suction cup base, characterized in that, Includes the following steps: Collect current attitude disturbance information of the object to be adsorbed. The object to be adsorbed includes lamps, glass, steel plates, plastic boxes and wooden boards. The attitude disturbance information includes pitch angle, roll angle, angular velocity and vertical acceleration. Estimate the offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array based on attitude perturbation information; Based on the offset, the target adsorption force of the object to be adsorbed is determined and distributed according to the set load distribution weight. Obtain the current adsorption force state of the object to be adsorbed, and calculate the corresponding air pump control input based on the adsorption response model; Based on a nonlinear predictive control algorithm, the control input of each suction cup is optimized so that the actual adsorption force gradually approaches the target adsorption force.

2. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The attitude perturbation information specifically includes: The pitch angle, roll angle, and attitude change rate of the object to be adsorbed in three-dimensional space can reflect the change in the center of gravity of the object to be adsorbed as a result of external disturbances. The acceleration of the object to be adsorbed in the vertical direction. This acceleration data can reflect the force situation of the object under gravity and external dynamic load, especially when the object vibrates and shakes.

3. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The offset specifically includes: The horizontal offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array is determined by the geometry and weight distribution of the object to be adsorbed. The vertical offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array is affected by the external forces and operational disturbances on the object to be adsorbed.

4. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The target adsorption force includes: The required suction force for each suction cup; Based on the center of gravity shift of the object to be adsorbed and the load distribution rules, the target adsorption force is dynamically changing and can be adjusted according to the real-time attitude changes of the object to be adsorbed.

5. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The load allocation weights include: The distance between the suction cup and the center of gravity of the object to be adsorbed is weighted; suction cups that are closer to the center of gravity will receive a higher weight value. The load capacity and stability of the suction cup; suction cups with stronger load capacity will be assigned more suction tasks. The interaction effect between suction cups necessitates adjusting the distribution strategy when multiple suction cups work together.

6. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The adsorption response model includes: The first-order kinetic relationship between the suction force of the suction cup and the control input; The adsorption force F i (t) depends on the control input u i The relationship changes with the change of (t), and the specific formula is: Among them, F i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time; Control input u i (t) can be adjusted according to real-time feedback to achieve dynamic adjustment of the suction cup adsorption force.

7. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The control inputs include: The air pump drive voltage for each suction cup; The suction force control signal for each suction cup is connected to the air pump through the control system, and the driving voltage is adjusted according to the suction force requirement. Where, α i The pressure response coefficient of suction cup i determines the degree of direct influence of the control input on the adsorption force; β i Let be the adsorption decay coefficient of suction cup i, which describes the natural decay of adsorption force over time; F is the target adsorption force of suction cup i, calculated through the center of gravity offset; i (t) represents the suction force of suction cup i at time t; u i (t) represents the control input; the function f is the suction cup control model, which is a control input function related to the adsorption force and is adjusted according to the response characteristics of the suction cup.

8. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The nonlinear predictive control algorithm includes: The suction force of each suction cup is optimized and adjusted by setting a control target function; The specific objective function is: Where J is the length of the prediction period; F i (t) represents the suction force of suction cup i at time t; λ is the target adsorption force of suction cup i, calculated through center of gravity offset; λ is the penalty coefficient for control input, used to limit excessive control input; u i (t) represents the control input; i is the index of the suction cup number; The objective function is optimized to ensure that the adsorption force error of the system is minimized in the future time period and that the control input is not too large.

9. The adaptive adjustment method for suction force based on a suction cup base according to claim 1, characterized in that, The actual adsorption force includes: The instantaneous suction force of each suction cup is measured using real-time feedback sensor data; The average suction force of each suction cup is used as an evaluation index for long-term stability.

10. A suction force adaptive adjustment system based on a suction cup base, and a suction force adaptive adjustment method based on a suction cup base according to any one of claims 1-9, characterized in that, include: The attitude sensing module is used to collect attitude disturbance information of the object to be adsorbed, including pitch angle, roll angle, angular velocity, and vertical acceleration. The offset calculation module estimates the offset of the center of gravity of the object to be adsorbed relative to the center of the suction cup array based on the attitude perturbation information. The adsorption force distribution module determines the target adsorption force of each suction cup based on the offset and the load distribution strategy. The adsorption control module is used to obtain real-time adsorption force feedback from the suction cup and adjust the control input of the suction cup through a nonlinear predictive control algorithm so that the actual adsorption force gradually approaches the target adsorption force.

Images