Vehicle control method, electronic equipment and vehicle

Abstract

The invention relates to the technical field of vehicle intelligent driving, and provides a vehicle control method, electronic equipment and a vehicle, the opening degree change of an accelerator pedal is monitored, historical environment data of the environment where the vehicle is located is obtained, and predicted temperature difference data is determined according to the historical environment data; the actual accelerator pedal opening degree is obtained, and the initial motor output power is determined according to the actual accelerator pedal opening degree and the predicted temperature difference data; acquiring actual motor output power, and determining a target power correction value according to the predicted temperature difference data, the initial motor output power and the actual motor output power; and the predicted temperature difference data, the initial motor output power and the target power correction value are input into a power determination model obtained through pre-training, target motor output power is obtained through processing of the power determination model, and vehicle driving is controlled based on the target motor output power. The power output parameters are adjusted in real time according to the working condition fluctuation caused by the day and night temperature difference, and the driving smoothness under the extreme climate is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of intelligent driving of vehicles, and in particular to a vehicle control method, an electronic device and a vehicle. BACKGROUND

[0002] With the rapid development of the vehicle technology field, vehicles have become an important means of transportation in people's daily life. The current vehicle power system generally lacks the ability to perceive and respond to the dynamic changes of day-night temperature difference in the control strategy, resulting in a significant deviation between power output and driving instructions during the early morning and evening when the temperature changes rapidly. Especially in the low-temperature, high-resistance or electric control limited power state of the battery, the power response is delayed, the output is insufficient, or the energy consumption is abnormal. SUMMARY

[0003] Therefore, the purpose of the present disclosure is to provide a vehicle control method, an electronic device and a vehicle to solve the problem that the current control system relies on static threshold or fixed power table for adjustment, cannot identify the change trend of temperature difference, and cannot predict the change of system response caused by temperature difference, resulting in strategy lag or insufficient correction, affecting the driving smoothness and system safety.

[0004] To achieve the above purpose, the first aspect of the present disclosure provides a vehicle control method, which comprises: monitoring the change of the accelerator pedal opening degree, obtaining historical environment data of the environment where the vehicle is located, and determining predicted temperature difference data according to the historical environment data; obtaining the actual accelerator pedal opening degree, and determining the initial motor output power according to the actual accelerator pedal opening degree and the predicted temperature difference data; obtaining the actual motor output power, and determining the target power correction value according to the predicted temperature difference data, the initial motor output power and the actual motor output power; inputting the predicted temperature difference data, the initial motor output power and the target power correction value into a power determination model trained in advance, processing via the power determination model to obtain the target motor output power, and controlling the vehicle to travel based on the target motor output power.

[0005] Based on the same inventive concept, the second aspect of the present disclosure provides a vehicle control device, which comprises: a data acquisition module configured to monitor the change of the accelerator pedal opening degree, obtain historical environment data of the environment where the vehicle is located, and determine predicted temperature difference data according to the historical environment data; an initial power determination module configured to obtain the actual accelerator pedal opening degree, and determine the initial motor output power according to the actual accelerator pedal opening degree and the predicted temperature difference data; The correction value determination module is configured to obtain an actual motor output power, and determine a target power correction value according to the predicted temperature difference data, the initial motor output power, and the actual motor output power. The target power determination module is configured to input the predicted temperature difference data, the initial motor output power, and the target power correction value into a power determination model trained in advance, obtain a target motor output power via processing of the power determination model, and control vehicle driving based on the target motor output power.

[0006] Based on the same inventive concept, a third aspect of the present disclosure provides an electronic device including a memory, a processor, and a computer program stored on the memory and executable by the processor, wherein the processor implements the vehicle control method as described above when executing the computer program.

[0007] Based on the same inventive concept, a fourth aspect of the present disclosure provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the vehicle control method as described above.

[0008] Based on the same inventive concept, a fifth aspect of the present disclosure provides a vehicle including the vehicle control device of the second aspect, the electronic device of the third aspect, or the storage medium of the fourth aspect.

[0009] It can be seen from the above that the present disclosure proposes a vehicle control method, an electronic device and a vehicle. The change in the accelerator pedal opening is monitored, historical environment data of the environment where the vehicle is located is obtained, and predicted temperature difference data is determined according to the historical environment data. The actual accelerator pedal opening is obtained, and the initial motor output power is determined according to the actual accelerator pedal opening and the predicted temperature difference data. The temperature difference of the environment where the vehicle is located is predicted according to the historical environment data, and then the influence of the temperature difference is considered when the initial motor output power of the vehicle is determined, so that the determination of the initial motor output power is more accurate. The actual motor output power is obtained, and the target power correction value is determined according to the predicted temperature difference data, the initial motor output power and the actual motor output power. The target power correction value represents the deviation value between the predicted initial motor output power and the actual motor output power. The predicted temperature difference data, the initial motor output power and the target power correction value are input into a power determination model trained in advance, and the target motor output power is obtained through processing of the power determination model. The initial motor output power represents the power demand level that should be had under the current driving behavior and environmental conditions. Therefore, when the target motor output power is determined, the environmental temperature difference, the ideal power demand under the environmental temperature difference and the actual deviation value are considered comprehensively, so that the determination of the target motor output power is more accurate. Then, the vehicle is controlled to travel based on the target motor output power, the power output parameter can be adjusted in real time according to the working condition fluctuation caused by the day and night temperature difference, and the driving smoothness in extreme climate is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present disclosure or the related art, the drawings needed to be used in the embodiments or the related art description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor.

[0011] Figure 1 Flow chart of the vehicle control method of the embodiments of the present disclosure; Figure 2 Structure block diagram of the vehicle control device of the embodiments of the present disclosure; Figure 3 Structure schematic diagram of the electronic device of the embodiments of the present disclosure. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, the present disclosure will be further described in detail below with reference to specific embodiments and drawings.

[0013] It should be noted that, unless otherwise defined, technical terms or scientific terms used in the embodiments of the present disclosure shall have the common meaning understood by one of ordinary skill in the art to which the present disclosure belongs. The terms "first", "second" and similar terms used in the embodiments of the present disclosure do not denote any order, quantity or importance, but are only used to distinguish different components. The terms "include" or "contain" and similar terms mean that the elements or objects before the terms encompass the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connected" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0014] The terms involved in the present disclosure are as follows: Prophet: Prophet is designed for business forecasting scenarios, which can automatically handle trend changes, seasonality and holiday effects, and is robust to missing values and outliers DeepFM: DeepFM (Deep Factorization Machine) is a recommendation system model that combines factorization machines (FM) and deep neural networks (DNN). It efficiently learns low-order linear and interaction relationships between features through the FM part, while capturing high-order nonlinear feature combinations using the DNN part.

[0015] SVR: SVR (Support Vector Regression) is a regression algorithm based on support vector machines (SVM), suitable for predicting continuous value problems. The core idea is to find a "boundary zone", allowing a small number of samples to fall outside the boundary during training, to balance model complexity and prediction accuracy, and to avoid overfitting.

[0016] SOC: State of Charge (SOC), is a core parameter to measure the ratio of remaining available capacity to fully charged state capacity of the battery, usually expressed as a percentage. The SOC value ranges from 0% to 100%, where 0% indicates a fully discharged battery and 100% indicates a fully charged battery.

[0017] With the rapid development of the vehicle technology field, vehicles have become an important means of transportation in people's daily life. The current vehicle power system generally lacks the ability to perceive and respond to the dynamic changes of day and night temperature differences in the control strategy, resulting in a significant deviation between power output and driving instructions during the early morning and evening periods when the temperature changes rapidly. Especially in the low-temperature, high-resistance or electric control limited power state of the battery, power response delay, insufficient output or abnormal energy consumption problems are prone to occur.

[0018] The occurrence of such problems is mainly due to the fact that the existing control system relies on static threshold or fixed power table for adjustment, which cannot identify the temperature difference change trend, nor can it predict the system response change caused by temperature difference, thereby leading to strategy lag or insufficient correction, affecting driving smoothness and system safety.

[0019] Therefore, the embodiment proposes a vehicle control method, as shown in the figure, the method comprises: Figure 1 As shown in the figure, the method comprises: Step 101, monitor the change of the accelerator pedal opening degree, obtain the historical environment data of the environment where the vehicle is located, and determine the predicted temperature difference data according to the historical environment data.

[0020] In specific implementation, when the change of the accelerator pedal opening degree is monitored, it indicates that the vehicle power will change. Because the temperature change has a great influence on the power output, such as in the morning and evening periods when the temperature changes rapidly, there is a significant deviation between the power output and the driving instruction. Therefore, in order to achieve accurate power output, the day and night temperature difference of the environment where the vehicle is located needs to be considered at this time.

[0021] In the embodiment, the historical environment data of the environment where the vehicle is located is obtained, and the predicted temperature difference data is determined according to the historical environment data. In order to ensure the accuracy of the predicted temperature difference data, the historical environment data is preferably the historical environment data in a preset time period before the current time. For example, the historical environment data is the day and night temperature difference data in the past 7 to 14 days before the current time, wherein each day corresponds to a day and night temperature difference data, and the day and night temperature difference data represents the temperature difference between day and night, i.e. the day and night temperature difference data is the average temperature difference between day and night in each daily cycle.

[0022] In the embodiment, the predicted temperature difference data is the temperature difference data in a preset time period after the current time, for example, the predicted temperature difference data is the hourly temperature difference prediction sequence in the next 12 hours.

[0023] Step 102, obtaining the actual accelerator pedal opening degree, determining the initial motor output power according to the actual accelerator pedal opening degree and the predicted temperature difference data.

[0024] In specific implementation, the actual accelerator pedal opening degree is obtained, and the initial motor output power is determined according to the actual accelerator pedal opening degree combined with the predicted temperature difference data, wherein the initial motor output power is used to describe the power response that the vehicle should have under ideal conditions corresponding to the actual accelerator pedal opening degree.

[0025] In this embodiment, a pre-trained power determination model is obtained, the actual accelerator pedal opening and the predicted temperature difference data are input into the power determination model, and an initial motor output power is obtained through processing of the power determination model.

[0026] In step 103, an actual motor output power is obtained, and a target power correction value is determined according to the predicted temperature difference data, the initial motor output power and the actual motor output power.

[0027] In specific implementation, the actual motor output power corresponding to the actual accelerator pedal opening is obtained, and the target power correction value is determined according to the predicted temperature difference data, the initial motor output power and the actual motor output power. The target power correction value is an estimated value of the deviation between the initial motor output power and the actual motor output power.

[0028] In this embodiment, a pre-trained deviation determination model is obtained, the predicted temperature difference data, the initial motor output power and the actual motor output power are input into the deviation determination model, and a target power correction value is obtained through processing of the deviation determination model.

[0029] In step 104, the predicted temperature difference data, the initial motor output power and the target power correction value are input into a pre-trained power determination model, a target motor output power is obtained through processing of the power determination model, and vehicle driving is controlled based on the target motor output power.

[0030] In specific implementation, a pre-trained power determination model is obtained, the predicted temperature difference data, the initial motor output power and the target power correction value are input into the pre-trained power determination model, a target motor output power is obtained through processing of the power determination model, and vehicle driving is controlled based on the target motor output power.

[0031] In this embodiment, the training process of the power determination model specifically includes: In step a, a first training data set and an initial power determination model are obtained, and the first training data set includes historical temperature difference data, historical initial motor output power, historical power correction value and historical motor output power.

[0032] In step b, training data in the first training data set is input into the initial power determination model for training, a first preset training end condition is met, and a power determination model is obtained.

[0033] In implementation, a first training data set and an initial power determination model are obtained, wherein the first training data set comprises historical temperature difference data, historical initial motor output power, historical power correction value and historical motor output power. Training data in the first training data set is input into the initial power determination model for training. When a first preset training end condition is met, a power determination model is obtained.

[0034] The first preset training end condition comprises at least one of the following: determining that all data in the first training data set is input into the initial power determination model for training, determining that a loss function of the initial power determination model converges to a first convergence threshold, or determining that the initial power determination model is iteratively trained to a first preset iteration number.

[0035] For example, the first preset training end condition is that all data in the first training data set is input into the initial power determination model for training. The first training data set comprises fifty groups of data, and each group of data comprises historical temperature difference data, historical initial motor output power, historical power correction value and historical motor output power. The first preset training end condition is that all data in the first training data set is input into the initial power determination model for training, that is, when the fifty groups of data are input into the initial power determination model, there is no training data in the first training data set that has not been input into the initial power determination model, and the initial power determination model training is determined to be completed, and a power determination model is obtained.

[0036] For another example, the first preset training end condition is that the loss function of the initial power determination model converges to the first convergence threshold. The training data in the first training data set is input into the initial power determination model for training, and a training result is output. According to the training result and the historical fault type, historical fault handling method and historical fault cause, a loss function is determined. The type of the loss function comprises at least one of the following: mean square error loss function, cross-entropy loss function, logarithmic loss function, exponential loss function, square loss function or absolute value loss function. When the loss function converges to the first convergence threshold, it is determined that the first preset training end condition is met, and a power determination model is obtained.

[0037] For another example, the first preset training end condition is that the initial power determination model is iteratively trained to the first preset iteration number.

[0038] The training data in the first training data set is input into the initial power determination model for iterative training, and the iteration number is recorded. When the iteration number is equal to the first preset iteration number, the first preset training end condition is met, and a power determination model is obtained.

[0039] By the above scheme, the training data in the first training data set includes historical temperature difference data, historical initial motor output power, historical power correction value and historical motor output power. The initial power determination model is trained by using the training data in the first training data set until the first preset training end condition is met, and the power determination model is obtained, so as to output the target motor output power corresponding to the predicted temperature difference data, the initial motor output power and the target power correction value by using the power determination model in the subsequent process. The determined target motor output power is more accurate.

[0040] By the above scheme, the change of the accelerator pedal opening is monitored, the historical environment data of the environment where the vehicle is located is obtained, and the predicted temperature difference data is determined according to the historical environment data. The actual accelerator pedal opening is obtained, and the initial motor output power is determined according to the actual accelerator pedal opening and the predicted temperature difference data. The temperature difference of the environment where the vehicle is located is predicted according to the historical environment data, and then the influence of the temperature difference is considered when the initial motor output power of the vehicle is determined, so that the determination of the initial motor output power is more accurate. The actual motor output power is obtained, and the target power correction value is determined according to the predicted temperature difference data, the initial motor output power and the actual motor output power. The target power correction value represents the deviation value between the predicted initial motor output power and the actual motor output power. The predicted temperature difference data, the initial motor output power and the target power correction value are input into the power determination model trained in advance, and the target motor output power is obtained by processing the power determination model. The initial motor output power represents the power demand level under the current driving behavior and environmental conditions. Therefore, when the target motor output power is determined, the environmental temperature difference, the ideal power demand under the environmental temperature difference and the actual deviation value are considered comprehensively, so that the determination of the target motor output power is more accurate. Then, the vehicle is controlled to travel based on the target motor output power, and the power output parameter can be adjusted in real time according to the working condition fluctuation caused by the day and night temperature difference, so as to significantly improve the driving smoothness in extreme climate.

[0041] In some embodiments, the predicted temperature difference data can be obtained by analyzing the historical environment data by a temperature difference determination model. The temperature difference determination model adopts a Prophet model based on additive time series decomposition, has an adaptive fitting capability for trend, seasonality and holiday influence, and is especially suitable for processing environmental variable change problems with periodicity and mutation characteristics. By using the time series modeling method, the day and night temperature difference change trend in the future several hours is predicted, so as to provide the deviation determination model with advanced temperature disturbance reference information.

[0042] The input data of the temperature difference determination model is derived from a subset of environmental variables, including the day and night temperature difference records of the past 7 to 14 consecutive days. The specific calculation method is the difference between the average temperature during the day and the average temperature during the night in each day cycle. Auxiliary variables are introduced to improve prediction accuracy, including humidity, wind speed, sunshine duration, weather label (sunny, rainy, snowy), sunshine start and end time, geographical location information (latitude, altitude), and solstice encoding to capture the temperature difference fluctuation characteristics caused by natural seasonal changes.

[0043] The model structure of the temperature difference determination model includes long-term trend items, seasonal items of day cycle and year cycle, and holiday items of abnormal or periodic changes. Inside the model, different types of changes are fitted as independent sub-models through logical addition, and the final prediction curve is obtained by superimposing the fitted sub-models. During the training process, the vehicle enterprise can constrain the upper limit of the model error by the historical temperature difference error distribution to avoid the problem of large deviation in the scene of severe temperature difference fluctuation. Automatic hyperparameter optimization is performed during the model training process, including trend change rate, rhythm adjustment factor, and regression window size.

[0044] The output of the temperature difference determination model is the predicted temperature difference data, which is a sequence of hourly temperature difference predictions within the next 12 hours, serving as the prior input of the power determination model. Each prediction point is accompanied by confidence interval information, which is used to judge the prediction credibility and drive the subsequent module weight adjustment.

[0045] In this embodiment, the temperature difference determination model is a Prophet model. In actual application, the Prophet model uses a sliding window mechanism for dynamic updating, automatically introducing the latest temperature difference records and external meteorological variables every hour to maintain the model's rapid response capability to environmental disturbances. At the same time, the model has high individualization support capability. Vehicles in different geographical locations can deploy their own trained localized models, and at the same time, according to the geographical and climatic labels of the area where the vehicle is located, match the corresponding historical samples to form a cluster of regional temperature difference prediction models. For specific cities with high temperature difference fluctuation scenarios, such as desert climate, plateau areas, etc., the priority of the temperature difference prediction model can be set separately to improve the pre-response of electric drive strategy adjustment in low temperature environment.

[0046] In some embodiments, the initial motor output power can be obtained by analyzing the actual accelerator pedal opening degree and the predicted temperature difference data by a power determination model. The power determination model is responsible for understanding the interaction between different features, and also describes the nonlinear dynamic response law through a deep neural network structure, providing a basic expected output power reference value for subsequent compensation models and fusion strategies. In this embodiment, the power determination model adopts a DeepFM model structure, aiming to use the DeepFM model to learn the dynamic system response characteristics under the disturbance of day-night temperature difference, so as to establish a multi-dimensional mapping relationship between temperature difference, driving behavior, vehicle state and power output.

[0047] The input data of the power determination model consists of two parts. The first part is sparse features, including vehicle platform number, drive motor type label, battery chemical system code, driver group label, road type label, operating environment code, etc., which are mapped to continuous vector space through embedding matrix. The second part is dense features, including current temperature difference, Prophet predicted future temperature difference, vehicle speed, accelerator pedal opening degree, power demand, electric drive current, voltage, SOC, battery temperature, motor temperature, power limitation state, vehicle thermal management operating state, etc., which are directly input into the model after normalization and standardization. In actual deployment of vehicle enterprises, these features are collected in real time by multi-domain controllers and aligned by time in the data platform, ensuring consistent model input structure.

[0048] The DeepFM model consists of FM and Deep parts. The FM part is responsible for processing the second-order interaction between features, and is particularly suitable for learning the combined influence of temperature difference and vehicle state parameters. For example, the interaction of temperature difference and SOC, temperature difference and battery temperature, temperature difference and accelerator pedal opening degree has a significant impact on the actual power output. These relationships are modeled in the FM module through feature vector inner product. The Deep part is a multi-layer feedforward neural network composed of multiple nonlinear activation layers, which is used to capture complex multi-dimensional nonlinear dynamic response patterns, such as torque decay law of electric drive system under high temperature difference, power reduction strategy trigger mode under low SOC, etc. The two modules are finally spliced at the output end to obtain the expected power output value of the comprehensive modeling capability.

[0049] In this embodiment, the output of the power determination model is the expected output power and expected torque of the vehicle under the current operating conditions, i.e., the initial motor output power, which is used to describe the power response that the system should have in an ideal state, and the estimated value of the system limit power, which provides a reference for the subsequent SVR error compensation module. Model training uses large-scale historical operation data built by the vehicle enterprise, and iteratively learns according to the deviation between the actual power output and the model prediction. In order to enhance the adaptability of the model to extreme temperature difference scenarios, a high temperature difference, low temperature difference, and sudden temperature difference special sample proportion weighting strategy is introduced in the training process to avoid overfitting of the model under normal temperature differences, which leads to unstable performance in extreme environments.

[0050] In the model, the internal structure further includes a vehicle enterprise self-defined feature gating mechanism for automatically selecting key features in different vehicle platforms. For example, in low-temperature areas, the temperature difference and battery temperature related feature weight will be automatically increased, and in humid southern areas, humidity related features will be given higher attention. All gating parameters are automatically learned through training to achieve regional adaptation capability without human intervention.

[0051] In this embodiment, after the power determination model is trained, DeepFM will be used as the core model in the vehicle power output prediction chain, deployed in the vehicle enterprise cloud and then distributed to the vehicle controller after being compressed. The output is not only used for real-time power prediction, but also as the input basis for the downstream error compensation model, which plays a decisive role in the accuracy of the final fused output. In the entire multi-model system, DeepFM plays a core role in building an ideal response model for the power system, enabling the entire temperature difference adaptive strategy to have sufficient behavior understanding ability and cross-model generalization ability.

[0052] In some embodiments, the target power correction value can be obtained by analyzing the predicted temperature difference data, the initial motor output power, and the actual motor output power through a deviation determination model. The deviation determination model uses a support vector regression model, i.e., the deviation determination model is an SVR model, which models and compensates for the output deviation of the power system caused by diurnal temperature difference disturbance in actual operation, thereby estimating and correcting the residual error between the predicted value of the power determination model and the actual power response. The deviation determination model performs error perception and boundary correction functions, which is an important link to improve the accuracy and robustness of the final target motor output power.

[0053] In this embodiment, the input of the deviation determination model mainly includes three types of features. One is the temperature difference feature, including the real-time temperature difference value at the current time, the future short-term temperature difference trend predicted by the temperature difference determination model, the temperature difference change rate, and the temperature difference fluctuation interval in the past one hour. The second is the output feature of the power determination model, mainly including the expected output power, the predicted torque, and the system power limit estimation value. The third is the real-time response data of the power system, including the actual electric drive output power at the current time, the electric control limit power signal, the torque actual value, the SOC, the system working mode, etc. These features jointly constitute the SVR input space, capturing the deviation mode between the environmental disturbance, the model prediction deviation and the actual system response.

[0054] In this embodiment, the output of the deviation determination model is the error compensation value, that is, the deviation estimation value between the model predicted power and the actual power, which finally serves as a correction term in the subsequent power fusion link to participate in the calculation of the final power output. In this way, the SVR model not only reflects the non-ideal state of the system operation, but also effectively makes up for the response lag problem of DeepFM in extreme temperature difference or dynamic rapid change scenarios, and plays a fine boundary correction role.

[0055] In this embodiment, when the deviation determination model is trained, based on the difference sequence between the DeepFM prediction result and the actual power output in the historical operation data set of the vehicle enterprise, the input sample is constructed by using the sliding window, and the data is layered according to the temperature difference distribution to ensure that the SVR has strong fitting ability in three typical scenarios of high temperature difference, low temperature difference and temperature difference fluctuation. During the training process, the samples are standardized, a radial basis kernel function is used to construct a nonlinear mapping space, and the optimal penalty coefficient and kernel width parameter are selected by using cross-validation to improve the generalization performance of the model.

[0056] In order to enhance the robustness of the model under boundary conditions, a data enhancement mechanism is designed to introduce noise disturbance samples or use time smoothing interpolation technology to generate approximate samples in the high temperature difference interval with insufficient sample size, so as to ensure that the SVR has sufficient learning ability in critical scenarios. The model output not only gives a compensation value, but also can be accompanied by a confidence interval range, which is used for dynamic weight adjustment in the subsequent fusion stage, to ensure that the correction process has a controllable range and a retreat mechanism.

[0057] The running characteristics of the deviation determination model determine that it is suitable to be deployed in the vehicle-mounted controller to form real-time linkage with DeepFM, and the compensation value is dynamically updated through the real-time input of the power response data and the prediction deviation, to ensure that the final power output is closer to the physical limit state of the system. At the same time, the model has fast response speed and simple structure, and is suitable for integration into the embedded control architecture, which takes into account the real-time performance and resource consumption on the basis of ensuring the prediction ability.

[0058] The above scheme constructs a high-precision, low-latency error boundary compensation model, which enables adaptive correction of the behavior deviation of the power system under complex nonlinear temperature disturbances. This effectively improves the stability of the vehicle's power output in extreme environments and avoids problems such as electric drive jitter, lag, or insufficient output caused by temperature difference misjudgment or model mismatch. It provides key error suppression support for multi-model fusion mechanisms.

[0059] In some embodiments, the predicted temperature difference data, the initial motor output power, and the target power correction value can be fused according to predetermined weights, and then the fused data can be input into the power determination model for processing. Specifically, step 104, which involves inputting the predicted temperature difference data, the initial motor output power, and the target power correction value into a pre-trained power determination model, and processing the data through the model to obtain the target motor output power, includes: Step 1041: Determine the target fusion weight, and fuse the predicted temperature difference data, the initial motor output power, and the target power correction value according to the target fusion weight to obtain the target fusion vector; Step 1042: Obtain the pre-trained power determination model, and process it to obtain the target motor output power.

[0060] In specific implementation, a target fusion weight is determined, wherein the target fusion weight is the weight value corresponding to each of the predicted temperature difference data, the initial motor output power, and the target power correction value when fusing the predicted temperature difference data, the initial motor output power, and the target power correction value.

[0061] Based on the determined target fusion weights, the predicted temperature difference data, the initial motor output power, and the target power correction value are fused according to their respective weights to obtain a target fusion vector. A pre-trained power determination model is then obtained, and the target fusion vector is processed by the power determination model to obtain the target motor output power.

[0062] In this embodiment, the predicted temperature difference data provides trend information on future temperature difference changes and acts as a look-ahead factor for temperature difference perturbations in the fusion process. Its main function is to provide prior scenario labels for other models, i.e., whether the current temperature difference is in a rapidly changing, high-low reversal, or stable range. The initial motor output power is the target power response under ideal conditions, representing the power demand level that the vehicle should have under the current driving behavior and environmental conditions. The target power correction value provides residual compensation information to correct the offset value of the power determination model in actual operation, which is particularly suitable for error correction under conditions of rapid temperature fluctuations, lag in electric drive system response, or boundary operating conditions.

[0063] In this embodiment, the process of fusing the predicted temperature difference data, the initial motor output power, and the target power correction value according to the target fusion weight to obtain the target fusion vector specifically includes: Step a: The predicted temperature difference data, the initial motor output power, and the target power correction value are mapped and aligned to obtain the temperature difference embedding vector, the power embedding vector, and the correction embedding vector. Step b involves inputting the target fusion weights, the temperature difference embedding vector, the power embedding vector, and the corrected embedding vector into a pre-trained vector fusion model. The vector fusion model then processes the data to output the target fusion vector.

[0064] In specific implementation, the predicted temperature difference data, the initial motor output power, and the target power correction value are mapped and aligned to obtain the temperature difference embedding vector, the power embedding vector, and the correction embedding vector.

[0065] Specifically, the predicted temperature difference data is linearly transformed and mapped to a preset semantic space to obtain a temperature difference embedding vector. The initial motor output power is linearly transformed and mapped to a preset semantic space to obtain a power embedding vector. The target power correction value is linearly transformed and mapped to a correction embedding vector to a preset semantic space to obtain a correction embedding vector.

[0066] The aligned temperature difference embedding vector, power embedding vector, and corrected embedding vector are further fused through a multi-head attention mechanism, enabling the model to identify the deep relationships and causal connections between them. Specifically, the target fusion weights, the temperature difference embedding vector, the power embedding vector, and the corrected embedding vector are input into a pre-trained vector fusion model, which outputs the target fusion vector.

[0067] In this embodiment, the target fusion weight represents the weight values ​​corresponding to the temperature difference embedding vector, power embedding vector, and corrected embedding vector when fusing the temperature difference embedding vector, power embedding vector, and corrected embedding vector to obtain the target fusion vector. In this embodiment, the target fusion weight can be preset or dynamically adjusted according to the accuracy of each of the temperature difference embedding vector, power embedding vector, and corrected embedding vector to ensure that the obtained target fusion vector is more consistent with the vehicle's environment and the driver's state.

[0068] In this embodiment, the vector fusion model employs a cross-modal contrastive loss function and is trained using historical pairing data to bring related samples closer together in the semantic space. The historical pairing data includes temperature difference embedding vectors, power embedding vectors, and corrected embedding vectors. Matched historical pairing data are used as positive samples, and mismatched data are used as negative samples to construct a contrastive learning model, i.e., the vector fusion model. This model can determine the correlation between the three modalities.

[0069] The above scheme fuses the temperature difference embedding vector, power embedding vector, and modified embedding vector to obtain the target fusion vector. This allows the subsequent parameter determination model to automatically identify the correlation between the three modes when analyzing and processing the target fusion vector, thereby improving the accuracy of the target screen parameters. Simultaneously, through cross-model information coordination and dynamic weighting, a balanced response to the driver's subjective perception and the objective environmental conditions is achieved.

[0070] In this embodiment, to enhance the stability of the fusion strategy, a time window smoothing mechanism is introduced. The fusion results from several consecutive time points are processed by moving average to reduce the impact of a single prediction error on the output. During extreme temperature disturbances or abnormal model fluctuations, a steady-state conservative strategy is activated, using the average value of past operating data for short-term coverage. Once the model stabilizes, it automatically switches back to the fused output, ensuring the vehicle's powertrain system possesses self-recovery and anomaly tolerance capabilities.

[0071] In some embodiments, if the predicted temperature difference between day and night is large, to ensure the accuracy of the final determined target motor output power, the weight value corresponding to the target power correction value can be increased accordingly. That is, the target fusion weight mentioned in step 1041 specifically includes: Step 10411: Obtain the current temperature difference data; Step 10412: In response to the current temperature difference data being greater than a preset temperature difference threshold and the actual motor output power being greater than a preset power threshold, determine the third target weight corresponding to the target power correction value based on the current temperature difference data and the actual motor output power. Step 10413: Determine the first target weight corresponding to the predicted temperature difference data and the second target weight corresponding to the initial motor output power based on the third target weight.

[0072] In practice, the current temperature difference data is acquired and compared with a preset temperature difference threshold, and the actual motor output power is compared with a preset power threshold. If the current temperature difference data is greater than the preset temperature difference threshold, and the actual motor output power is greater than the preset power threshold, then this is a nighttime low-temperature high-power request scenario. In this case, the target power correction value becomes dominant, and its weight will be increased. At this time, a third target weight corresponding to the target power correction value is determined based on the current temperature difference data and the actual motor output power.

[0073] In this embodiment, the correspondence between temperature difference data, motor output power and third weight can be preset. Then, when the current temperature difference data is greater than the preset temperature difference threshold and the actual motor output power is greater than the preset power threshold, the correspondence is found based on the current temperature difference data and the actual motor output power to obtain the third target weight corresponding to the target power correction value.

[0074] The first target weight corresponding to the predicted temperature difference data and the second target weight corresponding to the initial motor output power are determined based on the third target weight. Specifically, the process of determining the first target weight and the second target weight includes: Step A: Obtain the first initial weight corresponding to the predicted temperature difference data and the second initial weight corresponding to the initial motor output power; sum the first initial weight and the second initial weight to obtain the weight sum value. Step B: Determine the first weight ratio based on the first initial weight and the sum of the weights, and determine the second weight ratio based on the second initial weight and the sum of the weights; Step C: Determine the target remaining weight based on the third target weight, determine the first weight increment based on the first weight ratio and the target remaining weight, and determine the second weight increment based on the second weight ratio and the target remaining weight; Step D: Sum the first initial weight with the first weight increment to obtain the first target weight; Step E: Sum the second initial weight with the second weight increment to obtain the second target weight.

[0075] Specifically, a first initial weight, a second initial weight, and a third initial weight are preset. If the vehicle is in an environment with normal light intensity, the optimal weight combination is achieved when the confidence levels of the predicted temperature difference data, the initial motor output power, and the predicted light intensity data are all at the optimal confidence level, i.e., under ideal conditions.

[0076] The weight difference between the third target weight and the total weight is determined, i.e., the target remaining weight, where the total weight is the sum of the first initial weight, the second initial weight, and the third initial weight. The target remaining weight is then allocated to the first initial weight and the second initial weight. Specifically, the allocation can be performed according to the respective proportions of the first initial weight and the second initial weight. Preset first and second initial weights are obtained, and they are summed to obtain a weight sum value. A first weight proportion is determined based on the first initial weight and the weight sum value, and a second weight proportion is determined based on the second initial weight and the weight sum value.

[0077] The product of the first weight ratio and the target remaining weight is processed to obtain the first product. The first product is summed with the first initial weight to obtain the first target weight. The product of the second weight ratio and the target remaining weight is processed to obtain the second product. The second product is summed with the second initial weight to obtain the second target weight.

[0078] By using the above scheme, in high temperature difference disturbance scenarios such as low temperature and high power at night, the weight of the target power correction value is increased, and then the target motor output power is determined based on the target fusion vector determined by the adjusted weight, resulting in a more accurate target motor output power.

[0079] In some embodiments, the weight of the initial motor output power during fusion can be adjusted based on the confidence level corresponding to the initial motor output power. Specifically, determining the target fusion weight in step 1041 includes: Step 104A: Determine the second confidence level corresponding to the initial motor output power, and determine the temperature difference change rate based on the predicted temperature difference data; Step 104B: In response to the second confidence level being greater than the second preset confidence threshold, or the temperature difference change rate being greater than the preset change rate threshold, determine the second target weight corresponding to the initial motor output power; Step 104C: Determine the first target weight corresponding to the predicted temperature difference data and the third target weight corresponding to the target power correction value based on the second target weight.

[0080] In specific implementation, a second confidence level corresponding to the initial motor output power is determined, wherein the second confidence level is data obtained together with the initial motor output power. For example, the actual accelerator pedal opening, i.e., the predicted temperature difference data, is input into a pre-trained power determination model. Through processing by the power determination model, the initial motor output power and the second confidence level corresponding to the initial motor output power are obtained.

[0081] The rate of change of temperature difference is determined based on the predicted temperature difference data. Since the predicted temperature difference data comprises temperature difference data corresponding to multiple moments within a preset time period after the current moment, the rate of change of temperature difference represents the change in temperature difference data between adjacent moments within the predicted temperature difference data.

[0082] The second confidence level is compared with a preset confidence threshold, and the temperature difference change rate is compared with a preset change rate. If the second confidence level is greater than or equal to the preset confidence threshold, or the temperature difference change rate is greater than the preset change rate threshold, it indicates that the confidence level corresponding to the initial motor output power is high, or the future temperature difference change is large. Therefore, the weight of the initial motor output power in the fusion process can be appropriately increased.

[0083] Specifically, the second target weight corresponding to the initial motor output power can be obtained by searching the database based on the second confidence level or the temperature difference change rate.

[0084] After determining the second target weight, the first target weight corresponding to the predicted temperature difference data and the third target weight corresponding to the target power correction value are determined based on the second target weight. Specifically, the process of determining the first target weight and the third target weight based on the second target weight is the same as the method of determining the second target weight and the second target weight based on the third target weight, as described in steps A to D, and will not be repeated here.

[0085] If the second confidence level is less than a preset confidence threshold, and the temperature difference change rate is less than or equal to a preset change rate threshold, it indicates that the confidence level of the initial motor output power is within the normal range. In this case, the optimal weight combination can be used for the fusion weights. That is, the first target weight corresponding to the predicted temperature difference data is determined as the first initial weight, the second target weight corresponding to the initial motor output power is determined as the second initial weight, and the third target weight corresponding to the target power correction value is determined as the third initial weight.

[0086] In this embodiment, if the second confidence level is less than the minimum confidence level, it means that the confidence level of the initial motor output power is extremely low. The minimum confidence level is less than a preset confidence threshold. To reduce the impact on the subsequent determination of the target motor output power, the second target weight corresponding to the initial motor output power is set as the second minimum weight, and then the first and second target weights and the third target weight are determined based on the second target weight.

[0087] In some embodiments, the weights for fusion can be adjusted based on changes in throttle pedal opening and the temperature difference between predicted temperature difference data and current temperature difference data. Specifically, determining the target fusion weights in step 1041 includes: Step 104a: Obtain the current temperature difference data, and determine the temperature difference change value based on the predicted temperature difference data and the current temperature difference data; Step 104b: In response to the temperature difference change value being greater than a preset change threshold and the actual accelerator pedal opening being greater than a preset opening threshold, determine the second target weight corresponding to the initial motor output power and the third target weight corresponding to the target power correction value based on the temperature difference change value and the actual accelerator pedal opening. Step 104c: Determine the first target weight corresponding to the predicted temperature difference data based on the second target weight and the third target weight.

[0088] In practice, the current temperature difference data is acquired, and the temperature difference change value is determined based on the predicted temperature difference data and the current temperature difference data. The temperature difference change value represents the change between the future temperature difference and the current temperature difference.

[0089] The temperature difference change value is compared with a preset change threshold, and the actual accelerator pedal opening is compared with a preset opening threshold. If the temperature difference change value is greater than the preset change threshold, and the actual accelerator pedal opening is greater than the preset opening threshold, that is, the driver rapidly depresses the accelerator pedal and the predicted temperature difference value shows that it is in the cooling range, then the second target weight corresponding to the initial motor output power and the third target weight corresponding to the target power correction value are determined based on the temperature difference change value and the actual accelerator pedal opening.

[0090] In this embodiment, the database pre-stores the temperature difference change value and the actual accelerator pedal opening, along with the relationship between them and the second and third weights. Then, when the temperature difference change value is greater than a preset change threshold and the actual accelerator pedal opening is greater than a preset opening threshold, the database is searched based on the temperature difference change value and the actual accelerator pedal opening to obtain the second target weight corresponding to the initial motor output power and the third target weight corresponding to the target power correction value.

[0091] After determining the second target weight corresponding to the initial motor output power and the third target weight corresponding to the target power correction value, since the sum of the first target weight, the second target weight and the third target weight is a fixed value, the first target weight can be obtained by subtracting the sum from the second target weight and the third target weight in sequence.

[0092] With the above scheme, when the driver quickly and deeply presses the accelerator pedal but the temperature difference prediction value shows that it is in the cooling range, the system will actively reduce the second target weight corresponding to the initial motor output power and increase the third target weight corresponding to the target power correction value to prevent insufficient actual power or battery overload.

[0093] In some embodiments, when controlling vehicle movement based on the target motor output power, the target motor torque can be determined according to the target motor output power, and then the vehicle movement can be controlled according to the target torque. That is, controlling vehicle movement based on the target motor output power in step 104 specifically includes: Step 10A: Obtain the actual charge and actual battery temperature of the vehicle battery; determine a first correction coefficient based on the actual charge; and determine a second correction coefficient based on the actual battery temperature. Step 10B: Obtain the output power threshold, and use the first correction coefficient and the second correction coefficient to correct the output power threshold to obtain the first motor output power; Step 10C: Determine the output power of the second motor based on the target motor output power and the first motor output power; Step 10D: Obtain the motor speed and determine the first motor torque based on the motor speed and the output power of the second motor; Step 10E: Determine the target motor torque based on the first motor torque and the motor torque threshold, and control the vehicle to drive based on the target motor torque.

[0094] In practice, the actual charge level and actual battery temperature of the vehicle battery are obtained. A first correction coefficient is determined based on the actual charge level, and a second correction coefficient is determined based on the actual battery temperature. Specifically, the first correction coefficient corresponding to the actual charge level is determined by searching a database, and the second correction coefficient corresponding to the actual battery temperature is determined by searching a database.

[0095] For example, if the actual battery level is greater than 80%, the corresponding first correction factor is 1.0. If the actual battery level is less than 80% but greater than 20%, the corresponding first correction factor is 0.5. If the actual battery level is less than 20%, the corresponding first correction factor is 0.2. If the battery temperature is greater than 25 degrees Celsius but less than 50 degrees Celsius, the corresponding second correction factor is 1.0. If the battery temperature is greater than 50 degrees Celsius or less than 25 degrees Celsius, the corresponding second correction factor begins to linearly decrease to 0.5 or lower.

[0096] The output power threshold is obtained, and then corrected using the first correction coefficient and the second correction coefficient to obtain the output power of the first motor. Specifically, the first correction coefficient and the second correction coefficient, i.e., the output power threshold, are multiplied to obtain the output power of the first motor. The output power threshold is the physical or design limit of the system. Under no circumstances can the output power exceed this value.

[0097] The output power of the second motor is determined based on the output power of the target motor and the output power of the first motor. In this embodiment, the smaller value between the output power of the target motor and the output power of the first motor is selected as the output power of the second motor.

[0098] The motor speed is obtained, and the first motor torque is determined based on the motor speed and the output power of the second motor. Specifically, the motor speed is multiplied by the system efficiency coefficient to obtain a first product value. The second motor output power is multiplied by a preset conversion constant to obtain a second product value. The ratio of the second product value to the first product value is then calculated to obtain the first motor torque. The preset conversion constant is a unit conversion constant that links speed, torque, and power, typically set to 9549.

[0099] The target motor torque is determined based on the first motor torque and the motor torque threshold, that is, the smaller value between the first motor torque and the motor torque threshold is taken as the target motor torque, and the vehicle is controlled to drive based on the target motor torque.

[0100] By taking into account the impact of current battery charge and temperature, as well as the power limit, the above solution determines the target torque and then controls the vehicle's movement based on the target motor torque. This results in higher vehicle control stability and improves the user's driving experience.

[0101] In this embodiment, if the target motor output power exceeds the current maximum capacity of the battery or electric drive system, the power command is adjusted through a dynamic compression function to ensure that the hardware protection logic is not triggered while still satisfying the driver's intentions. For example, under conditions of low temperature in winter and increased battery internal resistance, even if the fusion model provides a higher power output, the controller will limit the target power to within the safe discharge threshold based on the current battery temperature.

[0102] In this embodiment, the target motor output power is converted into actual electric drive control parameters through a series of mapping and limiting mechanisms, including but not limited to the maximum torque request value, the maximum discharge power limit, the electronic control power reference input, the system torque request curve, the SOC dynamic discharge boundary, and the thermal management system load limit. Each parameter is constrained by multiple factors such as temperature, battery status, and safety logic. Therefore, during the adjustment process, it cannot be simply overridden but requires correction.

[0103] The aforementioned correction process can be executed by a parameter correction module within the vehicle. This module consists of multiple control channels, including a static limit channel, a temperature disturbance adjustment channel, a battery state constraint channel, and a driving intention analysis channel. Each channel independently participates in the final instruction determination based on specific conditions, and the output is coordinated by a priority mechanism.

[0104] For example, under rapid acceleration, the system prioritizes the response speed of the driving intention channel, while under conditions of extremely low SOC or drastic temperature fluctuations, the battery protection channel takes precedence. The control results output from each channel will be uniformly integrated at the scheduling layer to form the final torque request and power distribution commands.

[0105] In this embodiment, to ensure consistency between the algorithm output and the control system interface, a unified data structure protocol needs to be defined between the VMCU and the electronic control unit. Control parameter transmission includes control target values, fusion model confidence levels, parameter correction status flags, and model rollback flags, which are transmitted at high frequency to each execution module via CAN or Ethernet transmission protocols. Upon receiving parameters, the confidence level is verified, and a decision is made on whether to execute the current correction value or reuse the steady-state command from the previous cycle to ensure a smooth control response.

[0106] For example, when the thermal management system is in preheating mode and the battery temperature has not reached the target, the controller will reduce the maximum power command in advance based on the temperature difference prediction result to slow down the system load climb. Alternatively, when the temperature difference causes significant electronic control limitations, it can be combined with the HVAC load reduction command to release the thermal power budget and improve power performance.

[0107] In this embodiment, when the output power of the target motor experiences a drastic change, it will not immediately adjust to the target value. Instead, it will gradually and smoothly converge based on the first-order response function to avoid sudden changes in driving experience or power fluctuations caused by excessively rapid instantaneous response. During long-term vehicle operation, the control results and system response data at each stage will also be recorded and sent back to the model training center as samples for subsequent algorithm retraining and performance evaluation.

[0108] In some embodiments, to prevent over-adjustment, control instability, or abnormal vehicle response caused by model prediction inaccuracies, abnormal input data, extreme temperature disturbances, or system communication anomalies, a rollback mechanism should be executed when the confidence level of the target motor output power output by the power determination model is low. That is, after obtaining the target motor output power in step 104, the process further includes: Step 105: Determine the fourth confidence level corresponding to the output power of the target motor; Step 106: In response to the fourth confidence level being less than the fourth preset confidence threshold, the initial motor output power is used as the first motor output power, and the vehicle is controlled to drive based on the first motor output power.

[0109] In specific implementation, a fourth confidence level corresponding to the target motor output power is determined, wherein the fourth confidence level is data obtained together with the target motor output power. For example, the predicted temperature difference data, the initial motor output power, and the target power correction value are input into a pre-trained power determination model. The model processes the data to obtain the target motor output power and the corresponding fourth confidence level.

[0110] If the fourth confidence level is less than the fourth preset confidence threshold, the initial motor output power is used as the first motor output power, and the vehicle is controlled based on the first motor output power. That is, the target power compensation value is stopped, and the initial motor output power determined according to the actual accelerator pedal opening and the predicted temperature difference data is directly used as the target motor output power.

[0111] In this embodiment, a protective boundary limit is added to the initial motor output power to reduce the output gain and limit the rate of response change, thereby ensuring that basic performance is maintained under the premise of stable control.

[0112] In some embodiments, after step 106, in response to the fourth confidence level being less than a fourth preset confidence threshold, the method further includes: Step 10a: Obtain the ambient temperature of the vehicle's surroundings; Step 10b: In response to the predicted temperature difference data exceeding the preset temperature difference range and the ambient temperature being lower than the preset temperature threshold, obtain the actual charge and actual battery temperature of the vehicle battery. Step 10c: Determine the output power of the second motor based on the actual battery charge, the actual battery temperature, and the ambient temperature, and control the vehicle's movement based on the output power of the second motor.

[0113] In practice, the predicted temperature difference data of the vehicle's environment is compared with a preset temperature difference range, and the ambient temperature is compared with a preset temperature threshold. If the predicted temperature difference data exceeds the preset temperature difference range and the ambient temperature is lower than the preset temperature threshold, it indicates that the temperature difference prediction error has continuously exceeded the standard and the vehicle is in a low-temperature or extreme disturbance environment. The actual battery charge and actual battery temperature of the vehicle are then obtained.

[0114] The output power of the second motor is determined based on the actual battery charge, the actual battery temperature, and the ambient temperature. In other words, the power determination model is not relied upon at this time. Instead, the static power output mapping table preset by the car manufacturer is used. The corresponding output power of the second motor is determined only based on static parameters such as the current SOC, battery temperature, and ambient temperature. The vehicle driving is controlled based on the output power of the second motor, ensuring that the system does not cause safety hazards due to model link failure.

[0115] In this embodiment, to ensure the traceability of fault diagnosis, a logging function is set up to record key data in real time, such as model input, predicted output, residual evaluation, weight changes, compensation curves, and actual control values, for subsequent analysis and regression testing. The model must undergo complete regression testing and boundary stability verification before each OTA update to ensure that it remains within a controllable performance range after the update.

[0116] Meanwhile, to further improve the adaptability to model stability, a confidence-driven output fusion mechanism is designed. Each model's output is accompanied by a confidence score, calculated from multiple factors including historical prediction residuals, current input validity, and model state. When the confidence of a model decreases, its output automatically has a reduced weight in the fusion result, mitigating the impact of its instability on the overall output even if the model has not completely degraded. This mechanism is particularly effective in scenarios with drastic temperature fluctuations, sudden vehicle acceleration changes, or rapid battery state changes, improving the stability and controllability of the vehicle's response.

[0117] In this embodiment, physical reasonableness boundaries are set for each key input variable, especially for temperature difference, vehicle speed, SOC, electric drive voltage, and predicted power. Any input exceeding the defined threshold range will be automatically filtered out and the corresponding anomaly flag will be triggered. When the boundary is triggered, the system records the input features received by the current model, automatically stops providing abnormal data to the model, and instead uses the mean or median data from the historical sliding window to fill in the gaps, preventing prediction bias caused by abnormal inputs.

[0118] The real-time deviation monitoring module in the model continuously monitors the difference between the initial motor output power and the actual vehicle power for the power determination model. If the residual exceeds a set threshold for multiple consecutive cycles, the model is considered to have failed and an anomaly flag is issued. For the deviation determination model, if the output target power correction value is far from the center of the historical correction value distribution or changes too rapidly, it will also be judged as an unreliable output. The system will suppress the target power correction value by threshold reduction to prevent erroneous corrections from amplifying the overall vehicle response deviation.

[0119] The above solution ensures that the entire algorithm chain has the ability to self-monitor, self-diagnose, and self-degrade throughout its entire lifecycle, thus guaranteeing model accuracy without affecting vehicle safety and driving experience.

[0120] It should be noted that the method of this embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this embodiment, and the multiple devices will interact with each other to complete the method described.

[0121] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0122] Based on the same inventive concept, corresponding to any of the above-described embodiments, this disclosure also provides a vehicle control device.

[0123] refer to Figure 2 , Figure 2 The vehicle control device, as described in this embodiment, includes: The data acquisition module 201 is configured to monitor changes in the accelerator pedal opening, acquire historical environmental data of the vehicle's environment, and determine predicted temperature difference data based on the historical environmental data. The initial power determination module 202 is configured to acquire the actual accelerator pedal opening and determine the initial motor output power based on the actual accelerator pedal opening and the predicted temperature difference data. The correction value determination module 203 is configured to obtain the actual motor output power and determine the target power correction value based on the predicted temperature difference data, the initial motor output power and the actual motor output power. The target power determination module 204 is configured to input the predicted temperature difference data, the initial motor output power and the target power correction value into a pre-trained power determination model, process the power determination model to obtain the target motor output power, and control the vehicle driving based on the target motor output power.

[0124] In some embodiments, the target power determination module 204 is specifically configured as follows: Determine the target fusion weight, and fuse the predicted temperature difference data, the initial motor output power, and the target power correction value according to the target fusion weight to obtain the target fusion vector; A pre-trained power determination model is obtained, and the target motor output power is obtained by processing the power determination model.

[0125] In some embodiments, the target power determination module 204 is specifically configured as follows: Obtain the current temperature difference data; In response to the current temperature difference data being greater than a preset temperature difference threshold and the actual motor output power being greater than a preset power threshold, a third target weight corresponding to the target power correction value is determined based on the current temperature difference data and the actual motor output power. The first target weight corresponding to the predicted temperature difference data and the second target weight corresponding to the initial motor output power are determined based on the third target weight.

[0126] In some embodiments, the target power determination module 204 is specifically configured as follows: Determine the second confidence level corresponding to the initial motor output power, and determine the temperature difference change rate based on the predicted temperature difference data; In response to the second confidence level being greater than the second preset confidence threshold, or the temperature difference change rate being greater than the preset change rate threshold, the second highest weight corresponding to the initial motor output power is obtained, and the second highest weight is used as the second target weight; The first target weight corresponding to the predicted temperature difference data and the third target weight corresponding to the target power correction value are determined based on the second target weight.

[0127] In some embodiments, the target power determination module 204 is specifically configured as follows: Obtain the current temperature difference data, and determine the temperature difference change value based on the predicted temperature difference data and the current temperature difference data; In response to the temperature difference change value being greater than a preset change threshold and the actual accelerator pedal opening being greater than a preset opening threshold, a second target weight corresponding to the initial motor output power and a third target weight corresponding to the target power correction value are determined based on the temperature difference change value and the actual accelerator pedal opening. The first target weight corresponding to the predicted temperature difference data is determined based on the second target weight and the third target weight.

[0128] In some embodiments, the target power determination module 204 is specifically configured as follows: The actual charge level and actual battery temperature of the vehicle battery are obtained; a first correction factor is determined based on the actual charge level; and a second correction factor is determined based on the actual battery temperature. The output power threshold is obtained, and the output power threshold is corrected using the first correction coefficient and the second correction coefficient to obtain the output power of the first motor. The output power of the second motor is determined based on the target motor output power and the first motor output power; The motor speed is obtained, and the torque of the first motor is determined based on the motor speed and the output power of the second motor. The target motor torque is determined based on the first motor torque and the motor torque threshold, and the vehicle is controlled to move based on the target motor torque.

[0129] In some embodiments, the target power determination module 204 is specifically configured as follows: Determine the fourth confidence level corresponding to the output power of the target motor; In response to the fourth confidence level being less than the fourth preset confidence threshold, the initial motor output power is used as the first motor output power, and the vehicle is controlled to drive based on the first motor output power.

[0130] In some embodiments, the target power determination module 204 is specifically configured as follows: Obtain the ambient temperature of the vehicle's surroundings; In response to the predicted temperature difference data exceeding the preset temperature difference range and the ambient temperature being lower than the preset temperature threshold, the actual charge and actual battery temperature of the vehicle battery are obtained. The output power of the second motor is determined based on the actual battery charge, the actual battery temperature, and the ambient temperature, and the vehicle is controlled to move based on the output power of the second motor.

[0131] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this disclosure, the functions of each module can be implemented in one or more software and / or hardware.

[0132] The apparatus of the above embodiments is used to implement the corresponding vehicle control method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0133] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle control method described in any of the above embodiments.

[0134] Figure 3This embodiment illustrates a more specific hardware structure of an electronic device. The device may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0135] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0136] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0137] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0138] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0139] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0140] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0141] The electronic devices described above are used to implement the corresponding vehicle control methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0142] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to execute the vehicle control method as described in any of the above embodiments.

[0143] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0144] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the vehicle control method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0145] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a vehicle, including the vehicle control device, the electronic device, and the computer-readable storage medium in the above embodiments, wherein the vehicle device implements the vehicle control method described in any of the above embodiments.

[0146] The vehicles described in the above embodiments are used to implement the vehicle control method described in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0147] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0148] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0149] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0150] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0151] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this disclosure as described above, which are not provided in detail for the sake of brevity.

[0152] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this disclosure, the provided drawings may or may not show well-known power / ground connections to integrated circuit (IC) chips and other components. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this disclosure, and this also takes into account the fact that the details of implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this disclosure will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this disclosure, it will be apparent to those skilled in the art that the embodiments of this disclosure can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0153] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0154] This disclosure is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A vehicle control method, characterized in that, include: The system monitors changes in accelerator pedal opening, acquires historical environmental data of the vehicle's environment, and determines predicted temperature difference data based on the historical environmental data. Obtain the actual accelerator pedal opening, and determine the initial motor output power based on the actual accelerator pedal opening and the predicted temperature difference data; Obtain the actual motor output power, and determine the target power correction value based on the predicted temperature difference data, the initial motor output power, and the actual motor output power; The predicted temperature difference data, the initial motor output power, and the target power correction value are input into the pre-trained power determination model. The target motor output power is obtained through the power determination model, and the vehicle is controlled based on the target motor output power.

2. The method according to claim 1, characterized in that, The step of inputting the predicted temperature difference data, the initial motor output power, and the target power correction value into a pre-trained power determination model, and processing them through the power determination model to obtain the target motor output power, includes: Determine the target fusion weight, and fuse the predicted temperature difference data, the initial motor output power, and the target power correction value according to the target fusion weight to obtain the target fusion vector; A pre-trained power determination model is obtained, and the target motor output power is obtained by processing the power determination model.

3. The method according to claim 2, characterized in that, The target fusion weights include: Obtain the current temperature difference data; In response to the current temperature difference data being greater than a preset temperature difference threshold and the actual motor output power being greater than a preset power threshold, a third target weight corresponding to the target power correction value is determined based on the current temperature difference data and the actual motor output power. The first target weight corresponding to the predicted temperature difference data and the second target weight corresponding to the initial motor output power are determined based on the third target weight.

4. The method according to claim 2, characterized in that, The target fusion weights include: Determine the second confidence level corresponding to the initial motor output power, and determine the temperature difference change rate based on the predicted temperature difference data; In response to the second confidence level being greater than the second preset confidence threshold, or the temperature difference change rate being greater than the preset change rate threshold, the second highest weight corresponding to the initial motor output power is obtained, and the second highest weight is used as the second target weight; The first target weight corresponding to the predicted temperature difference data and the third target weight corresponding to the target power correction value are determined based on the second target weight.

5. The method according to claim 2, characterized in that, The target fusion weights include: Obtain the current temperature difference data, and determine the temperature difference change value based on the predicted temperature difference data and the current temperature difference data; In response to the temperature difference change value being greater than a preset change threshold and the actual accelerator pedal opening being greater than a preset opening threshold, a second target weight corresponding to the initial motor output power and a third target weight corresponding to the target power correction value are determined based on the temperature difference change value and the actual accelerator pedal opening. The first target weight corresponding to the predicted temperature difference data is determined based on the second target weight and the third target weight.

6. The method according to claim 1, characterized in that, The method of controlling vehicle movement based on the target motor output power includes: The actual charge level and actual battery temperature of the vehicle battery are obtained; a first correction factor is determined based on the actual charge level; and a second correction factor is determined based on the actual battery temperature. The output power threshold is obtained, and the output power threshold is corrected using the first correction coefficient and the second correction coefficient to obtain the output power of the first motor. The output power of the second motor is determined based on the target motor output power and the first motor output power; The motor speed is obtained, and the torque of the first motor is determined based on the motor speed and the output power of the second motor. The target motor torque is determined based on the first motor torque and the motor torque threshold, and the vehicle is controlled to move based on the target motor torque.

7. The method according to claim 1, characterized in that, After obtaining the target motor output power, the following is also included: Determine the fourth confidence level corresponding to the output power of the target motor; In response to the fourth confidence level being less than the fourth preset confidence threshold, the initial motor output power is used as the first motor output power, and the vehicle is controlled to drive based on the first motor output power.

8. The method according to claim 7, characterized in that, Following the response that the fourth confidence level is less than the fourth preset confidence threshold, the method further includes: Obtain the ambient temperature of the vehicle's surroundings; In response to the predicted temperature difference data exceeding the preset temperature difference range and the ambient temperature being lower than the preset temperature threshold, the actual charge and actual battery temperature of the vehicle battery are obtained. The output power of the second motor is determined based on the actual battery charge, the actual battery temperature, and the ambient temperature, and the vehicle is controlled to move based on the output power of the second motor.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in any one of claims 1 to 8.

10. A vehicle, characterized in that, The vehicle includes the electronic equipment as described in claim 9.

Images