Tractor control performance online compensation method, system, equipment and medium

Abstract

The invention relates to the field of agricultural machinery intelligent control, and particularly discloses a tractor control performance online compensation method, system and device and a medium, and the method comprises the steps: building a reference performance model of a control mechanism offline; online real-time monitoring is carried out, and a performance attenuation factor is calculated; generating a self-adaptive compensation signal based on the attenuation factor and the working condition decision; and closed-loop updating is carried out on the model or the threshold value by utilizing the compensated feedback data. According to the system, the processes are modularly integrated in a vehicle-mounted controller. By constructing a complete closed loop of'monitoring-decision-compensation-learning ', online autonomous perception and dynamic compensation of progressive performance degradation caused by abrasion and aging of steering and braking mechanisms and the like are realized, the consistency of control performance is effectively maintained, the operation precision and the system safety are improved, and the system reliability is improved. And meanwhile, a data basis is provided for predictive maintenance.

Description

Technical Field

[0001] This invention belongs to the field of intelligent control of agricultural machinery, specifically a method, system, device and medium for online compensation of tractor control performance. Background Technology

[0002] As a core piece of agricultural power machinery, the precision and reliability of tractor handling directly affect work quality and efficiency. With the in-depth development of mechatronics technology, modern tractors are increasingly adopting drive-by-wire technology (such as steer-by-wire and brake-by-wire) to replace traditional mechanical linkage mechanisms. In drive-by-wire systems, the driver's control commands are converted into electrical signals, processed by the controller, and then drive hydraulic or electric actuators to complete the action. While this architecture offers advantages such as flexible layout and the ability to integrate intelligent control, its performance is also more susceptible to changes in the condition of key components.

[0003] The core performance of a control system lies in the stability of the dynamic response relationship between its input and output. For a tractor's steer-by-wire system, the ideal response characteristic is a fast, accurate, and linear following relationship between the commanded steering angle and the actual wheel steering angle. However, in actual use, factors such as decreased volumetric efficiency of the hydraulic pump, wear and jamming of hydraulic valves, oil contamination, minor internal leakage caused by aging of pipeline seals, and sensor drift can all lead to a slow and gradual "attenuation" of the system's dynamic characteristics. This attenuation is mainly manifested as a sluggish response (decreased gain), slower reaction (increased delay), or nonlinearity, collectively referred to as "performance degradation." This degradation occurs gradually and is initially difficult for the driver to perceive, but it subtly reduces the steering's responsiveness and precision, potentially leading to a decline in work quality during delicate operations (such as sowing and ridging).

[0004] Currently, common approaches to addressing this type of performance degradation rely on two main strategies: one is regular manual inspection and maintenance, using specialized equipment to measure and calibrate system parameters at specific intervals; the other is pre-setting fixed fault alarm thresholds in the controller, triggering an alarm when a sensor signal (such as pressure) exceeds a certain range. However, the first approach is a passive and offline maintenance strategy, unable to address performance degradation occurring between maintenance intervals, and is also costly. The second approach can only detect sudden, severe faults (such as a sudden drop in pressure), and is extremely insensitive to the aforementioned slow, gradual performance degradation, failing to provide early warning or compensation before it affects operational quality. More importantly, existing methods generally lack an intrinsic mechanism for online, real-time evaluation of the overall dynamic performance of the system and adaptive adjustments accordingly. This makes it difficult for tractor control systems to maintain their optimal performance at the time of manufacture, significantly limiting their intelligence and self-maintenance capabilities. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method, system, device, and medium for online compensation of tractor control performance, thereby resolving the issue in the prior art of lacking online real-time evaluation and adaptive compensation capabilities for the gradual performance degradation of tractor control systems caused by long-term use.

[0006] A method for online compensation of tractor control performance includes the following steps:

[0007] S1. Offline benchmark establishment phase:

[0008] S1.1 Data Acquisition and Preprocessing: Under normal tractor conditions, sensor data and actuator response data of steering, braking or throttle control mechanisms under various preset commands are acquired;

[0009] S1.2 Benchmark Model Fitting: Based on the preprocessed data, linear or nonlinear regression methods are used to fit the ideal input-output response relationship of each control mechanism, establish a benchmark performance model, and solve for the benchmark performance parameters.

[0010] S2, Online Monitoring and Evaluation Phase:

[0011] S2.1 Real-time data acquisition: During tractor operation, the control command signals of each control mechanism and the actual response data fed back by the sensors are acquired in real time;

[0012] S2.2 Performance degradation factor calculation: Input the real-time data into the benchmark performance model, calculate the current performance parameters, compare them with the corresponding benchmark performance parameters, and calculate the degradation factor that characterizes the degree of performance degradation.

[0013] S3, Decision-making and Compensation Implementation Phase:

[0014] S3.1 Compensation Decision: Set a threshold for the attenuation factor. When the calculated attenuation factor exceeds the threshold, it is determined that the performance of the corresponding control mechanism has degraded, triggering the compensation mechanism and generating preliminary compensation parameters.

[0015] S3.2 Adaptive Compensation: Based on the preliminary compensation parameters and combined with the current operating conditions, the final compensation control signal is dynamically generated through the compensation algorithm, and the original control command is superimposed or corrected to compensate for the performance degradation of the actuator.

[0016] S4, Closed-loop update and recording phase:

[0017] S4.1 Model and parameter update: Adaptively adjust the benchmark performance model or the attenuation factor threshold using the actual response data of the compensated system;

[0018] S4.2 Historical trajectory recording: Continuously record the attenuation factor, compensation parameters and updated models or thresholds of each control mechanism to form a dynamically evolving performance profile.

[0019] Preferably, in step S1.1, the sensor data includes steering angle sensor, wheel speed sensor, and pressure sensor data; the actuator response data includes hydraulic cylinder displacement, brake pad pressure, and engine speed data.

[0020] Preferably, in step S1.2, the benchmark performance model is a linear transfer function model obtained by fitting using the least squares method.

[0021] Preferably, in step S2.2, the performance degradation factor is the ratio of the current response gain to the reference response gain, or the difference between the current response delay and the reference response delay.

[0022] Preferably, in step S3.1, the compensation decision process further includes: comparing and analyzing the current attenuation factor with the attenuation factor trend in the historical record; if it deviates from the long-term trend, it is determined to be a sudden performance degradation or sensor failure, and an early warning is triggered.

[0023] Preferably, in step S3.2, the compensation algorithm is a feedforward compensation algorithm or a PID compensation algorithm; the combination with the current working conditions includes taking into account the tractor load and travel speed.

[0024] A tractor control performance online compensation system for implementing the method described above, the system comprising:

[0025] The data acquisition and preprocessing module is used to execute steps S1.1 and S2.1;

[0026] The baseline modeling module is used to perform step S1.2;

[0027] The performance monitoring and evaluation module is used to perform step S2.2;

[0028] The compensation decision and execution module is used to execute steps S3.1 and S3.2;

[0029] The closed-loop update and data management module is used to execute steps S4.1 and S4.2.

[0030] Preferably, the system is integrated into the on-board controller of the tractor, and the compensation decision and execution module and the closed-loop update and data management module run in the real-time operating system of the controller.

[0031] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the online compensation method for tractor control performance as described above.

[0032] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the online compensation method for tractor control performance as described above.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] By establishing a complete methodology that includes offline benchmark modeling, online performance monitoring, intelligent decision compensation, and closed-loop self-updating, the method achieves autonomous perception and dynamic offsetting of performance degradation in key systems such as tractor steer-by-wire. This method can adaptively compensate for progressive performance degradation caused by mechanical wear or hydraulic leakage, thereby continuously maintaining the intuitiveness and consistency of control response and effectively improving driving experience and operational accuracy.

[0035] Meanwhile, the trend analysis and early warning mechanism embedded in this invention can keenly identify sudden failures, enhancing the safety and reliability of the system. In addition, the closed-loop update mechanism enables the system to automatically evolve benchmarks as the vehicle ages, ensuring the effectiveness of long-term compensation, while detailed performance digital profiles provide core data support for predictive maintenance, significantly improving the intelligent operation and maintenance level of the vehicle. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating the online compensation method for tractor control performance according to the present invention. Detailed Implementation

[0037] 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.

[0038] Example 1: This example describes in detail the online compensation method of the tractor steer-by-wire system of the present invention, such as... Figure 1 As shown:

[0039] Step S1: Offline benchmark establishment stage

[0040] This phase is performed when the tractor is in normal and healthy condition, and aims to establish a standard for subsequent comparisons.

[0041] S1.1 Data Acquisition and Preprocessing:

[0042] Preset command generation: When the tractor is stationary or traveling at low speed in a straight line on a flat road, a series of preset steering angle command signals are sent to the steering controller through a test program or manual operation. These commands may include sine waves with increasing amplitude, step signals, or frequency sweep signals of a specific frequency to stimulate the system's response under different operating conditions.

[0043] Data Synchronous Acquisition: Simultaneously acquire two key data streams:

[0044] Input command signal: the target steering angle (unit: degrees) sent to the steering controller.

[0045] Actuator response data: The actual response acquired through sensors. In this embodiment, it specifically includes:

[0046] Steering angle sensor data: used to measure the real-time actual turning angle of the steering wheel or steering wheel.

[0047] Pressure sensor data: measures the pressure difference between the two chambers of the steering hydraulic cylinder, reflecting the amount of steering assist.

[0048] Wheel speed sensor data: used to help determine vehicle dynamics and ensure test safety.

[0049] Preprocessing: The raw data is denoised (e.g., using a low-pass filter to remove high-frequency electrical noise), aligned (to compensate for the time difference between commands and responses caused by system delays), and normalized to form clean, synchronized command-response data pairs.

[0050] S1.2, Benchmark Model Fitting:

[0051] Model Selection: Considering that steering systems typically exhibit good linear characteristics under normal conditions, this embodiment selects a first- or second-order linear transfer function as the structure of the baseline performance model. For example, it can be modeled as follows: ,in This is the steady-state gain (i.e., the ratio between the commanded angle and the actual rotation angle). It is a time constant. This is pure delay.

[0052] Parameter identification: The preprocessed command-response data pairs are used for system identification using the least squares method; through fitting, a set of optimal baseline performance parameters is obtained, denoted as... , , This set of parameters characterizes the dynamic performance of the steering system under healthy conditions. This model and its parameters are permanently stored in the non-volatile memory of the vehicle controller.

[0053] Step S2, Online Monitoring and Evaluation Phase

[0054] This phase continues during the tractor's daily operations.

[0055] S2.1 Real-time data acquisition:

[0056] When the tractor is plowing, harvesting or other operations, the system captures the control command signal (target steering angle) issued by the driver through the steering wheel in real time, and simultaneously obtains the actual response data (actual steering angle, hydraulic pressure, etc.) through the sensor network. The sampling frequency is usually 10-100Hz.

[0057] S2.2 Calculation of performance degradation factor:

[0058] Current parameter estimation: Input the "command-response" data acquired in real time within a short time window (e.g., the first 5 seconds) into a model with the same structure as in S1.2 for online parameter identification (using recursive least squares method) to estimate the current performance parameters in real time. , .

[0059] Attenuation factor calculation: Compare the current parameters with the baseline parameters. Calculate the performance attenuation factor. :

[0060] For gain attenuation: ;when When this occurs, it indicates a decrease in steering sensitivity.

[0061] Regarding response delay: ;when When this occurs, it indicates that the steering response is slower.

[0062] In this embodiment, the main monitoring factor is the gain attenuation factor. .

[0063] Step S3: Decision-making and Compensation Implementation Phase

[0064] S3.1 Compensation Decision:

[0065] Threshold determination: Preset an attenuation threshold, for example When calculated If the value remains below 0.85 for an extended period (e.g., 3 calculation cycles), the system determines that the steering mechanism has experienced performance degradation (e.g., decreased hydraulic pump efficiency, slight internal leakage in the oil circuit).

[0066] Preliminary compensation parameter generation: Generate a preliminary feedforward compensation coefficient. This is used to theoretically compensate for the attenuated gain back to the reference level.

[0067] Early warning function: The system maintains a A historical queue (e.g., the most recent 1000 values). Calculate its moving average and standard deviation. If the current... If the value drops sharply (e.g., below the mean minus 3 standard deviations), significantly deviating from its long-term trend of slow decline, it is highly likely that the sensor has suddenly failed or the hydraulic line has burst. The system will immediately trigger the highest level of audible and visual warning and advise the driver to stop work immediately for inspection, rather than just performing compensation.

[0068] S3.2 Adaptive Compensation:

[0069] Operating condition correction: The system reads the current tractor load (estimated via the lifter pressure sensor) and travel speed (from the vehicle speed sensor). Under heavy loads or at low speeds, steering resistance increases, and the required compensation may differ; the system incorporates a two-dimensional lookup table to adjust the load and speed accordingly. Fine-tuning is performed to obtain the compensation coefficient after working condition correction. .

[0070] Compensation Execution: A feedforward compensation algorithm is employed. When the driver issues a target steering angle command... At that time, the compensation module multiplies it by a compensation coefficient to generate the corrected instruction. Then it is sent to the steering actuator; in this way, even if the system gain has decayed to The final actual steering angle output This restored the original handling feel.

[0071] Step S4, Closed-loop update and recording phase

[0072] S4.1 Model and Parameter Updates:

[0073] After the compensation takes effect, the system continues to collect new "command-response" data. Due to the introduction of compensation, the "equivalent" open-loop characteristics of the system are now close to the baseline. Using this new data, the true and more accurate current performance parameters of the system in the uncompensated state can be deduced in reverse.

[0074] Using recursive least squares or Kalman filtering algorithms, these newly estimated parameters are used to update the baseline performance model parameters online and slowly (e.g., Updated to ,in (with a forgetting factor close to 1), which enables the baseline model to adapt to the "new normal" performance drift of tractors caused by long-term, slow aging, and allows the baseline for compensation decisions to be dynamically adjusted accordingly, making it more intelligent.

[0075] S4.2 Historical Track Recording:

[0076] All key parameters, including those at every moment. , Updated All of these are timestamped and recorded in the performance profile of the vehicle's onboard storage; this profile can be used for:

[0077] Generate performance degradation trend reports to provide data support for preventative maintenance.

[0078] During maintenance, provide technicians with a detailed history of the fault's evolution.

[0079] As a foundation for further big data analysis, optimize thresholds and compensation strategies.

[0080] Example 2: This example describes in detail the online compensation system for tractor control performance:

[0081] The system of the present invention is integrated in the form of a software module into the on-board domain controller of the tractor or a dedicated intelligent control module.

[0082] Data acquisition and preprocessing module: Implemented by the underlying driver and signal processing thread, it is responsible for communicating with various sensors (angle, pressure, speed sensors) and actuator controllers to complete data acquisition, filtering and time alignment.

[0083] Benchmark modeling module: Runs in system initialization or maintenance mode, calls a mathematical library (such as the embedded ARMCMSIS-DSP library) to execute the least squares fitting algorithm and generate benchmark model parameters.

[0084] Performance monitoring and evaluation module: As a high-priority background task, it continuously runs online parameter identification and attenuation factor calculation.

[0085] Compensation Decision and Execution Module: Running as a high-priority control task within the Real-Time Operating System (RTOS), it ensures the real-time nature of compensation decisions and signal corrections (delay typically required to be less than 10ms). Based on the monitoring module's output and current operating conditions, it calculates and applies compensation in real time.

[0086] Closed-loop update and data management module: This module runs as a low-priority background task, responsible for the slow updating of the model and the recording and management of data. Its running cycle can be set to once every few minutes to avoid interfering with real-time control tasks.

[0087] All modules interact with each other through shared memory or message queues, working together to complete a closed loop from data to decision-making to learning.

[0088] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for online compensation of tractor control performance, characterized in that, Includes the following steps: S1. Offline benchmark establishment phase: S1.1 Data Acquisition and Preprocessing: Under normal tractor conditions, sensor data and actuator response data of steering, braking or throttle control mechanisms under various preset commands are acquired; S1.2 Benchmark Model Fitting: Based on the preprocessed data, linear or nonlinear regression methods are used to fit the ideal input-output response relationship of each control mechanism, establish a benchmark performance model, and solve for the benchmark performance parameters. S2, Online Monitoring and Evaluation Phase: S2.1 Real-time data acquisition: During tractor operation, the control command signals of each control mechanism and the actual response data fed back by the sensors are acquired in real time; S2.2 Performance degradation factor calculation: Input the real-time data into the benchmark performance model, calculate the current performance parameters, compare them with the corresponding benchmark performance parameters, and calculate the degradation factor that characterizes the degree of performance degradation. S3, Decision-making and Compensation Implementation Phase: S3.1 Compensation Decision: Set a threshold for the attenuation factor. When the calculated attenuation factor exceeds the threshold, it is determined that the performance of the corresponding control mechanism has degraded, triggering the compensation mechanism and generating preliminary compensation parameters. S3.2 Adaptive Compensation: Based on the preliminary compensation parameters and combined with the current operating conditions, the final compensation control signal is dynamically generated through the compensation algorithm, and the original control command is superimposed or corrected to compensate for the performance degradation of the actuator. S4, Closed-loop update and recording phase: S4.1 Model and parameter update: Adaptively adjust the benchmark performance model or the attenuation factor threshold using the actual response data of the compensated system; S4.2 Historical trajectory recording: Continuously record the attenuation factor, compensation parameters and updated models or thresholds of each control mechanism to form a dynamically evolving performance profile.

2. The online compensation method for tractor control performance according to claim 1, characterized in that, In step S1.1, the sensor data includes steering angle sensor, wheel speed sensor, and pressure sensor data; the actuator response data includes hydraulic cylinder displacement, brake pad pressure, and engine speed data.

3. The online compensation method for tractor control performance according to claim 1, characterized in that, In step S1.2, the benchmark performance model is a linear transfer function model obtained by fitting using the least squares method.

4. The online compensation method for tractor control performance according to claim 1, characterized in that, In step S2.2, the performance degradation factor is the ratio of the current response gain to the reference response gain, or the difference between the current response delay and the reference response delay.

5. The online compensation method for tractor control performance according to claim 1, characterized in that, In step S3.1, the compensation decision process further includes: comparing and analyzing the current attenuation factor with the attenuation factor trend in the historical records; if it deviates from the long-term trend, it is determined to be a sudden performance degradation or sensor failure, and an early warning is triggered.

6. The online compensation method for tractor control performance according to claim 1, characterized in that, In step S3.2, the compensation algorithm is either feedforward compensation or PID compensation algorithm; the combination with the current working conditions includes taking into account the tractor load and travel speed.

7. An online compensation system for tractor control performance, characterized in that, The system for implementing the method as described in any one of claims 1-6 comprises: A data acquisition and preprocessing module is used to perform steps S1.1 and S2.1 as described in claim 1; A baseline modeling module is used to perform step S1.2 as described in claim 1; A performance monitoring and evaluation module is used to perform step S2.2 as described in claim 1; The compensation decision and execution module is used to execute steps S3.1 and S3.2 as described in claim 1; The closed-loop update and data management module is used to perform steps S4.1 and S4.2 as described in claim 1.

8. The tractor control performance online compensation system according to claim 7, characterized in that, The system is integrated into the tractor's onboard controller, and the compensation decision and execution module and the closed-loop update and data management module run in the controller's real-time operating system.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the online compensation method for tractor control performance as described in any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the online compensation method for tractor control performance as described in any one of claims 1 to 6.

Images