A method, device and medium for full-domain control based on an intelligent vehicle control system

Abstract

This invention discloses a full-domain control method, device, and medium based on an intelligent vehicle control system, relating to the field of vehicle control technology. The method includes: dynamically adjusting the power source and energy recovery through full-domain control scheduling according to an energy management scheme; optimizing power source switching and energy recovery strategies to generate an optimized power source and energy recovery strategy; adjusting control parameters and optimizing the energy management scheme through real-time feedback and vehicle driving data; dynamically correcting the optimized power source and energy recovery strategy to generate a feedback-adjusted control strategy; and performing personalized optimization of driving behavior based on the feedback-adjusted control strategy, predicting driving behavior and adjusting energy scheduling to generate a personalized energy management scheme. This invention optimizes the vehicle's power distribution and energy recovery methods according to the driver's personalized needs, improving the overall vehicle performance and driving experience while reducing energy waste.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to a global control method, device and medium based on an intelligent vehicle control system. Background Technology

[0002] With the rapid development of new energy vehicles, intelligent vehicle control systems have gradually become an important technology for improving vehicle performance, energy efficiency, and driving experience. Existing vehicle control systems typically rely on preset rules and simple driving modes for energy management, based on traditional power source control and energy recovery mechanisms. In conventional methods, energy management strategies generally control driving behavior through fixed algorithms or rules, and combine external data such as vehicle speed and ambient temperature to adjust power output and energy recovery to achieve relatively ideal driving performance and fuel / electricity usage efficiency.

[0003] However, traditional vehicle control systems typically face two problems. First, energy management strategies based on preset rules or fixed algorithms lack accurate identification and real-time optimization of individual driver behaviors, resulting in insufficient adaptability to driver habits. Second, existing systems often rely on simplified models to respond to the external environment, failing to handle complex road conditions and dynamic environmental changes efficiently in real time, thus affecting the maximization of energy utilization. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a global control method based on an intelligent vehicle control system to solve the problems of energy efficiency optimization and real-time adaptive adjustment.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a full-domain control method based on an intelligent vehicle control system, comprising: collecting real-time vehicle data, analyzing driving behavior and external environmental information, and performing optimization calculations in conjunction with energy recovery strategies to generate an energy management scheme; dynamically adjusting the power source and energy recovery through full-domain control scheduling according to the energy management scheme, optimizing the power source switching and energy recovery strategies, and generating optimized power source and energy recovery strategies; adjusting control parameters and optimizing the energy management scheme through real-time feedback and vehicle driving data, dynamically correcting the optimized power source and energy recovery strategies, and generating a feedback-adjusted control strategy; performing personalized optimization of driving behavior according to the feedback-adjusted control strategy, predicting driving behavior and adjusting energy scheduling to generate a personalized energy management scheme; and combining the personalized energy management scheme with the collaborative work requirements of each power source to intelligently adjust the collaborative work between each power source, generating a full-power source collaborative scheduling scheme.

[0008] As a preferred embodiment of the global control method based on an intelligent vehicle control system described in this invention, the specific steps for generating the energy management scheme are as follows:

[0009] The system collects current vehicle status data, including vehicle speed, battery level, and engine operating status, as well as external environmental information, including road conditions, weather conditions, and traffic conditions, and integrates them to form real-time vehicle data.

[0010] Based on real-time vehicle data, the K-means clustering method is used to perform pattern recognition and trend analysis on driving behavior and external environment information, generating driving behavior and environment analysis results.

[0011] Based on driving behavior, external environmental information, and the current vehicle status, optimize energy recovery calculations and generate an energy recovery strategy.

[0012] Based on the energy recovery strategy, the analysis results of driving behavior and the environment are optimized and calculated to generate an energy management plan.

[0013] As a preferred embodiment of the global control method based on an intelligent vehicle control system described in this invention, the specific steps for generating the optimized power source and energy recovery strategy are as follows:

[0014] Based on the energy management scheme, the power source and energy recovery strategy of the vehicle are dynamically adjusted through global control and scheduling to generate global control and scheduling results;

[0015] By using the results of global control and scheduling, the timing of power source switching and the intensity of energy recovery are optimized, and the power source is automatically adjusted under different driving scenarios to generate optimized power source and energy recovery strategies.

[0016] As a preferred embodiment of the full-domain control method based on the intelligent vehicle control system described in this invention, the real-time feedback and vehicle driving data are collected in real time by on-board sensors, cameras, radar and GPS, including vehicle speed, acceleration, engine operating status, battery power, road conditions, traffic flow and weather conditions, and are transmitted to the vehicle control system for analysis and processing via the on-board network.

[0017] As a preferred embodiment of the global control method based on an intelligent vehicle control system described in this invention, the specific steps for generating the feedback-adjusted control strategy are as follows:

[0018] Based on real-time feedback and vehicle driving data, analyze and adjust the vehicle's control parameters in real time, optimize power output and energy recovery strategies, and generate adjusted control parameters.

[0019] By adjusting the control parameters, the energy management scheme is dynamically optimized using the random forest method, configuring the power source, energy recovery intensity and power allocation method, and generating the optimized energy management scheme.

[0020] Based on the optimized energy management scheme, the power source and energy recovery strategy are dynamically adjusted. By sensing driving behavior and external environmental information in real time, the energy recovery and power distribution methods are corrected, and the corrected power source and energy recovery strategy are generated.

[0021] Based on the modified power source and energy recovery strategy, the control strategy is continuously adjusted and optimized using the deep Q-network method and real-time feedback to generate a feedback-adjusted control strategy.

[0022] As a preferred embodiment of the global control method based on an intelligent vehicle control system described in this invention, the specific steps for generating a personalized energy management scheme are as follows:

[0023] Based on feedback, the control strategy is adjusted and the driving behavior is optimized in a personalized way to generate personalized driving behavior.

[0024] Based on personalized driving behavior, predict the driver's future behavior patterns, and adjust energy scheduling through real-time feedback and vehicle driving data to generate an adjusted energy scheduling strategy.

[0025] Based on the adjusted energy dispatch strategy, optimize the allocation of power sources, the intensity of energy recovery, and the mode of energy use to generate personalized energy management solutions.

[0026] As a preferred embodiment of the global control method based on the intelligent vehicle control system described in this invention, the collaborative working requirements of each power source are obtained by using machine learning algorithms and real-time feedback to analyze personalized energy management schemes, driver behavior, traffic flow, road conditions, and vehicle driving data.

[0027] As a preferred embodiment of the full-domain control method based on an intelligent vehicle control system described in this invention, the specific steps for generating a full-power source collaborative scheduling scheme are as follows:

[0028] Based on the personalized energy management plan and the collaborative work requirements of each power source, intelligent analysis and real-time vehicle data are used to identify the collaborative work requirements between each power source, formulate scheduling rules, and generate a collaborative work requirements analysis report.

[0029] Based on the collaborative work requirements analysis report, reinforcement learning methods are used to adjust the collaborative work requirements between various power sources and generate a collaborative work mode for power sources.

[0030] By combining the collaborative working mode of power sources, driving behavior, traffic flow, and vehicle driving data, a comprehensive power source collaborative scheduling scheme is generated.

[0031] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the global control method based on an intelligent vehicle control system as described in the first aspect of the present invention.

[0032] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the global control method based on an intelligent vehicle control system as described in the first aspect of the present invention.

[0033] The beneficial effects of this invention are as follows: By analyzing the driver's driving behavior in real time and combining it with external environment and vehicle driving data, the vehicle control system can predict the driver's future driving mode and optimize energy scheduling accordingly, generate personalized energy management solutions, and accurately adjust the power source and energy recovery strategy. This not only improves energy efficiency but also optimizes the vehicle's power distribution and energy recovery methods according to the driver's personalized needs, thereby enhancing the overall performance and driving experience of the vehicle while reducing energy waste. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of a full-domain control method based on an intelligent vehicle control system.

[0036] Figure 2 A flowchart generated for an energy management solution.

[0037] Figure 3 A flowchart generated for the optimized power source and energy recovery strategy.

[0038] Figure 4 A flowchart generated for the coordinated scheduling scheme of all power sources. Detailed Implementation

[0039] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0040] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0041] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0042] Reference Figures 1-4 This is one embodiment of the present invention, which provides a global control method based on an intelligent vehicle control system, including the following steps:

[0043] S1. Collect real-time vehicle data, analyze driving behavior and external environmental information, and perform optimization calculations in conjunction with energy recovery strategies to generate an energy management plan.

[0044] S1.1 Collect current vehicle status data, including vehicle speed, battery level, and engine operating status, and external environmental information, including road conditions, weather conditions, and traffic conditions, and integrate them to form real-time vehicle data.

[0045] Specifically, the system collects current vehicle status data, including vehicle speed, battery level, and engine operating status, which is acquired in real time through vehicle speed sensors, battery management sensors, and engine sensors. External environmental information is collected through sensors, including road conditions, weather conditions, and traffic conditions. Road conditions are collected through road monitoring sensors or onboard environmental perception sensors (such as radar and lidar). Weather conditions are acquired through onboard weather sensors, such as temperature, humidity, and wind speed. Traffic conditions are obtained through traffic flow sensors, cameras, or GPS data to acquire real-time traffic information. All vehicle speed, battery level, engine operating status, road conditions, weather conditions, and traffic conditions are integrated to form real-time vehicle data.

[0046] S1.2 Based on real-time vehicle data, the K-means clustering method is used to perform pattern recognition and trend analysis on driving behavior and external environment information, generating driving behavior and environment analysis results.

[0047] Specifically, based on real-time vehicle data, the K-means clustering method is used to perform pattern recognition and trend analysis on driving behavior and external environment information. Driving behavior data is collected in real-time through vehicle speed sensors, accelerator and brake pedal sensors, including the driver's acceleration, braking, steering, and accelerator operations. The K-means clustering algorithm is used to perform pattern recognition on driving behavior and external environment information. The parameters of the K-means clustering algorithm are set as follows: the number of clusters K=3 is determined by the elbow rule (the elbow rule determines the optimal number of clusters by calculating the sum of squared squares (WCSS) within each cluster corresponding to different K values); input features include vehicle speed, acceleration, and braking frequency. All features related to vehicle speed, acceleration, and braking frequency are standardized (vehicle speed is normalized to [0,1], and acceleration is normalized to [-1,1]); The K-means++ algorithm is used for initializing cluster centers, with a maximum of 100 iterations. In practical applications, if the data volume is large or the number of clusters is large, the number of iterations needs to be increased. However, for small datasets or low-dimensional data, 100 iterations are usually sufficient. Euclidean distance is used as the distance metric, and the convergence threshold is 0.001, based on the balance between accuracy and efficiency. A smaller convergence threshold can improve clustering accuracy but may increase computation time. For example, the convergence threshold ranges from 0.0001 to 0.01. The output clusters correspond to smooth driving, normal driving, and aggressive driving modes, generating driving behavior and environment analysis results. The driving behavior and environment analysis results include driver driving style analysis, future behavior pattern prediction, assessment of the impact of the external environment on driving, and suggestions for dynamically adjusting energy management strategies.

[0048] S1.3. Based on driving behavior, external environmental information and the current state of the vehicle, optimize energy recovery calculations and generate an energy recovery strategy.

[0049] Specifically, the K-means clustering method is used to process driving behavior, external environmental information, and the vehicle's current state. Based on the vehicle's acceleration, braking, and other driving behavior characteristics, as well as the influence of the external environment, the energy recovery intensity is calculated. The optimization process considers the vehicle's actual power demand and dynamically adjusts the energy recovery intensity. The calculation expression is as follows:

[0050] ;

[0051] in, It is the intensity of energy recovery. Vehicle speed acceleration and time step The function, It refers to energy recovery efficiency. It refers to the external ambient temperature; A multinomial regression model was constructed and fitted based on historical driving data, for example, with a value range of [0,1]. The temperature-dependent efficiency coefficient is obtained through a standard thermodynamic model, for example, with a value in the range of 0.8–0.95;

[0052] By calculating the energy recovery intensity and combining it with the vehicle's actual power demand, driving behavior (such as acceleration and braking), and external environmental factors (such as temperature and road conditions), the K-means clustering method dynamically adjusts the energy recovery intensity, standardizing vehicle speed and acceleration data; K-means clustering is used to form, for example, three driving mode clusters; the energy recovery intensity is adjusted according to the cluster (for example, the recovery intensity of the smooth driving cluster is increased by 20%); the energy recovery efficiency is optimized in real time based on the vehicle's current state, such as vehicle speed, acceleration, and time step, and the energy recovery intensity is corrected according to external factors such as ambient temperature to generate an energy recovery strategy.

[0053] S1.4. Based on the energy recovery strategy, optimize the analysis results of driving behavior and environment to generate an energy management plan.

[0054] Specifically, using the K-means clustering method, combined with acceleration, braking, and throttle operation characteristics in driving behavior and changes in the external environment (such as traffic flow, road conditions, and weather conditions), the energy recovery intensity is adjusted in real time. By evaluating the impact of driver behavior on energy recovery and power demand, the evaluation method includes calculating the variance of acceleration (a variance below, for example, 0.5 m² / s³ is considered smooth driving). Simultaneously, considering the impact of external environmental factors (such as temperature, humidity, and road conditions) on energy efficiency, the intensity of the energy recovery strategy is dynamically adjusted. Combining the vehicle's actual power demand and the optimized energy recovery intensity, an energy management scheme is generated. The energy management scheme includes energy recovery and power output strategies adjusted for driving behavior and the external environment.

[0055] S2. Based on the energy management scheme, dynamically adjust the power source and energy recovery through global control and scheduling, optimize the power source switching and energy recovery strategies, and generate optimized power source and energy recovery strategies.

[0056] S2.1. Based on the energy management scheme, dynamically adjust the vehicle's power source and energy recovery strategy through global control and scheduling to generate global control and scheduling results.

[0057] Specifically, the process of dynamically adjusting the vehicle's power source and energy recovery strategy through full-domain control and scheduling begins with collecting real-time vehicle data, including vehicle speed, battery charge, engine operating status, and external environmental information such as road conditions, weather conditions, and traffic conditions. This provides a foundation for further analysis of driving behavior and the external environment. Based on pattern recognition of driving behavior and analysis of the external environment, the energy recovery strategy is optimized to improve energy recovery efficiency. For example, during braking, the energy recovery intensity is adjusted according to the driver's braking habits, and the power source strategy is dynamically adjusted to ensure that power output and battery energy use are balanced under different driving conditions such as acceleration, deceleration, or climbing. By combining the optimized power source and energy recovery strategies, a full-domain control and scheduling result is generated. This result ensures that the vehicle's power source switching and energy recovery process are optimally configured under different driving scenarios, improving overall energy efficiency and driving performance.

[0058] S2.2. Based on the results of global control and scheduling, optimize the timing of power source switching and the intensity of energy recovery, automatically adjust the power source under different driving scenarios, and generate optimized power source and energy recovery strategies.

[0059] Specifically, based on the results of the overall control and scheduling, the vehicle's current driving status is analyzed in real time, including vehicle speed, battery charge, and engine operating status. Combined with external environmental information, such as road conditions and weather conditions, the current driving scenario is determined. Based on the current driving scenario, the timing of power source switching is precisely adjusted. When driving at high speeds, engine output is increased and energy recovery intensity is reduced to ensure the vehicle receives sufficient power. When driving at low speeds or decelerating, engine output is reduced and energy recovery intensity is increased to improve energy recovery efficiency. An optimized power source and energy recovery strategy is generated to ensure that the vehicle can automatically adjust power output and energy recovery under different driving scenarios to achieve optimal energy efficiency and driving performance.

[0060] S3. By using real-time feedback and vehicle driving data, adjust control parameters and optimize energy management schemes, dynamically correct the optimized power source and energy recovery strategy, and generate a feedback-adjusted control strategy.

[0061] S3.1 Real-time feedback and vehicle driving data are collected in real time by on-board sensors, cameras, radar and GPS, including vehicle speed, acceleration, engine operating status, battery level, road conditions, traffic flow and weather conditions, and are transmitted to the vehicle control system for analysis and processing via the on-board network.

[0062] Specifically, real-time feedback and vehicle driving data are collected in real time through onboard sensors, cameras, radar, and GPS. The process of assessing environmental factors includes: using sensor data fusion algorithms to combine radar ranging with camera images to assess the risk level of road obstacles; the data analysis process includes: the data is first filtered and denoised, then Kalman filtering is used to predict vehicle trajectory and match it with environmental data for analysis, providing data support for subsequent energy management; all vehicle speed, acceleration, engine operating status, battery charge, road conditions, traffic flow, and weather conditions are transmitted to the vehicle control system through the onboard network for real-time updates to reflect the vehicle's operating status and changes in the external environment, providing data support for subsequent energy management and control strategy adjustments.

[0063] S3.2 Based on real-time feedback and vehicle driving data, analyze and adjust the vehicle's control parameters in real time, optimize power output and energy recovery strategies, and generate adjusted control parameters.

[0064] Specifically, a rule-based vehicle control system (marked as rapid acceleration if acceleration > 2 m / s²) is used to evaluate vehicle speed and acceleration to understand the current vehicle dynamics. Combined with battery charge and engine operating status, the system analyzes the power source usage, calculating the power output ratio of the engine and battery to analyze energy efficiency deviations and ensure a balance between energy consumption and recovery. Based on road conditions and traffic flow information, it determines whether the energy recovery intensity needs adjustment, such as increasing energy recovery during traffic congestion or braking periods. Simultaneously, considering weather conditions, the energy management strategy is adjusted to adapt to temperature changes or wind speed effects, identifying the most suitable power output and energy recovery strategy. Control parameters are adjusted in real-time during driving, including throttle response parameters (PID controller proportional coefficient Kp, Kp range 0.5-2.0 determined based on PID controller stability analysis and optimized through real-vehicle testing), braking force parameters (maximum braking pressure, pressure range 50-200 bar set according to braking safety standards), and energy recovery intensity parameters (trigger threshold range 0.1-0.5g), generating adjusted control parameters.

[0065] It should also be noted that the most suitable power output and energy recovery strategy is to comprehensively consider the vehicle's current state, driving behavior, external environment (such as traffic flow and road conditions), battery charge, and weather conditions, and adjust the power source intensity and energy recovery intensity in real time to achieve the best energy efficiency and driving performance.

[0066] S3.3 By adjusting the control parameters, the energy management scheme is dynamically optimized using the random forest method, configuring the power source, energy recovery intensity and power distribution method, and generating the optimized energy management scheme.

[0067] Specifically, based on real-time feedback and vehicle driving data, the current driving state is determined, and initial adjustments are made to the power source, energy recovery intensity, and power distribution method. For example, when going uphill, engine output is increased to provide sufficient power; while when going downhill, battery energy recovery is increased and engine output is reduced. Using a random forest method, the power source output method is adjusted based on historical driving data and the current vehicle operating state. For example, engine output is increased when driving at high speeds, or battery power is used more when driving at low speeds. Regarding the energy recovery intensity, it is dynamically adjusted based on battery charge, acceleration, braking status, and road condition information. The intensity of energy recovery is adjusted to recover more energy during braking or deceleration and less during smooth driving to improve driving comfort and energy efficiency. The random forest method continuously learns from driver habits and environmental changes, updates control parameters in real time, and optimizes the coordinated operation of the power source. For example, when going uphill, the engine output power is increased to ensure the vehicle climbs smoothly; while when going downhill, the energy recovery intensity is increased to make full use of the kinetic energy of the downhill for charging. Based on all adjustments and optimizations, an optimized energy management scheme is generated. The random forest method, for example, uses 100 trees with a maximum depth of 10, and the training data includes 100,000 historical driving records.

[0068] S3.4. Based on the optimized energy management scheme, dynamically adjust the power source and energy recovery strategy. By sensing driving behavior and external environmental information in real time, correct the energy recovery and power distribution methods, and generate the corrected power source and energy recovery strategy.

[0069] Specifically, based on the optimized energy management scheme, the system senses driving behavior and external environmental information in real time, including vehicle speed, acceleration, engine operating status, battery charge, road conditions, traffic flow, and weather conditions. By acquiring driving behavior and external environmental information in real time, the system analyzes changes in the current driving state and external environment to determine whether adjustments to the power source and energy recovery strategy are needed. For example, during rapid acceleration, power output is increased while energy recovery intensity is reduced to ensure rapid vehicle response; during braking or deceleration, energy recovery intensity is increased to recover more energy through braking. Based on the real-time sensed driving behavior and external environmental information, the system corrects the power distribution method and energy recovery strategy. When external environmental factors such as road gradient and traffic flow change, the system automatically adjusts the use of the power source, for example, increasing engine output and reducing energy recovery when going uphill, and enhancing energy recovery and reducing engine output when going downhill. These adjustments ensure that the vehicle's power source and energy recovery strategy achieve the optimal balance under different driving scenarios. Through real-time adjustments and corrections, a revised power source and energy recovery strategy is generated to ensure that the vehicle can dynamically optimize power output and energy recovery under different driving conditions, improving overall energy efficiency and driving experience.

[0070] S3.5 Based on the modified power source and energy recovery strategy, the control strategy is continuously adjusted and optimized using the deep Q-network method and real-time feedback to generate a feedback-adjusted control strategy.

[0071] Specifically, based on the revised power source and energy recovery strategies, a deep Q-network method is used for dynamic evaluation and optimization. The process includes: the deep Q-network uses real-time vehicle driving data (e.g., vehicle speed, acceleration, battery level, and road conditions) as its state space and power source switching options (e.g., pure electric mode, hybrid mode, and energy recovery mode) as its action space, quantifying the strategy execution effect through a reward function; the reward function evaluates the effectiveness of the current control strategy, for example, optimizing energy utilization efficiency by selecting an appropriate power source mode when the battery level is low; the deep Q-network is trained through experience replay and target network updates, with experience replay using stored data... The deep Q-network stores historical experience data and randomly reuses it to stabilize the training process, while the target network reduces the fluctuation of Q-value estimation by periodically updating fixed parameters, ensuring the stability of the training process. The deep Q-network continuously learns and optimizes the strategy, combining real-time feedback information (such as vehicle speed, acceleration, battery level, and road conditions) to adjust the use of the power source and the intensity of energy recovery. For example, in a certain driving scenario, the deep Q-network may find that the current energy recovery intensity is insufficient to optimize battery charging efficiency, and thus automatically adjust the energy recovery intensity to improve recovery efficiency; in another scenario, the deep Q-network may adjust the timing of power source switching to optimize vehicle energy efficiency.

[0072] Through real-time feedback, the Deep Q-Network method can respond in real time to changes in driver behavior and the external environment. For example, if the driver's driving habits change (such as switching from smooth driving to aggressive acceleration) or external environmental conditions change (such as weather changes or traffic flow), the Deep Q-Network will adjust the control strategy based on the new environmental data to optimize the timing of power source switching and the intensity of energy recovery. Through continuous optimization of the Deep Q-Network, a feedback-adjusted control strategy is generated to ensure that the vehicle's power source and energy recovery strategy remain in optimal state under various driving scenarios, thereby improving energy efficiency, driving performance, and comfort. The state space of the Deep Q-Network includes vehicle speed and battery charge, and the reward function is designed based on maximizing energy efficiency.

[0073] It should be noted that by combining real-time feedback and vehicle driving data, the system can continuously learn and adjust the vehicle's control strategy to ensure optimal energy efficiency under different driving conditions. It can self-correct based on real-time changes in driver behavior, road conditions, and vehicle status, flexibly respond to various complex environments, provide personalized energy efficiency optimization, improve energy efficiency and driving experience, and avoid maladaptation and energy loss in complex dynamic environments.

[0074] S4. Adjust the control strategy based on feedback, optimize driving behavior in a personalized way, predict driving behavior and adjust energy scheduling to generate a personalized energy management plan.

[0075] S4.1 Adjust the control strategy based on feedback, optimize driving behavior in a personalized way, and generate personalized driving behavior.

[0076] Specifically, the control strategy is adjusted based on feedback, and the driver's driving behavior is monitored in real time. By analyzing data such as vehicle speed, acceleration, braking operations, and accelerator pedal usage, the driver's driving habits and preferences are identified. Combined with the vehicle's operating status and external environmental information, the power source and energy recovery strategies are adjusted in a personalized way to better adapt to the driver's driving style. For example, when the driver prefers smooth driving, the power output during acceleration is optimized to reduce the intensity of rapid acceleration, and the energy recovery intensity is enhanced during braking. If the driver is accustomed to rapid acceleration, the power source configuration during acceleration is optimized to improve response speed and generate personalized driving behavior. This ensures that the vehicle provides an experience that matches the driver's habits in every drive, while improving energy efficiency and driving comfort. The generation of personalized driving behavior ensures a perfect match between the control strategy and the driver's needs, improving the overall driving performance and experience.

[0077] S4.2 Based on personalized driving behavior, predict the driver's future behavior patterns, and adjust energy scheduling through real-time feedback and vehicle driving data to generate an adjusted energy scheduling strategy.

[0078] Specifically, based on personalized driving behavior, the system analyzes the driver's historical driving patterns, including acceleration, braking, and driving habits, to predict future driving behavior patterns. Combined with current vehicle data, such as speed, acceleration, battery charge, and road conditions, the system predicts the driver's upcoming actions. For example, if the driver tends to brake frequently when there is heavy traffic ahead, the system predicts the driver will continue to maintain this driving behavior pattern. Real-time feedback is used to adjust energy scheduling by dynamically adjusting the distribution of power sources and the intensity of energy recovery to adapt to the driver's predicted behavior. For example, if a sudden braking is predicted, the energy recovery intensity is adjusted in advance to maximize energy recovery; if acceleration is predicted, engine output is appropriately increased to provide the necessary power. Finally, an adjusted energy scheduling strategy is generated, enabling the vehicle to automatically adjust energy distribution based on the driver's future behavior patterns, ensuring optimized power output and energy recovery in various driving scenarios to improve energy efficiency and adapt to the driver's personalized needs.

[0079] S4.3 Based on the adjusted energy dispatch strategy, optimize the allocation of power sources, the intensity of energy recovery, and the mode of energy use to generate a personalized energy management plan.

[0080] Specifically, based on the adjusted energy dispatch strategy, the allocation of power sources is optimized to ensure that the use of power sources maximizes driving performance and energy efficiency under different driving scenarios. For example, during rapid acceleration, engine output is increased to provide sufficient power; while during low-speed driving or deceleration, engine output is reduced, and the battery is used to provide power instead. The intensity of energy recovery is adjusted, increasing the intensity of energy recovery during braking or deceleration to convert the vehicle's kinetic energy into electrical energy stored in the battery, while reducing energy recovery during acceleration or driving to ensure driving smoothness and comfort. The energy usage is optimized by comprehensively assessing the current battery charge, road conditions, and traffic flow. Factors such as power source and battery usage are considered to optimize the power source switching timing when the battery charge is low, avoiding excessive battery consumption and ensuring continuous vehicle operation. When the battery charge is sufficient, the battery output is appropriately increased (e.g., when the battery charge is above 80%, the battery output power is increased to 90% of the rated value) to reduce the engine's workload and improve energy efficiency. Based on these adjustments, a personalized energy management plan is generated to ensure that the power source allocation, energy recovery intensity, and energy usage mode can be optimally configured according to the driver's personalized needs and changes in the external environment under various driving conditions, thereby improving energy efficiency and driving experience.

[0081] S5 combines personalized energy management solutions with the collaborative work requirements of each power source, intelligently adjusts the collaborative work between each power source, and generates a full power source collaborative scheduling solution.

[0082] S5.1 The collaborative working requirements of each power source are obtained by using machine learning algorithms and real-time feedback to analyze personalized energy management schemes, driver behavior, traffic flow, road conditions and vehicle driving data.

[0083] Specifically, machine learning algorithms assess power source usage based on personalized energy management plans, analyze driver acceleration, braking, and steering behaviors, and predict driver needs in different driving scenarios. For example, when a driver accelerates rapidly, the algorithm analyzes the need for more power source support and predicts the timing of power source switching based on driving habits. Combining real-time feedback on traffic flow and road conditions, such as traffic congestion and gradient changes, the algorithm adjusts the collaborative working methods of the power sources to balance power output and energy recovery needs in complex road conditions. Machine learning algorithms also optimize the coordination of various power sources in real time based on vehicle driving data, such as vehicle speed and battery charge, ensuring that power sources can work efficiently together in different driving scenarios to maximize energy efficiency and vehicle performance. By performing multi-source data fusion and feature extraction on driver behavior, traffic flow, road conditions, and vehicle driving data, the algorithm generates collaborative working requirements for each power source, ensuring that in each drive, the power sources can accurately coordinate based on real-time vehicle data and driving needs, thereby improving the overall efficiency of the vehicle and the driving experience.

[0084] It should also be noted that the training process of the machine learning algorithm includes: Data preparation and labeling: Collecting historical driving data covering various scenarios such as urban congestion, highway cruising, and hill driving, including vehicle speed, acceleration, battery level, engine status, GPS trajectory, camera and radar data, and labeling the data segments with ideal operating mode labels for each power source (such as engine and battery) according to the principle of optimal energy efficiency, forming a supervised learning sample set. The ideal operating mode is based on the real-time state of the vehicle (such as vehicle speed, acceleration, and battery level), driving environment (such as road slope and traffic conditions), and energy efficiency optimization principles to ensure that the power source utilization efficiency and energy recovery effect are maximized in each driving scenario; Training and validation: Using ensemble learning methods mainly based on random forests or gradient boosting decision trees, the ideal operating mode is used as the output label, with vehicle speed, acceleration, battery level, engine status, GPS trajectory, camera and radar data as input features and the labeled ideal operating mode as the output label. The system trains the system by adjusting hyperparameters (e.g., maximum tree depth and learning rate) through cross-validation to prevent overfitting, and verifies prediction accuracy (>95%) using an independent test set. Online learning and adaptive updates are implemented: real-time data is continuously collected during actual vehicle operation, and the machine learning algorithm is fine-tuned using a deep Q-network. If the energy efficiency of the current strategy falls below the expected threshold (determined through statistical analysis of historical driving data and real-vehicle testing, typically set to ±10% to ±15% of the energy efficiency ratio benchmark), parameter updates are automatically triggered to ensure adaptation to changes in driver habits and new road environments. The optimal energy efficiency principle is based on a comprehensive analysis of historical driving data, real-time vehicle status (e.g., vehicle speed and battery charge), and the external environment (e.g., traffic flow and road conditions). Machine learning algorithms optimize the collaborative work of the power source to maximize energy recovery and power output efficiency.

[0085] S5.2 Based on the personalized energy management plan and the collaborative work requirements of each power source, use intelligent analysis and real-time vehicle data to identify the collaborative work requirements between each power source, formulate scheduling rules, and generate a collaborative work requirements analysis report.

[0086] Specifically, based on personalized energy management solutions and the collaborative needs of various power sources, fuzzy logic is used to fuzzify parameters such as vehicle speed, battery charge, and energy recovery intensity. A pre-defined rule base is used to infer the collaborative tendencies of each power source. Simultaneously, multi-sensor data fusion is combined to perform weighted analysis of real-time vehicle speed, road gradient, and traffic flow, comprehensively evaluating the optimal usage of power sources such as the engine and battery in the current driving scenario. Through real-time vehicle data, such as vehicle speed, battery charge, and road conditions, the collaborative needs between various power sources are identified in real time. For example, during rapid acceleration, the engine needs to provide more power output, while the battery may need to reduce its load; at the same time, when the vehicle... When driving downhill, the collaborative working requirements of the power sources change, with engine output decreasing and energy recovery intensity increasing. Based on real-time vehicle data, combined with factors such as driver behavior patterns, traffic flow, and road conditions, corresponding scheduling rules are formulated. These rules help coordinate the cooperation between the engine, battery, and other power sources under different driving conditions, ensuring effective collaboration among power sources to achieve optimal energy efficiency and driving performance. Through the combination of intelligent analysis and real-time vehicle data, a collaborative working requirements analysis report is generated. This report details the collaborative needs between each power source and provides specific scheduling rules to guide the efficient collaborative work of power sources in different driving scenarios.

[0087] It should also be noted that the establishment of the preset rule base is mainly based on the statistical analysis of a large amount of historical driving data. By performing cluster analysis on tens of thousands of driving behavior data in different driving scenarios (such as urban congestion, highway cruising and hill driving), the correspondence between typical driving modes and optimal power configurations is extracted. At the same time, the knowledge of power source characteristics (such as the engine's optimal efficiency range and battery charging and discharging characteristics) is solidified to form fuzzy rules in the form of "IF-THEN".

[0088] S5.3 Based on the collaborative work requirements analysis report, reinforcement learning methods are used to adjust the collaborative work requirements between various power sources and generate a collaborative work mode for power sources.

[0089] Specifically, based on the collaborative work requirements analysis report, reinforcement learning methods are used to adjust the collaborative work requirements between various power sources. Specific applications of reinforcement learning include: states such as vehicle speed and road conditions; actions such as power source switching; and reward functions designed based on energy consumption and comfort. The iterative process dynamically adjusts strategies by continuously learning from environmental changes. During acceleration, the reinforcement learning method may increase engine output power and reduce energy recovery intensity; while during deceleration or braking, it will enhance energy recovery intensity and reduce engine output to achieve optimal energy efficiency and driving experience. In each iteration, the reinforcement learning method adjusts the collaborative work requirements between various power sources by analyzing feedback information. For example, in complex road conditions, such as uphill driving, the reinforcement learning method may increase engine output and reduce battery load to ensure the vehicle can climb smoothly; while downhill driving, it may reduce engine output and increase battery energy recovery to improve overall energy efficiency. Based on real-time feedback and the collaborative work requirements analysis report, the reinforcement learning method adjusts and generates collaborative work modes for power sources, ensuring that each power source can work efficiently in different driving scenarios to achieve optimized power output and energy recovery strategies.

[0090] S5.4 Combine the power source collaborative working mode, driving behavior, traffic flow and vehicle driving data to generate a full power source collaborative scheduling scheme.

[0091] Specifically, based on adjustments to the power source collaborative working mode, considering acceleration, braking, and driving habits, and combining traffic flow and road condition information, the power source demand of the vehicle in specific driving environments is determined. For example, in high-traffic conditions, the vehicle's acceleration demand is lower, while its energy recovery demand is higher; in this case, power source output will be reduced and energy recovery increased. On open roads, the vehicle's acceleration demand is higher; the power source collaborative working mode will increase engine output, while the battery load will be moderately increased. This is combined with vehicle driving data, such as vehicle speed, battery charge, engine operating status, and external environmental factors, such as weather conditions, through real-time... Feedback is provided to adjust the timing of power source switching. At this time, the power source collaborative working mode is further optimized to adapt to different driving scenarios. For example, when driving uphill or at high speed, engine output is enhanced and battery usage is reduced; while when driving downhill or at low speed in the city, energy recovery intensity is increased and engine output is reduced. Based on the power source collaborative working mode, driving behavior, traffic flow and vehicle driving data, a full power source collaborative scheduling scheme is generated. By dynamically adjusting the working status of each power source, it is ensured that the power sources can work effectively together in various driving situations to maximize energy efficiency and optimize driving performance and comfort.

[0092] This embodiment also provides a computer device applicable to the full-domain control method based on an intelligent vehicle control system, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the full-domain control method based on an intelligent vehicle control system as proposed in the above embodiment.

[0093] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0094] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the full-domain control method based on an intelligent vehicle control system as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0095] In summary, this invention, through real-time analysis of the driver's driving behavior and in conjunction with external environment and vehicle driving data, enables the vehicle control system to predict the driver's future driving mode and optimize energy scheduling accordingly. This generates a personalized energy management plan, precisely adjusts the power source and energy recovery strategy, not only improving energy efficiency but also optimizing the vehicle's power distribution and energy recovery methods based on the driver's individual needs, thereby enhancing the overall vehicle performance and driving experience while reducing energy waste.

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A global control method based on an intelligent vehicle control system, characterized in that: include, Collect real-time vehicle data, analyze driving behavior and external environmental information, and combine energy recovery strategies for optimization calculations to generate energy management solutions; According to the energy management plan, the power source and energy recovery are dynamically adjusted through full-domain control and scheduling, the power source switching and energy recovery strategies are optimized, and the optimized power source and energy recovery strategies are generated. By using real-time feedback and vehicle driving data, control parameters are adjusted and energy management schemes are optimized, dynamic corrections are made to the optimized power source and energy recovery strategies, and feedback-adjusted control strategies are generated. Based on feedback, the control strategy is adjusted to optimize driving behavior in a personalized way, predict driving behavior and adjust energy scheduling to generate a personalized energy management plan. By combining personalized energy management solutions with the collaborative work requirements of various power sources, the collaborative work between power sources is intelligently adjusted to generate a comprehensive power source collaborative scheduling solution.

2. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The specific steps for generating the energy management scheme are as follows: The system collects current vehicle status data, including vehicle speed, battery level, and engine operating status, as well as external environmental information, including road conditions, weather conditions, and traffic conditions, and integrates them to form real-time vehicle data. Based on real-time vehicle data, the K-means clustering method is used to perform pattern recognition and trend analysis on driving behavior and external environment information, generating driving behavior and environment analysis results. Based on driving behavior, external environmental information, and the current vehicle status, optimize energy recovery calculations and generate an energy recovery strategy. Based on the energy recovery strategy, the analysis results of driving behavior and the environment are optimized and calculated to generate an energy management plan.

3. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The specific steps for generating the optimized power source and energy recovery strategy are as follows: Based on the energy management scheme, the power source and energy recovery strategy of the vehicle are dynamically adjusted through global control and scheduling to generate global control and scheduling results; By using the results of global control and scheduling, the timing of power source switching and the intensity of energy recovery are optimized, and the power source is automatically adjusted under different driving scenarios to generate optimized power source and energy recovery strategies.

4. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The real-time feedback and vehicle driving data are collected in real time by on-board sensors, cameras, radar and GPS, including vehicle speed, acceleration, engine operating status, battery level, road conditions, traffic flow and weather conditions, and are transmitted to the vehicle control system for analysis and processing via the on-board network.

5. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The specific steps for generating the feedback-adjusted control strategy are as follows. Based on real-time feedback and vehicle driving data, analyze and adjust the vehicle's control parameters in real time, optimize power output and energy recovery strategies, and generate adjusted control parameters. By adjusting the control parameters, the energy management scheme is dynamically optimized using the random forest method, configuring the power source, energy recovery intensity and power allocation method, and generating the optimized energy management scheme. Based on the optimized energy management scheme, the power source and energy recovery strategy are dynamically adjusted. By sensing driving behavior and external environmental information in real time, the energy recovery and power distribution methods are corrected, and the corrected power source and energy recovery strategy are generated. Based on the modified power source and energy recovery strategy, the control strategy is continuously adjusted and optimized using the deep Q-network method and real-time feedback to generate a feedback-adjusted control strategy.

6. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The specific steps for generating a personalized energy management plan are as follows: Based on feedback, the control strategy is adjusted and the driving behavior is optimized in a personalized way to generate personalized driving behavior. Based on personalized driving behavior, predict the driver's future behavior patterns, and adjust energy scheduling through real-time feedback and vehicle driving data to generate an adjusted energy scheduling strategy. Based on the adjusted energy dispatch strategy, optimize the allocation of power sources, the intensity of energy recovery, and the mode of energy use to generate personalized energy management solutions.

7. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The collaborative working requirements of the various power sources are obtained by using machine learning algorithms and real-time feedback to analyze personalized energy management schemes, driver behavior, traffic flow, road conditions, and vehicle driving data.

8. The full-domain control method based on an intelligent vehicle control system as described in claim 1, characterized in that: The specific steps for generating a comprehensive power source coordinated scheduling scheme are as follows: Based on the personalized energy management plan and the collaborative work requirements of each power source, intelligent analysis and real-time vehicle data are used to identify the collaborative work requirements between each power source, formulate scheduling rules, and generate a collaborative work requirements analysis report. Based on the collaborative work requirements analysis report, reinforcement learning methods are used to adjust the collaborative work requirements between various power sources and generate a collaborative work mode for power sources. By combining the collaborative working mode of power sources, driving behavior, traffic flow, and vehicle driving data, a comprehensive power source collaborative scheduling scheme is generated.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the global control method based on the intelligent vehicle control system as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the global control method based on the intelligent vehicle control system as described in any one of claims 1 to 8.

Images