ZR axial force closing control method and related equipment thereof

Abstract

The invention provides a ZR axial force closed control method and related equipment thereof. The method comprises the following steps: acquiring mechanical data of at least one force sensor in real time; determining correction parameters of a force closing control algorithm based on the mechanical data and corresponding target mechanical data; on the basis of the correction parameters, correction control output used for force closing control is generated in real time through the force closing control algorithm; and driving corresponding target equipment to perform force closing correction control based on the correction control output. The current value, the historical cumulant and the change rate of a force value error signal are subjected to quantitative calculation and synthesis control output in a control period, and the internal state of the algorithm is dynamically corrected in combination with an amplitude limiting and integral saturation resisting strategy, so that low-delay, high-impedance-disturbance and high-robustness force closing control is realized, and the control precision is improved. And the intellectualization of force closing control is improved.

Description

Technical Field

[0001] This invention relates to the field of automation control, and in particular to a ZR axial force closed-loop control method, device, electronic device and its storage medium. Background Technology

[0002] Currently, in force control applications for the ZR axis of robots, traditional PID control algorithms or simple threshold judgment methods are commonly used. The control signals are usually calculated by a separate motion controller or PLC and then sent to the servo driver or amplifier for execution.

[0003] Traditional solutions often rely on upper-level controllers or fixed-parameter PID controllers, which suffer from closed-loop delays caused by communication round trips. They are difficult to balance the rapid response and steady-state phase of the contact phase. At the same time, they are not robust enough to sensor noise and nonlinear disturbances (friction, load changes, coupling interference), and are prone to jitter and force fluctuations. When the control output reaches the physical limit of the actuator, the saturation problem caused by the accumulation of integral terms will slow down the recovery speed and amplify overshoot, affecting process consistency and yield.

[0004] Therefore, existing ZR axial force closed-loop control methods suffer from problems such as large closed-loop time delay, poor robustness to noise and changes in operating conditions (friction / load / attitude coupling), overshoot and oscillation caused by fixed-parameter PID and easy integral saturation, lack of effective protection against over-limits, and excessive reliance on the host controller, resulting in high system complexity and cost. Summary of the Invention

[0005] This invention provides a ZR axial force closed-loop control method to solve the problems of existing ZR axial force closed-loop control methods, such as large closed-loop time delay, poor robustness to noise and operating condition changes (friction / load / attitude coupling), overshoot and oscillation caused by fixed parameter PID and easy integral saturation, lack of effective protection against over-limit, and excessive reliance on the host controller leading to high system complexity and cost.

[0006] In a first aspect, the present invention provides a ZR axial force closed-loop control method, the method comprising the following steps: Real-time acquisition of mechanical data from at least one force sensor; Based on the aforementioned mechanical data and the corresponding target mechanical data, the correction parameters for the force closure control algorithm are determined. Based on the correction parameters, the correction control output for force closure control is generated in real time through the force closure control algorithm; Based on the corrected control output, the corresponding target device is driven to perform force closure correction control.

[0007] Optionally, before acquiring the mechanical data from at least one force sensor in real time, the method further includes: Based on the current target device, determine the corresponding sensing parameters and control parameters; Based on the sensing parameters and control parameters, the force sensor corresponding to the current target device is calibrated and zeroed to determine the initial state of the target device.

[0008] Optionally, the real-time acquisition of mechanical data from at least one force sensor includes: Within a preset sampling period, the raw data of the at least one force sensor is read in real time via an analog-to-digital converter; The original data is written into the sampled data buffer, and the corresponding index is updated; Based on the data in the sampling data buffer, a filtering operation is performed using a preset filtering algorithm to determine a smooth output, which is then used as the mechanical data.

[0009] Optionally, the step of performing filtering operations based on the data in the sampled data buffer using a preset filtering algorithm to determine the smooth output includes: The corresponding storage location is located in the sampled data buffer based on the updated index; The sampled data at the storage location are accumulated and averaged according to the preset window length to obtain the current smooth output.

[0010] Optionally, determining the correction parameters of the force closure control algorithm based on the mechanical data and the corresponding target mechanical data includes: Based on the aforementioned mechanical data and the corresponding target mechanical data, the real-time force value and the corresponding target force value are determined respectively. Based on the real-time force value and the target force value, the corresponding force error signal is calculated; Based on the current value, historical cumulative amount, and rate of change of the force error signal, the response component, cumulative component, and prediction component are calculated respectively, and the original amplitude control output is synthesized. When the original amplitude control output exceeds the preset amplitude range, the original amplitude control output is limited according to the preset amplitude limiting strategy, and the cumulative component is adjusted by inverse integration to obtain the amplitude control output after limiting. The cumulative component adjusted by inverse integration is used as the correction parameter of the force closure control algorithm.

[0011] Optionally, the step of generating a corrected control output for force closure control in real time based on the corrected parameters using the force closure control algorithm includes: Based on the correction parameters, the internal state parameters in the force closure control algorithm are corrected to obtain the updated force closure control algorithm; Based on the updated force-closed control algorithm, the response component, cumulative component and prediction component are calculated for the force error signal determined by the target force value and the real-time force value in the next control cycle, and then the corrected control output is synthesized.

[0012] Optionally, the step of driving the corresponding target device to perform force closure correction control based on the corrected control output includes: The correction control output is converted by digital-to-analog conversion to generate a control signal for driving the corresponding target device to perform force closure correction control. The control signal is sent to the corresponding target device to drive the actuator to move; When the target device drives the actuator to move, the actuator is continuously adjusted according to the correction control output so as to maintain the force closure state corresponding to the target force value.

[0013] Secondly, the present invention also provides a ZR axial force closure control device, the ZR axial force closure control device comprising: The first acquisition module is used to acquire mechanical data from at least one force sensor in real time. The first determining module is used to determine the correction parameters of the force closure control algorithm based on the mechanical data and the corresponding target mechanical data. The first correction module is used to generate a correction control output for force closure control in real time based on the correction parameters and the force closure control algorithm. The first driving module is used to drive the corresponding target device to perform force closure correction control based on the correction control output.

[0014] Thirdly, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the ZR axial force closure control method provided by the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the ZR axial force closure control method provided by the invention.

[0016] This invention acquires mechanical data from at least one force sensor in real time; based on the mechanical data and the corresponding target mechanical data, it determines the correction parameters for a force closure control algorithm; based on the correction parameters, it generates a corrected control output for force closure control in real time using the force closure control algorithm; and based on the corrected control output, it drives the corresponding target device to perform force closure correction control. By quantizing and calculating the current value, historical cumulative amount, and rate of change of the force error signal within the control cycle and synthesizing the control output, and by combining amplitude limiting and anti-integral saturation strategies, the internal state of the algorithm is dynamically corrected, thereby achieving low-latency, high-impedance disturbance, and high-robust force closure control, and improving the intelligence of force closure control. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a ZR axial force closed-loop control method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of another ZR axial force closing control device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of 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.

[0020] like Figure 1 As shown, Figure 1 This is a flowchart of a ZR axial force closure control method provided by an embodiment of the present invention. The ZR axial force closure control method includes the following steps: 101. Real-time acquisition of mechanical data from at least one force sensor.

[0021] In this embodiment of the invention, the ZR axial force closure control method described above can be applied to a ZR axial force closure control platform. The ZR axial force closure control platform has functions such as force closure control data processing, force closure control data transmission and reception, and force closure control data memory storage. It can be built based on a server or server cluster. The server or server cluster can be an electronic device with force closure control data processing capabilities.

[0022] The aforementioned mechanical data can refer to the real-time force data obtained from the raw sampling data acquired from at least one force sensor within a preset sampling period, which is then read by an analog-to-digital converter and stored in the sampling data buffer, and smoothed according to a preset filtering algorithm. It can be understood that the aforementioned mechanical data can be used as input to the control algorithm to reflect the actual force situation of the current actuator.

[0023] The aforementioned preset sampling period can be a periodic parameter for sampling force sensor data at predetermined time intervals, used to determine the readout frequency of the analog-to-digital converter (ADC) and the operation cycle of the control algorithm. It can be understood that within one sampling period, the aforementioned ZR-axis force closure control platform completes one force sensor data acquisition, filtering, and control operation update. More specifically, a fixed value within the range of 1 ms to 10 ms can be selected according to the requirements of control accuracy and response speed, or it can be dynamically adjusted according to the working conditions. For example, a shorter sampling period can improve the real-time performance of force value feedback, but it places higher demands on the processing power and communication bandwidth of the aforementioned ZR-axis force closure control platform; a longer sampling period is beneficial for stability and anti-interference capabilities.

[0024] The aforementioned sampling data buffer can refer to a cache structure used to store the sampling data of the force sensor in the ZR axial force closed control platform, which can adopt storage forms such as a circular queue, a circular array, or double buffering.

[0025] In one possible embodiment, the ZR axial force closure control platform described above writes the raw force sensor data read by the analog-to-digital converter into the current index position of the sampling data buffer during each sampling cycle, and updates the index pointer after writing; when the index reaches the upper limit of the buffer, it automatically wraps back to the starting position to form a sliding window-style data update mechanism.

[0026] In each cycle of filtering calculation, the aforementioned control algorithm calls upon the latest N sampled values ​​in the sampling data buffer, performs accumulation and averaging operations on them, and obtains a smooth real-time force value. This allows it to reflect the sensor output trend in real time and suppress transient noise without increasing storage burden.

[0027] 102. Based on the mechanical data and the corresponding target mechanical data, determine the correction parameters of the force closure control algorithm.

[0028] In this embodiment of the invention, the aforementioned target mechanical data may refer to the target force value sequence pre-set by the ZR axial force closure control platform according to the operation process or control stage (such as contact, lifting, holding, unloading), which is used to compare with the real-time collected mechanical data to form a force value error signal, providing a basis for subsequent control algorithm calculation.

[0029] The aforementioned force-closed-loop control algorithm can refer to an algorithm that performs composite feedback regulation based on the force error signal. This includes, but is not limited to, quantizing and calculating the current value, historical cumulative amount, and rate of change of the error to form response components, cumulative components, and prediction components, and then weighting and synthesizing the three to achieve the control output. It is understood that the aforementioned force-closed-loop control algorithm is executed within each control cycle, and is combined with amplitude limiting processing and anti-integral saturation strategies to achieve stable closed-loop control.

[0030] The aforementioned correction parameter can refer to the correction value obtained by the ZR axial force closing control platform in reverse adjusting the cumulative component (i.e. integral component) in the algorithm while performing amplitude limiting processing when the control output exceeds the preset amplitude range. It can be used to update the internal state of the algorithm to suppress integral saturation and improve the convergence speed.

[0031] 103. Based on the correction parameters, the correction control output for force closure control is generated in real time through the force closure control algorithm.

[0032] In this embodiment of the invention, the ZR axial force closed-loop control platform calculates the response component, cumulative component and prediction component based on the current force error signal and the corrected internal state parameters within the control cycle, and synthesizes the control output, which is then sent out as a corrected control output after being limited.

[0033] The aforementioned corrected control output refers to the control quantity generated after executing the amplitude limiting and anti-integral saturation strategies. It represents the execution signal of the ZR shaft force closure control platform in this control cycle and is used to drive the target equipment to achieve force closure control.

[0034] 104. Based on the corrected control output, drive the corresponding target device to perform force closure corrected control.

[0035] In this embodiment of the invention, the target device may refer to the execution end connected to the ZR axial force closing control platform, including but not limited to the servo driver and the mechanical actuator it drives. It is understood that the ZR axial force closing control platform converts the correction control output into a drive signal after digital-to-analog conversion and sends it to the target device to perform the action.

[0036] In one possible embodiment, the ZR axial force closed-loop control platform described above uses a modified control output to drive the actuator, continuously adjusting the force state in a closed loop, so that the actual output force value quickly approaches and stabilizes at the target force value, maintaining a constant force operation state.

[0037] In another possible embodiment, the ZR axis force closure control platform acquires force data in real time from the force sensor installed on the ZR axis. The sampling results are written into the sampling data buffer after analog-to-digital conversion, and a smooth output value is calculated by the moving average filtering algorithm as the real-time mechanical data for the current cycle. The target force value corresponding to the current control cycle is read from the target mechanical data, and the difference between the target force value and the real-time force value is calculated to obtain the force error signal. Based on the current value, historical cumulative amount and rate of change of the force error signal, response component, cumulative component and prediction component are formed respectively, and the original control output is synthesized.

[0038] When the original control output exceeds the preset amplitude range, the ZR axial force closing control platform performs amplitude limiting processing and reverses the cumulative component to prevent integral saturation, thereby obtaining correction parameters. These correction parameters are recorded in the algorithm's internal state and used to update the calculation benchmark for the next cycle.

[0039] In the updated algorithm state, the ZR axial force closing control platform recalculates the three types of components and synthesizes the corrected control output based on the new force error signal in the subsequent control cycle. The output is then converted into an analog drive signal by the digital-to-analog converter circuit and sent to the servo driver to drive the actuator.

[0040] During execution, the ZR axis force closure control platform monitors the feedback force value in real time and continuously adjusts and corrects the control output, so that the ZR axis can maintain a stable target force value in the contact, lifting, holding and unloading stages, thereby realizing force closure correction control.

[0041] By employing the above methods and steps, sensor noise and integral saturation can be suppressed while maintaining a rapid response, and force overshoot and oscillation amplitude can be reduced, thereby improving the constant force control accuracy and dynamic tracking performance of the ZR axial force closed-loop control platform.

[0042] In this embodiment of the invention, mechanical data from at least one force sensor is acquired in real time; based on the mechanical data and the corresponding target mechanical data, correction parameters for the force closure control algorithm are determined; based on the correction parameters, a correction control output for force closure control is generated in real time through the force closure control algorithm; based on the correction control output, the corresponding target device is driven to perform force closure correction control. By quantizing and calculating the current value, historical cumulative amount, and rate of change of the force error signal within the control cycle and synthesizing the control output, and by combining amplitude limiting and anti-integral saturation strategies, the internal state of the algorithm is dynamically corrected, thereby achieving low-latency, high-impedance disturbance and high-robust force closure control, and improving the intelligence of force closure control.

[0043] Optionally, in the steps prior to real-time acquisition of mechanical data from at least one force sensor, the corresponding sensing parameters and control parameters can be determined based on the current target device; based on the sensing parameters and control parameters, the force sensor corresponding to the current target device can be calibrated and zeroed to determine the initial state of the target device.

[0044] In this embodiment of the invention, the aforementioned sensing parameters may refer to a set of data describing the characteristics of the force sensor on the target device, used to determine the sensor's output characteristics under different forces, including but not limited to sensitivity coefficient, full-scale output, voltage zero-point offset, linear compensation coefficient, and temperature drift correction coefficient. For example, if the sensor's sensitivity is 2.0 mV / V, the output signal should be 2.0 mV when 100 N is applied. The ZR axial force closure control platform uses this parameter to convert the voltage signal into the actual force value. If the temperature rises and causes output drift, the ZR axial force closure control platform can automatically correct the error based on the compensation coefficient.

[0045] The aforementioned control parameters may refer to the basic variables used to configure the force-closed-loop control algorithm, including but not limited to the proportional coefficient K. p Integral coefficient K i Differential coefficient K d Sampling period T s And output limiting values ​​(Umax / Umin), etc. Generally speaking, K p This can be used to describe the response speed of the aforementioned ZR axial force closed-loop control platform to the current error, K i Determines steady-state accuracy, K d Used to suppress overshoot; if the goal is "rapidly adhere to the workpiece without impact", then K is usually increased. p Decrease K i Appropriately increase K d By setting appropriate control parameters, a balance can be achieved between response speed and stability.

[0046] The aforementioned force sensor can be installed at the execution end of the target equipment to convert mechanical force into a measurable electrical signal. Generally, when the strain gauge force sensor on the ZR shaft clamping mechanism is subjected to clamping force, the resistance of its bridge circuit will change slightly, and the output voltage signal is amplified and input to the aforementioned ZR shaft force closure control platform; the aforementioned ZR shaft force closure control platform can read this signal to know the current actual force magnitude of the gripper.

[0047] In one possible embodiment, the ZR axial force closed-loop control platform described above can correct the deviation between the sensor output and the actual force through standard force or no-load testing, so that the measured value is consistent with the actual force value. For example, after leaving the factory or after installation, the ZR axial force closed-loop control platform will collect the output voltage at multiple standard force points such as 0N, 50N, and 100N, and establish a voltage-force value comparison table; if there is a residual 0.02V in the output at 0N, the deviation will be eliminated by calibration coefficient.

[0048] The initial state mentioned above refers to the static steady state where the control variables and sensor outputs have returned to their reference values ​​after the ZR axial force closed-loop control platform has completed calibration and zeroing. Understandably, at this point, the sensor output is zero, the controller's integral term is cleared, the servo driver does not output voltage, and the actuator remains stationary and unloaded. This state is equivalent to the "starting point of the control loop," ensuring that the ZR axial force closed-loop control platform begins closed-loop control from a zero-force equilibrium state, avoiding any initial shocks.

[0049] Optionally, the step of acquiring mechanical data from at least one force sensor in real time further includes reading the raw data from at least one force sensor in real time via an analog-to-digital converter within a preset sampling period; writing the raw data into a sampling data buffer and updating the corresponding index; and performing filtering operations based on the data in the sampling data buffer using a preset filtering algorithm to determine a smooth output and using it as mechanical data.

[0050] In this embodiment of the invention, the preset sampling period can refer to a fixed time interval set by the ZR axial force closure control platform for periodically collecting force sensor data. It is understood that the shorter the preset sampling period is, the more sensitive the ZR axial force closure control platform is, but the higher the computational load; the longer the period, the less noise impact there is, but the response is slightly slower. Therefore, it can be automatically adjusted according to the specific implementation plan.

[0051] The aforementioned raw data may refer to the unfiltered or uncorrected digital signal obtained after the force sensor is sampled by the analog-to-digital converter (ADC). It can be used to reflect the instantaneous voltage value or digital quantity output by the sensor and often contains a certain amount of noise or jitter.

[0052] The aforementioned sampling data buffer can be a cache structure within the ZR axial force closure control platform used to store raw data collected over several cycles. It can typically be implemented as a circular array with a capacity equal to the filter window length (e.g., 10-20 sampling points). For example, if the window length is 10, the buffer always stores data from the most recent 10 cycles. When new data is written, the oldest data is overwritten to achieve real-time sliding updates. This sampling data buffer allows filtering operations to be performed continuously within limited memory, avoiding data loss or delays.

[0053] In one possible embodiment, after each new sampled data is written, the storage position pointer (index) of the buffer in the ZR axial force closure control platform is automatically incremented to point to the next write position. When the index reaches the upper limit of the buffer, the ZR axial force closure control platform automatically wraps back to the starting position (i.e., "circular queue" mode).

[0054] The aforementioned preset filtering algorithm can be an algorithm rule used to smooth the signal of the original data in the sampling data buffer, including but not limited to the moving average method, weighted average method, or low-pass filtering method, which are used to reduce short-term noise fluctuations and improve signal stability.

[0055] The smooth output mentioned above can refer to the result after processing by a preset filtering algorithm, that is, the stable force value data obtained after removing high-frequency noise. The smooth output can more accurately reflect the actual force condition of the ZR axis and serve as the input signal for subsequent control algorithms (such as PID control).

[0056] In another possible embodiment, the ZR axial force closure control platform described above acquires raw data from the force sensor in real time through an analog-to-digital converter in each preset sampling period, writes the sampling results into a circular sampling data buffer, and then automatically updates the corresponding index position. Based on the continuous data in the buffer, a preset filtering algorithm is executed to obtain a smooth mechanical output value.

[0057] By employing the above methods and steps, sampling noise and instantaneous fluctuations can be suppressed, ensuring that the feedback force signal is continuous, stable, and can truly reflect the force state of the actuator. This provides accurate input data for subsequent force closure algorithms, enabling the ZR axis to have higher response accuracy and stability in dynamic fitting, constant force holding, and other control processes.

[0058] Optionally, in the step of determining the smooth output by performing filtering operations on the data in the sampled data buffer using a preset filtering algorithm, the method further includes finding the corresponding storage location in the sampled data buffer based on the updated index; and performing accumulation and averaging operations on the sampled data at the storage location according to a preset window length to obtain the current smooth output.

[0059] In this embodiment of the invention, the aforementioned storage location may refer to a specific index unit within the aforementioned sampling data buffer used to store the original data within the current sampling period. Specifically, the aforementioned ZR axial force closure control platform can indicate the buffer address to which each sampling result should be written or read using the index number in the index. For example, when the sampling data buffer capacity is 10 and the index is 7, new sampling data will be stored in the 7th unit; the index increments to 8 during the next update, and wraps back to 0 when the upper limit is reached, achieving cyclic coverage.

[0060] The aforementioned preset window length can refer to the number of consecutive sampled data that the ZR axial force closure control platform takes at one time when performing filtering operations. It can be understood that the aforementioned preset window length can usually be determined according to the required response characteristics and noise frequency of the ZR axial force closure control platform (for example, N=10 represents the most recent 10 sampling points). The larger the preset window length, the more obvious the filtering and smoothing effect, but the slightly slower the response speed; the smaller the preset window length, the faster the response but the lower the smoothness.

[0061] For example, when the sampling period Ts = 5ms and the window length N = 10, the time range covered by the filter window is 50ms, thereby suppressing short-period noise.

[0062] The aforementioned sampled data refers to the raw force signal sequence acquired by the ZR axis force closure control platform from the force sensor via an analog-to-digital converter. Generally, the sampled data can be stored in a sampling data buffer in chronological order to reflect the force changes of the ZR axis end actuator within a continuous sampling period. For example, when the robotic arm enters the contacting stage, the continuously acquired force values ​​may be [9.7, 9.8, 10.0, 10.2, 9.9] N, which are the sampled data to be filtered.

[0063] In one possible embodiment, the ZR axial force closure control platform accumulates and averages the data at the corresponding storage locations according to a preset window length to obtain a smooth output for the current cycle. Specifically, a moving average or weighted average algorithm can be used.

[0064] For example: when the window length N=5, the corresponding sampled data is [9.8,10.0,10.2,9.9,10.1]N, then the smoothed output F_filtered=(9.8+10.0+10.2+9.9+10.1) / 5=10.0N.

[0065] Alternatively, the current filtered output value y_output can be obtained by dividing the updated sum Sum by the window length N, and the calculated average value y_output can be sent to the ZR axial force closure control platform mentioned above for further processing.

[0066] In another possible embodiment, the ZR axial force closure control platform determines the storage location of the current data in the sampling data buffer based on the updated index. From this storage location, it traces back a preset window length N (e.g., 10 sampling points), reads the corresponding sampling data sequence, and performs cumulative summation and averaging operations to obtain the smooth output force value for the current cycle. In the specific implementation, the sampling data buffer adopts a circular structure design, automatically wrapping around when the index reaches the upper limit of the buffer, ensuring the continuity and real-time nature of data updates.

[0067] Moving average or weighted average algorithms enable smoothed output values ​​to dynamically track the trend of force changes, while filtering out instantaneous spikes or random noise, making the feedback signal stable and usable.

[0068] Optionally, the step of determining the correction parameters of the force-closed-loop control algorithm based on the mechanical data and the corresponding target mechanical data further includes: determining the real-time force value and the corresponding target force value based on the mechanical data and the corresponding target mechanical data; calculating the corresponding force error signal based on the real-time force value and the target force value; calculating the response component, cumulative component, and prediction component based on the current value, historical cumulative amount, and rate of change of the force error signal, and synthesizing the original amplitude control output; when the original amplitude control output exceeds the preset amplitude range, limiting the original amplitude control output according to the preset limiting strategy, and performing inverse integration adjustment on the cumulative component to obtain the amplitude control output after limiting; and using the cumulative component adjusted by inverse integration as the correction parameter of the force-closed-loop control algorithm.

[0069] In this embodiment of the invention, the aforementioned real-time force value refers to the actual force value obtained by the ZR axis force closure control platform within the current control cycle, based on force sensor data and after filtering. This value reflects the true force state of the ZR axis actuator at the current moment. For example, when the actuator is in contact with the workpiece surface, the real-time force value may be 9.8N, representing the current force magnitude at the axis end.

[0070] The target force value mentioned above can be the expected output force preset by the ZR shaft force closed control platform or issued by the upper control logic. It can be compared with the real-time force value in real time, so as to dynamically adjust according to different working conditions. In other words, the target force value is used as a reference benchmark for the control algorithm to compare with the real-time force value to generate control error.

[0071] In this embodiment, the ZR axial force closure control platform described above can derive the control output by performing multiple processes such as error signal acquisition, PID component calculation, amplitude synthesis, and amplitude limiting correction on the input signal.

[0072] The aforementioned force error signal can refer to the difference between the target force value and the real-time force value, i.e.:

[0073] Where K represents the current sampling time, For the target force value, This is the real-time force value.

[0074] The aforementioned historical cumulative value refers to the integral value of the error signal from the start-up control to each cycle, used to reflect the degree of long-term deviation accumulation and to eliminate system steady-state error. It is understandable that if the platform is consistently 0.1 N lower than the target force value, the historical cumulative value will continue to increase, thereby driving an increase in the output force.

[0075] The aforementioned rate of change can refer to the speed at which the force error signal changes between two adjacent sampling periods. It is used to predict the trend of error change and help the platform correct the output in advance when the error rises or falls rapidly.

[0076] The above response components can be the proportional term (P term) in the corresponding PID algorithm, which is used to generate an instantaneous response based on the current error. Its calculation formula is: P=Kp×e(k). The larger the P term, the faster the platform responds, but the risk of overshoot also increases. The aforementioned cumulative component can be the integral term (I term) in the corresponding PID algorithm, used to correct long-term residual deviations. Its calculation formula is: I = I_prev + Ki × Ts × e(k). The I term can eliminate steady-state error, but if it accumulates excessively, it may lead to integral saturation. The aforementioned predictive component can be the differential term (D term) in the corresponding PID algorithm, used to predict the trend of error change and generate a damping effect. Its calculation formula is: D=Kd×(e(k)-e(k-1)) / Ts. The introduction of the D term can suppress the overshoot and oscillation of the output.

[0077] The aforementioned raw amplitude control output can refer to the uncorrected control signal obtained by the linear superposition of the response component, cumulative component, and prediction component, i.e.: Uraw = P + I + D The above-mentioned original amplitude control output is used to reflect the theoretical output calculated by the current control algorithm, but does not take into account the physical limitations of the actuator.

[0078] In this embodiment, the ZR axial force closure control platform superimposes the response component, cumulative component, and prediction component according to their weight ratios to form the original amplitude control output.

[0079] The aforementioned preset amplitude range may refer to the safety limit range set by the ZR axial force closing control platform to prevent overload or abnormal output of the actuator, such as ±10 V or ±100%. When the original amplitude control output exceeds this range, the ZR axial force closing control platform needs to perform amplitude limiting and integral correction operations to ensure that the force output is safe and controllable.

[0080] The aforementioned preset limiting strategy can refer to the logic rules used to correct the control signal when the output exceeds the limit, including but not limited to: limiting the output to a physically feasible range (U_max or U_min); performing a "reverse integration" operation on the integral term when limiting is triggered to prevent integral saturation; maintaining output continuity and avoiding abrupt changes or reverse oscillations.

[0081] In this embodiment, when the ZR axial force closing control platform detects that the output signal exceeds the limit, it performs reverse correction on the cumulative component. Specifically, when U_raw exceeds the upper limit U_max, the I component is reduced in reverse to offset the excessively accumulated integral. The above-mentioned reverse integration mechanism can prevent hysteresis and oscillation caused by integral saturation.

[0082] Optionally, in the step of generating a corrected control output for force closure control in real time based on the corrected parameters using a force closure control algorithm, the method further includes correcting the internal state parameters in the force closure control algorithm based on the corrected parameters to obtain an updated force closure control algorithm; and based on the updated force closure control algorithm, calculating the response component, cumulative component, and predicted component of the force error signal determined by the target force value and the real-time force value in the next control cycle, and synthesizing the corrected control output.

[0083] In this embodiment of the invention, the ZR-axis force closed-loop control platform dynamically corrects the internal state parameters in the force closed-loop control algorithm based on the correction parameters determined in the previous control cycle. This includes, but is not limited to, the proportional coefficient, integral cumulative value, and differential estimate, which are directly related to the response characteristics and reflect the control bias and dynamic response state of the current algorithm. For example, if an output overshoot trend is detected in the previous cycle, the platform will appropriately reduce the proportional coefficient Kp or increase the differential coefficient Kd according to the correction parameters to enhance the damping characteristics of the system. If a long-term steady-state deviation is detected, the residual error is eliminated by correcting the integral cumulative value I_prev.

[0084] After completing the internal state parameter correction, the ZR axis force closure control platform enters the next control cycle based on the updated force closure control algorithm. In the new cycle, the platform calculates the force error signal e(k) based on the currently sampled real-time force value and the target force value, and calculates the response component (P term), cumulative component (I term), and prediction component (D term) according to the updated algorithm parameters. The response component is used to correct the current deviation in real time, the cumulative component is used to offset long-term errors, and the prediction component is used to predict the error change trend and suppress oscillations in advance. These three types of components are synthesized into a corrected control output U_out through linear superposition, and then output to the actuator after digital-to-analog conversion.

[0085] In practical implementation, the aforementioned ZR axial force closed-loop control platform can also update the algorithm state cache in real time based on the corrected control output to ensure the time consistency and computational continuity between components. When a sudden change in external load or fluctuation in sensor signal is detected, the platform can immediately recalculate the corrected parameters and update the internal state in the next sampling period, thereby achieving dynamic adaptive adjustment and enabling the control algorithm to maintain stable convergence and rapid response under different operating conditions.

[0086] Through the above methods and steps, the algorithm can achieve adaptive adjustment and fast and stable response. Through the closed-loop process of "state correction - algorithm update - real-time calculation - output synthesis", the system can automatically optimize the control gain and integral bias during long-term operation, avoiding the lag, overshoot and saturation problems commonly found in fixed parameter PID control.

[0087] Optionally, in the step of driving the corresponding target device to perform force closure correction control based on the correction control output, the method further includes converting the correction control output through digital-to-analog conversion to generate a control signal for driving the corresponding target device to perform force closure correction control; sending the control signal to the corresponding target device to drive the actuator to move; and continuously adjusting the actuator according to the correction control output when the target device drives the actuator to move, so as to maintain the force closure state corresponding to the target force value.

[0088] In this embodiment of the invention, the aforementioned corrected control output may refer to the final control quantity obtained by the ZR axial force closed-loop control platform after comprehensive response component, cumulative component and predicted component, and after amplitude limiting and reverse integral adjustment processing. It represents the optimal adjustment result calculated by the force closed-loop control algorithm in the current cycle and includes comprehensive correction information on the deviation between the target force value and the real-time force value.

[0089] The aforementioned control signal can refer to the analog drive signal obtained after the correction control output is processed by digital-to-analog conversion (DAC), which is used to directly drive the target device or its actuator. It is the direct interaction medium between the ZR axial force closure control platform and the target device, and its amplitude and waveform reflect the output result of the control algorithm.

[0090] In one possible embodiment, after generating the correction control output, the ZR axial force closing control platform converts the digital control quantity into an analog drive signal through a digital-to-analog converter (DAC). For example, when the target device is an electric gripper in a precision assembly mechanism, the control signal output by the platform will directly control the current of the gripper drive motor, thereby achieving the adjustment of the clamping force.

[0091] During this process, the ZR shaft force closure control platform continuously updates the control signal amplitude based on the correction control output and adjusts the action state of the actuator in real time, so that the actuator maintains the target force value throughout the entire force closure process. When a small fluctuation in the real-time force value is detected (such as ±0.05 N), the platform will immediately recalculate the correction control output and adjust the control signal synchronously to offset the force deviation caused by external disturbances or mechanical relaxation.

[0092] Through the above methods and steps, the force output of the actuator can be stably controlled, thereby ensuring that the ZR axis maintains a constant force closed state during the micro-motion, contact, or holding stages.

[0093] like Figure 2 As shown, this embodiment of the invention also provides a ZR axial force closure control device 200, which includes: The first acquisition module 201 is used to acquire mechanical data from at least one force sensor in real time. The first determining module 202 is used to determine the correction parameters of the force closure control algorithm based on the mechanical data and the corresponding target mechanical data. The first correction module 203 is used to generate a correction control output for force closure control in real time based on the correction parameters and the force closure control algorithm. The first drive module 204 is used to drive the corresponding target device to perform force closure correction control based on the correction control output.

[0094] Optionally, the above-mentioned device further includes: The second determining module is used to determine the corresponding sensing parameters and control parameters based on the current target device; The third determining module is used to perform calibration and zeroing operations on the force sensor corresponding to the current target device based on the sensing parameters and control parameters, so as to determine the initial state of the target device.

[0095] Optionally, the first acquisition module 201 mentioned above includes: The first acquisition submodule is used to read the raw data of the at least one force sensor in real time through an analog-to-digital converter within a preset sampling period. The update submodule is used to write the original data into the sampling data buffer and update the corresponding index; The filtering submodule is used to perform filtering operations based on the data in the sampling data buffer using a preset filtering algorithm, determine a smooth output, and use it as the mechanical data.

[0096] Optionally, the above filtering submodule includes: The first lookup unit is used to find the corresponding storage location in the sampled data buffer based on the updated index; The first arithmetic unit is used to perform accumulation and averaging operations on the sampled data at the storage location according to a preset window length to obtain the current smooth output.

[0097] Optionally, the first determining module 202 mentioned above includes: The first determining submodule is used to determine the real-time force value and the corresponding target force value based on the mechanical data and the corresponding target mechanical data, respectively. The first calculation submodule is used to calculate the corresponding force error signal based on the real-time force value and the target force value. The first synthesis submodule is used to calculate the response component, cumulative component and prediction component based on the current value, historical cumulative amount and rate of change of the force error signal, and synthesize the original amplitude control output. The first acquisition submodule is used to limit the original amplitude control output according to a preset amplitude limiting strategy when the original amplitude control output exceeds the preset amplitude range, and to perform reverse integration adjustment on the cumulative component to obtain the amplitude control output after limiting. The first adjustment submodule is used to use the cumulative component adjusted by reverse integration as the correction parameter of the force closure control algorithm.

[0098] Optionally, the first correction module 203 mentioned above includes: The first correction submodule is used to correct the internal state parameters in the force closure control algorithm based on the correction parameters, so as to obtain an updated force closure control algorithm. The second correction submodule is used to calculate the response component, cumulative component and prediction component of the force error signal determined by the target force value and the real-time force value in the next control cycle based on the updated force closure control algorithm, and synthesize the correction control output.

[0099] Optionally, the first driving module 204 mentioned above includes: The first conversion submodule is used to convert the correction control output through digital-to-analog conversion to generate a control signal for driving the corresponding target device to perform force closure correction control. The first drive submodule is used to send the control signal to the corresponding target device to drive the actuator to move; The maintenance submodule is used to continuously adjust the actuator according to the correction control output when the target device drives the actuator to move, so as to maintain the force closure state corresponding to the target force value.

[0100] like Figure 3 As shown, this embodiment of the invention also provides an electronic device 300, including a processor, which can execute any of the above-described ZR axial force closure control methods.

[0101] Specifically, it includes a processor 301 and a memory 302, as well as a computer program stored in the memory 302 and capable of running on the processor 301 to execute the ZR axial force closure control method, wherein: The processor 301 executes the calculator program for the ZR axial force closed-loop control method stored in the memory 302, and performs the following steps: Real-time acquisition of mechanical data from at least one force sensor; Based on the aforementioned mechanical data and the corresponding target mechanical data, the correction parameters for the force closure control algorithm are determined. Based on the correction parameters, the correction control output for force closure control is generated in real time through the force closure control algorithm; Based on the corrected control output, the corresponding target device is driven to perform force closure correction control.

[0102] Optionally, before the processor 301 performs the real-time acquisition of mechanical data from at least one force sensor, the method further includes: Based on the current target device, determine the corresponding sensing parameters and control parameters; Based on the sensing parameters and control parameters, the force sensor corresponding to the current target device is calibrated and zeroed to determine the initial state of the target device.

[0103] Optionally, the processor 301 performs the real-time acquisition of mechanical data from at least one force sensor, including: Within a preset sampling period, the raw data of the at least one force sensor is read in real time via an analog-to-digital converter; The original data is written into the sampled data buffer, and the corresponding index is updated; Based on the data in the sampling data buffer, a filtering operation is performed using a preset filtering algorithm to determine a smooth output, which is then used as the mechanical data.

[0104] Optionally, the processor 301 executes the filtering operation based on the data in the sampled data buffer, using a preset filtering algorithm, to determine a smooth output, including: The corresponding storage location is located in the sampled data buffer based on the updated index; The sampled data at the storage location are accumulated and averaged according to the preset window length to obtain the current smooth output.

[0105] Optionally, the processor 301 executes the step of determining the correction parameters of the force closure control algorithm based on the mechanical data and the corresponding target mechanical data, including: Based on the aforementioned mechanical data and the corresponding target mechanical data, the real-time force value and the corresponding target force value are determined respectively. Based on the real-time force value and the target force value, the corresponding force error signal is calculated; Based on the current value, historical cumulative amount, and rate of change of the force error signal, the response component, cumulative component, and prediction component are calculated respectively, and the original amplitude control output is synthesized. When the original amplitude control output exceeds the preset amplitude range, the original amplitude control output is limited according to the preset amplitude limiting strategy, and the cumulative component is adjusted by inverse integration to obtain the amplitude control output after limiting. The cumulative component adjusted by inverse integration is used as the correction parameter of the force closure control algorithm.

[0106] Optionally, the processor 301 executes the step of generating a corrected control output for force closure control in real time based on the corrected parameters and through the force closure control algorithm, including: Based on the correction parameters, the internal state parameters in the force closure control algorithm are corrected to obtain the updated force closure control algorithm; Based on the updated force-closed control algorithm, the response component, cumulative component and prediction component are calculated for the force error signal determined by the target force value and the real-time force value in the next control cycle, and then the corrected control output is synthesized.

[0107] Optionally, the processor 301 executes the force closure correction control based on the corrected control output, driving the corresponding target device to perform force closure correction control, including: The correction control output is converted by digital-to-analog conversion to generate a control signal for driving the corresponding target device to perform force closure correction control. The control signal is sent to the corresponding target device to drive the actuator to move; When the target device drives the actuator to move, the actuator is continuously adjusted according to the correction control output so as to maintain the force closure state corresponding to the target force value.

[0108] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the ZR axial force closure control method or the application-side ZR axial force closure control method provided in this invention, and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0109] Those skilled in the art will understand that implementing all or part of the processes in the above embodiments can be done by a computer program instructing related hardware, and can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0110] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A ZR axial force closed-loop control method, characterized in that, include: Real-time acquisition of mechanical data from at least one force sensor; Based on the aforementioned mechanical data and the corresponding target mechanical data, the correction parameters for the force closure control algorithm are determined. Based on the correction parameters, the correction control output for force closure control is generated in real time through the force closure control algorithm; Based on the corrected control output, the corresponding target device is driven to perform force closure correction control.

2. The ZR axial force closed-loop control method as described in claim 1, characterized in that, Before acquiring mechanical data from at least one force sensor in real time, the method further includes: Based on the current target device, determine the corresponding sensing parameters and control parameters; Based on the sensing parameters and control parameters, the force sensor corresponding to the current target device is calibrated and zeroed to determine the initial state of the target device.

3. The ZR axial force closed-loop control method as described in claim 1, characterized in that, The real-time acquisition of mechanical data from at least one force sensor includes: Within a preset sampling period, the raw data of the at least one force sensor is read in real time via an analog-to-digital converter; The original data is written into the sampled data buffer, and the corresponding index is updated; Based on the data in the sampling data buffer, a filtering operation is performed using a preset filtering algorithm to determine a smooth output, which is then used as the mechanical data.

4. The ZR axial force closed-loop control method as described in claim 3, characterized in that, The step of performing filtering operations on the data in the sampling data buffer using a preset filtering algorithm to determine a smooth output includes: The corresponding storage location is located in the sampled data buffer based on the updated index; The sampled data at the storage location are accumulated and averaged according to the preset window length to obtain the current smooth output.

5. The ZR axial force closed-loop control method as described in claim 1, characterized in that, The step of determining the correction parameters for the force closure control algorithm based on the mechanical data and the corresponding target mechanical data includes: Based on the aforementioned mechanical data and the corresponding target mechanical data, the real-time force value and the corresponding target force value are determined respectively. Based on the real-time force value and the target force value, the corresponding force error signal is calculated; Based on the current value, historical cumulative amount, and rate of change of the force error signal, the response component, cumulative component, and prediction component are calculated respectively, and the original amplitude control output is synthesized. When the original amplitude control output exceeds the preset amplitude range, the original amplitude control output is limited according to the preset amplitude limiting strategy, and the cumulative component is adjusted by inverse integration to obtain the amplitude control output after limiting. The cumulative component adjusted by inverse integration is used as the correction parameter of the force closure control algorithm.

6. The ZR axial force closed-loop control method as described in claim 5, characterized in that, The step of generating a corrected control output for force closure control in real time based on the corrected parameters using the force closure control algorithm includes: Based on the correction parameters, the internal state parameters in the force closure control algorithm are corrected to obtain the updated force closure control algorithm; Based on the updated force-closed control algorithm, the response component, cumulative component and prediction component are calculated for the force error signal determined by the target force value and the real-time force value in the next control cycle, and then the corrected control output is synthesized.

7. The ZR axial force closed-loop control method as described in claim 1, characterized in that, The step of driving the corresponding target device to perform force closure correction control based on the corrected control output includes: The correction control output is converted by digital-to-analog conversion to generate a control signal for driving the corresponding target device to perform force closure correction control. The control signal is sent to the corresponding target device to drive the actuator to move; When the target device drives the actuator to move, the actuator is continuously adjusted according to the correction control output so as to maintain the force closure state corresponding to the target force value.

8. A ZR axial force closed-loop control device, characterized in that, include: The first acquisition module is used to acquire mechanical data from at least one force sensor in real time. The first determining module is used to determine the correction parameters of the force closure control algorithm based on the mechanical data and the corresponding target mechanical data. The first correction module is used to generate a correction control output for force closure control in real time based on the correction parameters and the force closure control algorithm. The first driving module is used to drive the corresponding target device to perform force closure correction control based on the correction control output.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the ZR axial force closure control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the ZR axial force closure control method as described in any one of claims 1 to 7.

Images