Adaptive adjustment method and device of vehicle torque, vehicle terminal and storage medium

Abstract

The application provides a vehicle torque adaptive adjustment method and device, a vehicle terminal and a storage medium, and relates to the field of intelligent driving. The method comprises the following steps: determining an initial working quality through a vehicle state flag bit, predicting a load change trend according to a historical working quality, and configuring calculation parameters according to different load change trends, so that the mass estimation value can quickly and stably converge to the real working quality in different scenes, the vehicle can output accurate torque matched with different working conditions, the problem of abrupt acceleration or power deficiency caused by fixed calibration is solved, and driving smoothness and driving safety are considered.

Description

Technical Field

[0001] This application relates to the field of intelligent driving, and in particular to a method, device, vehicle terminal, and storage medium for adaptive adjustment of vehicle torque. Background Technology

[0002] Pure electric light commercial vehicles (including light trucks and mini-trucks) are widely equipped with L2 and above level intelligent assisted driving systems. These vehicles mainly operate in scenarios such as short-distance urban delivery, frequent start-stop, and multi-point loading and unloading, where the vehicle load changes extremely frequently, and the driving torque required for acceleration varies significantly under different loads.

[0003] Currently, existing intelligent driving systems generally use a fixed mass as the initial mass for acceleration torque calibration, and use an online mass estimation scheme to correct the initial mass. However, since the load change characteristics of loading and unloading scenarios are different, but the relevant technologies use uniform calculation parameters, there is a problem of insufficient accuracy in the process of converging to the actual working mass, resulting in inaccurate acceleration torque requests. Summary of the Invention

[0004] Therefore, it is necessary to provide an adaptive adjustment method, device, vehicle terminal, and storage medium for vehicle torque to address the aforementioned technical problems and improve the accuracy of acceleration request torque.

[0005] In a first aspect, embodiments of this application provide an adaptive adjustment method for vehicle torque, comprising: determining the initial working mass of the vehicle based on a vehicle status flag, wherein the vehicle status flag indicates whether the vehicle mass has been reset; determining the load change trend of the vehicle based on historical working masses, and configuring calculation parameters for the working mass based on the load change trend; the load change trend includes a decreasing trend, an increasing trend, and a stable trend; calculating the working mass based on the initial working mass, the vehicle longitudinal dynamics model, and the calculation parameters when the vehicle is in an effective driving condition; and calculating the acceleration request torque based on the working mass, wherein the acceleration request torque is adjusted in the same direction as the working mass changes.

[0006] The above method determines the initial working mass by using vehicle status flags and predicts load change trends based on historical working mass. Calculation parameters are configured for different load change trends, enabling the estimated mass to quickly and stably converge to the actual working mass in different scenarios. This ensures that the vehicle can output precise torque that matches the actual working mass under different operating conditions, solving the problem of abrupt acceleration or insufficient power caused by fixed calibration, and balancing driving smoothness and driving safety.

[0007] In some embodiments, determining the initial working quality of a vehicle based on a vehicle status flag includes: if the vehicle status is any one of cargo storage / retrieval status, first parking status, or power-down restart status, then triggering a quality reset operation and marking the status flag with a first status value; if the vehicle status is a second parking status or driving status, then not triggering a quality reset operation and marking the status flag with a second status value; if the status flag is the first status value, then using the half-load calibration quality as the initial working quality of the vehicle; if the status flag is the second status value, then using the historical working quality as the initial working quality.

[0008] By using the above method, the mass reset is triggered by joint determination of multi-dimensional vehicle status, accurately identifying load changes caused by loading and unloading, and using the half-load calibrated mass as the initial value for reset, so that the convergence path at both the empty and full load ends is the shortest, and the convergence time is greatly shortened.

[0009] In some embodiments, obtaining vehicle status includes: obtaining vehicle signals, which include at least one of: gear position signal, parking gear duration, power status signal, and cargo door open / close signal; if the vehicle is in parking gear and the parking gear duration is greater than or equal to a first preset duration threshold, the vehicle status is determined to be a first parking state; if the vehicle is in parking gear, the parking gear duration is less than or equal to a second preset duration threshold, and the cargo door open / close signal changes from closed to open, the vehicle status is determined to be a cargo storage / retrieval state; the first preset duration threshold is greater than the second preset duration threshold; if the power status changes from power-off to power-on, the vehicle status is determined to be a power-off restart state; if the vehicle is in parking gear, the parking gear duration is less than the first preset duration threshold, and the cargo door open / close signal is a closed signal, the vehicle status is determined to be a second parking state; if the vehicle is not in parking gear, the vehicle status is determined to be a driving state.

[0010] By combining the above methods and integrating multi-dimensional information such as parking gear duration, cargo box door status, and power status, loading and unloading behaviors and parking scenarios can be accurately identified.

[0011] In some embodiments, determining the load change trend based on historical work quality includes: if the historical work quality is within the unloaded quality range, then the load change trend is determined to be an upward trend, i.e., a loading scenario; if the historical work quality is within the fully loaded quality range, then the load change trend is determined to be a downward trend, i.e., a unloading scenario; if the historical work quality is within the intermediate quality range, then the load change trend is determined to be a stable trend, i.e., a stable load scenario.

[0012] Using the above method, the current load change trend can be automatically identified based on historical work quality, thereby determining loading, unloading, and load stability scenarios according to different load change trends, and then calculating the vehicle work quality under different scenarios.

[0013] In some embodiments, if the calculation parameters include the forgetting factor of the recursive least squares method and the convergence determination period, the calculation parameters of the working quality are configured according to the load change trend, including: if the load is determined to be a stable scenario, a default forgetting factor and a default convergence determination period are configured; if the load is determined to be an unloading scenario, a first forgetting factor and a first convergence determination period are configured, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period; if the load is determined to be a loading scenario, a second forgetting factor and a second convergence determination period are configured, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

[0014] Using the above method, differentiated forgetting factors and convergence periods are configured for three scenarios: loading, unloading, and load stabilization. During unloading, a smaller forgetting factor and shorter convergence period are used to ensure the estimated value quickly follows the quality decline. During loading, a larger forgetting factor and longer convergence period are used to suppress estimation fluctuations caused by stepped loading. During load stabilization, default parameters are used to maintain steady-state accuracy. All three scenarios can adapt to their own varying convergence characteristics, ensuring that quality estimation balances convergence speed and stability under different operating conditions.

[0015] In some embodiments, obtaining the working quality includes: collecting vehicle operating data under effective driving conditions, the operating data including at least: vehicle speed, driving torque, and longitudinal acceleration; iteratively updating the initial working quality based on the vehicle longitudinal dynamics model, operating data, and calculation parameters of the corresponding scenario to obtain an estimated quality; if the state flag is a first state value, determining the estimated quality obtained in the current iteration as the working quality after smoothing; if the state flag is a second state value, calculating the acceleration request torque based on the historical working quality and keeping it unchanged, while continuously iteratively updating the estimated quality; comparing the continuously iteratively estimated quality with the historical working quality, and if the quality update conditions are met, determining the estimated quality as the new working quality after smoothing.

[0016] Using the above method, in the first state value, the estimated mass is directly determined as the working mass after smoothing, realizing a rapid response of torque after load change; in the second state value, it can reduce the risk of torque output change caused by data jitter, while retaining the system's adaptive ability to small load changes, thus taking into account both stability and adaptability.

[0017] In some embodiments, the quality update conditions include: the deviation between the estimated quality and the working quality is greater than a preset deviation threshold, and remains there for a preset number of consecutive effective driving cycle periods, and the vehicle is not in a braking or rapid acceleration / deceleration state during this period.

[0018] Using the above method, switching is only performed when the deviation between the estimated quality and the working quality is real, continuous, and reliable, thereby reducing the interference of occasional jumps caused by sensor noise or transient fluctuations on the working quality.

[0019] Secondly, embodiments of this application provide an adaptive adjustment device for vehicle torque, comprising: a mass reset module, used to determine the initial working mass of the vehicle based on a vehicle status flag, the vehicle status flag indicating whether the vehicle mass has been reset; a load trend module, used to determine the load change trend of the vehicle based on historical working masses, and configure calculation parameters for the working mass based on the load change trend; the load change trend includes a decreasing trend, an increasing trend, and a stable trend; a mass calculation module, used to calculate the working mass based on the initial working mass, the vehicle longitudinal dynamics model, and calculation parameters when the vehicle is in an effective driving condition; and a torque request module, used to calculate the acceleration request torque based on the working mass, the acceleration request torque being adjusted in the same direction as the working mass changes.

[0020] In some embodiments, the quality reset module is used to determine the initial working quality of the vehicle based on the vehicle status flag. If the vehicle status is any one of cargo storage / retrieval status, first parking status, or power-down restart status, a quality reset operation is triggered, and the status flag is marked with a first status value. If the vehicle status is a second parking status or a driving status, a quality reset operation is not triggered, and the status flag is marked with a second status value. If the status flag is the first status value, the half-load calibration quality is used as the initial working quality of the vehicle. If the status flag is the second status value, the historical working quality is used as the initial working quality.

[0021] In some embodiments, the quality reset module is used to acquire vehicle signals, which include at least one of: a gear position signal, a parking gear duration, a power status signal, and a cargo door opening / closing signal; if the vehicle is in the parking gear and the parking gear duration is greater than or equal to a first preset duration threshold, the vehicle state is determined to be a first parking state; if the vehicle is in the parking gear, the parking gear duration is less than or equal to a second preset duration threshold, and the cargo door opening / closing signal changes from closed to open, the vehicle state is determined to be a cargo storage / retrieval state; the first preset duration threshold is greater than the second preset duration threshold; if the power status changes from power-off to power-on, the vehicle state is determined to be a power-off restart state; if the vehicle is in the parking gear, the parking gear duration is less than the first preset duration threshold, and the cargo door opening / closing signal is a closed signal, the vehicle state is determined to be a second parking state; if the vehicle is not in the parking gear, the vehicle state is determined to be a driving state.

[0022] In some embodiments, the load trend module is used to determine the load change trend as an upward trend, i.e., loading scenario, if the historical working quality is within the unloaded quality range; to determine the load change trend as a downward trend, i.e. unloading scenario, if the historical working quality is within the full-loaded quality range; and to determine the load change trend as a stable trend, i.e., load stable scenario, if the historical working quality is within the intermediate quality range.

[0023] In some embodiments, the load trend module is used to configure a default forgetting factor and a default convergence determination period if the load is determined to be a stable load scenario; to configure a first forgetting factor and a first convergence determination period if the load is determined to be an unloading scenario, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period; and to configure a second forgetting factor and a second convergence determination period if the load is determined to be a loading scenario, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

[0024] In some embodiments, the mass calculation module is used to collect vehicle operating data under effective driving conditions, the operating data including at least: vehicle speed, driving torque and longitudinal acceleration; iteratively update the initial working mass based on the vehicle longitudinal dynamics model, operating data and calculation parameters of the corresponding scenario to obtain an estimated mass; if the state flag is a first state value, the estimated mass obtained by the current iteration is smoothed and determined as the working mass; if the state flag is a second state value, the acceleration request torque is calculated based on the historical working mass and kept unchanged, while the estimated mass is continuously iteratively updated; the continuously iterative estimated mass is compared with the historical working mass, and if the mass update condition is met, the estimated mass is smoothed and determined as the new working mass.

[0025] In some embodiments, the quality calculation module is further configured to determine the estimated quality as the new working quality after smoothing if the deviation between the estimated quality and the working quality is greater than a preset deviation threshold and remains so for a consecutive preset number of effective driving condition cycles, and the vehicle is not in a braking or rapid acceleration / deceleration state during this period.

[0026] Thirdly, embodiments of this application provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the first aspect and any possible implementation method.

[0027] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of the first aspect and any possible implementation.

[0028] The beneficial effects of this application are as follows: the initial working mass is determined by the vehicle status flag, and the load change trend is predicted based on the historical working mass. Calculation parameters are configured for different load change trends, so that the mass estimate can quickly and stably converge to the actual working mass in different scenarios. This ensures that the vehicle can output accurate torque that matches the actual working mass under different working conditions, solves the problem of abrupt acceleration or insufficient power caused by fixed calibration, and balances driving smoothness and driving safety. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0030] Figure 1 A schematic flowchart of an adaptive adjustment method for vehicle torque provided in an embodiment of this application; Figure 2 A schematic diagram of an adaptive adjustment device for vehicle torque provided in an embodiment of this application; Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0032] The terms "first" and "second" in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the term "comprising" and any variations thereof are intended to cover non-exclusive protection. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. The term "multiple" in this application can mean at least two, for example, two, three, or more, and the embodiments of this application do not impose limitations.

[0033] To facilitate understanding of the technical solutions provided in the embodiments of this application, the design concept of the embodiments of this application will be introduced first below: Pure electric light commercial vehicles (such as light trucks and mini trucks) are widely equipped with L2 and above level intelligent assisted driving systems. These vehicles mainly operate in scenarios such as short-distance urban delivery, frequent start-stop, and multi-point loading and unloading, where the vehicle load changes extremely frequently. Taking a typical light truck as an example, its unloaded weight is about 3.5T, and its fully loaded weight is about 4.5T, a weight range of 1T. The driving torque required for acceleration varies significantly under different loads.

[0034] Currently, the acceleration torque calibration of existing intelligent driving systems generally uses a fixed mass as the sole benchmark (usually calibrated with full load mass).

[0035] However, the core operating scenarios for pure electric light commercial vehicles are short-distance urban delivery, frequent start-stop operations, and multi-point loading and unloading. Therefore, the load changes frequently and drastically: a single delivery route may involve multiple loading and unloading operations, with the vehicle weight repeatedly switching between 3.5T and 4.5T, and each switch occurring within a very short period of time. However, the actual load of the vehicle is dynamically changing, while the calculated parameters in the torque calculation remain fixed. Consequently, the torque under acceleration requests will not match the actual demand. For example, when calibrated with a full load, the torque output of the vehicle is too high when unloaded, resulting in abrupt and jerky acceleration, which seriously affects driving comfort; when calibrated with an unloaded load, the torque output is insufficient when fully loaded or on a slope, resulting in insufficient vehicle power and potential driving safety hazards.

[0036] To address the issue of adapting to operating conditions caused by fixed calibration, some solutions introduce online mass estimation to dynamically correct the mass parameters in torque calculation.

[0037] However, existing estimation schemes still have the following shortcomings: Firstly, the load change characteristics of loading and unloading scenarios are fundamentally different. During unloading, the mass continuously decreases, while during loading, the mass increases discontinuously in a step-like manner. However, existing schemes use a uniform forgetting factor and convergence criteria, which cannot simultaneously meet the convergence requirements of both scenarios. In the unloading scenario, a large forgetting factor in the uniform parameters causes the algorithm to excessively retain old data, resulting in slow convergence of the estimated value, long-term deviation from the true mass, and a persistently large torque output. In the loading scenario, a small forgetting factor causes the algorithm to be overly sensitive to each loading, with the estimated value fluctuating drastically and repeatedly jumping with each batch of goods loaded, making it impossible to stably converge to the true mass. Secondly, once the estimated value is updated, it is directly used for torque control, lacking verification and filtering of sensor noise and transient operating condition fluctuations. Jumps in the estimated value directly cause unexpected fluctuations in torque output, further reducing convergence accuracy and ride consistency. The above factors together result in the existing solution having slow convergence speed, large steady-state error, and weak anti-interference ability in actual operation scenarios. It cannot provide reliable quality input for intelligent driving acceleration torque control, and the accuracy of acceleration request torque is difficult to guarantee.

[0038] The adaptive torque adjustment method for vehicles provided in this application can be applied to pure electric light commercial vehicles equipped with intelligent driver assistance systems, as well as other electric vehicles operating under conditions of frequent load changes. This method is entirely based on the existing signals of the vehicle's CAN bus and requires no additional hardware sensors.

[0039] The adaptive adjustment method for vehicle torque proposed in this application will be described in detail below: In some embodiments, such as Figure 1 As shown, an adaptive adjustment method for vehicle torque is provided. Taking the application of this method to an in-vehicle terminal (such as an intelligent driving domain controller) as an example, the method includes the following steps: Step S101: Determine the initial working quality of the vehicle based on the vehicle status flag bit, which indicates whether the vehicle quality has been reset.

[0040] To clearly distinguish system states in the software implementation, this application introduces an internal mode flag, a status flag. If any reset condition is triggered, the status flag is set to the first state value: reset_occurred = TRUE. This enters a fast follow mode to prioritize convergence speed, allowing the mass estimate to quickly approach the actual load. If no reset condition is triggered, the status flag is set to the second state value: reset_occurred = FALSE. This enters a cautious update mode to prioritize the smoothness of the output torque and avoid mass value fluctuations.

[0041] The initial working mass is the starting value calculated by the vehicle mass estimation algorithm, and this value will serve as the benchmark for subsequent iterative updates using the recursive least squares method. The specific value of the initial working mass varies depending on the state flag. When the state flag is in the first state value, the initial working mass can be the half-load calibration mass. When the state flag is in the second state value, the most recent historical working mass is used as the initial working mass.

[0042] In some embodiments, determining the initial working quality of the vehicle based on the vehicle status flag includes: acquiring vehicle signals, which include at least one of: a gear position signal, a parking gear duration, a power status signal, and a cargo door opening / closing signal; if the vehicle is in the parking gear and the parking gear duration is greater than or equal to a first preset duration threshold (e.g., greater than or equal to 5 minutes), then the vehicle status is determined to be a first parking state; if the vehicle is in the parking gear, the parking gear duration is less than or equal to a second preset duration threshold (e.g., less than or equal to 2 minutes), and the cargo door opening / closing signal changes from closed to open, then the vehicle status is determined to be a cargo storage / retrieval state; the first preset duration threshold is greater than the second preset duration threshold; if the power status changes from a power-off state to a power-on state, then the vehicle status is determined to be a power-off restart state; if the vehicle is in the parking gear, the parking gear duration is less than the first preset duration threshold, and the cargo door opening / closing signal is a closed signal, then the vehicle status is determined to be a second parking state; if the vehicle is not in the parking gear, then the vehicle status is determined to be a driving state.

[0043] By combining the above methods and integrating multi-dimensional information such as parking gear duration, cargo box door status, and power status, loading and unloading behaviors and parking scenarios can be accurately identified.

[0044] If the vehicle is in any of the following states: cargo storage / retrieval, first parking, or power-down / restart, a quality reset operation is triggered, and the status flag is marked with the first state value. If the vehicle is in the second parking or driving state, a quality reset operation is not triggered, and the status flag is marked with the second state value. If the status flag is the first state value, the half-load calibration mass is used as the initial working mass of the vehicle. The half-load calibration mass is half of the full-load calibration mass of the vehicle, ensuring that the convergence path of the algorithm is the shortest regardless of whether the actual load tends towards empty or full load. If the status flag is the second state value, the historical working mass is used as the initial working mass.

[0045] Working mass refers to the vehicle mass used to calculate the requested acceleration torque.

[0046] By using the above method, the mass reset is triggered by joint determination of multi-dimensional vehicle status, accurately identifying load changes caused by loading and unloading, and using the half-load calibrated mass as the initial value for reset, so that the convergence path at both the empty and full load ends is the shortest, and the convergence time is greatly shortened.

[0047] Step S102: Determine the load change trend of the vehicle based on historical work quality, and configure the calculation parameters of work quality according to the load change trend; the load change trend includes a downward trend, an upward trend, and a stable trend.

[0048] In some embodiments, determining the load change trend based on historical work quality includes: if the historical work quality is within the unloaded quality range, then the load change trend is determined to be an upward trend, i.e., a loading scenario; if the historical work quality is within the fully loaded quality range, then the load change trend is determined to be a downward trend, i.e., a unloading scenario; if the historical work quality is within the intermediate quality range, then the load change trend is determined to be a stable trend, i.e., a stable load scenario.

[0049] Using the above method, the current load change trend can be automatically identified based on historical work quality, thereby determining loading, unloading, and load stability scenarios according to different load change trends, and then calculating the vehicle work quality under different scenarios.

[0050] In some embodiments, if the calculation parameters include the forgetting factor of the recursive least squares method and the convergence determination period, the calculation parameters of the working quality are configured according to the load change trend, including: if the load is determined to be a stable scenario, a default forgetting factor and a default convergence determination period are configured; if the load is determined to be an unloading scenario, a first forgetting factor and a first convergence determination period are configured, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period; if the load is determined to be a loading scenario, a second forgetting factor and a second convergence determination period are configured, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

[0051] Using the above method, differentiated forgetting factors and convergence periods are configured for three scenarios: loading, unloading, and load stabilization. During unloading, a smaller forgetting factor and shorter convergence period are used to ensure the estimated value quickly follows the quality decline. During loading, a larger forgetting factor and longer convergence period are used to suppress estimation fluctuations caused by stepped loading. During load stabilization, default parameters are used to maintain steady-state accuracy. All three scenarios can adapt to their own varying convergence characteristics, ensuring that quality estimation balances convergence speed and stability under different operating conditions.

[0052] Step S103: Under the condition that the vehicle is in an effective driving condition, the working mass is calculated based on the initial working mass, the vehicle longitudinal dynamics model and calculation parameters to obtain the working mass.

[0053] In some embodiments, obtaining the working quality includes: collecting vehicle operating data under effective driving conditions, the operating data including at least: vehicle speed, driving torque, and longitudinal acceleration; iteratively updating the initial working quality based on the vehicle longitudinal dynamics model, operating data, and calculation parameters of the corresponding scenario to obtain an estimated quality; if the state flag is a first state value, determining the estimated quality obtained in the current iteration as the working quality after smoothing; if the state flag is a second state value, calculating the acceleration request torque based on the historical working quality and keeping it unchanged, while continuously iteratively updating the estimated quality; comparing the continuously iteratively estimated quality with the historical working quality, and if the quality update conditions are met, determining the estimated quality as the new working quality after smoothing.

[0054] Using the above method, in the first state value, the estimated mass is directly determined as the working mass after smoothing, realizing a rapid response of torque after load change; in the second state value, it can reduce the risk of torque output change caused by data jitter, while retaining the system's adaptive ability to small load changes, thus taking into account both stability and adaptability.

[0055] For example, in the intelligent driving activation state, the following are continuously collected via the CAN bus: real-time vehicle speed v, actual output drive torque T of VCU, and longitudinal acceleration a of the vehicle; and factory parameters are retrieved: wheel radius r, transmission system (reducer) efficiency η, rolling resistance coefficient f, air resistance coefficient Cd, frontal area A, air density ρ, and gravitational acceleration g.

[0056] Data can only be used for mass estimation under effective driving conditions, and the following conditions must be met simultaneously: pure driving state, VCU drive torque T>0, no braking signal; longitudinal acceleration within a reasonable range: 0.1m / s². 2 ≤ a ≤ 2m / s 2 The vehicle speed is within a stable range: 10km / h ≤ v ≤ 120km / h.

[0057] A simplified model is established based on the longitudinal dynamic equilibrium equation of a pure electric vehicle: ; in: The driving force at the wheel end, For the real-time quality of the whole vehicle, For the longitudinal acceleration of the vehicle, For rolling resistance, For air resistance, This represents the slope resistance; under flat road conditions, this value is 0, while under slope conditions, it is offset by robust estimation by the algorithm. After rearranging the above formula, we can obtain the estimated mass value for a single test: ; in, : Estimated vehicle mass for a single test, i.e., the preliminary value of the vehicle mass calculated under a single effective working condition; The drive torque output by the VCU is collected in real time on the CAN bus; Transmission system efficiency, a fixed calibration at the vehicle's factory; Wheel radius, a fixed calibration at the vehicle's factory; air density; Air drag coefficient, a fixed calibration at the vehicle's factory; Vehicle frontal area, fixed calibration at the factory; The vehicle's current speed is collected in real time via the CAN bus; Vehicle longitudinal acceleration is collected or calculated in real time. The acceleration due to gravity is fixed at 9.8 m / s². 2 ; This is the rolling resistance coefficient, which is fixed and calibrated at the factory when the vehicle leaves the factory.

[0058] Robust design for ramp conditions: For flat roads or gently sloping ramps, Rolling resistance has already been covered. For the ramp resistance component introduced by transients or large slopes, robust estimation is performed using the filtering characteristics of recursive least squares with a forgetting factor. By treating ramp resistance as a low-frequency or transient disturbance and appropriately setting and dynamically adjusting the forgetting factor according to the scenario, its influence can be effectively suppressed, obtaining accurate quality estimates without relying on ramp sensors.

[0059] In some embodiments, if a reset is not triggered (i.e., reset_occurred = FALSE), the front-end work quality is updated using a front-end-back-end dual-track system when the work quality update conditions are met. This front-end-back-end dual-track system includes a front-end maintenance layer and a back-end iteration layer, preventing sudden changes in torque output caused by single noise or instantaneous fluctuations in operating conditions. For example, this mechanism includes the following parallel operating layers: At the front-end maintenance level, a stable front-end work quality value is always used to directly participate in the real-time calculation of acceleration request torque, ensuring absolutely smooth driving control. This front-end work quality value remains unchanged until it passes a rigorous update judgment, thereby shielding the torque output from any interference caused by factors such as instantaneous sensor noise, road bumps, or occasional signal jumps, ensuring consistent driving quality.

[0060] At the background iteration level, based on effective driving condition data, the latest background estimated mass value is continuously generated iteratively using a recursive least squares method with a forgetting factor. This background estimated mass value iteration process is completely decoupled from the front-end torque control, meaning that whether or not the background estimated mass is updated does not affect the torque output of the current cycle, thus continuously tracking possible changes in the actual load while ensuring smooth driving.

[0061] Within each effective operating cycle, the front-end work quality value and the back-end estimated quality value are continuously compared. The back-end estimated quality value is smoothly replaced with the new front-end work quality value only when all of the following conditions are met simultaneously: 1) If the relative deviation between the two exceeds the preset threshold (e.g., ±15%), the backend estimated value has significantly deviated from the current frontend work quality and there is a need for an update; 2) The over-limit state is maintained without interruption for N (e.g., 20) consecutive effective working condition cycles, rather than an occasional single jump, in order to eliminate non-continuous interference factors such as sensor noise and transient impacts on the road surface, and to ensure the authenticity and continuity of the deviation. 3) No abnormal operating conditions such as braking, idling, or rapid acceleration / deceleration were detected in the vehicle during this period, ensuring the reliability of the iterative data source and the credibility of the quality estimation.

[0062] When all three checks pass, it is determined that the estimated quality in the background has stabilized and converged to the new actual load level. If any condition is not met, the original work quality remains unchanged.

[0063] While ensuring absolute smoothness of output torque, it also achieves adaptive correction to small and gradual load changes during driving (such as slight displacement of cargo, rainwater accumulation, fuel consumption, etc.), thereby obtaining more precise requested acceleration torque.

[0064] In some embodiments, the operation of a pure electric light truck in an urban delivery scenario is taken as an example: Initial state: The vehicle is powered on for the first time today, and the status flags are initially empty. The vehicle is currently in an unloaded state, and the historical work quality records are empty.

[0065] Scenario 1: The power state changes from OFF to ON, indicating a vehicle restart. Since this is one of the three states that trigger a quality reset, the status flag is set to the first state value (e.g., reset_occurred = TRUE), and the half-load calibration mass is used as the initial working mass. Assuming the vehicle's full-load calibration mass is 4500kg, the initial working mass is 2250kg. Then, using a fast follow mode, the vehicle quickly converges to the actual mass during subsequent driving.

[0066] Scenario 2: Normal driving, no reset. The vehicle exits the parking lot, shifts from parking to drive, and begins normal driving. If the vehicle is not detected to be in parking, the vehicle is determined to be in a driving state. This state does not trigger a mass reset, and the status flag retains the second state value (e.g., reset_occurred = FALSE). The most recent historical working mass (i.e., the initial working mass of 2250kg mentioned above, which has converged and been updated to approximately 3500kg of actual unloaded mass after driving a certain distance) is used as the initial working mass, and iterative updates continue. At this time, a cautious update mode is used, and a front-end-back-end dual-track mechanism ensures smooth torque output.

[0067] Scenario 3: Long-term parking (parking position ≥ 5 minutes) triggers a reset. The vehicle travels to the first delivery point, starts timing, and if the parking position remains in the parking position for 5 minutes (the first preset time threshold is 5 minutes), the vehicle is determined to be in the first parking state (long-term parking state). This state triggers a quality reset, the status flag is marked with the first state value (reset_occurred = TRUE), and the initial working quality is set to a half-load calibrated quality of 2250kg. At this point, through the fast follow mode, the vehicle quickly converges to the actual quality during subsequent travel.

[0068] Simultaneously, the most recent historical work quality (the empty load weight that had converged to approximately 3500 kg before unloading) is retrieved and determined to fall within the empty load weight range, with the load change trend predicted to be upward (loading scenario). Based on this prediction, a second forgetting factor (e.g., λ=0.995) and a second convergence determination period (e.g., 30 consecutive effective working condition cycles) are configured to address the characteristic of discontinuous, step-like increases in quality during subsequent loading processes, filtering out fluctuations in single-batch loading.

[0069] Scenario 4: Cargo Storage / Retrieval (Short-term Parking + Cargo Door Opening / Closing) Triggers Reset. The vehicle temporarily stops at the second delivery point, and the vehicle is in parking gear for 1 minute (less than or equal to the second preset duration threshold of 2 minutes). At this time, the system detects that the cargo door switch signal changes from closed to open (rising edge transition), and immediately determines that the vehicle status is cargo storage / retrieval. This status triggers a quality reset, and the status flag is marked with the first status value (reset_occurred = TRUE), using a half-load calibrated weight of 2250kg as the initial working weight.

[0070] At this point, the most recent historical work quality (the empty load weight that has converged to about 3500kg after unloading at the previous delivery point) is retrieved. It is determined that it belongs to the empty load weight range, and the load change trend is predicted to be an upward trend (loading scenario). A second forgetting factor (λ=0.995) and a second convergence judgment period (30 consecutive effective working condition cycles) are configured to smoothly track the loading process.

[0071] Scenario 5: Short-term temporary parking (parking gear < 5 minutes + cargo door closed). No reset is triggered. If the vehicle is in parking gear for 30 seconds (less than the first preset time threshold of 5 minutes), the cargo door open / close signal remains closed (no loading / unloading action). The vehicle status is determined to be the second parking state (temporary parking state). This state does not trigger a quality reset; the status flag remains at the second state value (reset_occurred = FALSE), and operation continues with the most recent historical work quality as the initial work quality. If the vehicle is in driving mode (e.g., in D gear), a reset is also not triggered.

[0072] Based on the vehicle's longitudinal dynamics model, operational data, and calculation parameters for the corresponding scenario, the initial working mass is iteratively updated to obtain an estimated mass. If the state flag is the first state value, the estimated mass obtained in the current iteration is smoothed and determined as the working mass. If the state flag is the second state value, the acceleration request torque is calculated based on the most recent working mass and kept unchanged, while the estimated mass is continuously iteratively updated. The continuously iterated estimated mass is compared with the most recent working mass. If the mass update condition is met, the estimated mass is smoothed and determined as the new working mass.

[0073] Using the above method, in the first state value, the estimated mass is directly determined as the working mass after smoothing, realizing a rapid response of torque after load change; in the second state value, it can reduce the risk of torque output change caused by data jitter, while retaining the system's adaptive ability to small load changes, thus taking into account both stability and adaptability.

[0074] In some embodiments, the quality update conditions include: the deviation between the estimated quality and the working quality is greater than a preset deviation threshold, and remains there for a preset number of consecutive effective driving cycle periods, and the vehicle is not in a braking or rapid acceleration / deceleration state during this period.

[0075] In some embodiments, after loading, the vehicle is driven to estimate the convergence process. After loading is completed, the vehicle is switched from parking gear to drive (D) and resumed driving. Effective driving condition data is collected (e.g., drive torque T > 0 and no braking signal, longitudinal acceleration a = 0.5 m / s²). 2The vehicle speed is v=40km / h. Iterative calculations are performed based on the vehicle's longitudinal dynamics model and a pre-configured second forgetting factor (λ=0.995). Since loading is done in staggered batches, the large forgetting factor prevents the algorithm from reacting drastically to each batch, and the estimated mass gradually and steadily increases from 2250kg. After each batch is loaded, the estimated mass increases slightly but does not jump. After 30 consecutive effective operating cycles with the deviation remaining constant, the estimated mass finally converges to approximately 4500kg (full load mass), which is then smoothed using a low-pass filter and determined as the new working mass. =4500kg.

[0076] During unloading, the system determines the unloading scenario and achieves rapid convergence. After the vehicle reaches the unloading point and completes unloading, it resumes its journey. The system retrieves the most recent historical working mass (the full load mass converged in the previous driving cycle was 4500kg) and determines that it falls within the full load mass range. The predicted load change trend is downward (unloading scenario), and a first forgetting factor (λ=0.95) and a first convergence determination period (10 consecutive effective working condition cycles) are configured. After unloading, the mass continuously decreases. The system collects effective driving condition data, and the small forgetting factor allows the algorithm to quickly "forget" the old full load data, converging to an empty load mass of approximately 3500kg within a shorter convergence period (10 cycles). After smoothing, this empty load mass is determined as the new working mass. =3500kg. At this time, the system calculates the acceleration request torque based on 3500kg, avoiding the abrupt acceleration problem caused by calculating torque based on full load weight.

[0077] In a dual-track update without a reset, during normal operation, if the status flag is the second status value (reset_occurred = FALSE), the foreground work quality... =3500kg calculates the acceleration request torque and maintains stable output. Meanwhile, the backend continuously iterates and updates the estimated mass. If road bumps or sensor noise cause a certain moment... The weight jumped to 3800 kg (deviation of approximately 8.6%, not exceeding the ±15% deviation threshold), which does not meet the quality update requirements. Maintaining a constant weight of 3500 kg effectively filters out occasional noise. If the weight increases slowly due to rainwater accumulation or slight cargo movement, The weight remained stable at 3700 kg (with a deviation of approximately 5.7%), which did not exceed the preset deviation threshold (±15%). Keeping the weight constant at 3500 kg, the system maintains its current torque output within this deviation range. If If the weight remains consistently at 4100 kg (deviation of approximately 17.1%, exceeding the ±15% deviation threshold), and this out-of-tolerance condition persists for 20 consecutive effective operating cycles without the vehicle undergoing braking or rapid acceleration / deceleration, then the quality update conditions are met. The new working quality was determined after low-pass filtering and smoothing. =4100kg, enabling adaptive correction for minute load changes.

[0078] Using the above method, switching is only performed when the deviation between the estimated quality and the working quality is real, continuous, and reliable, thereby reducing the interference of occasional jumps caused by sensor noise or transient fluctuations on the working quality.

[0079] Step S104: Calculate the acceleration request torque based on the working quality, and adjust the acceleration request torque in the same direction as the working quality changes.

[0080] In some embodiments, calculating the acceleration request torque based on the working mass includes: replacing the fixed full-load calibration mass in the conventional acceleration torque calibration formula with the working mass to obtain an adaptive acceleration request torque; and applying a smoothing process based on a rate of change limit or low-pass filtering to the acceleration request torque before finally outputting the acceleration request torque to ensure a smooth transition of vehicle acceleration at any mass switching instant.

[0081] In some embodiments, the formula for calculating the torque at a fixed full-load mass calibration is: ; in, This refers to the torque requested for intelligent driving acceleration under the original calibration logic, i.e., the torque request value output under the traditional fixed calibration method. The full-load calibration mass is a fixed value, that is, the fixed and unchanging full-load vehicle mass calibration value used in the prior art; The target acceleration set for the intelligent driving system is the acceleration effect that the intelligent driving system expects the vehicle to achieve. In the above formula Replace with the output front-end work quality This results in adaptive torque output: ; in, The adjusted intelligent driving acceleration request torque is the adaptive torque request value output based on the real-time mass estimation value. This is the converged real-time vehicle mass estimate, which is the final vehicle mass value obtained through iterative optimization. The target acceleration set for the intelligent driving system is consistent with the target acceleration in the original calibration logic; to avoid [further issues]... The update caused a step change in the output torque, affecting the system's final output. Applying smoothing processing based on rate of change limits or low-pass filtering ensures a smooth transition in vehicle acceleration at any mass shift, comprehensively enhancing the driving experience.

[0082] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0083] Based on the same inventive concept, embodiments of this application provide an adaptive adjustment device for vehicle torque, comprising: a mass reset module 210, a load trend module 220, a mass calculation module 230, and a torque request module 240, wherein: The quality reset module 210 is used to determine the initial working quality of the vehicle based on the vehicle status flag bit, which indicates whether the vehicle quality has been reset.

[0084] In some embodiments, the quality reset module is further configured to: acquire vehicle signals, the vehicle signals including at least one of: gear position signal, parking gear duration, power status signal, and cargo box door opening / closing signal; if the vehicle is in parking gear and the parking gear duration is greater than or equal to a first preset duration threshold, then determine the vehicle state as a first parking state; if the vehicle is in parking gear, the parking gear duration is less than or equal to a second preset duration threshold, and the cargo box door opening / closing signal changes from closed to open, then determine the vehicle state as a cargo storage / retrieval state; the first preset duration threshold is greater than the second preset duration threshold; if the power status changes from power-off to power-on, then determine the vehicle state as a power-off restart state; if the vehicle is in parking gear, the parking gear duration is less than the first preset duration threshold, and the cargo box door opening / closing signal is a closed signal, then determine the vehicle state as a second parking state; if the vehicle is not in parking gear, then determine the vehicle state as a driving state.

[0085] In some embodiments, the quality reset module 210 is further configured to: trigger a quality reset operation and mark the status flag as a first status value if the vehicle status is any one of cargo storage / retrieval status, first parking status, or power-down restart status; not trigger a quality reset operation and mark the status flag as a second status value if the vehicle status is a second parking status or driving status; if the status flag is the first status value, use the half-load calibration quality as the initial working quality of the vehicle; if the status flag is the second status value, use the most recent historical working quality as the initial working quality.

[0086] The load trend module 220 is used to determine the load change trend of the vehicle based on historical work quality, and to configure the calculation parameters of work quality according to the load change trend; the load change trend includes a downward trend, an upward trend and a stable trend.

[0087] In some embodiments, the load trend module 220 is further configured to: determine the load change trend as an upward trend if the historical working quality is within the unloaded quality range, i.e., loading scenario; determine the load change trend as a downward trend if the historical working quality is within the fully loaded quality range, i.e. unloading scenario; and determine the load change trend as a stable trend if the historical working quality is within the intermediate quality range, i.e., load stable scenario.

[0088] In some embodiments, the load trend module 220 is further configured to: if the load is determined to be a stable scenario, configure a default forgetting factor and a default convergence determination period; if the load is determined to be an unloading scenario, configure a first forgetting factor and a first convergence determination period, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period; if the load is determined to be a loading scenario, configure a second forgetting factor and a second convergence determination period, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

[0089] The mass calculation module 230 is used to calculate the working mass based on the initial working mass, the vehicle's longitudinal dynamics model, and calculation parameters when the vehicle is in an effective driving condition.

[0090] In some embodiments, the mass calculation module 230 is further configured to: collect vehicle operating data under effective driving conditions, the operating data including at least: vehicle speed, driving torque and longitudinal acceleration; iteratively update the initial working mass based on the vehicle longitudinal dynamics model, operating data and calculation parameters of the corresponding scenario to obtain an estimated mass; if the state flag is a first state value, determine the estimated mass obtained by the current iteration as the working mass after smoothing; if the state flag is a second state value, calculate the acceleration request torque based on the historical working mass and keep it unchanged, while continuously iteratively updating the estimated mass; compare the continuously iterative estimated mass with the historical working mass, and if the mass update condition is met, determine the estimated mass as the new working mass after smoothing.

[0091] In some embodiments, the quality update conditions include: the deviation between the estimated quality and the working quality is greater than a preset deviation threshold, and remains there for a preset number of consecutive effective driving cycle periods, and the vehicle is not in a braking or rapid acceleration / deceleration state during this period.

[0092] The torque request module 240 is used to calculate the acceleration request torque based on the work quality, and the acceleration request torque is adjusted in the same direction as the work quality changes.

[0093] In some embodiments, the torque request module 240 is further configured to: replace the fixed full-load calibration mass in the conventional acceleration torque calibration formula with the working mass to obtain an adaptive acceleration request torque; and apply a smoothing process based on a rate of change limit or low-pass filtering to the acceleration request torque before finally outputting the acceleration request torque.

[0094] Specific limitations regarding the adaptive torque adjustment device for vehicles can be found in the limitations of the adaptive torque adjustment method for vehicles described above, and will not be repeated here. Each module in the aforementioned adaptive torque adjustment device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0095] Based on the same inventive concept, embodiments of this application provide a computer device, which may be a vehicle-mounted terminal, and its internal structure diagram may be as follows. Figure 3As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When executed by the processor, the computer program implements an adaptive torque adjustment method for a vehicle.

[0096] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0097] Based on the same inventive concept, embodiments of this application provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: determining the initial working mass of the vehicle based on a vehicle status flag bit, the vehicle status flag bit indicating whether the vehicle mass has been reset; determining the load change trend of the vehicle based on historical working masses, and configuring calculation parameters for the working mass based on the load change trend; the load change trend includes a decreasing trend, an increasing trend, and a stable trend; calculating the working mass based on the initial working mass, the vehicle longitudinal dynamics model, and the calculation parameters when the vehicle is in an effective driving condition; and calculating the acceleration request torque based on the working mass, the acceleration request torque being adjusted in the same direction as the working mass changes.

[0098] In some embodiments, when the processor executes the computer program, it further implements the following steps: acquiring vehicle signals, the vehicle signals including at least one of: gear position signal, parking gear duration, power status signal, and cargo door opening / closing signal; if the vehicle is in parking gear and the parking gear duration is greater than or equal to a first preset duration threshold, then the vehicle state is determined to be a first parking state; if the vehicle is in parking gear, the parking gear duration is less than or equal to a second preset duration threshold, and the cargo door opening / closing signal changes from closed to open, then the vehicle state is determined to be a cargo storage / retrieval state; the first preset duration threshold is greater than the second preset duration threshold; if the power status changes from power-off to power-on, then the vehicle state is determined to be a power-off restart state; if the vehicle is in parking gear, the parking gear duration is less than the first preset duration threshold, and the cargo door opening / closing signal is a closed signal, then the vehicle state is determined to be a second parking state; if the vehicle is not in parking gear, then the vehicle state is determined to be a driving state.

[0099] In some embodiments, when the processor executes the computer program, it further implements the following steps: if the vehicle state is any one of cargo storage / retrieval state, first parking state, or power-down restart state, a quality reset operation is triggered, and the status flag is marked with a first status value; if the vehicle state is a second parking state or driving state, a quality reset operation is not triggered, and the status flag is marked with a second status value; if the status flag is the first status value, the half-load calibration quality is used as the initial working quality of the vehicle; if the status flag is the second status value, the historical working quality is used as the initial working quality.

[0100] In some embodiments, when the processor executes the computer program, it further implements the following steps: if the historical working quality is within the unloaded quality range, the load change trend is determined to be an upward trend, i.e., a loading scenario; if the historical working quality is within the full-load quality range, the load change trend is determined to be a downward trend, i.e., a unloading scenario; if the historical working quality is within the intermediate quality range, the load change trend is determined to be a stable trend, i.e., a stable load scenario.

[0101] In some embodiments, when the processor executes the computer program, it further implements the following steps: if the load is determined to be a stable scenario, a default forgetting factor and a default convergence determination period are configured; if the unloading scenario is determined to be a scenario, a first forgetting factor and a first convergence determination period are configured, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period; if the loading scenario is determined to be a scenario, a second forgetting factor and a second convergence determination period are configured, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

[0102] In some embodiments, when the processor executes the computer program, it further implements the following steps: collecting vehicle operating data under effective driving conditions, the operating data including at least: vehicle speed, driving torque, and longitudinal acceleration; iteratively updating the initial working mass based on the vehicle longitudinal dynamics model, operating data, and calculation parameters of the corresponding scenario to obtain an estimated mass; if the state flag is a first state value, determining the estimated mass obtained in the current iteration as the working mass after smoothing; if the state flag is a second state value, calculating the acceleration request torque based on the historical working mass and keeping it unchanged, while continuously iteratively updating the estimated mass; comparing the continuously iteratively estimated mass with the historical working mass, and if the mass update condition is met, determining the estimated mass as the new working mass after smoothing.

[0103] In some embodiments, the quality update conditions include: the deviation between the estimated quality and the working quality is greater than a preset deviation threshold, and remains there for a preset number of consecutive effective driving cycle periods, and the vehicle is not in a braking or rapid acceleration / deceleration state during this period.

[0104] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the following steps: determining the initial working mass of the vehicle based on a vehicle status flag bit, the vehicle status flag bit indicating whether the vehicle mass has been reset; determining the load change trend of the vehicle based on historical working masses, and configuring calculation parameters for the working mass based on the load change trend; the load change trend includes a decreasing trend, an increasing trend, and a stable trend; calculating the working mass based on the initial working mass, the vehicle longitudinal dynamics model, and calculation parameters when the vehicle is in an effective driving condition; calculating the acceleration request torque based on the working mass, the acceleration request torque being adjusted in the same direction as the working mass changes.

[0105] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: acquiring vehicle signals, the vehicle signals including at least one of: gear position signal, parking gear duration, power status signal, and cargo box door opening / closing signal; if the vehicle is in parking gear and the parking gear duration is greater than or equal to a first preset duration threshold, then the vehicle state is determined to be a first parking state; if the vehicle is in parking gear, the parking gear duration is less than or equal to a second preset duration threshold, and the cargo box door opening / closing signal changes from closed to open, then the vehicle state is determined to be a cargo storage / retrieval state; the first preset duration threshold is greater than the second preset duration threshold; if the power status changes from a power-off state to a power-on state, then the vehicle state is determined to be a power-off restart state; if the vehicle is in parking gear, the parking gear duration is less than the first preset duration threshold, and the cargo box door opening / closing signal is a closed signal, then the vehicle state is determined to be a second parking state; if the vehicle is not in parking gear, then the vehicle state is determined to be a driving state.

[0106] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: if the vehicle state is any one of cargo storage / retrieval state, first parking state, or power-down restart state, a quality reset operation is triggered, and the status flag is marked with a first status value; if the vehicle state is a second parking state or driving state, a quality reset operation is not triggered, and the status flag is marked with a second status value; if the status flag is the first status value, the half-load calibration quality is used as the initial working quality of the vehicle; if the status flag is the second status value, the historical working quality is used as the initial working quality.

[0107] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: if the historical working quality is within the unloaded quality range, the load change trend is determined to be an upward trend, i.e., a loading scenario; if the historical working quality is within the full-load quality range, the load change trend is determined to be a downward trend, i.e., a unloading scenario; if the historical working quality is within the intermediate quality range, the load change trend is determined to be a stable trend, i.e., a stable load scenario.

[0108] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: if the scenario is determined to be a stable load scenario, a default forgetting factor and a default convergence determination period are configured; if the scenario is determined to be an unloading scenario, a first forgetting factor and a first convergence determination period are configured, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period; if the scenario is determined to be a loading scenario, a second forgetting factor and a second convergence determination period are configured, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

[0109] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: collecting vehicle operating data under effective driving conditions, the operating data including at least: vehicle speed, driving torque, and longitudinal acceleration; iteratively updating the initial working mass based on the vehicle longitudinal dynamics model, operating data, and calculation parameters of the corresponding scenario to obtain an estimated mass; if the state flag is a first state value, determining the estimated mass obtained in the current iteration as the working mass after smoothing; if the state flag is a second state value, calculating the acceleration request torque based on the historical working mass and keeping it unchanged, while continuously iteratively updating the estimated mass; comparing the continuously iterative estimated mass with the historical working mass, and if the quality update condition is met, determining the estimated mass as the new working mass after smoothing.

[0110] In some embodiments, the quality update conditions include: the deviation between the estimated quality and the working quality is greater than a preset deviation threshold, and remains therefore for a preset number of consecutive effective driving cycle periods, and during this period the vehicle is not in a braking or rapid acceleration / deceleration state. Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0111] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0112] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method of adaptive adjustment of vehicle torque, characterized in that, include: The initial working quality of the vehicle is determined based on the vehicle status flag, which indicates whether the vehicle quality has been reset. The load change trend of the vehicle is determined based on the historical work quality, and the calculation parameters of the work quality are configured according to the load change trend. When the vehicle is in an effective driving condition, the working mass is calculated based on the initial working mass, the vehicle longitudinal dynamics model, and the calculation parameters. The acceleration request torque is calculated based on the work quality, and the acceleration request torque is adjusted in the same direction as the work quality changes.

2. The method of claim 1, wherein, The initial operating quality of the vehicle is determined based on the vehicle status flags, including: If the vehicle status is any one of cargo storage / retrieval status, first parking status, or power-down / restart status, a quality reset operation is triggered, and the status flag is marked with the first status value. If the vehicle status is the second parking status or the driving status, the quality reset operation is not triggered, and the status flag is marked with the second status value. If the status flag is the first status value, then the half-load calibration mass is used as the initial working mass of the vehicle; if the status flag is the second status value, then the historical working mass is used as the initial working mass.

3. The method of claim 2, wherein, The method further includes: Acquire vehicle signals, the vehicle signals including at least one of: gear position signal, parking gear duration, power status signal, and cargo box door open / close signal; If the vehicle is in the parking position and the duration of the parking position is greater than or equal to a first preset duration threshold, then the vehicle state is determined to be the first parking state. If the vehicle is in the parking position, and the duration of the parking position is less than or equal to the second preset duration threshold, and the cargo door switch signal changes from closed to open, then the vehicle status is determined to be cargo storage / retrieval status; the first preset duration threshold is greater than the second preset duration threshold. If the power state changes from power-off to power-on, then the vehicle state is determined to be a power-off restart state. If the vehicle is in the parking position, the duration of the parking position is less than the first preset duration threshold, and the cargo box door switch signal is a closed signal, then the vehicle state is determined to be the second parking state. If the vehicle is not in the parking gear, then the vehicle is determined to be in a driving state.

4. The method according to claim 1, characterized in that, Determine the load change trend based on historical work quality, including: If the historical working quality falls within the no-load quality range, then the load change trend is determined to be an upward trend. If the historical work quality falls within the full load range, then the load change trend is determined to be a downward trend. If the historical work quality falls within the intermediate quality range, then the load change trend is determined to be a stable trend.

5. The method according to claim 4, characterized in that, The calculation parameters include the forgetting factor of the recursive least squares method and the convergence determination period; The calculation parameters for configuring the working quality based on the load change trend include: If the load change trend is a stable trend, then configure the default forgetting factor and the default convergence determination period; If the load change is a downward trend, then a first forgetting factor and a first convergence determination period are configured, wherein the first forgetting factor is less than the default forgetting factor and the first convergence determination period is less than the default convergence determination period. If the load change trend is the upward trend, then a second forgetting factor and a second convergence determination period are configured, wherein the second forgetting factor is greater than the default forgetting factor and the second convergence determination period is greater than the default convergence determination period.

6. The method according to claim 1, characterized in that, The working mass is calculated based on the initial working mass, the vehicle longitudinal dynamics model, and the calculation parameters, and includes: Collect the vehicle's operating data under the effective driving conditions, and the operating data includes at least: vehicle speed, driving torque, and longitudinal acceleration; The initial working quality is iteratively updated based on the vehicle longitudinal dynamics model, the operating data, and the calculation parameters of the corresponding scenario to obtain the estimated quality; If the status flag is the first status value, the estimated quality obtained in the current iteration is smoothed and then determined as the working quality; If the status flag is the second status value, the acceleration request torque is calculated based on the historical work quality and remains unchanged, while the estimated quality is continuously iterated and updated. The estimated quality obtained through continuous iteration is compared with the historical work quality. If the quality update conditions are met, the estimated quality is smoothed and determined as the new work quality.

7. The method according to claim 6, characterized in that, The quality update conditions include: The deviation between the estimated quality and the working quality is greater than a preset deviation threshold and remains so for a preset number of consecutive effective driving cycle periods, and the vehicle is not in a braking or rapid acceleration / deceleration state during this period.

8. An adaptive torque adjustment device for a vehicle, characterized in that, include: The quality reset module is used to determine the initial working quality of the vehicle based on the vehicle status flag bit, wherein the vehicle status flag bit indicates whether the vehicle quality has been reset. The load trend module is used to determine the load change trend of the vehicle based on historical work quality, and to configure the calculation parameters of work quality based on the load change trend. The mass calculation module is used to calculate the working mass based on the initial working mass, the vehicle longitudinal dynamics model, and the calculation parameters when the vehicle is in an effective driving condition. The torque request module is used to calculate the acceleration request torque based on the working quality, and the acceleration request torque is adjusted in the same direction as the working quality changes.

9. A vehicle terminal, 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 computer program, it implements the adaptive adjustment method for vehicle torque according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor and implementing the adaptive adjustment method for vehicle torque as claimed in any one of claims 1 to 7.

Images