Commercial vehicle adaptive torque control method, device, equipment and system

Abstract

This application provides a method, device, equipment, and system for adaptive torque control of commercial vehicles. The method includes: acquiring target driver operation datasets under different offline modes, the target driver operation datasets containing excellent driving samples under different loads and gradients; grouping the target driver operation datasets to generate an acceleration mapping table and a torque slope mapping table; and dynamically adjusting and outputting vehicle torque control commands based on real-time operating data, the acceleration mapping table, and the torque slope mapping table in online mode of the commercial vehicle. This solves the problems of existing commercial vehicles, which, under complex operating conditions with significant load fluctuations, gradient changes, and driver style differences, struggle to balance energy consumption and power performance by relying on fixed calibration rules, suffer from insufficient scenario adaptability, and have high manual calibration and maintenance costs. This improves the overall vehicle energy efficiency, torque response, and driving smoothness.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology for commercial vehicles, and in particular to a method, device, equipment and system for controlling adaptive torque in commercial vehicles. Background Technology

[0002] In the commercial vehicle sector, vehicle energy consumption is significantly affected by driver behavior. Existing power and torque control strategies for commercial vehicles are mostly based on fixed calibrations or preset rules, making it difficult to dynamically adapt to different driver styles and complex, ever-changing operating scenarios. Therefore, there is an urgent need to propose a method that can effectively capture driver operational characteristics and convert them into reusable control strategies to achieve online vehicle control.

[0003] Most existing commercial vehicles use fixed torque MAP, or predict information such as road curvature and slope through high-precision maps, or group vehicle operating data according to energy consumption, and then dynamically adjust control parameters.

[0004] However, the above control strategies cannot adapt to differences in driver style and complex operating conditions, making it difficult for vehicles to balance fuel efficiency and driving performance under complex conditions. Summary of the Invention

[0005] This application proposes a method, device, equipment, and system for controlling adaptive torque in commercial vehicles, which enables adaptive adjustment of torque output based on changes in the vehicle's operating state.

[0006] Firstly, this application provides a method for controlling adaptive torque in a commercial vehicle, including:

[0007] Obtain target driver operation datasets under different offline modes, wherein the target driver operation datasets contain excellent driving samples under different loads and slopes;

[0008] The target driver operation dataset is grouped to generate an acceleration mapping table and a torque slope mapping table.

[0009] Based on the real-time operating data of the commercial vehicle in online mode, the acceleration mapping table, and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output.

[0010] In one possible implementation, the target driver operation dataset is grouped to generate an acceleration mapping table and a torque slope mapping table, including:

[0011] Based on vehicle speed and throttle opening, the target driver operation dataset is discretized and grouped to obtain multiple vehicle speed intervals and multiple throttle opening intervals.

[0012] Collect triplet data from the multiple vehicle speed ranges and the multiple throttle opening ranges to generate an acceleration mapping table. The triplet data includes vehicle speed, throttle opening, and acceleration.

[0013] Calculate the torque change rate of the target driver operation dataset and generate a torque slope mapping table.

[0014] In one possible implementation, triplet data of the multiple vehicle speed ranges and the multiple throttle opening ranges are collected to generate an acceleration mapping table, including:

[0015] Collect triplet data for the multiple vehicle speed ranges and the multiple throttle opening ranges;

[0016] The upper limit of acceleration is obtained by performing quantile statistics on the triplet data;

[0017] An acceleration mapping table is generated based on the multiple vehicle speed ranges, the multiple throttle opening ranges, and the upper limit of acceleration.

[0018] In one possible implementation, the torque change rate of the target driver operation dataset is calculated, and a torque slope mapping table is generated, including:

[0019] Calculate the rate of change of torque at each sampling point in the target driver operation dataset;

[0020] Based on the multiple vehicle speed ranges and the multiple throttle opening ranges, the quantile statistics of the torque change rate are performed to obtain the upper limit of the torque change rate.

[0021] A torque slope mapping table is generated based on the multiple vehicle speed ranges, the multiple throttle opening ranges, and the upper limit of torque change rate.

[0022] In one possible implementation, based on real-time operating data of the commercial vehicle in online mode, the acceleration mapping table, and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output, including:

[0023] Collect real-time operating data of commercial vehicles in online mode, including current vehicle speed, throttle opening, load and gradient;

[0024] The real-time running data is matched with the acceleration mapping table to determine the target acceleration;

[0025] Based on the target acceleration and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output.

[0026] In one possible implementation, based on the target acceleration and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output, including:

[0027] Calculate the required torque limit for the commercial vehicle based on the vehicle dynamics model;

[0028] The required torque limit is matched with the torque slope mapping table to determine the target torque change rate;

[0029] Based on the target acceleration and the target torque change rate, the required torque limit is dynamically adjusted, and a vehicle torque control command is output.

[0030] In one possible implementation, the target driver operation dataset under different offline modes is obtained, including:

[0031] Collect historical operating data of commercial vehicles in offline mode;

[0032] The historical operation data is preprocessed to obtain cleaned historical operation data;

[0033] Based on load and slope, the working conditions are divided into scenarios, and excellent driver samples that meet the vehicle energy consumption requirements are selected from the historical operating data after cleaning.

[0034] Based on the excellent driver samples, obtain target driver operation datasets under different offline modes.

[0035] Secondly, this application provides a control device for adaptive torque in commercial vehicles, comprising:

[0036] The operation data acquisition module is used to acquire target driver operation datasets under different offline modes. The target driver operation datasets contain excellent driving samples under different loads and slopes.

[0037] The data mapping module is used to group the target driver operation dataset to generate an acceleration mapping table and a torque slope mapping table.

[0038] The torque control module is used to dynamically adjust and output vehicle torque control commands based on real-time operating data of the commercial vehicle in online mode, the acceleration mapping table, and the torque slope mapping table.

[0039] Thirdly, embodiments of this application provide a control device for adaptive torque in a commercial vehicle, comprising: a memory and a processor;

[0040] The memory stores computer-executed instructions;

[0041] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0042] Fourthly, this application provides a control system for adaptive torque of a commercial vehicle, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0043] This application provides a method, device, equipment, and system for adaptive torque control of commercial vehicles. The method includes: acquiring target driver operation datasets under different offline modes, the target driver operation datasets containing excellent driving samples under different loads and gradients; grouping the target driver operation datasets to generate an acceleration mapping table and a torque slope mapping table; and dynamically adjusting and outputting vehicle torque control commands based on real-time operating data, the acceleration mapping table, and the torque slope mapping table in online mode of the commercial vehicle. This solves the problems of existing commercial vehicles, which, under complex operating conditions with significant load fluctuations, gradient changes, and driver style differences, struggle to balance energy consumption and power performance by relying on fixed calibration rules, and suffer from insufficient scenario adaptability and high manual calibration maintenance costs. This method improves the overall vehicle's energy efficiency, torque response, and driving smoothness. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0045] Figure 1 A flowchart illustrating an embodiment of the adaptive torque control method for commercial vehicles provided in this application;

[0046] Figure 2 A flowchart illustrating Embodiment 2 of the adaptive torque control method for commercial vehicles provided in this application;

[0047] Figure 3 A flowchart illustrating Embodiment 3 of the adaptive torque control method for commercial vehicles provided in this application;

[0048] Figure 4 A schematic diagram of the structure of the adaptive torque control device for commercial vehicles provided in this application;

[0049] Figure 5 A schematic diagram of the structure of the adaptive torque control device for commercial vehicles provided in this application.

[0050] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application.

[0052] This application applies to energy-saving control scenarios for commercial vehicles (such as coal transportation and express delivery). These vehicles operate under complex conditions (such as frequent starts and stops, inclines and declines, and load fluctuations), and driver behavior varies significantly. This solution learns from the operational data of experienced drivers offline via the cloud, generates control strategies tailored to the specific scenario, and deploys them to the vehicle control unit (VCU). The VCU collects real-time data on vehicle speed, throttle opening, load, and gradient, calculates the required torque using a dynamic model, and dynamically adjusts the output torque through dual constraints of acceleration and torque slope, thereby optimizing energy consumption while ensuring driving performance. This solution supports continuous optimization driven by factory calibration and after-sales data, and is suitable for the differentiated needs of different market segments (such as coal transportation and express delivery).

[0053] Existing commercial vehicle control strategies have the following problems:

[0054] (1) Limitations of fixed calibration MAP: Traditional methods divide the load into multiple intervals, and design a fixed torque MAP for each interval. This requires manual calibration, which takes a long time and cannot adapt to changes in driver style or working conditions (such as load fluctuations or slope changes).

[0055] (2) Dependence on high-precision maps: Some solutions use high-precision maps to predict road information (such as slope and curvature) and dynamically adjust control parameters, but they are limited by the map coverage and the parameters still need to be manually calibrated, making it difficult to achieve full-scene adaptation.

[0056] (3) Insufficient scenario adaptability of data-driven methods: Existing data-driven methods do not combine scenario factors such as load and slope for detailed analysis. The generated MAP is fixed and lacks adaptability, which limits the effectiveness of control strategies under complex working conditions.

[0057] (4) Insufficient physical interpretability of control strategies: Some data-driven methods generate control rules that lack clear physical meaning and are difficult to implement efficiently in vehicle controllers.

[0058] To address the aforementioned technical challenges, this application employs a two-layer architecture combining data-driven offline learning and online control. This architecture transforms the operational characteristics of skilled drivers into adaptive acceleration and torque slope constraint strategies, achieving a dynamic balance between energy consumption optimization and driving performance in commercial vehicles. The core of this concept lies in: 1) Extracting the behavioral characteristics of skilled drivers hierarchically by scenario (load × slope) based on real-world operational data, generating a high-precision acceleration and torque slope constraint map; 2) In the online control phase, combining vehicle dynamics models and real-time operating conditions, dynamically adjusting torque output, and achieving coordinated optimization of energy consumption and power performance through dual constraints (acceleration feedforward constraint + slope constraint). This concept overcomes the limitations of traditional fixed calibration or reliance on high-precision maps, achieving scenario-adaptive control strategies and engineering feasibility.

[0059] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0060] Figure 1 This is a flowchart illustrating an embodiment of the adaptive torque control method for commercial vehicles provided in this application. Figure 1 As shown, it includes:

[0061] S101. Obtain the target driver operation dataset under different offline modes.

[0062] In this application, the offline mode refers to the working mode in which data processing and sample screening of batch real vehicle historical data are carried out by the cloud or back-end platform during the non-vehicle driving online control stage.

[0063] The target driver operation dataset serves as the foundational dataset for offline learning. It includes excellent driving samples under various operating conditions and contains various operating parameters related to power output, which are then used for subsequent grouping and processing to generate acceleration and torque slope maps.

[0064] Excellent driving samples are driving segments of vehicles with the best energy consumption performance under the same load and slope conditions. The samples can retain operating parameters related to power output to reflect the driver's acceleration habits and torque change characteristics under specific conditions.

[0065] In addition, different offline modes correspond to different ways of forming historical data, which can be collection batches at different time periods, different transportation routes, different vehicle usage statuses, or sets of operational data under different seasonal environments, specifically including excellent driving samples under different loads and slopes.

[0066] In this application, the acquisition of target driver operation datasets under different offline modes can be performed by a cloud server, a back-end analysis platform, or a corporate fleet data center. Specifically, it can be performed by a cloud server, a back-end analysis platform, or a corporate fleet data center, collecting historical operating data for no less than 300 vehicles and no less than 6 months from the target market segment (such as coal transportation).

[0067] The data fields of historical operation data include, but are not limited to: vehicle speed, accelerator pedal opening, engine or motor torque, engine or motor speed, cumulative energy consumption, longitudinal acceleration, load, gear, etc.

[0068] Next, the collected historical operational data undergoes data cleaning. This includes format checks, integrity checks, and rationality checks, removing abnormal data caused by sensor jumps and communication interruptions, resulting in offline training base samples under various load and slope conditions. This step yields offline training base samples covering multiple vehicle loads and road slope conditions, triggering subsequent mapping relationship generation processing.

[0069] Next, based on the cleaned historical operating data, the operating segments were divided into scenarios according to load range and gradient range. The load could be estimated from the vehicle's loaded mass, and the gradient could be obtained from high-precision maps, inertial navigation information, or gradient sensor information. For the operating segments within each operating scenario, segments with better operating performance were selected based on energy consumption per unit mileage, energy consumption per unit time, or comprehensive energy consumption indicators. The driving operation trajectory, speed changes, throttle response, and torque changes corresponding to these segments were used as samples of excellent drivers.

[0070] Finally, the excellent driver samples selected under different load and slope conditions are summarized to obtain target driver operation datasets covering a variety of working conditions under different offline modes; the offline mode can correspond to different time period collection batches, different transportation routes, different vehicle usage statuses or different seasonal environments.

[0071] The aforementioned target driver operation dataset is constructed based on scenario-based grouping. Cleaned historical operating data is compared and filtered under the same load and gradient conditions, ensuring that samples with superior energy consumption and stable operating characteristics are retained. These samples are then aggregated to form the data source required for subsequent mapping table generation. In this way, the target driver operation dataset maintains a correspondence with actual operating conditions, providing a consistent data foundation for the subsequent generation of acceleration and torque slope mapping tables.

[0072] This step extracts representative excellent driving behaviors from historical operating data in offline mode, reduces the interference of abnormal data and invalid samples on the construction of mapping relationships, and enables subsequent offline learning results to reflect the optimal operating characteristics under different load and slope conditions, and provides a reliable basis for the dynamic matching of vehicle torque control commands in online mode.

[0073] S102. Group the target driver operation dataset to generate an acceleration mapping table and a torque slope mapping table.

[0074] In this step, the acceleration mapping table is a lookup table that stores the correspondence between vehicle speed, throttle opening, and target acceleration, providing a reference for determining the target acceleration in online mode.

[0075] The torque slope mapping table is used to establish the correspondence between operating states and torque change characteristics, so as to limit the rate of torque change in online mode.

[0076] Group processing refers to classifying and statistically analyzing the target driver operation dataset according to preset parameter dimensions, and outputting lookup table results that can be directly accessed by the vehicle controller.

[0077] In this application, after the target driver operation dataset is formed, it can be grouped and statistically processed according to different operating scenarios or operating states to generate an acceleration mapping table and a torque slope mapping table. For example, feature extraction, statistical analysis, smoothing processing, or table item organization can be performed on relevant parameters in the target driver operation dataset to improve the stability and usability of the mapping results. After the acceleration mapping table and torque slope mapping table are generated, they can be distributed and stored in the vehicle controller for use during the vehicle's online control phase.

[0078] Based on the above analysis, this step involves grouping the target driver's operation dataset and generating acceleration and torque slope mapping tables that can be directly accessed, enabling the offline learning results to distinguish between different scenarios. The acceleration mapping table defines the target acceleration level corresponding to the driving intention under different operating conditions, while the torque slope mapping table defines the boundary of the torque build-up speed under similar operating conditions. Together, they constitute the constraint source for online control, allowing the controller to output torque not solely based on fixed calibration values ​​under complex transportation conditions.

[0079] S103. Based on the real-time operating data, acceleration mapping table, and torque slope mapping table of the commercial vehicle in online mode, dynamically adjust and output vehicle torque control commands.

[0080] In this application, the online mode is the working mode in which the vehicle controller collects vehicle signals in real time and executes torque closed-loop control during the actual driving process of the vehicle.

[0081] Real-time operating data in the online mode of commercial vehicles consists of operating parameters that are periodically collected during vehicle operation and represent the vehicle's current real-time operating conditions.

[0082] The vehicle torque control command is a power execution command directly issued by the vehicle controller to the engine controller, motor controller, or hybrid power coordination control module. Its value is determined comprehensively based on the operating status in online mode and the matching results of the mapping table.

[0083] Dynamic adjustment refers to the controller continuously updating the lookup table results and torque output according to the sampling cycle during vehicle operation, so that the torque command changes with the real-time operating conditions.

[0084] In this application, after the vehicle controller enters the online control mode, it receives real-time operating data according to a fixed control cycle. The vehicle controller matches the real-time operating data with an acceleration mapping table to obtain the corresponding target acceleration, and dynamically adjusts the vehicle torque output process in conjunction with a torque slope mapping table, thereby outputting vehicle torque control commands.

[0085] In one possible implementation, the vehicle controller can interpolate, transform, or limit the lookup table results based on the current operating conditions to improve the continuity and adaptability of torque output. During continuous vehicle operation, the vehicle controller repeatedly performs real-time acquisition, matching, and adjustment processing to achieve continuous updates of torque commands.

[0086] Based on the above analysis, this application achieves this by jointly matching real-time operating data with offline-generated acceleration mapping tables and torque slope mapping tables during the online control phase. This ensures that the vehicle torque control command is simultaneously constrained by the target acceleration and torque change characteristics. Therefore, even in scenarios with continuous fluctuations in vehicle load, frequent changes in road gradient, and differences in driver operation, the controller can still output matched vehicle torque control commands with unified data constraint logic, reducing the problems of stiff or slow response and discontinuous switching caused by fixed rules under complex working conditions.

[0087] The adaptive torque control method for commercial vehicles provided in this application converts excellent driving samples under different vehicle load and road gradient conditions from offline historical operating data into acceleration constraints and torque slope constraints that can be directly invoked at the vehicle end. The vehicle controller then combines this with real-time operating data to perform matching and torque adjustment processing, generating continuously updated vehicle torque control commands that change with operating conditions. This allows the commercial vehicle's power output to be based on the statistical results of scenario-specific samples. Based on this processing path, online control no longer relies solely on fixed mapping relationships but uses the operating condition characteristics learned offline as the control basis. This enables the same control framework to cover operational processes with significant changes in vehicle load and road gradient, such as coal transportation, express delivery, and long-haul logistics.

[0088] Figure 2 This is a flowchart illustrating Embodiment Two of the adaptive torque control method for commercial vehicles provided in this application. Figure 2 As shown, in Figure 1 Based on the implementation examples, the target driver operation dataset is grouped to generate an acceleration mapping table and a torque slope mapping table, including:

[0089] S201. Based on vehicle speed and throttle opening, the target driver operation dataset is discretized and grouped to obtain multiple vehicle speed intervals and multiple throttle opening intervals.

[0090] In this application, discretization grouping divides the continuously changing physical quantities such as vehicle speed and throttle opening into several discrete bin intervals to adapt to the lookup table index logic.

[0091] Vehicle speed ranges can be divided according to fixed speed spans, and throttle opening ranges can be divided according to the percentage of pedal opening, so that data within the same range remain similar in terms of power demand.

[0092] In the specific implementation process of this step, the vehicle speed and throttle opening fields are first extracted from the target driver operation dataset, and the vehicle speed is discretely segmented according to the preset interval width. At the same time, the throttle opening is discretely segmented to form several vehicle speed intervals and throttle opening intervals.

[0093] S202. Collect triplet data from multiple vehicle speed ranges and multiple throttle opening ranges to generate an acceleration mapping table.

[0094] In this application, the ternary data includes a combination of data consisting of vehicle speed, throttle opening, and longitudinal acceleration, which is used to construct an entry in the acceleration mapping table.

[0095] In practice, samples falling into the same speed range and throttle opening range are aggregated, and the combination of speed, throttle opening and acceleration at the corresponding time is statistically analyzed to form triplet data and written into the acceleration mapping table.

[0096] The acceleration mapping table can be stored in the form of a lookup table, where each entry corresponds to a vehicle speed range and a throttle opening range, and records the representative acceleration value or acceleration distribution value within that range.

[0097] In one possible implementation, multiple discrete intervals are first divided according to the vehicle speed sampling range and the throttle opening sampling range, and the corresponding operating data are aggregated in each vehicle speed interval and each throttle opening interval to form a triplet data set.

[0098] Subsequently, quantiles are calculated for acceleration samples within the same vehicle speed range and the same throttle opening range. The quantile threshold used can be set as a fixed percentile based on the sample dispersion to obtain the upper limit of acceleration under this combination of conditions.

[0099] For intervals with a small number of samples, supplementary samples from adjacent intervals can be introduced for joint statistics to improve the stability of the upper limit of acceleration.

[0100] After the statistics are completed, the vehicle speed range identifier, throttle opening range identifier and the corresponding acceleration upper limit value are written into the mapping table to form an acceleration mapping table that can be called by the vehicle controller.

[0101] The acceleration mapping table is used to quickly find the upper limit of the target acceleration according to the current vehicle speed and throttle opening interval during operation, and uses it as the constraint basis for torque adjustment, so that the control output maintains a corresponding relationship with the current operating conditions.

[0102] Since the upper limit of acceleration is obtained through quantile statistics, the mapping table can reflect the typical acceleration capability within different ranges and suppress the influence of outliers. Therefore, when the vehicle calls this mapping table under complex operating conditions, it can obtain more stable target acceleration constraint values ​​and generate more consistent vehicle torque control commands accordingly.

[0103] This step uses quantile statistics to obtain the upper limit of acceleration corresponding to excellent drivers and suppresses the interference of abnormal samples; the adjacent interval sample supplementation strategy solves the problem of insufficient samples in sparse operating condition intervals, ensuring that the output of each table item of the whole table is a stable and usable target acceleration constraint value.

[0104] S203. Calculate the torque change rate of the target driver's operation dataset and generate a torque slope mapping table.

[0105] In this application, the torque slope mapping table can record the allowable torque change slope or statistical slope in each discrete interval, which is used for smoothness constraints during subsequent torque control.

[0106] In the specific implementation, the torque difference is first calculated for adjacent sampling points in the target driver operation dataset, and the torque change rate corresponding to each sampling point is obtained by combining the sampling timestamp; for the first sampling point, the difference between the next sampling point and the point can be used as the initial value.

[0107] Subsequently, each sampling point is mapped to the corresponding vehicle speed range and throttle opening range according to its vehicle speed and throttle opening, respectively, and all torque change rate samples are aggregated within the same range. When performing quantile statistics on the aggregated torque change rate samples, a preset quantile threshold can be set as the statistical position reflecting the upper limit of the range. For example, the high quantile value of the samples in the range can be taken as the upper limit of the torque change rate to characterize the characteristic boundary of torque change under the conditions of vehicle speed and throttle opening. For ranges with insufficient sample size, samples from adjacent ranges can be combined for statistical analysis to ensure the stability and table lookupability of the upper limit of the torque change rate.

[0108] For example, the torque change rate is calculated for each sampling point. For historical operating data, the formula for calculating the torque change rate is: .

[0109] Then, the segments were grouped according to speed range and throttle opening range. For each group, the best segments were counted. The 90th percentile (or other higher percentile) is denoted as .

[0110] Finally, when generating the torque slope mapping table, a one-to-one correspondence is established between each vehicle speed range, each throttle opening range, and the corresponding upper limit of torque change rate, and stored in the form of a table, matrix, or hash mapping for subsequent online control to retrieve based on real-time vehicle speed and throttle opening.

[0111] In this application, the torque slope mapping table can be deployed in the vehicle-side controller or the back-end computing unit. The query results serve as the torque slope constraint input, used to limit the magnitude boundary of torque variation. If a specific vehicle model uses different torque units or sampling frequencies, unit unification and time scale normalization can be performed before generating the mapping table. In practical applications, other models of this component can also be selected, and this application does not impose any limitations on this.

[0112] The torque slope mapping table obtained in this step can reflect the torque change characteristics of excellent driving samples under different vehicle speeds and throttle opening conditions, so that the torque slope constraint is consistent with the operating habits of the target driver. When called online, the upper limit of the torque change rate of the corresponding interval can be directly obtained, thereby providing a stable table lookup basis for vehicle torque control commands and improving the matching and consistency of torque adjustment under complex operating conditions.

[0113] In this embodiment, the acceleration mapping table and torque slope mapping table are output as calibration data that can be accessed by the vehicle. During online operation, the vehicle controller locates the corresponding interval based on the current vehicle speed and throttle opening, reads the corresponding acceleration mapping value and torque slope value, and generates vehicle torque control commands accordingly, enabling the driving characteristics in the offline samples to be mapped to the real-time control process. Using this method, vehicle speed and throttle opening are uniformly discretized into fixed intervals, the statistical objects of the mapping table are clearer, and acceleration and torque change characteristics can be stored according to operating conditions, thereby improving the consistency between the vehicle torque control commands and the current driving state and reducing reliance on manual calibration experience.

[0114] Figure 3 The flowchart of Embodiment 3 of the adaptive torque control method for commercial vehicles provided in this application is shown below. Figure 3 As shown, based on real-time operating data, acceleration mapping table, and torque slope mapping table in the online mode of commercial vehicles, the vehicle torque control commands are dynamically adjusted and output, including:

[0115] S301: Collects real-time operating data of commercial vehicles in online mode.

[0116] In this application, real-time operating data is periodically collected by the vehicle controller in online mode, including current vehicle speed, throttle opening, load, and gradient. The current vehicle speed can be obtained from wheel speed sensors or a vehicle speed signal interface; the throttle opening is obtained from a pedal position sensor; the load can be obtained from an onboard weighing module or based on axle load estimation results; and the gradient can be jointly determined by an inertial measurement unit, attitude sensors, or map gradient information.

[0117] This step fully acquires all the operating condition inputs required for online control, enabling the control strategy to perceive the actual load, gradient, vehicle speed, and pedal status, providing complete raw inputs for table lookup, dynamic resistance calculation, and slope adaptive correction.

[0118] S302. Match the real-time running data with the acceleration mapping table to determine the target acceleration.

[0119] In this application, the acceleration mapping table can be stored in the non-volatile memory of the controller. Its entries are established according to the vehicle speed range, throttle opening range, load range and gradient range. The acceleration value in the table can represent the upper limit of acceleration that is allowed or the target can be achieved under the current working conditions.

[0120] The target acceleration is obtained from offline excellent driving samples and is the upper limit of vehicle acceleration allowed under the current load-gradient-speed-throttle conditions.

[0121] In this step, the vehicle controller inputs the real-time operating data into the mapping table, first looks up the corresponding interval based on the vehicle speed and throttle opening, and then corrects the lookup result by combining the load and slope to obtain the target acceleration that matches the current operating state.

[0122] This step outputs a target acceleration that adapts to the current real-world working conditions based on the offline learning results, replacing the fixed calibration acceleration value. The target acceleration is consistent with the acceleration habits of excellent drivers under the same load and gradient scenarios.

[0123] S303. Based on the target acceleration and torque slope mapping table, dynamically adjust and output vehicle torque control commands.

[0124] In this step, after determining the target acceleration, the vehicle controller continues to read the torque slope mapping table corresponding to the same or adjacent operating condition intervals. The torque slope mapping table is used to limit the incremental range of torque command changes over time.

[0125] In practice, the vehicle controller calculates the required torque change direction based on the deviation between the target acceleration and the current power output state, and constrains the torque increase or decrease rate according to the torque slope mapping table, thereby forming continuously updated vehicle torque control commands.

[0126] The vehicle torque control command can be written into the input interface of the engine control unit or motor control unit to ensure that the power output changes smoothly while meeting the target acceleration requirements. If the real-time operating data changes, the vehicle controller will rematch the mapping table and update the target acceleration and torque commands in the next control cycle, so that the output is adjusted synchronously with changes in vehicle speed, throttle opening, load, and gradient.

[0127] In one possible implementation, the vehicle controller collects the current vehicle speed, throttle opening, load, and gradient, and inputs these data into the vehicle dynamics model to calculate the required torque limit. The vehicle dynamics model can establish a torque balance relationship based on driving force, rolling resistance, slope resistance, and acceleration resistance, and then combine this with the transmission ratio, transmission efficiency, and effective wheel radius to calculate the torque limit.

[0128] Subsequently, the speed range and throttle opening range corresponding to the limit are indexed and matched with the torque slope mapping table to read the allowable target torque change rate in the current scenario, and torque change constraints are formed accordingly.

[0129] In this application, the vehicle's current total resistance must first be considered. After determining the target acceleration, the vehicle controller reads the torque slope mapping table for the corresponding operating range to obtain the allowable unit-time increment range of the torque command; it determines the direction of torque change by combining the deviation between the target acceleration and the current power output state, and generates continuously updated torque control commands to be sent to the engine control unit or motor control unit according to the torque slope constraint of the torque rise / fall rate. If the real-time operating data changes, the vehicle controller will rematch the mapping table and update the target acceleration and torque command in the next control cycle, so that the output is adjusted synchronously with changes in vehicle speed, throttle opening, load, and gradient.

[0130] In practice, the vehicle controller collects the current vehicle speed, throttle opening, load, and gradient, and substitutes these data into the vehicle dynamics model to calculate the required torque limit. For example, the total resistance is calculated based on the vehicle dynamics model using the following formula:

[0131]

[0132] Among them, F res The total resistance is the sum of rolling resistance, gradient resistance, and air resistance during vehicle operation.

[0133] f represents the rolling resistance coefficient (e.g., 0.01); at room temperature .

[0134] Then, combined with the target acceleration a des The formula for calculating demand drivers is as follows:

[0135]

[0136] in, This represents the rotational mass coefficient, which can be simply taken as 1.08.

[0137] Further calculation of the required motor torque is performed using the following formula:

[0138]

[0139] in, For the wheel radius, The total transmission ratio is... For transmission efficiency.

[0140] Subsequently, the vehicle speed range and throttle opening range corresponding to this limit are indexed and matched with the torque slope mapping table to read the basic torque slope coefficient TrqGrad. base Among them, the basic torque slope coefficient TrqGrad base This is the upper limit of the basic torque change rate obtained by looking up the torque slope mapping table.

[0141] Based on the current total resistance T res With the nominal reference resistance T res_nom Calculate the resistance correction factor λ, where the nominal reference resistance T res_nom λ represents the vehicle's driving resistance under the preset benchmark conditions (flat road, 60 kph, constant speed of 49 t). λ is a correction factor that adaptively scales the base torque slope based on the deviation of the current driving resistance from the benchmark nominal resistance. The specific formula for calculating λ is as follows:

[0142]

[0143] Ultimately, the upper limit of the actual torque slope obtained is .

[0144] Among them, when in uphill, heavy load conditions, i.e. When λ is greater than 1, the slope increases, allowing for a larger torque variation slope.

[0145] When in downhill, light-load conditions, i.e. When λ is less than 1, the slope decreases, limiting the slope of torque change.

[0146] Based on the above analysis, after obtaining the target acceleration and the target torque change rate after resistance correction, the vehicle controller continuously corrects the required torque limit to ensure that the corrected torque output simultaneously meets the acceleration target and the change rate constraint. Specifically, the vehicle controller determines the torque correction direction based on the deviation between the target acceleration and the current actual acceleration, then limits the torque increment or decrement within a unit control cycle based on the target torque change rate, and finally outputs the vehicle torque control command to the engine control unit or motor control unit. This command is recursively updated within the continuous control cycle and is adjusted synchronously with changes in real-time operating data.

[0147] This step allows the vehicle's torque output to be simultaneously constrained by dynamic requirements and torque variation characteristics, ensuring that the vehicle's torque control command remains consistent with the current operating conditions and that acceleration response and torque switching are smoother, thereby improving control matching under complex load and gradient conditions.

[0148] This application embodiment periodically collects real-time operating data such as vehicle speed, throttle opening, load, and gradient through the vehicle controller. It then obtains the target acceleration adapted to the current operating conditions by looking up a table and matching it. Based on the torque slope mapping table and introducing a resistance correction coefficient, it adaptively scales the basic torque slope. Relying on the vehicle dynamics model, it calculates the required torque limit from the target acceleration, realizing dual constraint control of acceleration feedforward torque limiting and torque slope limiting. This not only outputs torque that conforms to the operating habits of excellent drivers, suppresses aggressive driving behaviors such as sudden acceleration and deceleration, and ensures a smooth transition of torque output, but also adapts to changes in load and gradient, improving the matching of torque control commands with actual operating scenarios under complex conditions.

[0149] Figure 4 This is a schematic diagram of the structure of the adaptive torque control device for commercial vehicles provided in this application. Figure 4 As shown, the commercial vehicle adaptive torque control device 40 provided in this application includes:

[0150] The operation data acquisition module 401 is used to acquire target driver operation datasets under different offline modes. The target driver operation datasets contain excellent driving samples under different loads and slopes.

[0151] Data mapping module 402 is used to group the target driver operation dataset to generate an acceleration mapping table and a torque slope mapping table;

[0152] The torque control module 403 is used to dynamically adjust and output vehicle torque control commands based on real-time operating data, acceleration mapping table and torque slope mapping table in the online mode of commercial vehicles.

[0153] In one possible implementation, the data mapping module 402 is further specifically used for:

[0154] Based on vehicle speed and throttle opening, the target driver operation dataset is discretized and grouped to obtain multiple vehicle speed intervals and multiple throttle opening intervals.

[0155] Collect triplet data from multiple vehicle speed ranges and multiple throttle opening ranges to generate an acceleration mapping table. The triplet data includes vehicle speed, throttle opening, and acceleration.

[0156] Calculate the torque change rate of the target driver's operation dataset and generate a torque slope mapping table.

[0157] In one possible implementation, the data mapping module 402 is further specifically used for:

[0158] Collect ternary data for multiple vehicle speed ranges and multiple throttle opening ranges;

[0159] Quantile statistics are performed on the triplet data to obtain the upper limit of acceleration.

[0160] An acceleration mapping table is generated based on multiple vehicle speed ranges, multiple throttle opening ranges, and acceleration upper limit values.

[0161] In one possible implementation, the data mapping module 402 is further specifically used for:

[0162] Calculate the rate of torque change at each sampling point in the target driver's operation dataset;

[0163] Based on multiple vehicle speed ranges and multiple throttle opening ranges, the quantile statistics of the torque change rate are performed to obtain the upper limit of the torque change rate.

[0164] A torque slope mapping table is generated based on multiple vehicle speed ranges, multiple throttle opening ranges, and the upper limit of torque change rate.

[0165] In one possible implementation, the torque control module 403 is further specifically used for:

[0166] Collect real-time operating data of commercial vehicles in online mode. The real-time operating data includes current vehicle speed, throttle opening, load and gradient.

[0167] The target acceleration is determined by matching real-time running data with an acceleration mapping table.

[0168] Based on the target acceleration and torque slope mapping table, the vehicle torque control command is dynamically adjusted and output.

[0169] In one possible implementation, the torque control module 403 is further specifically used for:

[0170] Calculate the required torque limit for commercial vehicles based on the vehicle dynamics model;

[0171] Match the required torque limit with the torque slope mapping table to determine the target torque change rate;

[0172] Based on the target acceleration and the target torque change rate, the required torque limit is dynamically adjusted, and the vehicle torque control command is output.

[0173] In one possible implementation, the operation data acquisition module 401 is further specifically used for:

[0174] Collect historical operating data of commercial vehicles in offline mode;

[0175] Historical operational data is preprocessed to obtain cleaned historical operational data;

[0176] Based on load and slope, the working conditions are divided into scenarios, and excellent driver samples that meet the vehicle energy consumption requirements are selected from the cleaned historical operating data.

[0177] Based on excellent driver samples, obtain target driver operation datasets under different offline modes.

[0178] The adaptive torque control device for commercial vehicles provided in this embodiment extracts excellent driving samples under different load and gradient conditions through an operational data acquisition module. This transforms actual energy-saving and smooth driving behaviors into a usable data foundation. The data mapping module then groups these samples by scenario to form acceleration and torque slope mapping tables, providing a structured representation of acceleration habits and torque variation patterns from offline samples. This provides corresponding control basis for different operating conditions. The torque control module dynamically adjusts the torque output by invoking the corresponding mapping relationships in conjunction with online real-time operating data. This allows the torque output to adaptively match changes in load, gradient, and operating status, thereby improving issues such as stiff, sluggish, or disjointed responses under fixed calibration. Therefore, it helps to balance power, smoothness, and energy consumption, and reduces reliance on manual calibration and high-precision road information in complex scenarios.

[0179] Figure 5 This is a schematic diagram of the structure of the adaptive torque control device for commercial vehicles provided in this application. Figure 5 As shown, the adaptive torque control device 50 for commercial vehicles provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the adaptive torque control device 50 for commercial vehicles further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0180] In the specific implementation process, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to execute the above-mentioned adaptive torque control method for commercial vehicles.

[0181] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0182] The memory stores the scene mapping relationships formed during offline learning, the constraint results of excellent driving samples, and torque control parameters. After the processor calls the computer to execute instructions, it can combine real-time vehicle speed, pedal opening, load, and gradient data to execute the corresponding adaptive torque control method at the vehicle end, thereby stably deploying the control data obtained from the cloud or backend to the vehicle controller. During operation, the processor outputs torque control instructions based on the acceleration mapping table and torque slope mapping table, enabling the torque output to be dynamically adjusted according to changes in current load, gradient, and driving operation. This improves the problem of fixed calibration rules responding too stiffly, too slowly, or with inconsistent switching under complex conditions. Therefore, it can balance energy consumption, power performance, and smoothness in high-frequency, high-energy-consumption scenarios such as coal transportation, express delivery, and long-haul logistics, while reducing manual calibration and subsequent maintenance costs.

[0183] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0184] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0185] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0186] This application also provides a control system for adaptive torque in a commercial vehicle, including a computer program that, when executed by a processor, implements the aforementioned control method for adaptive torque in a commercial vehicle.

[0187] This computer program can be deployed in the vehicle controller or the processor platform that communicates with it. When executed, it calls upon the acceleration mapping and torque slope mapping relationships formed through offline learning, and combines them with real-time operating parameters such as vehicle speed, pedal opening, load, and gradient to dynamically match the current operating conditions and output torque control commands. This allows torque adjustment to no longer rely solely on fixed calibration rules. Because the program can transform excellent driving samples under different load and gradient conditions into executable control criteria, it can more accurately match actual power requirements in scenarios with frequent starts and stops, continuous climbing, and load fluctuations. This results in a better balance between torque response, energy consumption, and overall vehicle smoothness, thereby reducing the burden of manual calibration and improving applicability in complex transportation conditions.

[0188] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the aforementioned adaptive torque control method for commercial vehicles.

[0189] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0190] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0191] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0192] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0193] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0194] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0195] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0196] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for controlling adaptive torque in a commercial vehicle, characterized in that, include: Obtain target driver operation datasets under different offline modes, wherein the target driver operation datasets contain excellent driving samples under different loads and slopes; The target driver operation dataset is grouped to generate an acceleration mapping table and a torque slope mapping table. Based on the real-time operating data of the commercial vehicle in online mode, the acceleration mapping table, and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output.

2. The control method according to claim 1, characterized in that, The target driver operation dataset is grouped and processed to generate an acceleration mapping table and a torque slope mapping table, including: Based on vehicle speed and throttle opening, the target driver operation dataset is discretized and grouped to obtain multiple vehicle speed intervals and multiple throttle opening intervals. Collect triplet data from the multiple vehicle speed ranges and the multiple throttle opening ranges to generate an acceleration mapping table. The triplet data includes vehicle speed, throttle opening, and acceleration. Calculate the torque change rate of the target driver operation dataset and generate a torque slope mapping table.

3. The control method according to claim 2, characterized in that, Collect triplet data from the multiple vehicle speed ranges and the multiple throttle opening ranges to generate an acceleration mapping table, including: Collect triplet data for the multiple vehicle speed ranges and the multiple throttle opening ranges; The upper limit of acceleration is obtained by performing quantile statistics on the triplet data; An acceleration mapping table is generated based on the multiple vehicle speed ranges, the multiple throttle opening ranges, and the upper limit of acceleration.

4. The control method according to claim 2, characterized in that, Calculate the torque change rate of the target driver operation dataset and generate a torque slope mapping table, including: Calculate the rate of change of torque at each sampling point in the target driver operation dataset; Based on the multiple vehicle speed ranges and the multiple throttle opening ranges, the quantile statistics of the torque change rate are performed to obtain the upper limit of the torque change rate. A torque slope mapping table is generated based on the multiple vehicle speed ranges, the multiple throttle opening ranges, and the upper limit of torque change rate.

5. The control method according to claim 1, characterized in that, Based on real-time operating data of the commercial vehicle in online mode, the acceleration mapping table, and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output, including: Collect real-time operating data of commercial vehicles in online mode, including current vehicle speed, throttle opening, load and gradient; The real-time running data is matched with the acceleration mapping table to determine the target acceleration; Based on the target acceleration and the torque slope mapping table, the vehicle torque control command is dynamically adjusted and output.

6. The control method according to claim 5, characterized in that, Based on the target acceleration and the torque slope mapping table, dynamically adjust and output vehicle torque control commands, including: Calculate the required torque limit for the commercial vehicle based on the vehicle dynamics model; The required torque limit is matched with the torque slope mapping table to determine the target torque change rate; Based on the target acceleration and the target torque change rate, the required torque limit is dynamically adjusted, and a vehicle torque control command is output.

7. The control method according to claim 1, characterized in that, Obtain target driver operation datasets under different offline modes, including: Collect historical operating data of commercial vehicles in offline mode; The historical operation data is preprocessed to obtain cleaned historical operation data; Based on load and slope, the working conditions are divided into scenarios, and excellent driver samples that meet the vehicle energy consumption requirements are selected from the historical operating data after cleaning. Based on the excellent driver samples, obtain target driver operation datasets under different offline modes.

8. A control device for adaptive torque in commercial vehicles, characterized in that, include: The operation data acquisition module is used to acquire target driver operation datasets under different offline modes. The target driver operation datasets contain excellent driving samples under different loads and slopes. The data mapping module is used to group the target driver operation dataset to generate an acceleration mapping table and a torque slope mapping table. The torque control module is used to dynamically adjust and output vehicle torque control commands based on real-time operating data of the commercial vehicle in online mode, the acceleration mapping table, and the torque slope mapping table.

9. A control device for adaptive torque in commercial vehicles, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A control system for adaptive torque in a commercial vehicle, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images