Torque temperature drift compensation method and device and computer readable storage medium

Abstract

The invention discloses a torque temperature drift compensation method and device and a computer readable storage medium. The method comprises the steps that sensor data of a to-be-detected position is acquired, the to-be-detected position is used for generating torque in the operation process, and the sensor data is used for representing the operation state and the external force condition of the to-be-detected position; determining whether the to-be-detected position is in a zero moment state based on the sensor data, the zero moment state being used for indicating that the to-be-detected position is not subjected to external force; if the to-be-detected position is not in the zero moment state, predicting the predicted state information of the to-be-detected position at the current moment according to the state information of the to-be-detected position at the previous moment; and determining a target structural torque of the to-be-detected position based on the structural torque at the current moment, the predicted structural torque and the predicted temperature drift deviation. The technical problem of low torque measurement precision caused by poor compensation effect of a temperature drift compensation method of a torque sensor in the prior art is solved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and more specifically, to a method, apparatus, and computer-readable storage medium for torque temperature drift compensation. Background Technology

[0002] Torque sensors are widely used in industrial automation, precision measurement, and other fields, and their accuracy directly affects the performance and reliability of the system. However, torque sensors often face the challenge of temperature drift during use, meaning that even without external force, the output reading will deviate when the temperature changes, due to the temperature sensitivity of the materials and manufacturing processes. Temperature drift not only reduces measurement accuracy but also increases the difficulty of system debugging and maintenance.

[0003] Currently, temperature drift compensation methods for torque sensors mainly include table lookup correction based on static models and software filtering strategies. The former relies on a preset temperature-torque offset relationship table, while the latter utilizes digital filtering techniques for signal processing. However, these methods are limited by the limitations of static models and insufficient dynamic adaptability, especially in environments with rapid temperature changes or high temperatures, where the compensation effect is poor, leading to a decrease in torque measurement accuracy.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method, apparatus, and computer-readable storage medium for torque temperature drift compensation, in order to at least solve the technical problem in the related art where the temperature drift compensation method for torque sensors has poor compensation effect, resulting in low torque measurement accuracy.

[0006] According to one aspect of the embodiments of this application, a method for temperature drift compensation of torque is provided, comprising: acquiring sensor data of a location to be detected, wherein the location to be detected is used to generate torque during operation, and the sensor data is used to characterize the operating state and external force conditions of the location to be detected; determining whether the location to be detected is in a zero-torque state based on the sensor data, wherein a zero-torque state indicates that the location to be detected is not subjected to external force; if the location to be detected is not in a zero-torque state, predicting the predicted state information of the location to be detected at the current moment based on the state information of the location to be detected at the previous moment, wherein the predicted state information includes a predicted structural torque and a predicted temperature drift deviation; and determining the target structural torque of the location to be detected based on the structural torque at the current moment, the predicted structural torque, and the predicted temperature drift deviation.

[0007] Furthermore, a torque sensor and a temperature sensor are installed at the location to be detected. The method also includes: if the location to be detected is in a zero torque state, obtaining structural torque based on the torque sensor and obtaining temperature data based on the temperature sensor; storing the structural torque and temperature data in a preset storage area, and updating the temperature drift deviation of the filter based on the value of the structural torque, wherein the filter is used to filter the sensor data.

[0008] Furthermore, the location to be detected is equipped with an encoder, a current sensor, and a torque sensor. The sensor data includes angular acceleration, driving torque, and structural torque. The sensor data for the location to be detected includes: calculating the angular acceleration based on the angular position collected by the encoder; calculating the driving torque based on the current data collected by the current sensor; and obtaining the structural torque based on the torque sensor.

[0009] Furthermore, determining whether the position to be detected is in a zero-torque state based on sensor data includes: in response to the sensor data satisfying the zero-torque condition, starting a timer and recording the duration, wherein the zero-torque condition includes: angular acceleration less than a first threshold, absolute value of driving torque less than a second threshold, and absolute value of structural torque less than a third threshold, and the duration is used to represent the duration for which the sensor data continuously satisfies the zero-torque condition; if the duration is greater than a preset duration, it is determined that the position to be detected is in a zero-torque state; if the duration is less than or equal to the preset duration, it is determined that the position to be detected is not in a zero-torque state.

[0010] Furthermore, based on the state information of the position to be detected at the previous moment, the predicted state information of the position to be detected at the current moment includes: obtaining the predicted structural moment and the predicted temperature drift deviation of the position to be detected at the previous moment; and predicting the predicted structural moment and the predicted temperature drift deviation based on the predicted structural moment and the predicted temperature drift deviation of the previous moment.

[0011] Furthermore, the method also includes: obtaining the previous time covariance matrix, wherein the previous time covariance matrix is ​​used to reflect the uncertainty of the predicted value at the previous time; and predicting the current time covariance matrix based on the previous time covariance matrix and the process noise covariance matrix.

[0012] Furthermore, the method also includes: adjusting the gain of the filter based on the predicted covariance matrix in response to the temperature at the detection location being higher than a preset temperature.

[0013] Furthermore, determining the target structural moment at the detection location based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation includes: calculating innovation based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation, where the innovation is used to represent the difference between the measured value and the predicted value; and adjusting the current structural moment based on the innovation and the adjusted gain to obtain the target structural moment.

[0014] According to another aspect of the embodiments of this application, a torque temperature drift compensation device is also provided, comprising: an acquisition module, configured to acquire sensor data of a location to be detected, wherein the location to be detected is used to generate torque during operation, and the sensor data is used to characterize the operating state and external force conditions of the location to be detected; a first determination module, configured to determine whether the location to be detected is in a zero torque state based on the sensor data, wherein a zero torque state indicates that the location to be detected is not subjected to external force; a prediction module, configured to, if the location to be detected is not in a zero torque state, predict the predicted state information of the location to be detected at the current moment based on the state information of the location to be detected at the previous moment, wherein the predicted state information includes a predicted structural torque and a predicted temperature drift deviation; and a second determination module, configured to determine the target structural torque of the location to be detected based on the structural torque at the current moment, the predicted structural torque, and the predicted temperature drift deviation.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0017] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the methods in various embodiments of this application.

[0018] In this embodiment, the torque temperature drift compensation method of this application forms a closed-loop compensation mechanism by acquiring sensor data in real time, accurately identifying and calibrating the zero torque state, intelligently predicting state information, and finally determining a more accurate target torque. Specifically, by monitoring temperature changes in real time, this application provides timely data input for subsequent temperature drift compensation. By treating temperature drift deviation and true torque as independent state variables, it effectively decouples the two, avoiding interference from measurement errors caused by temperature drift in the measurement of true torque. Simultaneously, the mechanism for predicting structural torque and temperature drift deviation can adaptively adjust with temperature changes, ensuring the stability of torque measurement even during rapid temperature fluctuations. Finally, by fusing real-time measured values ​​with predicted values, the system can immediately correct measurement errors, especially compensating for temperature-induced deviations. Therefore, this application not only effectively overcomes the limitations of traditional solutions under dynamic temperature changes, achieving real-time temperature drift compensation under all operating conditions, but also significantly improves the measurement accuracy of the torque sensor and the stability of the system. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a schematic flowchart of a torque temperature drift compensation method according to an embodiment of this application;

[0021] Figure 2 This is a flowchart illustrating a method for compensating for temperature drift of a torque sensor at a robot joint according to an embodiment of this application.

[0022] Figure 3 This is a structural block diagram of a torque temperature drift compensation device according to one embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] Torque sensors are widely used in industrial automation, precision measurement, and other fields, and their accuracy directly affects the performance and reliability of the system. Taking torque sensors used in robot joints as an example, robot joint torque sensors have inherent temperature drift problems due to the characteristics of their materials and structures.

[0026] Current mainstream temperature drift compensation techniques suffer from core flaws: limitations of static models and insufficient dynamic adaptability, making it difficult to meet the high-precision control requirements of robot joints. Traditional lookup table methods or polynomial fitting rely on preset temperature drift curves, which cannot adapt to individual sensor differences and temperature drift characteristics changes caused by aging, and do not consider nonlinear responses during rapid temperature changes. Traditional software filtering schemes produce hysteresis when temperatures fluctuate rapidly, and existing methods treat temperature drift as independent noise, failing to establish a dynamic coupling model between it and the actual torque, thus amplifying the temperature drift compensation error.

[0027] According to an embodiment of this application, a method embodiment for torque temperature drift compensation is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0028] This application provides a method. The torque temperature drift compensation method can be used to provide torque temperature drift compensation functionality for preset application scenarios. The preset application scenarios may include: torque temperature drift compensation scenarios for intelligent robots (such as cleaning robots, service robots, delivery robots, etc.), torque temperature drift compensation scenarios for drones, and scenarios in the vehicle field that require torque temperature drift compensation, which are not limited here.

[0029] Figure 1 This is a flowchart illustrating a torque temperature drift compensation method according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes the following steps:

[0030] Step S11: Obtain sensor data of the position to be detected, wherein the position to be detected is used to generate torque during operation, and the sensor data is used to characterize the operating state and external force of the position to be detected.

[0031] In this embodiment of the application, the location to be detected is used to generate torque during operation. The location to be detected refers to a specific point or component in the device that needs to monitor the change in torque. The device can be any mechanical or electronic component that involves torque measurement and is affected by temperature, such as a robotic arm joint, an engine shaft, or the torque point of a precision measuring instrument. There are no restrictions here.

[0032] Sensor data is used to characterize the operating status and external forces acting on the location to be detected, and serves as the data basis for subsequent torque and temperature drift compensation. For example, sensor data includes, but is not limited to, temperature sensor data, torque sensor data, and angle sensor data.

[0033] As can be seen, this application acquires real-time temperature, torque, and related sensor data by integrating multiple sensors at the detection location. This sensor data is the basis for subsequent evaluation of temperature drift effect, determination of zero torque state, and temperature drift compensation.

[0034] Therefore, by acquiring real-time sensor data, the temperature and torque status of the location to be detected can be grasped instantly, providing accurate data support for subsequent temperature drift compensation.

[0035] Step S12: Determine whether the position to be detected is in a zero-torque state based on sensor data, where the zero-torque state indicates that the position to be detected is not subjected to external force.

[0036] In this embodiment, the zero-torque state refers to a state where the detected position is not subjected to any external force or is subjected to only a negligible external force at the current moment; that is, the actual torque at the detected position is zero or extremely close to zero. In the zero-torque state, the theoretical torque reading should be zero or close to zero. This zero-torque state is the ideal time to perform sensor temperature drift calibration and compensation.

[0037] As can be seen, this application determines whether the location to be detected is in a zero-torque state by using the aforementioned sensor data. Therefore, by using multiple sensors in a coordinated manner, the accuracy of zero-torque state identification is improved, avoiding potential misjudgments caused by relying on a single indicator.

[0038] Step S13: If the position to be detected is not in a zero-torque state, predict the predicted state information of the position to be detected at the current moment based on the state information of the position to be detected at the previous moment. The predicted state information includes the predicted structural torque and the predicted temperature drift deviation.

[0039] In this embodiment of the application, if the position to be detected is not in a zero torque state, it means that the position to be detected is being subjected to an external force and there is a non-zero structural torque. At this time, the temperature drift phenomenon may affect the torque measurement, that is, affect the measurement accuracy.

[0040] The state information of the previous moment refers to the state information obtained from the most recent measurement of the position to be detected, denoted as For example, the status information may include parameters such as the actual torque value, temperature information, and temperature drift deviation.

[0041] Predicted state information includes predicted structural moments and predicted temperature drift deviations. This means that based on the state information from the previous moment, the estimated structural moments and temperature drift deviations for the current moment are used to guide real-time moment measurement and temperature drift compensation. This is denoted as... .

[0042] Furthermore, predicting the structural moment means predicting the structural moment (actual moment) that the detected location will experience at the current moment or the next upcoming moment, denoted as . This prediction takes into account the continuity and dynamic characteristics of the torque, which helps to more accurately track the changing trend of the actual torque.

[0043] Predicting temperature drift deviation means predicting the temperature drift deviation caused by temperature fluctuations, denoted as By observing the dynamic characteristics of temperature drift deviation, the influence of temperature changes on torque measurement can be reflected in real time, especially the exponential growth law of temperature drift deviation under high temperature conditions.

[0044] As can be seen, in this application, if it is determined that the location to be detected is not in a zero-torque state, the prediction stage begins. Using the state information from the previous moment, combined with dynamic characteristics (dynamic model), predictions are made to forecast the current structural torque and temperature drift deviation. Therefore, this prediction method based on a dual-state (structural torque and temperature drift deviation) not only considers the dynamic changes in torque but also accurately estimates the immediate impact of temperature drift deviation on the torque reading, thus providing a reference basis for subsequent temperature drift compensation.

[0045] In other words, this application proposes a dynamic prediction mechanism that uses structural moment and temperature drift deviation as dual-state variables to construct a prediction model that can dynamically adapt to actual moment changes and temperature fluctuations. This dynamic prediction mechanism allows the compensation algorithm to move beyond a static model and adjust according to actual temperature and moment changes. Therefore, it can flexibly respond to real-time changes in temperature and moment, ensuring the real-time performance and accuracy of the compensation algorithm. Especially in situations with rapid temperature fluctuations, it avoids compensation lag and error amplification caused by a static model.

[0046] Therefore, through prediction, this application can anticipate the changing trends of structural moment and temperature drift deviation in advance, preparing for subsequent real-time compensation and thus enhancing the timeliness and accuracy of compensation. Furthermore, this application dynamically adjusts the temperature drift compensation strategy according to actual conditions, especially in high-temperature environments, prioritizing accurate estimation of temperature drift moment to prevent amplification of measurement errors. In addition, the introduction of dual-state variables in this application separates moment error and temperature drift deviation, avoiding mutual interference between the two and contributing to improved overall accuracy of moment measurement.

[0047] Step S14: Determine the target structural moment at the location to be detected based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation.

[0048] In this embodiment, the structural torque at the current moment is the torque value directly read from the torque sensor, which is the original torque value before temperature compensation, denoted as... The structural moment at the current moment will include the actual moment and the temperature drift caused by temperature.

[0049] The target structural moment is the structural moment obtained after temperature drift deviation correction, which is closer to the actual structural moment.

[0050] As can be seen, this application, under non-zero torque conditions, can correct the current structural torque reading by comparing it with the predicted structural torque and incorporating the predicted temperature drift deviation, thereby obtaining a target structural torque that is closer to the actual value. For example, this process can be implemented through observation updates using a Kalman filter algorithm to ensure high-precision torque measurement results under various temperature conditions.

[0051] Therefore, by applying the predicted temperature drift deviation to the current structural torque reading, measurement deviations caused by temperature can be corrected in real time, improving the accuracy of the torque reading. Through dynamic temperature drift compensation of the current structural torque reading, torque measurement with extremely low delay and extremely low residual error is achieved under all operating conditions, significantly improving the performance and application range of the torque sensor in complex temperature environments.

[0052] In summary, the torque temperature drift compensation method of this application forms a closed-loop compensation mechanism by acquiring sensor data in real time, accurately identifying and calibrating the zero torque state, intelligently predicting state information, and ultimately determining a more accurate target torque. Specifically, by monitoring temperature changes in real time, this application provides timely data input for subsequent temperature drift compensation. By treating temperature drift deviation and true torque as independent state variables, it effectively decouples the two, avoiding interference from measurement errors caused by temperature drift in the measurement of true torque. Simultaneously, the mechanism for predicting structural torque and temperature drift deviation can adaptively adjust with temperature changes, ensuring the stability of torque measurement even during rapid temperature fluctuations. Finally, by fusing real-time measured values ​​with predicted values, the system can immediately correct measurement errors, especially compensating for temperature-induced deviations. Therefore, this application not only effectively overcomes the limitations of traditional solutions under dynamic temperature changes, achieving real-time temperature drift compensation under all operating conditions, but also significantly improves the measurement accuracy of the torque sensor and the stability of the system.

[0053] In this embodiment, sensor data of the location to be detected is acquired. The location to be detected generates torque during operation, and the sensor data characterizes the operating state and external forces acting on the location. Based on the sensor data, it is determined whether the location to be detected is in a zero-torque state, where zero torque indicates that the location is not subjected to external forces. If the location to be detected is not in a zero-torque state, the predicted state information of the location at the current moment is predicted based on the state information of the location at the previous moment. This predicted state information includes predicted structural torque and predicted temperature drift deviation. Based on the current structural torque, predicted structural torque, and predicted temperature drift deviation, the target structural torque of the location to be detected is determined. This achieves accurate compensation for temperature drift deviation, thereby improving the accuracy of torque measurement and solving the technical problem in related technologies where the temperature drift compensation method for torque sensors has poor compensation effect, resulting in low torque measurement accuracy.

[0054] In an optional embodiment, a torque sensor and a temperature sensor are equipped at the location to be detected, and the method further includes the following steps:

[0055] Step S151: If the position to be detected is in a zero torque state, obtain the structural torque based on the torque sensor and obtain the temperature data based on the temperature sensor.

[0056] Step S152: Store the structural moment and temperature data in a preset storage area, and update the temperature drift deviation of the filter based on the value of the structural moment. The filter is used to filter the sensor data.

[0057] In this embodiment, a torque sensor and a temperature sensor are installed at the location to be detected. The torque sensor measures the torque applied at the location, and its reading is affected by temperature, resulting in temperature drift. The temperature sensor measures the temperature inside the joint and around the torque sensor, providing temperature data for analysis and compensation of temperature drift deviation.

[0058] If the location to be detected is in a zero-torque state, the structural torque is obtained based on a torque sensor, and the temperature data is obtained based on a temperature sensor. Since the location to be detected is in a zero-torque state, the applied external torque at that location is close to zero, meaning that theoretically there is no influence of external torque on the sensor readings.

[0059] It can be seen that if the location to be tested is in a zero-torque state, it indicates that no external force is applied at the location, making it an ideal time to perform temperature drift calibration. Collecting the torque sensor readings (i.e., structural torque) and temperature sensor data (i.e., temperature data) in the zero-torque state reveals that the obtained structural torque reading actually reflects the deviation caused by temperature changes, i.e., temperature drift, rather than the actual external torque value. Therefore, in the zero-torque state, the torque sensor readings almost entirely reflect the temperature drift deviation, providing data reference for subsequent calibration.

[0060] The structural moment and temperature data are then stored in a preset access area, and the temperature drift deviation of the filter is updated based on the structural moment value. The filter is used to filter the sensor data; here, the filter can be a dual-state variable-parameter Kalman filter, used for real-time processing and correction of the moment sensor readings to reduce the impact of temperature drift deviation.

[0061] As can be seen, this application stores the acquired structural moment and temperature data, and uses the structural moment to update the state estimate of temperature drift deviation in the filter. That is, the filter of this application uses the stored structural moment as the new observation value. The observation value only reflects the temperature drift deviation and has no real moment component. Therefore, the observation value can be directly used to update the temperature drift deviation.

[0062] Therefore, through direct observation under zero torque conditions, the temperature drift bias is estimated and updated more accurately, further improving the compensation effect. Furthermore, this method enables periodic updates of the temperature drift bias under zero torque conditions, thus helping to reduce the accumulation of errors caused by temperature drift during long-term operation and maintaining the long-term stability of torque measurements. Even under conditions of significant ambient temperature variations, continuous updates to the temperature drift bias allow the system to better adapt to temperature changes, enhancing its robustness.

[0063] In other words, if this application determines that the location to be detected is in a zero-torque state, it stores the current torque sensor reading and temperature sensor reading, and forcibly resets the temperature drift estimate to make it equal to the torque sensor value. This not only improves the real-time accuracy of torque measurement, but also reduces the accumulation of errors during long-term operation, enhancing the overall stability and robustness of the system.

[0064] In one optional embodiment, an encoder, a current sensor, and a torque sensor are provided at the location to be detected. The sensor data includes angular acceleration, driving torque, and structural torque. In step S11, the sensor data at the location to be detected is acquired, including the following method steps:

[0065] Step S111: Angular acceleration is calculated based on the angular position acquired by the encoder.

[0066] Step S112: Calculate the driving torque based on the current data collected by the current sensor.

[0067] Step S113: Obtain structural torque based on torque sensor.

[0068] In this embodiment, an encoder, a current sensor, and a torque sensor are installed at the location to be detected. The sensor data includes angular acceleration, driving torque, and structural torque. The encoder measures the angular position and angular velocity at the location to be detected. Angular acceleration is a physical quantity describing the rate of change of rotational speed at the location to be detected, and is a parameter used to determine whether the location is stationary or in a low-speed state (i.e., a preliminary condition for a zero-torque state).

[0069] Current sensors are used to measure the magnitude of the current passing through a driving device. The current data reflects the load condition of the driving device. For example, if the driving device is a motor, the current data reflects the motor's load condition. Driving torque is the torque output by the driving device to rotate the detected position. It is directly related to the driving device's current and is another parameter for determining the zero-torque state.

[0070] Torque sensors are used to directly measure the torque applied at the location to be detected. The readings of torque sensors are prone to deviation under temperature changes, a phenomenon known as temperature drift.

[0071] As can be seen, when acquiring sensor data at the location to be detected, angular acceleration is calculated based on the angular position collected by the encoder. For example, angular acceleration is calculated by reading the angular position information acquired by the encoder and performing differentiation. This process typically involves data processing algorithms, such as digital differentiation or sliding window averaging, to ensure the accuracy of the calculation and reduce the impact of noise. Therefore, the calculation of angular acceleration helps the system identify the motion state of the robot joints and determine whether they are close to a stationary or low-speed motion state.

[0072] Simultaneously, the driving torque is calculated based on the current data collected by the current sensor. For example, the driving torque is calculated based on the current data collected by the current sensor, combined with the motor's characteristic parameters (such as torque constant). This process may also involve data processing algorithms, such as filtering or correction algorithms, to eliminate external interference and sensor noise. Therefore, the calculation of the driving torque helps determine whether there is an external torque acting at the detection location.

[0073] Simultaneously, structural torque is acquired based on torque sensors. For example, structural torque is read directly from the torque sensor; this is the raw data for torque measurement, containing the actual torque signal and possible temperature drift deviations. Therefore, the direct reading of structural torque provides the raw data input for torque measurement, serving as the data foundation for subsequent temperature drift compensation and torque calculation.

[0074] In summary, this application, by simultaneously monitoring angular acceleration, driving torque, and structural torque, can comprehensively understand the motion state and torque status of the detected position from multiple dimensions, thereby improving the system's accuracy in identifying zero-torque states.

[0075] In an optional embodiment, in step S12, determining whether the position to be detected is in a zero-torque state based on sensor data includes the following method steps:

[0076] Step S121: In response to the sensor data satisfying the zero torque condition, start a timer and record the duration. The zero torque condition includes: angular acceleration less than a first threshold, absolute value of driving torque less than a second threshold, and absolute value of structural torque less than a third threshold. The duration is used to represent the duration for which the sensor data continuously satisfies the zero torque condition.

[0077] Step S122: If the duration is longer than the preset duration, determine that the position to be detected is in a zero torque state.

[0078] Step S123: If the duration is less than or equal to the preset duration, it is determined that the position to be detected is not in a zero torque state.

[0079] In this embodiment, when determining whether the location to be detected is in a zero-torque state based on sensor data, if the sensor data meets the zero-torque condition, a timer is started and the duration is recorded. The zero-torque condition is a set of preset standards used to determine whether the location to be detected is in a state with almost no external torque, a state suitable for calibrating temperature drift deviation.

[0080] The zero torque condition includes: angular acceleration less than a first threshold, that is, angular acceleration less than a threshold value for angular acceleration. For example, the first threshold value can be 0.5 rad / s². 2The absolute value of the driving torque is less than a second threshold, meaning the absolute value of the driving torque is less than a threshold value for the absolute value of the driving torque. For example, the second threshold value can be 0.05 × the rated torque of the motor. The absolute value of the structural torque is less than a third threshold, meaning the absolute value of the structural torque is less than a threshold value for the absolute value of the structural torque. For example, the third threshold value can be 2 × the maximum deviation value provided in the manual.

[0081] Duration is used to indicate the duration for which sensor data continuously satisfy the zero torque condition.

[0082] It can be seen that if the absolute values ​​of angular acceleration, driving torque, and structural torque are lower than their respective first, second, and third thresholds, a timer is started to record the duration of the sensor data in this state. This duration is used to confirm that the sensor data does not accidentally meet the zero-torque condition, but has indeed lasted for a period of time, to ensure the validity and reliability of the zero-torque state.

[0083] If the duration exceeds a preset duration, the detected position is determined to be in a zero-torque state. If the duration is less than or equal to the preset duration, the detected position is determined not to be in a zero-torque state. The preset duration is a pre-set time threshold used to determine whether the sensor data continuously satisfies the zero-torque condition for a sufficiently long time to confirm the zero-torque state. For example, the preset duration can be 500ms.

[0084] It can be seen that if the duration exceeds the preset duration, meaning the sensor data continuously meets the zero-torque condition for more than a set threshold, then the detected position is determined to have entered the zero-torque state, ensuring that the system's judgment is based on continuous stability. If the duration does not exceed the preset duration, meaning the sensor data fails to continuously meet the zero-torque condition for the set time, then the detected position is determined not to have entered the zero-torque state, indicating that the detected position is still subject to a certain external torque, and routine torque measurement and temperature drift compensation operations need to continue.

[0085] In an optional embodiment, in step S13, the predicted state information of the position to be detected at the current time is predicted based on the state information of the position to be detected at the previous time, including the following method steps:

[0086] Step S131: Obtain the predicted structural moment and the predicted temperature drift deviation of the previous moment at the location to be detected.

[0087] Step S132: Based on the predicted structural moment and the predicted temperature drift deviation at the previous moment, the predicted structural moment and the predicted temperature drift deviation are predicted.

[0088] In this embodiment of the application, when predicting the predicted state information of the position to be detected at the current time based on the state information of the position to be detected at the previous time, the predicted structural moment and the predicted temperature drift deviation of the position to be detected at the previous time can be obtained. Then, based on the predicted structural moment and the predicted temperature drift deviation of the previous time, the predicted structural moment and the predicted temperature drift deviation are predicted.

[0089] The predicted structural moment at the previous moment is the predicted value of the structural moment (i.e., the moment before the influence of the actual moment and the temperature drift deviation) at the previous moment. The predicted temperature drift deviation at the previous moment is the prediction of the possible temperature drift deviation of the moment sensor at the previous moment.

[0090] As can be seen, this application first extracts the predicted structural moment and the predicted temperature drift deviation from the previous time step's execution results. Because the Kalman filter algorithm uses a recursive approach, it utilizes the best estimate from the previous time step to predict the current state information, thereby reducing prediction errors and improving system stability. Then, using the predicted structural moment and the predicted temperature drift deviation from the previous time step, combined with the current temperature data, it predicts the predicted structural moment and the predicted temperature drift deviation for the current time step. This process also incorporates the state equation, which typically considers the impact of process noise to reflect the uncertainties in the actual system.

[0091] Therefore, this application ensures that the algorithm can continuously iterate based on historical data, forming a complete state estimation chain, improving prediction accuracy. Furthermore, the prediction information from the previous moment includes the latest state estimate of the Kalman filter, helping the system make more reasonable predictions at the current moment and reducing the impact of sudden changes. Through prediction, the system can estimate the torque and temperature drift deviation at the current moment in advance, preparing for subsequent measurement updates, reducing response time. Moreover, the prediction process considers the impact of temperature changes, enabling the system to dynamically adjust the readings of temperature-sensitive sensors, improving the effectiveness of temperature drift compensation.

[0092] In an optional embodiment, the method further includes the following method steps:

[0093] Step S161: Obtain the covariance matrix of the previous time step, where the covariance matrix of the previous time step is used to reflect the uncertainty of the predicted value at the previous time step.

[0094] Step S162: Based on the covariance matrix of the previous time step and the process noise covariance matrix, the prediction covariance matrix of the current time step is predicted.

[0095] In this embodiment, the covariance matrix of the previous time step is also obtained. The covariance matrix of the previous time step reflects the uncertainty of the predicted value at the previous time step and is denoted as... In the Kalman filter algorithm, the previous time step covariance matrix represents the error distribution between the predicted and actual values ​​of the state variables at the previous time step, reflecting the uncertainty level of the system state prediction. For the dual-state variable parameter Kalman filter of this application, the previous time step covariance matrix is ​​typically a 2×2 matrix, used to represent the prediction uncertainty of the two state variables: the actual torque and the temperature drift deviation.

[0096] As can be seen, this application extracts the previous time covariance matrix from the previous time operation data. The previous time covariance matrix contains the uncertainty measure of the predicted values ​​of the two state variables, the true torque and the temperature drift deviation, from the previous time.

[0097] Then, based on the covariance matrix of the previous time step and the process noise covariance matrix, the prediction covariance matrix of the current time step is predicted. In the Kalman filter, the process noise covariance matrix describes the uncertainty of the system's inherent stochastic process, i.e., a statistical description of how the system state changes without observation, denoted as […]. In this application, the process noise covariance matrix reflects the uncertainty of torque and temperature drift deviation changes and is adaptively adjusted with temperature changes.

[0098] As can be seen, based on the covariance matrix of the previous time step and the process noise covariance matrix of the current time step, this application predicts the prediction covariance matrix of the current time step, which is denoted as... .Right now, The prediction process involves matrix multiplication.

[0099] Therefore, by predicting the covariance matrix, this application can quantify how the uncertainty of the predicted value changes over time, especially how the uncertainty increases under the influence of external factors such as temperature changes. Furthermore, as the temperature changes, the value of the predicted covariance matrix adjusts accordingly, thereby ensuring that the predicted covariance matrix can reflect the uncertainty in the actual environment and improving the algorithm's dynamic adaptability and anti-interference capability.

[0100] Therefore, the generation of the prediction covariance matrix in this application ensures that the algorithm can more accurately predict the state variables (true torque and temperature drift deviation) at the current moment, reducing prediction errors caused by temperature changes. By accurately assessing the uncertainty of the predicted values, the Kalman filter can optimally combine the predicted and measured data, minimizing the error in state estimation, thereby improving the accuracy and stability of torque measurement.

[0101] In an optional embodiment, the method further includes the following method steps:

[0102] Step S171: In response to the temperature at the location to be detected being higher than a preset temperature, the gain of the filter is adjusted based on the predicted covariance matrix.

[0103] In this embodiment, if the temperature at the detection location is higher than a preset temperature, the gain of the filter is adjusted based on the predicted covariance matrix. The preset temperature is a pre-defined temperature threshold. When the actual temperature exceeds this preset temperature, the influence of temperature drift is considered to have increased significantly, requiring measures to improve the compensation effect.

[0104] In the Kalman filter algorithm, the filter gain is a parameter used to weigh the contribution of predicted values ​​to the state update compared to actual measurements, denoted as […]. The greater the gain, the greater the impact of the actual measurement on the state update.

[0105] As can be seen, when the temperature at the detection location is higher than the preset temperature, the system adjusts the filter gain based on the prediction covariance matrix. Under high-temperature conditions, the uncertainty of temperature drift bias typically increases, thus the relevant elements in the prediction covariance matrix (reflecting the variance of the temperature drift bias prediction error) also increase. This application increases the weight of the actual measured value in the updated state by adjusting the filter gain, thereby enabling faster response and correction of temperature drift bias.

[0106] Therefore, when the temperature at the detection location is higher than the preset temperature, by increasing the gain, the system can update the state more quickly using the actual measured value, effectively reducing the impact of temperature drift and improving the accuracy of torque measurement. This temperature-based gain adjustment mechanism allows the filtering algorithm to dynamically optimize its parameters according to changes in ambient temperature, enhancing its adaptability to temperature drift.

[0107] In an optional embodiment, in step S14, the target structural moment at the location to be detected is determined based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation, including the following method steps:

[0108] Step S141: Calculate the innovation based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation, where the innovation is used to represent the difference between the measured value and the predicted value.

[0109] Step S142: Adjust the structural moment at the current moment based on the innovation and the adjusted gain to obtain the target structural moment.

[0110] In this embodiment, when determining the target structural moment at the detection location based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation, innovation is calculated based on these three parameters. Innovation represents the difference between the measured value and the predicted value. In a Kalman filter, innovation represents the difference between the sensor's measured value at the current moment and the value predicted based on the state at the previous moment. For example, innovation = .

[0111] As can be seen, this application calculates innovation by comparing the structural torque directly read from the torque sensor at the current moment with the predicted structural torque and predicted temperature drift deviation. The calculation of innovation can reflect the deviation between the sensor reading and the system prediction in real time, which helps to dynamically correct the error in the prediction and provides a quantitative basis for the adjustment of subsequent algorithms.

[0112] Then, based on the innovation and the adjusted gain, the structural moment at the current moment is adjusted to obtain the target structural moment. In this application, the gain is dynamically adjusted according to the temperature to optimize compensation for temperature drift.

[0113] As can be seen, this application combines the calculated innovation with the adjusted gain to adjust the structural moment at the current moment, thereby obtaining a target structural moment that is closer to the true moment. Thus, by adjusting the gain, the predicted structural moment can be appropriately corrected according to the magnitude of the innovation, reducing errors, especially those caused by temperature drift. Furthermore, the dynamic adjustment of the gain ensures that temperature drift deviation is prioritized under different temperature conditions, improving the accuracy and reliability of moment measurement.

[0114] Figure 2 This is a flowchart illustrating a method for compensating for temperature drift of a torque sensor at a robot joint according to an embodiment of this application. Figure 2 As shown, this application first acquires sensor data at each robot joint. Specifically, it reads joint encoder data and calculates angular acceleration. Read current sensor data and calculate motor torque. And read torque sensor data to calculate joint torque. For example, the weights of the angular acceleration, motor torque, and joint torque mentioned above are 40%, 40%, and 20%, respectively. If the zero torque condition is determined to be met based on the data from each sensor, then the zero torque calibration mode is entered. That is, if the following conditions are met simultaneously: <0.5rad / s 2 , Absolute value < 0.05 × motor rated torque If the absolute value is less than 2 × the maximum deviation value provided in the manual, and the duration is greater than 500 ms, then the current state is determined to be zero torque. In zero torque calibration mode, the current torque sensor reading and temperature sensor reading are stored, the temperature drift estimate is forcibly reset to be equal to the torque sensor value, and the true torque is returned to zero.

[0115] If the zero-torque condition is not met based on the data from each sensor, the system enters a two-state variable-parameter Kalman filter mode. The implementation steps of this mode are as follows:

[0116] Step 1: State Variables definition.

[0117] .in, k is a positive number. Temperature drift deviation at time k (the deviation in sensor readings caused by temperature).

[0118] Step 2: Process noise covariance matrix definition.

[0119] .in, This indicates the uncertainty in the prediction of the actual torque (e.g., errors caused by sudden load changes). This indicates the uncertainty in temperature drift prediction, which varies with temperature. Exponential growth. This represents the reference noise. This represents the temperature sensitivity coefficient. The zero off-diagonal term is used to assume that the torque error is independent of the temperature error.

[0120] Step 3: Define the measurement model.

[0121] Among them, sensor readings =True torque + Temperature deviation + Random noise. H is the measurement matrix, representing how sensor readings are linearly combined from state variables. For measuring noise. The variance R is determined by the sensor accuracy.

[0122] Step 4: Initialization (k=0).

[0123] .in, It is the initial state estimate (usually set to zero or a known calibration value). It is the initial covariance matrix, and large values ​​on the diagonal indicate high uncertainty.

[0124] Step 5: State prediction step.

[0125] .in, This is the predicted state vector at the current time (k). This is the predicted state vector at the previous time step (k-1). Let be the prediction covariance matrix at the current time (k). Let be the prediction covariance matrix of the previous time step (k-1). Let be the process noise covariance matrix at the current time (k).

[0126] It can be seen that the predicted state Assume the state remains unchanged over a short period of time (system dynamics matrix A = identity matrix I). Predict covariance. Superposition process noise This reflects the increased uncertainty in forecasts. At high temperatures... Increase, then If the temperature drift is increased, the prediction of temperature drift becomes more uncertain.

[0127] Step Six: Kalman Gain calculate.

[0128] Among them, the molecule: This indicates how the state prediction error is propagated to the measurement space. Denominator: This represents the total uncertainty in the predicted measurement (including sensor noise). That is, the gain. The weighting of error correction is determined if High (high temperature) Increase the value of the temperature drift estimate first.

[0129] Furthermore, Let be the prediction covariance matrix of the Kalman filter at time (k). This is the transpose of the measurement matrix H. R represents the measurement noise covariance matrix.

[0130] Step 7: Status Update.

[0131] Where, new information = This refers to the difference between the measured value and the predicted value. The update rule is that if the innovation is positive (measured value > predicted value), the Kalman gain for temperature drift bias is adjusted. If the temperature drift is considered insufficient, the temperature drift bias will be increased. When the Kalman gain is relative to the true torque If the actual torque is underestimated, then the actual torque is increased. .

[0132] Step 8: Output the actual torque .

[0133] In other words, traditional torque sensor temperature drift compensation methods rely on offline calibration, which cannot respond to temperature changes and individual differences in real time, and ignores the coupling between temperature drift torque and true torque, leading to a decrease in the force control accuracy of robot joints. This application proposes a torque temperature drift compensation algorithm combining "dual-state Kalman filtering + temperature adaptive parameter tuning + zero torque online calibration" to achieve real-time temperature drift compensation under all working conditions. First, the zero torque state is determined by a joint encoder angular acceleration, motor current, force sensor, and duration criteria. If the state is zero torque, the algorithm enters the zero torque calibration mode and updates the temperature drift torque covariance. Otherwise, it enters the dual-state parameter Kalman filter mode, ultimately outputting the true torque of the joint.

[0134] Therefore, it can be seen that the advantages of this application compared with the traditional solution are as follows:

[0135] Beneficial Effect 1: A dual-state dynamic coupling compensation model is proposed, which treats the temperature drift torque as a state variable with equal weight to the real torque and observes it in real time through a Kalman filter, effectively solving the dynamic limitation of treating temperature drift as independent noise in traditional schemes.

[0136] Beneficial Effect 2: A temperature adaptive mechanism is proposed, which is based on the temperature drift covariance exponential adjustment law. The process noise covariance is dynamically scaled according to the actual temperature. At high temperatures, the Kalman gain is forcibly increased to realize the dynamic improvement of the priority of temperature drift estimation, which effectively solves the limitations of traditional schemes with single and fixed parameters.

[0137] Beneficial Effect 3: A zero-torque self-calibration closed-loop mode is proposed, which is based on multi-source criteria (angular acceleration, motor current, duration) to autonomously trigger calibration, effectively solving the problems of cumulative error and high maintenance cost of traditional solutions.

[0138] In summary, this application proposes a two-dimensional variable-parameter Kalman filter algorithm, including: 1) a dynamically coupled model establishment, creating two-dimensional state variables of temperature drift torque and true torque to reduce errors caused by model inaccuracies; 2) a covariance temperature adaptive mechanism, dynamically adjusting the covariance according to temperature to improve the accuracy of temperature drift compensation; and 3) a zero-torque self-calibration closed-loop algorithm, automatically identifying and calibrating the zero-torque state to reduce accumulated errors and maintenance costs. The proposed scheme can achieve adaptive temperature drift compensation with extremely low latency and extremely low residual error.

[0139] Furthermore, in this application, an H∞ robust filter can be used instead of a Kalman filter, and the filter is designed according to the worst-case scenario so that the error between the filtered result and the expected result is within a certain range. However, the H∞ robust filter has a high computational complexity and is prone to overcompensation when used for sensor temperature drift compensation.

[0140] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0141] According to an embodiment of this application, an apparatus embodiment for a torque temperature drift compensation method is provided. It should be noted that the apparatus can be used to perform the above-mentioned torque temperature drift compensation.

[0142] Figure 3 This is a structural block diagram of a torque temperature drift compensation device according to one embodiment of this application, as shown below. Figure 3 As shown, the torque temperature drift compensation device 300 includes: an acquisition module 301, used to acquire sensor data of the position to be detected, wherein the position to be detected is used to generate torque during operation, and the sensor data is used to characterize the operating state and external force conditions of the position to be detected; a first determination module 302, used to determine whether the position to be detected is in a zero torque state based on the sensor data, wherein a zero torque state indicates that the position to be detected is not subjected to external force; a prediction module 303, used to predict the predicted state information of the position to be detected at the current moment based on the state information of the position to be detected at the previous moment if the position to be detected is not in a zero torque state, wherein the predicted state information includes the predicted structural torque and the predicted temperature drift deviation; and a second determination module 304, used to determine the target structural torque of the position to be detected based on the structural torque at the current moment, the predicted structural torque, and the predicted temperature drift deviation.

[0143] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0144] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0145] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0146] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the methods in various embodiments of this application.

[0147] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0149] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0150] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0151] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0152] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for compensating for torque temperature drift, characterized in that, include: Acquire sensor data at a location to be detected, wherein the location to be detected is used to generate torque during operation, and the sensor data is used to characterize the operating state and external forces acting on the location to be detected. Based on the sensor data, it is determined whether the position to be detected is in a zero-torque state, wherein the zero-torque state is used to indicate that the position to be detected is not subjected to external force; If the position to be detected is not in the zero torque state, the predicted state information of the position to be detected at the current moment is predicted based on the state information of the position to be detected at the previous moment, wherein the predicted state information includes the predicted structural torque and the predicted temperature drift deviation. The target structural moment at the location to be detected is determined based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation.

2. The method according to claim 1, characterized in that, The location to be detected is equipped with a torque sensor and a temperature sensor, and the method further includes: If the location to be detected is in the zero torque state, the structural torque is obtained based on the torque sensor, and the temperature data is obtained based on the temperature sensor; The structural moment and the temperature data are stored in a preset access area, and the temperature drift deviation of the filter is updated based on the value of the structural moment, wherein the filter is used to filter the sensor data.

3. The method according to claim 1, characterized in that, The location to be detected is equipped with an encoder, a current sensor, and a torque sensor. The sensor data includes angular acceleration, driving torque, and structural torque. Acquiring the sensor data at the location to be detected includes: The angular acceleration is calculated based on the angular position acquired by the encoder; The driving torque is calculated based on the current data collected by the current sensor. The structural torque is obtained based on the torque sensor.

4. The method according to claim 3, characterized in that, Determining whether the position to be detected is in a zero-torque state based on the sensor data includes: In response to the sensor data satisfying the zero torque condition, a timer is started and the duration is recorded. The zero torque condition includes: the angular acceleration is less than a first threshold, the absolute value of the driving torque is less than a second threshold, and the absolute value of the structural torque is less than a third threshold. The duration is used to represent the duration for which the sensor data continuously satisfies the zero torque condition. If the duration is longer than the preset duration, the position to be detected is determined to be in the zero torque state; If the duration is less than or equal to the preset duration, it is determined that the position to be detected is not in the zero torque state.

5. The method according to claim 1, characterized in that, The step of predicting the predicted state information of the target location at the current time based on the state information of the target location at the previous time includes: The predicted structural moment and predicted temperature drift deviation of the detected position at the previous time are obtained. Based on the predicted structural moment and the predicted temperature drift deviation at the previous moment, the predicted structural moment and the predicted temperature drift deviation are predicted.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the covariance matrix of the previous time step, wherein the covariance matrix of the previous time step is used to reflect the uncertainty of the predicted value at the previous time step; Based on the covariance matrix of the previous time step and the process noise covariance matrix, the prediction covariance matrix of the current time step is predicted.

7. The method according to claim 6, characterized in that, The method further includes: In response to the temperature at the location to be detected being higher than a preset temperature, the gain of the filter is adjusted based on the predicted covariance matrix.

8. The method according to claim 7, characterized in that, The determination of the target structural moment at the location to be detected based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation includes: Information is calculated based on the current structural moment, the predicted structural moment, and the predicted temperature drift deviation, wherein the information is used to represent the difference between the measured value and the predicted value; The structural moment at the current moment is adjusted based on the new information and the adjusted gain to obtain the target structural moment.

9. A torque temperature drift compensation device, characterized in that, include: The acquisition module is used to acquire sensor data of the location to be detected, wherein the location to be detected is used to generate torque during operation, and the sensor data is used to characterize the operating state and external force conditions of the location to be detected. The first determining module is used to determine whether the position to be detected is in a zero torque state based on the sensor data, wherein the zero torque state is used to indicate that the position to be detected is not subjected to external force; The prediction module is used to predict the predicted state information of the position to be detected at the current moment based on the state information of the position to be detected at the previous moment if the position to be detected is not in the zero torque state. The predicted state information includes the predicted structural torque and the predicted temperature drift deviation. The second determining module is used to determine the target structural moment at the location to be detected based on the structural moment at the current moment, the predicted structural moment, and the predicted temperature drift deviation.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the torque temperature drift compensation method as described in any one of claims 1 to 8 when run on a computer or processor.

11. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the torque temperature drift compensation method as described in any one of claims 1 to 8.

Images