Vehicle driving style adjusting method and system

Abstract

The embodiment of the invention provides a vehicle driving style adjusting method and system. The method comprises the following steps: acquiring traffic flow state information broadcasted by a vehicle-road cooperative communication network, and determining a recommendation style coefficient according to the traffic flow state information; according to the recommended style coefficient and a reference style parameter of the user, determining an execution style parameter; and controlling the driving state of the vehicle according to the execution style parameters. The method is used for improving the adjusting effect of the vehicle driving style in different scenes.

Description

Technical Field

[0001] This application relates to the field of vehicle driver assistance, and in particular to a method and system for adjusting vehicle driving style. Background Technology

[0002] Driving style refers to a driver's behavior and habits when operating a vehicle, including acceleration, braking, and steering techniques. Different driving styles affect a vehicle's fuel consumption, comfort, and safety. Generally, they are categorized into three types: aggressive, smooth, and conservative. Aggressive driving often leads to increased fuel consumption and danger, while smooth driving helps improve fuel efficiency and extend vehicle life. Understanding driving style helps improve the driving experience and promotes energy conservation and environmental protection.

[0003] In existing technologies, the adjustment of vehicle driving style is mainly based on the driver's subjective choice. That is, the driver determines the vehicle's driving style during driving by interacting with the vehicle.

[0004] However, existing methods do not take into account the actual driving conditions of the vehicle, which leads to poor adjustment results when adjusting the vehicle's driving style. Summary of the Invention

[0005] The vehicle driving style adjustment method and system provided in this application are used to improve the adjustment effect of vehicle driving style in different scenarios.

[0006] In a first aspect, embodiments of this application provide a method for adjusting vehicle driving style, applied to a vehicle, including:

[0007] Obtain traffic flow status information broadcast by the vehicle-road cooperative communication network, and the recommendation style coefficient determined based on the traffic flow status information;

[0008] The execution style parameters are determined based on the recommended style coefficients and the user's baseline style parameters;

[0009] The vehicle's driving status is controlled based on the execution style parameters.

[0010] In one possible implementation, acquiring traffic flow state information broadcast by the vehicle-to-infrastructure (V2I) communication network and recommendation style coefficients determined based on the traffic flow state information includes:

[0011] Obtain traffic flow status information broadcast by the vehicle-road cooperative communication network and suggested style coefficients sent from the cloud; the suggested style coefficients are determined by the cloud based on the traffic flow status information.

[0012] The recommended style coefficient is obtained based on traffic flow status information, suggested style coefficient, and vehicle status.

[0013] In one possible implementation, the recommended style coefficient is obtained based on traffic flow state information, the suggested style coefficient, and the vehicle's own state, including:

[0014] The corresponding scene processing model is determined based on the scene tags sent from the cloud; the scene tags are determined by the cloud based on traffic flow status information.

[0015] Traffic flow status information, suggested style coefficients, and vehicle status are input into the scene processing model to obtain suggested style coefficients.

[0016] In one possible implementation, after determining the execution style parameters based on the recommended style coefficients and the user's baseline style parameters, the method further includes:

[0017] The execution style parameters and vehicle driving data are sent to the cloud so that the cloud can train the prediction model based on the execution style parameters, vehicle driving data and traffic flow state information; the prediction model is used to output suggested style coefficients based on traffic flow state information.

[0018] In one possible implementation, the execution style parameters are determined based on the recommended style coefficients and the user's baseline style parameters, including:

[0019] Based on the recommended style coefficient and the baseline style parameters, the user's adjustment intention is obtained;

[0020] If the intention to adjust indicates that the user agrees to the adjustment, the execution style parameters are determined based on the recommended style coefficient and the user's baseline style parameters.

[0021] If the user refuses to adjust the intended style, the baseline style parameter will be used as the executed style parameter.

[0022] The vehicle's driving status is controlled based on the execution style parameters.

[0023] In one possible implementation, the user's adjustment intent is obtained based on the recommended style coefficient and the baseline style parameter, including:

[0024] If the difference between the recommended style coefficient and the baseline style parameter is greater than the preset difference value, a query message will be output to the user.

[0025] By responding to user responses to inquiries, we can understand the user's intention to make adjustments.

[0026] In one possible implementation, the execution style parameters are determined based on the recommended style coefficients and the user's baseline style parameters, including:

[0027] Dynamic weights are determined based on the confidence level of traffic flow status information and the urgency of traffic scenarios;

[0028] The execution style parameters are determined based on dynamic weights, recommended style coefficients, and baseline style parameters.

[0029] In one possible implementation, the execution style parameters are determined based on dynamic weights, recommended style coefficients, and baseline style parameters, including:

[0030] The theoretical style correction coefficient is obtained based on the dynamic weights and the recommended style coefficient.

[0031] Based on the dynamic weights and the baseline style parameters, the baseline style correction parameters are obtained;

[0032] The execution style parameters are obtained based on the theoretical style correction coefficient and the baseline style correction parameter.

[0033] In one possible implementation, controlling the vehicle's driving state based on execution style parameters includes:

[0034] The driving parameters are determined based on the execution style parameters; the type of driving parameters is determined based on the baseline style parameters.

[0035] The vehicle's driving status is smoothly controlled based on driving parameters.

[0036] In one possible implementation, when the driving parameter is characterized as the desired following distance, the driving parameter is determined based on the desired following distance and the execution style parameter.

[0037] And / or,

[0038] When the driving parameters are represented as acceleration limits, the driving parameters are determined based on the preset boundary of acceleration and the adjustment function of the execution style parameters;

[0039] And / or,

[0040] When the driving parameter is represented as energy recovery torque, the driving parameter is determined based on the maximum recovery torque and the execution style parameter;

[0041] And / or,

[0042] When the driving parameters are represented as lateral decision thresholds, the driving parameters are determined based on theoretical safety gaps, scenario coefficients, and execution style parameters.

[0043] Secondly, embodiments of this application provide a vehicle driving style adjustment device, comprising:

[0044] The acquisition module is used to acquire traffic flow status information broadcast by the vehicle-road cooperative communication network, as well as recommendation style coefficients determined based on the traffic flow status information;

[0045] The determination module is used to determine the execution style parameters based on the recommended style coefficients and the user's baseline style parameters;

[0046] The control module is used to control the vehicle's driving status based on the execution style parameters.

[0047] Thirdly, embodiments of this application provide a controller, including: a memory and a processor;

[0048] The memory stores instructions that the computer executes;

[0049] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0050] The fourth aspect includes: the vehicle body and the controller installed in the vehicle body.

[0051] Fifthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0052] Sixthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0053] The vehicle driving style adjustment method and system provided in this application obtain traffic flow state information broadcast by the vehicle-road cooperative communication network and a recommended style coefficient determined based on the traffic flow state information to determine a recommended style coefficient that matches the vehicle's driving style in the current scenario. Based on the recommended style coefficient and the user's baseline style parameters, execution style parameters are obtained that adapt to the traffic flow state while respecting the driver's subjective choice. This method controls the vehicle's driving state according to the execution style parameters, avoiding the problem of the vehicle's inability to dynamically adapt to different traffic flow demands due to the driver's subjective choice of driving style. This effectively improves the vehicle's adaptive adjustment capability for driving style in diverse traffic scenarios, thereby enhancing the adjustment effect of vehicle driving style in different scenarios. Attached Figure Description

[0054] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0055] Figure 1 A schematic diagram illustrating a scenario for the vehicle driving style adjustment method provided in this application;

[0056] Figure 2 Flowchart of the method for adjusting vehicle driving style provided in this application Figure 1 ;

[0057] Figure 3 Flowchart of the method for adjusting vehicle driving style provided in this application Figure 2 ;

[0058] Figure 4 A schematic diagram of the architecture of the vehicle driving style adjustment system provided in this application;

[0059] Figure 5 A schematic diagram of the vehicle driving style adjustment device provided in this application;

[0060] Figure 6 A schematic diagram of the controller provided in this application.

[0061] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0062] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0063] First, let me explain the terms used in this application:

[0064] An On-Board Unit (OBU) can refer to a communication and computing device installed on a vehicle, used to realize V2X communication such as vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I), and to support traffic information reception, local decision-making, and control command issuance.

[0065] Roadside units (RSUs) can refer to fixed communication infrastructure deployed along roads to interact with vehicles in real time and broadcast collaborative sensing data such as traffic flow status, traffic light phases, and event warnings.

[0066] The cloud can refer to a remote data center built on a cloud computing platform, which has the capabilities of global traffic situation perception, big data analysis, driving strategy generation and model training, providing high-dimensional intelligent support for vehicle-road cooperative systems.

[0067] V2X communication refers to wireless information interaction technology between vehicles and any entity that may affect their driving (including other vehicles, roadside facilities, pedestrians, networks, etc.). It covers V2V, V2I, V2N, V2P and other modes, and is the basic communication means to realize intelligent connected driving.

[0068] With the increasing penetration of V2X communication, roadside units can broadcast real-time information on road traffic flow, density, speed, and events such as accidents / construction to vehicles. How to utilize this macro-level information to optimize individual vehicle driving behavior has become a new research direction for improving road capacity and traffic safety. However, existing methods remain in an open-loop stage of "human-selected mode, vehicle-executed," failing to incorporate traffic flow status into a closed loop, leading to the following contradictions:

[0069] When merging into a smooth ramp, an overly conservative fixed time interval turns vehicles into "moving roadblocks," reducing zipper efficiency.

[0070] In high-density synchronous flow, aggressive acceleration and deceleration are not suppressed, amplifying velocity fluctuations and inducing "phantom traffic jams";

[0071] Simply increasing the distance between vehicles in the event zone can actually create a "black hole" where vehicles are forced to join a queue, resulting in an overall longer transit time.

[0072] Therefore, there is an urgent need for a dynamic harmonization mechanism for connected driving styles that can be activated only in specific scenarios and can balance "global efficiency" and "individual style experience".

[0073] Currently, existing methods offer the following solutions to address the aforementioned problems:

[0074] Option 1: Dynamic fleet segmentation and rolling time-domain optimization can improve traffic efficiency while ensuring safety. However, this option relies on a relatively long prediction time (5 seconds), which places high demands on onboard computing resources. Ordinary processors are prone to computational bottlenecks, leading to control command delays and affecting real-time performance. In addition, the system makes decisions entirely autonomously, without providing an interface for drivers to intervene in driving style weights, resulting in weak human-machine collaboration capabilities and difficulty in meeting personalized driving needs. Furthermore, its algorithm focuses on intersection scenarios and lacks adaptability to typical scenarios such as highway ramp merging and high-density synchronous flow, limiting its versatility.

[0075] Option 2, aimed at economical driving, constructs an optimal control model with the goal of minimizing energy consumption and uses a two-layer rolling distance domain strategy to generate an energy-saving vehicle speed spectrum by combining traffic light information. However, its optimization objective is too singular, focusing only on local vehicle energy consumption and neglecting system-level indicators such as regional traffic flow stability, which may lead to a conflict between individual energy saving and overall efficiency. Furthermore, Option 2 does not perform full-process physical feasibility and safety boundary verification for key parameters such as acceleration / deceleration gradients and following distance, posing a potential rear-end collision risk. In addition, the driving style fusion weights are fixed by the system, allowing drivers to only select a basic mode and preventing real-time adjustment of preference intensity, resulting in a rigid human-computer interaction. Moreover, the algorithm is mainly applicable to conventional urban and highway scenarios, failing to consider style adaptation issues in complex environments such as rain, snow, and nighttime.

[0076] Option 3: This option models lane-changing interactions using a Bayesian game theory framework, estimating the driving style of adjacent vehicles using prior statistics and posterior observations, and then solving for the lane-changing probability to improve decision-making intelligence. However, Bayesian inference and game theory solutions are computationally complex, resulting in a latency of approximately 200ms on ordinary in-vehicle platforms, which is insufficient to meet the real-time control requirements of highly dynamic scenarios. Furthermore, its prior style distribution is highly dependent on historical big data, exhibiting poor generalization ability under non-steady-state traffic conditions such as newly opened roads or holidays. More importantly, Option 3 only focuses on the micro-game between the two vehicles, lacking perception and coordination of regional traffic flow states, making it prone to getting trapped in local optima. Additionally, the driving style preference weight of the vehicle itself is fixed at 0.5, unable to be dynamically adjusted according to user intent, limiting the in-depth optimization of individualized experience.

[0077] Therefore, the existing technical solution has one or more of the following defects:

[0078] 1) Coarse adjustment granularity: mostly forced and discrete mode switching or fixed ratio parameter adjustment, lacking continuous and smooth gradual change capability, which impairs the driving experience;

[0079] 2) One-sided decision-making basis: either focusing only on the micro-level behavior of vehicles ahead, or responding only to a single event, failing to comprehensively utilize real-time macro-level traffic flow as the core decision input;

[0080] 3) Local optimization objectives: The adjustment logic is not directly related to system-level objectives such as "maximizing global capacity" and "optimizing traffic flow stability";

[0081] 4) Stiff human-machine collaboration: The failure to intelligently and dynamically integrate the driver's subjective style preferences with the system's overall efficiency goals leads to either excessive intervention or ineffective regulation.

[0082] Based on the deficiencies of existing technical solutions, the vehicle driving style adjustment method provided in this application aims to solve the following technical problems:

[0083] 1. Intelligent Integration of Driver's Personal Preferences and Global System Optimization: This addresses the contradiction between the binary opposition of "driver selection" and "system suggestions" in existing technologies. Specifically, this application designs a dynamic weight allocation mechanism that, while ensuring no strong intervention for the driver, subtly integrates macro-level traffic flow efficiency targets with reasonable weights into the driver's original style settings in specific key scenarios, achieving synergistic optimization of both "individual experience" and "collective interests."

[0084] 2. Using macroscopic traffic flow status as the core decision-making basis to achieve predictive and scenario-based regulation: This addresses the limitations of existing technologies that rely solely on microscopic vehicle behavior or single point events, enabling regulation strategies to respond to complex macroscopic traffic scenarios such as ramp merging, high-density synchronous flow, traffic oscillations, and event bottlenecks, and to provide differentiated treatment.

[0085] 3. Establish a closed-loop link between individual vehicle regulation and overall traffic efficiency goals: This addresses the problem that existing regulation strategies only optimize individual vehicle energy consumption or following comfort without considering their impact on overall traffic flow. Specifically, this application establishes a mapping model between individual vehicle driving parameters (such as following distance and acceleration) and macroscopic traffic flow performance indicators, enabling regulation behavior to directly serve to improve the overall capacity of road segments.

[0086] 4. To achieve smooth and gradual adjustment of driving parameters in specific scenarios, thereby improving user experience and acceptance: This addresses the sudden changes in driving experience and the feeling of "being deprived of control" caused by forced and discrete mode switching in existing technologies, and provides a parameter fusion adjustment mechanism that is imperceptible or slightly imperceptible and continuously gradual.

[0087] In summary, the vehicle driving style adjustment method provided in this application acquires macroscopic traffic flow status (such as flow rate, density, and speed) perceived by roadside units in real time via V2X communication, as well as scenario-based instructions issued from the cloud, and inputs them into a model with the dual objectives of "maximizing global traffic capacity" and "ensuring driving comfort." This model outputs a continuous recommended style coefficient. The system does not force the vehicle to perform this action. Instead, it uses a dynamic weighted fusion algorithm to... To match the baseline style parameters manually set by the driver Perform smooth fusion to generate execution style parameters to be executed. ,Right now: Among them, the fusion weight The system dynamically adjusts based on the confidence level of V2X data and the urgency of the traffic scenario. Therefore, the system... By fine-tuning underlying control parameters such as following distance and acceleration limits in real time and continuously, the vehicle's behavior can intelligently adapt to the current traffic flow while respecting the driver's preferences. The vehicle also transmits actual behavior data back to the cloud for model iteration, thereby achieving the synergistic evolution of individual experience and overall efficiency.

[0088] Figure 1 A schematic diagram illustrating the scenario of the vehicle driving style adjustment method provided in this application, such as... Figure 1 As shown, the specific application scenario of this application can be a vehicle driving style adjustment system, which is suitable for dynamic adjustment scenarios of driving style based on vehicle-road-cloud collaboration. This vehicle driving style adjustment system can be a server, such as an onboard computer. The vehicle driving style adjustment system of this application does not restrict the executing entity, as long as it can obtain traffic flow state information broadcast by the vehicle-road cooperative communication network and the recommended style coefficient determined based on the traffic flow state information; determine the execution style parameters based on the recommended style coefficient and the user's baseline style parameters; and control the vehicle's driving state based on the execution style parameters.

[0089] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0090] Figure 2 Flowchart of the method for adjusting vehicle driving style provided in this application Figure 1 ,like Figure 2 As shown, the method includes:

[0091] S201. Obtain traffic flow status information broadcast by the vehicle-road cooperative communication network, and the recommended style coefficient determined based on the traffic flow status information.

[0092] Among them, the vehicle-road cooperative communication network can refer to a communication system that interacts with the roadside unit (RSU) and the cloud through the on-board unit (OBU). This communication system can be a V2X-based system that can be used to broadcast traffic flow status information.

[0093] Traffic flow status information can refer to three traffic flow parameters: flow rate, density, and speed. In some embodiments, this information may also include traffic event status (such as accidents or construction) and traffic light timing schemes to describe the overall traffic status of a road segment.

[0094] In this embodiment of the application, RSUs deployed at key road nodes (such as ramp entrances and weaving areas) can detect and broadcast traffic flow status information within their coverage area at a preset frequency (such as a 100ms cycle).

[0095] The recommended style coefficient can refer to a set of parameters used to quantify preset driving behavior characteristics in intelligent driving or vehicle-road cooperative systems. In the embodiments of this application, the parameter is used to characterize the driving style benchmark recommended or to be adopted under the current traffic environment. The recommended style coefficient can be a continuous value between 0 and 1, where 0 represents extremely conservative (e.g., maximum time distance, minimum acceleration) and 1 represents extremely aggressive.

[0096] After obtaining traffic flow status information, recommended style coefficients can be dynamically generated using preset driving strategy mapping rules or optimization models. In some embodiments, the recommended style coefficients can be generated by the vehicle after receiving traffic flow status information and combining it with its own driving data, or they can be generated by the cloud or other servers based on traffic flow status information and vehicle driving data, and then sent to the vehicle.

[0097] S202. Determine the execution style parameters based on the recommended style coefficients and the user's baseline style parameters.

[0098] Among them, the user can refer to the driver of the vehicle or a passenger with operating authority, that is, the subject who can set, confirm or intervene in the vehicle's driving style, control mode or system behavior through the human-machine interface (such as the central control screen, voice system or physical knob).

[0099] The baseline style parameter can refer to the initial driving style coefficient manually set by the user to characterize their preferred driving behavior characteristics. In some embodiments, the baseline style parameter can also be the baseline style initially set for the vehicle.

[0100] The execution style parameter refers to the control parameters of the final driving behavior generated when adjusting the vehicle's driving style, after comprehensively considering the recommended style coefficient generated from traffic flow state information and the baseline style parameter manually set by the driver. In some embodiments, it can be determined through weighted fusion, rule-based judgment, or optimization algorithms.

[0101] S203. Control the vehicle's driving status according to the execution style parameters.

[0102] Among them, the driving state of a vehicle can refer to the dynamic operating characteristics of a vehicle at a specific moment and under specific conditions, which include, but are not limited to, physical quantities such as speed, acceleration, heading angle, position, lane departure, and steering angle.

[0103] In this embodiment of the application, after determining the execution style parameters, the key driving behaviors that need to be dynamically adjusted, such as acceleration, deceleration, following distance, lane change timing and steering response, can be determined through preset parameter mapping rules. This ensures that the vehicle's driving state meets the current traffic flow requirements while also reflecting the driver's personalized style preferences, under the premise of ensuring safety and efficiency.

[0104] The vehicle driving style adjustment method provided in this application integrates the baseline style parameters set by the driver with the recommended style coefficients generated based on real-time traffic flow to dynamically generate execution style parameters that are adapted to the current scenario. Based on these parameters, the vehicle's speed, acceleration, steering, and other driving states are adjusted, so that driving behavior can take into account both safety and efficiency and personalized preferences, thereby improving the vehicle's adaptability and driving experience in diverse traffic environments.

[0105] Optionally, traffic flow state information broadcast by the vehicle-road cooperative communication network and recommendation style coefficients determined based on the traffic flow state information are obtained, including:

[0106] Obtain traffic flow status information broadcast by the vehicle-road cooperative communication network and suggested style coefficients sent from the cloud; the suggested style coefficients are determined by the cloud based on the traffic flow status information.

[0107] The recommended style coefficient is obtained based on traffic flow status information, suggested style coefficient, and vehicle status.

[0108] Among them, the suggested style coefficient can refer to the recommended parameters generated by the cloud based on traffic flow status information broadcast by the vehicle-road cooperative network, through big data analysis or model prediction, to guide vehicles to adjust their driving behavior.

[0109] In some embodiments, the cloud integrates multiple RSU data to perform higher-dimensional traffic scenario diagnosis (such as determining whether the current road segment is in "free flow," "synchronous flow," or "oscillating flow") and predict short-term traffic state changes (such as a surge in merging traffic). The cloud then sends the diagnostic results and suggested style coefficients calculated for the specific scenario to the vehicles.

[0110] The vehicle's own state can refer to the vehicle's current real-time state, which can include the vehicle's location and speed.

[0111] In some embodiments, a prediction model can be set up in the cloud, which can be trained based on different traffic flow state sample information and sample labels. This model can be a machine learning or deep learning model, such as a Support Vector Machine (SVM), Random Forest, Gradient Boosting Tree (e.g., XGBoost), Multilayer Perceptron (MLP), or more complex temporal modeling structures such as Long Short-Term Memory Network (LSTM) or Transformer.

[0112] In this embodiment, the suggested style coefficient is a general recommendation parameter generated by the cloud based on the traffic flow status information broadcast by the vehicle-road cooperative network. It does not take into account the real-time status of the vehicle itself. If used directly, it may cause the adjusted vehicle driving style to be mismatched with the actual situation or safety boundary of the vehicle. Therefore, the vehicle needs to correct and adjust the suggested style coefficient based on the traffic flow status information and the vehicle's own status to obtain the recommended style coefficient.

[0113] In some embodiments, after obtaining the suggested style coefficients, the suggested style coefficients can be corrected by constraint mapping, weight correction or optimization algorithms (such as model predictive control, fuzzy logic or rule engine) to obtain the recommended style coefficients.

[0114] In some embodiments, the vehicle model may have an embedded policy library for typical scenarios. For example, for a "ramp merging" scenario, the model may tend to output higher values. (e.g., 0.7-0.9) to improve inflow efficiency; for "high-density synchronous stream" scenarios, the output approaches the median value. (e.g., 0.4-0.6) to maintain flow stability.

[0115] Therefore, by associating the recommended style coefficient with traffic flow state information, suggested style coefficient and vehicle's own state, refined and contextualized control of driving style can be achieved, ensuring that vehicle control not only conforms to the overall traffic coordination intention but also meets the constraints of vehicle dynamics and comfort, thereby improving the practicality and feasibility of vehicle-road cooperative decision-making.

[0116] In this embodiment of the application, the recommended style coefficient is obtained based on traffic flow state information, suggested style coefficient, and vehicle status, including:

[0117] The corresponding scene processing model is determined based on the scene tags sent from the cloud; the scene tags are determined by the cloud based on traffic flow status information.

[0118] Traffic flow status information, suggested style coefficients, and vehicle status are input into the scene processing model to obtain suggested style coefficients.

[0119] Among them, scene label refers to the result of the cloud identifying the traffic state of the current vehicle based on traffic flow status information. This result can characterize the macroscopic operating state of the traffic flow of the current road segment, which can correspond to unobstructed flow (free flow), low speed but high and relatively stable flow (synchronous flow), and unstable congestion state with drastic flow fluctuations and frequent sudden drops in speed (wide-amplitude oscillation flow).

[0120] A scene processing model can refer to an optimization objective function and its constraints customized for different scene labels. For example:

[0121] In the scenario of ramp merging, the scenario processing model can aim to maximize the traffic efficiency of the merging area by coordinating traffic flow by minimizing the total delay of vehicles on the main line and ramps in the prediction time domain.

[0122] In high-density synchronous flow scenarios, the scenario processing model can improve traffic flow stability by steering, setting the goal to minimize speed fluctuations, and introducing a constraint that the rate of change of vehicle acceleration (jerk) does not exceed the comfort threshold, thereby suppressing congestion spread while ensuring driving comfort.

[0123] In this embodiment, the corresponding scene processing model can be activated based on the scene tags provided by the cloud. This model can serve as an optimization or mapping mechanism customized for specific traffic scenarios. It comprehensively considers the suggested style coefficient, traffic flow state information, and vehicle state, and performs fusion calculation on the input parameters through its embedded objective function and constraints. Thus, it can output a recommended style coefficient that is suitable for the current scenario and is safe and feasible.

[0124] For example, in some embodiments, the scene processing model can be a weighted fusion formula that satisfies:

[0125]

[0126] In this case, the weight coefficients a, b, and c are different for different scenario processing models.

[0127] Therefore, by calling the scene processing model that matches the scene label, the suggested style coefficient is dynamically modified in a contextualized manner. This not only ensures that the driving style is highly consistent with the macro traffic situation, but also takes into account the physical constraints and comfort requirements of the vehicle, making the control commands more precise and feasible, thereby realizing the transformation from general suggestions to optimal execution in the scene.

[0128] In this embodiment of the application, after determining the execution style parameters based on the recommended style coefficient and the user's baseline style parameters, the method further includes:

[0129] The execution style parameters and vehicle driving data are sent to the cloud so that the cloud can train the prediction model based on the execution style parameters, vehicle driving data and traffic flow state information; the prediction model is used to output suggested style coefficients based on traffic flow state information.

[0130] The prediction model can refer to a model that outputs suggested style coefficients based on traffic flow state information. In some embodiments, the model can be a general prediction model applicable to multiple scenarios and vehicle types, or a special model customized for specific users, vehicle characteristics, or typical traffic scenarios.

[0131] Vehicles can send execution style parameters and vehicle driving data (such as vehicle trajectory, speed profile, etc.) to the cloud. The cloud uses a large amount of data returned by vehicles to continuously train and iterate the parameters of the prediction model, making its prediction of suggested style coefficients more and more accurate, thus forming an enhanced closed loop of data-driven optimization.

[0132] Therefore, by continuously training and iterating the parameters of the prediction model in the cloud, the prediction accuracy of the suggested style coefficient is continuously optimized, thereby constructing a data-driven enhancement closed loop with actual traffic data and driving feedback as input and style strategy optimization as output, so as to realize the continuous evolution and adaptive improvement of the driving decision-making ability of the vehicle-road cooperative system.

[0133] Optionally, the vehicle's driving state is controlled according to the execution style parameters, including:

[0134] The driving parameters are determined based on the execution style parameters; the type of driving parameters is determined based on the baseline style parameters.

[0135] The vehicle's driving status is smoothly controlled based on driving parameters.

[0136] Among them, driving parameters can refer to control commands that can be directly called and executed by the vehicle's underlying controller (such as longitudinal acceleration controller, steering actuator or ADAS module), such as target vehicle speed, desired acceleration, brake pressure request, throttle opening, front wheel angle or trajectory curvature, etc. These parameters are transformed from execution style parameters to concretize the abstract driving style (such as aggressive, smooth or conservative) into underlying control quantities that conform to vehicle dynamics constraints and can be executed in real time, thereby achieving driving behavior consistent with the target driving style at the physical level.

[0137] In some embodiments, the type of driving parameters (such as emphasizing speed tracking, following distance control, or highlighting acceleration smoothness and trajectory stability) can be determined by the driver's preferences represented by the baseline style parameters. For example, when the baseline style parameters are aggressive, driving parameters centered on target speed and rapid response are prioritized; while when the coefficients are more comfortable or conservative, parameter types characterized by low jerk, large following distance, and soft steering are more commonly used, thereby ensuring that the underlying control commands align with the user's preset driving style. Figure 1 To.

[0138] After determining the driving parameters, the driving parameters are input into the smooth control module at the bottom layer of the vehicle. Under the premise of meeting the vehicle dynamics constraints and comfort boundaries, the control commands are continuously optimized and smoothed in the time domain to generate throttle, braking and steering execution signals without sudden changes and with low vibration. This enables smooth and precise control of the vehicle's speed, position and heading, thus avoiding driving shocks or instability caused by style switching or external disturbances.

[0139] In the embodiments of this application, the driving parameters include, but are not limited to, driving parameters characterized as desired following distance, driving parameters characterized as acceleration limits, driving parameters characterized as energy recovery torque, and driving parameters characterized as lateral decision thresholds.

[0140] Among them, the expected following distance can refer to the time interval between the vehicle and the vehicle in front, which is set to maintain safe and smooth driving when the vehicle is following another vehicle.

[0141] Acceleration limits refer to the upper limit of a vehicle's acceleration and deceleration, used to constrain the intensity of longitudinal motion and ensure driving comfort and safety.

[0142] Energy recovery torque can refer to the energy recovery torque during coasting or braking;

[0143] Lateral decision threshold can refer to the minimum safe clearance required for actions such as lane changing.

[0144] In the embodiments of this application, if To execute the style parameters, then:

[0145] When the driving parameter is represented as the expected following distance, the driving parameter is determined based on the expected following distance and the execution style parameter;

[0146] The desired following distance for the vehicle is set as follows: .Should The mapping rules satisfy:

[0147] ;

[0148] in, and These correspond to the conservative (maximum) and aggressive (minimum) following distance boundaries allowed by the system, respectively.

[0149] When the driving parameters are represented as acceleration limits, the driving parameters are determined based on the preset boundary of acceleration and the adjustment function of the execution style parameters;

[0150] Among them, the upper limit of the acceleration and deceleration range of the vehicle (i.e., the acceleration limit) is set as follows: ;Should The mapping rules satisfy:

[0151] ;

[0152] in, and The preset minimum and maximum boundaries for acceleration. This is an adjustment function used to achieve a gradual increase in acceleration over an aggressive interval. In some embodiments, it can be a function relating to... The nonlinear adjustment function, for example, can be a sigmoid function, a piecewise linear function, or a power function with saturation properties, such that... As the acceleration increases, the rate of increase gradually slows down.

[0153] When the driving parameter is represented as energy recovery torque, the driving parameter is determined based on the maximum recovery torque and the execution style parameter;

[0154] The energy recovery torque during coasting or braking is set as follows: ; The mapping rules satisfy:

[0155] ;

[0156] in, This represents the maximum regenerative torque. The mapping rule for this regenerative torque allows for a more aggressive driving style, resulting in a gliding feel closer to that of a traditional gasoline-powered vehicle.

[0157] When the driving parameters are represented as lateral decision thresholds, the driving parameters are determined based on theoretical safety gaps, scenario coefficients, and execution style parameters.

[0158] The minimum safety gap (i.e., the lateral decision threshold) required for actions such as lane changing is set as follows: ,Should The mapping rules satisfy:

[0159] ;

[0160] in, The hyperparameter represents the scene coefficients. and These are the preset maximum and minimum safe clearances, respectively. This mapping rule allows for more aggressive lane-changing decisions with a more aggressive style.

[0161] In some embodiments, after obtaining the driving parameters through mapping rules, each driving parameter can be smoothed by first-order low-pass filtering or ramp transition, thereby ensuring asymptotic output within several control cycles, so that the smoothed parameters can be sent to the underlying controllers such as ACC, energy management, and steering for execution.

[0162] In some embodiments, the adjustment of driving parameters can either coordinate and optimize multiple parameters to achieve consistency in the overall driving style, or independently and finely adjust a certain type of parameter according to specific scenario requirements or control priorities, thereby flexibly responding to changes in dynamic traffic environment and user preferences while ensuring safety and comfort.

[0163] Figure 3 Flowchart of the method for adjusting vehicle driving style provided in this application Figure 2 ,like Figure 3 As shown, in this embodiment... Figure 2 Based on the examples, the steps for determining the execution style parameters according to the recommended style coefficients and the user's baseline style parameters are described in detail. The method includes:

[0164] S301. Obtain the user's adjustment intention based on the recommended style coefficient and the baseline style parameters;

[0165] S302. If the intention to adjust indicates that the user agrees to the adjustment, the execution style parameters shall be determined based on the recommended style coefficient and the user's baseline style parameters.

[0166] Among them, the intention to adjust can refer to the user's willingness to actively intervene or correct the driving style through the human-computer interaction interface based on the current driving experience, changes in road conditions, or personal preferences.

[0167] When the recommended style coefficients and baseline style parameters do not match, the recommended driving style conflicts with the driver's preset preferences, potentially affecting driving comfort, acceptability, or collaborative efficiency. In this case, an interactive query can be sent to the user. If the user's feedback indicating an intention to adjust based on the query shows agreement with the system's suggestion, the recommended style coefficients and baseline style parameters are merged to generate execution style parameters that take into account both traffic collaboration needs and user preferences, for subsequent vehicle control.

[0168] In this embodiment of the application, obtaining the user's adjustment intention based on the recommended style coefficient and the baseline style parameter includes:

[0169] If the difference between the recommended style coefficient and the baseline style parameter is greater than the preset difference value, a query message will be output to the user.

[0170] By responding to user responses to inquiries, we can understand the user's intention to make adjustments.

[0171] Among them, the difference between the recommended style coefficient and the baseline style parameter being greater than the preset difference value can indicate that the driving style recommended by the vehicle deviates from the driver's preset preference style in terms of quantitative indicators, exceeding the acceptable fusion tolerance range. This indicates that if the driving style is directly implemented, it may cause problems such as driving discomfort, operation conflict or decreased collaborative efficiency. It is necessary to send an inquiry to the user to determine whether to adjust the driving style.

[0172] In this embodiment of the application, the preset difference value can be based on a value pre-set for the user. For example, the threshold can be 0.3.

[0173] Inquiry information refers to the interactive prompts proactively sent to the user when the system detects a difference between the recommended driving style in the current traffic scenario and the driver's set baseline style. This information includes, but is not limited to, a description of the scenario and the intention to suggest adjustments. In some embodiments, inquiry information can be output through voice, images, or other means.

[0174] In this embodiment, the query information can be output via HMI (such as a pop-up window on the central control screen or voice), and this information may include a scenario clearly explained to the driver and the intention to make the suggested adjustments. For example, "The system suggests slightly shortening the following distance to improve overall traffic efficiency when merging at the next ramp. Is this adjustment allowed?"

[0175] The response information can refer to the answer to the inquiry, which can include two types of user intent regarding the adjustment: agreement to the adjustment and disagreement with the adjustment. In some embodiments, users can provide feedback to the vehicle through various natural interaction methods, including voice commands, selecting preset options on the in-vehicle interaction platform (such as the central control screen or HUD), and non-contact behaviors such as gestures, body movements, or facial expressions. The vehicle uses multimodal perception technology to identify and analyze these inputs, thereby accurately obtaining the user's intent to confirm or reject the driving style adjustment suggestion.

[0176] In this embodiment of the application, when the user agrees to the adjustment, the execution style parameters are determined based on the recommended style coefficient and the user's baseline style parameters, including:

[0177] Dynamic weights are determined based on the confidence level of traffic flow status information and the urgency of traffic scenarios;

[0178] The execution style parameters are determined based on dynamic weights, recommended style coefficients, and baseline style parameters.

[0179] The dynamic weights are used to weight and fuse the recommended style coefficients and the baseline style parameters. These parameters can be determined based on the confidence level of traffic flow state information and the urgency of the traffic scenario.

[0180] The confidence level of traffic flow status information can be used as a quantitative indicator to characterize the completeness of traffic flow status information, and its value ranges from [0, 1]. For example, when V2X broadcast data is missing, the confidence level of traffic flow status information can be 0.5.

[0181] In some embodiments, the confidence level of traffic flow status information can be quantified by the completeness, timeliness, and consistency of the traffic flow status information. For example, the confidence level of complete V2X broadcast data with a latency of less than 100ms is set to 1.0. If some lane data is missing or there is a brief packet loss (such as 1 to 2 cycles), the confidence level decreases linearly to 0.6 to 0.8 according to the proportion of missing data. When relying entirely on historical extrapolation or single-vehicle perception to complete the data (such as RSU not broadcasting for a long time), the confidence level drops to 0.3 to 0.5. If the data from multiple sources (such as RSU, cameras, and floating cars) are mutually verified to be consistent, the confidence level can be increased to above 0.9.

[0182] Traffic scenario urgency is a quantitative indicator that represents the urgency of traffic scenario response to adjustment needs, with a value range of [0, 1]. For example, the urgency parameter for a ramp merging into a bottleneck scenario is 0.8.

[0183] In some embodiments, the urgency level of a traffic scenario can be preset with corresponding urgency parameters based on different traffic scenario labels.

[0184] In this embodiment of the application, the confidence level of traffic flow status information and / or the urgency of traffic scenarios can be determined by the cloud based on the traffic flow status information, or by other servers that have obtained the traffic flow status information.

[0185] In some embodiments, the dynamic weights can be constrained to the range [0, 1]. satisfy:

[0186] ;

[0187] ;

[0188] in, is the basic weight; C is the confidence level of traffic flow state information; The urgency level is determined by the traffic situation.

[0189] In some embodiments:

[0190] Based on confidence level: When V2X data quality is high and the confidence level for scene diagnosis is >90%, A higher value can be taken (such as 0.7-0.9).

[0191] Based on the urgency of the scenario: In special safety scenarios such as "emergency vehicles behind", the system can temporarily override the weight settings, which can then be forced. That is, to fully adopt (For aggressive styles with a value >0.8) Perform urgent actions such as yielding.

[0192] Therefore, when traffic flow status information is reliable and the traffic scenario is critical, the system's recommendations will have a higher weight.

[0193] Therefore, by quantifying the confidence parameter and the scenario urgency parameter, the dynamic weight parameter can be adaptively adjusted, making the calculation logic of the dynamic weight parameter more scenario-adaptable, thus accurately matching the adjustment needs of traffic flow status while respecting driver preferences.

[0194] In this embodiment of the application, the execution style parameters are determined based on dynamic weights, recommended style coefficients, and baseline style parameters, including:

[0195] The theoretical style correction coefficient is obtained based on the dynamic weights and the recommended style coefficient.

[0196] Based on the dynamic weights and the baseline style parameters, the baseline style correction parameters are obtained;

[0197] The execution style parameters are obtained based on the theoretical style correction coefficient and the baseline style correction parameter.

[0198] Among them, execution style parameters satisfy:

[0199] ;

[0200] in, This is the recommended style coefficient; This is a theoretical style correction coefficient; Used as the baseline style parameter; Adjust parameters for the baseline style; For dynamic weights.

[0201] The vehicle driving style adjustment method provided in this application integrates data reliability, scenario criticality, and driver authorization willingness in a quantitative manner, thereby achieving a refined trade-off in driving style adjustment strategies. In low-risk or high-confidence scenarios, the system prioritizes maintaining the user's preferred baseline style to avoid unnecessary intervention; while in high-risk or low-confidence scenarios, the system increases the probability of adopting the recommended style coefficient to ensure safety as the primary goal, thus achieving a dynamic balance between user preferences and traffic collaboration needs.

[0202] Optionally, after obtaining the user's adjustment intention based on the recommended style coefficient and the user's baseline style parameters, the method further includes:

[0203] If the user refuses to adjust the intended style, the baseline style parameter will be used as the executed style parameter.

[0204] The vehicle's driving status is controlled based on the execution style parameters.

[0205] When a user receives a system inquiry about changing their driving style, they can clearly express their desire to maintain their current driving style (i.e., refuse to adjust) through voice, touch selection, gestures, or other interactive methods. At this point, the baseline style parameter is used as the execution style parameter, and the driver can drive according to their original preference.

[0206] The vehicle driving style adjustment method provided in this application effectively protects the driver's control over vehicle behavior by using the baseline style parameter as the execution style parameter when the user explicitly refuses style adjustment. This enhances the trust in human-computer interaction and the consistency of the driving experience, while avoiding the abruptness of operation and psychological burden caused by unnecessary intervention.

[0207] Figure 4 A schematic diagram of the architecture of the vehicle driving style adjustment system provided in this application is shown below. Figure 4 As shown, this system architecture can be implemented in the electronic control unit (ECU), domain controller, or onboard computing platform of an intelligent connected vehicle through software, hardware, or a combination of both, and is deeply integrated with the vehicle's longitudinal / lateral control system, human-machine interface (HMI), and V2X communication module (such as a T-Box). The methods of its use include:

[0208] I. System Initialization and Driver Preference Settings;

[0209] When the vehicle is started or a function is activated for the first time, the system enters an initialization state. The in-vehicle HMI provides the driver with a preset driving style interface, with the style based on continuous baseline style parameters. This indicates that the range is typically [0, 1]. For example, the driver can use a slider to select "Comfort" (corresponding to...). =0.2), "standard" ( =0.5), "movement" ( d The selection can be made steplessly between 0.8 and 0.8, or a specific value can be set directly. The value is stored as the driver's original preference baseline for this trip. At the same time, the driver can initially set the interaction preference of the "Dynamic Style Fusion" function (such as selecting the default interaction mode), and the system loads the built-in default control parameter mapping table and the initial parameters of the traffic flow benefit assessment model;

[0210] II. Real-time perception of macroscopic traffic flow information based on V2X;

[0211] During operation, the vehicle continuously listens to broadcast information from the RSU via its OBU (On-Board Unit) and maintains a connection with the cloud (cloud-based traffic brain) server via cellular network (4G / 5G).

[0212] 1) Directly sensed information from RSUs: Vehicles receive raw traffic flow parameters (q, density, etc.) broadcast in real time from nearby RSUs. Spatial average velocity v, as well as event identifiers (such as accidents, construction, weather warnings) and local scene identifiers (such as "upstream of ramp merging area").

[0213] 2) Cloud-based fusion and diagnostic information: The cloud fuses wide-area multi-RSU data to perform traffic state identification (e.g., determining whether the current road segment is in "free flow," "synchronous flow," or "wide-amplitude oscillating flow" based on qk graphs or speed variance) and makes short-term predictions. The cloud then labels the diagnosed traffic scene (e.g., "high-density synchronous flow," "ramp merging bottleneck") and the suggested style coefficients calculated based on global optimization. Issued to vehicles within this area. Among them, It can serve as an important input reference for traffic flow benefit assessment models.

[0214] III. Online solution of recommendation style coefficient for traffic flow revenue assessment model ;

[0215] The local (i.e., vehicle-side) traffic flow revenue assessment model (core decision engine) is triggered at fixed intervals (e.g., every 1 second). Its workflow is as follows:

[0216] 1) Input processing: The model receives the current macroscopic state (q) from the V2X. v), scene tags distributed from the cloud and This is combined with the real-time status of the vehicle (such as speed and location).

[0217] 2) Scene matching and objective function construction: Based on the scene labels, the model calls a pre-defined optimization objective function. The core objective is to maximize the traffic capacity of the road segment.

[0218] "Ramp merging" scenario: The objective function tends to maximize the throughput of the merging area, and its mathematical expression may be to minimize the overall delay between the mainline and ramp traffic flow in the prediction time domain;

[0219] In the "high-density synchronous flow" scenario, the objective function focuses on minimizing traffic flow speed fluctuations to improve stability, with the constraint that the rate of change of vehicle acceleration (jerk) does not exceed the comfort threshold.

[0220] 3) Rolling optimization solution: The model adopts the MPC framework. Within the prediction time domain of T (e.g., 10 seconds), the recommended style coefficient k (the vehicle's own state) is used as the optimization variable to perform extrapolation and solution, thereby obtaining the optimal recommended style coefficient that maximizes the system objective. .

[0221] 4) Output: The model outputs continuous recommendation style coefficients. (range [0, 1]), and the confidence level C of the current solution (calculated based on the integrity and consistency of the input data, 0 ≤ C ≤ 1).

[0222] IV. Dynamic Style Fusion Calculation and Execution Style Coefficient ;

[0223] Dynamic style fusion unit receives and and based on and Calculate the final execution The process is as follows:

[0224] 1) Judgment and Interaction: Calculation ,like If the threshold is exceeded, an inquiry or automatic decision will be initiated based on the driver's preset interaction protocol.

[0225] 2) Weighting and fusion calculation:

[0226] If the decision result is "reject adjustment", then ;

[0227] If the decision result is "adjustment allowed", then the dynamic weights are calculated based on the current data confidence level C and the scenario urgency level S. Then through the formula Computational fusion .

[0228] 3) Output: Output Zhizhi Driving Strategy Execution Unit.

[0229] V. Smooth mapping and execution of driving control parameters;

[0230] In practical implementation, the intelligent driving strategy execution unit operates according to the following steps to achieve smooth issuance of control commands:

[0231] Step 1: Parameter Mapping Calculation. In each control cycle, the unit calculates the parameter mapping based on the latest... The instantaneous theoretical values ​​of all target control parameters are calculated using a preset set of mapping functions. For example:

[0232] Calculate the instantaneous expected time interval (unit: seconds): ;

[0233] Calculate the upper limit of instantaneous acceleration (unit: ):

[0234] ;

[0235] Step Two: Smoothing Filtering. Input the instantaneous theoretical value calculated in Step One into the smoothing filter. Taking the desired time interval as an example, a first-order low-pass filtering algorithm is used:

[0236] ;

[0237] in, For filter coefficients (0 < <1), the value of which determines the speed of the transition, for example, a transition time of 2-3 seconds. (n) is the filtered output value at time n of this cycle.

[0238] Step 3: Instruction Issuance and Execution. The final parameter values ​​after smoothing filtering ( , (etc.) are sent to the corresponding controller in the vehicle in real time:

[0239] Towards adaptive cruise control: as a new target for following distance and speed control;

[0240] To the vehicle controller / power domain controller: as a new frontier for torque response and energy recovery strategies;

[0241] To the driver assistance decision-making module: as a sensitive parameter for decision-making on behaviors such as lane changing and merging.

[0242] By repeating the above three steps in a loop, the vehicle... The guided behavioral changes are continuous and gradual, and the driver perceives only subtle and smooth adaptive adjustments to the vehicle style, thereby improving traffic efficiency while ensuring excellent driving comfort and functional acceptability.

[0243] VI. Data closed-loop feedback and model iteration;

[0244] 1) Data Feedback: After anonymization, the vehicle periodically uploads its data packets to the cloud via cellular network. The data packets include: timestamp, location, ... Values, executed control parameters, actual vehicle trajectories (speed, acceleration), and locally perceived microscopic traffic conditions (such as actual vehicle spacing).

[0245] 2) Cloud Aggregation and Model Training: Massive amounts of anonymized vehicle feedback data are collected in the cloud and spatiotemporally aligned with macro-level traffic flow data to construct a "driving style-traffic flow response" database. The strategy library and prediction model parameters in the traffic flow benefit assessment model are periodically trained offline and updated online to ensure their effectiveness. The predictions are more accurate, forming a "data-driven" continuous optimization loop;

[0246] 3) Software Update (OTA): Optimized model parameters or strategy libraries are delivered to vehicles via OTA to complete the evolution of the entire system's capabilities.

[0247] Therefore, through the cyclical execution of the above six steps, this application achieves dynamic, smooth, and intelligent adjustment of a single vehicle's driving style in specific traffic scenarios, effectively serving the overall goal of improving road traffic efficiency while minimizing driver intervention.

[0248] For example, when the vehicle driving style adjustment method of this application is applied to the merging area scenario of an urban expressway ramp, the scenario settings include:

[0249] I. Scene Setting;

[0250] Road conditions: A three-lane urban expressway with moderate to dense traffic on the main line. ,speed There is a merging ramp about 1 kilometer ahead. Vehicles on the ramp must merge into the rightmost lane.

[0251] Vehicle status: Vehicle A is traveling in the rightmost lane with adaptive cruise control activated. Driver A prefers a smooth and comfortable driving experience and set the baseline style parameters to [value missing] during system initialization. (This corresponds to "Comfort" mode, which defaults to a larger following distance and gentler acceleration.)

[0252] Question: If vehicle A... The correspondingly large fixed time interval when passing through the merging area may lead to an overly conservative approach when facing merging vehicles, resulting in insufficient utilization of the merging gap, turning it into a "moving roadblock," reducing the overall merging efficiency of the ramp, and potentially triggering a chain reaction of braking by vehicles behind.

[0253] Its workflow includes:

[0254] S1, Perception Stage;

[0255] Vehicle A's OBU received a broadcast message from the upstream RSU: "The merging area is 800 meters ahead. Current mainline traffic flow..." ,density ,speed Ramp flow detection Meanwhile, based on multi-source data fusion, the cloud platform diagnoses the area as a "moderate congestion active merging bottleneck" scenario, and sends scenario labels and cloud-calculated suggested style coefficients to vehicles (including vehicle A) within the area. .

[0256] S2, Decision-making stage (local model calculation in the vehicle);

[0257] Local revenue assessment model calculation: The local traffic flow revenue assessment model for vehicle A is triggered. The model inputs received macro data, cloud-based suggestions, and vehicle status into a pre-set "ramp cooperative merging" optimization sub-model. The core objective of this model is to maximize the throughput of the merging area within the next 10 seconds (prediction time domain). After MPC rolling optimization, the model outputs the locally considered optimal style coefficient. (That is, the recommendation style coefficient, which is higher than the cloud suggestion because the local model perceives that the car behind it is close enough), and the confidence level of this decision is calculated to be C=0.85 (data is complete and the scene matching is high).

[0258] S3, Integration Phase (Interpersonal Collaborative Decision Making);

[0259] a. Triggering Interaction: Calculating Differences (Interaction threshold) triggers human-computer interaction.

[0260] b. Driver response: The HMI pops up a prompt: "Merging onto the ramp ahead. The system suggests slightly shortening the following distance to improve overall traffic efficiency. Do you allow the adjustment?"

[0261] In some embodiments, the driver may select from the following options:

[0262] 0 (each query): The system queries every time the difference exceeds the threshold.

[0263] 1 (Full Agreement): In the current itinerary, for all similar scenarios, the system will automatically adopt its suggested adjustments (without needing to ask again).

[0264] 2 (Complete Rejection): During the current trip, the system will no longer automatically adjust the driving style and will always maintain [the specified style]. .

[0265] 3 (Agreed for this time only): Adjustments are permitted only for this current system suggestion.

[0266] 4 (Rejection Only for This Time): Only the current adjustment suggestion is rejected.

[0267] The system will record the driver's choices and manage subsequent interaction logic accordingly, aiming to reduce unnecessary disturbance to experienced users while preserving full awareness and control for new users.

[0268] After the user makes a selection:

[0269] Scenario 1 (Driver selects to allow): Driver A selects "3 (Agree only for this time)".

[0270] Scenario 2 (Driver not responding): Mr. A is focused on driving and does not operate the system. After 5 seconds, the system will process the request according to the preset default behavior (such as "default consent").

[0271] c. Calculate the execution coefficient: Subject to adjustments:

[0272] Calculate the base weights: The scenario urgency level S is preset to 0.8 for this scenario.

[0273] Determine the fusion weights: .

[0274] Calculate the execution style coefficient:

[0275] .

[0276] If the driver selects an option such as "Refuse for the entire journey," then the settings will be directly configured. .

[0277] S4, Execution Phase;

[0278] Received by the intelligent driving strategy execution unit hour:

[0279] Parameter mapping calculation:

[0280] Instantaneous expected time interval: .

[0281] Upper limit of instantaneous acceleration: .

[0282] Smoothing filtering: The instantaneous value above is passed through a first-order low-pass filter (with a set transition time of approximately 3 seconds) to produce a smoothly changing target value. and .

[0283] Control Execution: The smoothed parameters are sent to controllers such as ACC. Vehicle A begins to imperceptibly and gradually shorten the time distance with the vehicle in front (smoothly transitioning from about 1.84s to 1.43s), and moderately increases the upper limit of acceleration response to create a better gap for merging vehicles.

[0284] S5, Feedback Phase;

[0285] After vehicle A passes through the merging area, it will send an anonymous data packet (containing timestamp, location, ... Data such as actual time distance and speed curves are uploaded to the cloud. The cloud aggregates similar data from hundreds or thousands of vehicles to analyze different scenarios involving "ramp merging bottlenecks." The relationship between the value and key efficiency indicators such as the merging success rate and the main line speed maintenance rate is used to iteratively optimize the benefit evaluation model (i.e., the prediction model) in the "ramp merging" scenario.

[0286] Therefore, the vehicle driving style adjustment method provided in this application embodiment can:

[0287] 1. It achieves intelligent and flexible human-machine collaboration, fundamentally solving the feeling of "being usurped": the system does not forcibly switch modes, but instead, through clear scenario-based prompts ("Merging at the next ramp...") and diverse options (such as "Agree only this time"), it gives the final decision-making power to the driver. Only after the driver authorizes it does it allow the system to consider the driver's personal preferences ( ) and system efficiency recommendations ( ) through dynamic weights ( Intelligent fusion is performed to obtain a balanced solution. This design fully respects the driver's autonomy and individual needs, significantly improving the acceptability of the functionality.

[0288] 2. Based on the predictive adjustment of macro traffic flow status, the limitations of micro perception are overcome: the decision-making basis is no longer the instantaneous emergency response when merging vehicles cut in, but the scenario diagnosis of "merging bottleneck" in the cloud based on macro traffic flow and density data provided by V2X in advance. This makes the adjustment forward-looking and can smoothly adjust vehicle behavior before potential problems occur, changing from passive response to active optimization.

[0289] 3. Directly link individual vehicle behavior with the global traffic efficiency objective in a closed loop: Solve using the local model. The optimization objective is directly to "maximize the throughput of the merging area," a system-level performance indicator. This ensures that every style adjustment (shortening the time distance) of vehicle A has a clear global positive significance, guaranteeing that the intelligent behavior of a single vehicle serves the overall traffic flow optimization, rather than isolated vehicle comfort or energy saving.

[0290] 4. It achieves smooth adjustment with "unnoticeable" or "minimal" sensation, greatly improving driving comfort and experience: by using k e The system continuously maps to control parameters and uses a first-order low-pass filter for smooth transitions. All changes to control parameters are completed linearly and gradually within 2-3 seconds, thus improving the user's driving experience.

[0291] Based on this, the embodiments of this application can dynamically reconcile individual driving styles with collective traffic interests in a user-friendly, globally efficient, and smooth manner, effectively resolving typical contradictions such as mobile roadblocks and phantom traffic jams, and providing a practical and feasible technical solution for realizing the overall efficiency of the transportation system in an intelligent connected environment.

[0292] Figure 5 A schematic diagram of the structure of the vehicle driving style adjustment device provided in this application is shown below. Figure 5 As shown, the vehicle driving style adjustment device 50 provided in this embodiment includes:

[0293] The acquisition module 501 is used to acquire traffic flow status information broadcast by the vehicle-road cooperative communication network, as well as the recommended style coefficient determined based on the traffic flow status information.

[0294] The determination module 502 is used to determine the execution style parameters based on the recommended style coefficients and the user's baseline style parameters.

[0295] The control module 503 is used to control the driving status of the vehicle according to the execution style parameters.

[0296] In one possible implementation, the acquisition module 501 can also be specifically used for:

[0297] Obtain traffic flow status information broadcast by the vehicle-road cooperative communication network and suggested style coefficients sent from the cloud; the suggested style coefficients are determined by the cloud based on the traffic flow status information.

[0298] The recommended style coefficient is obtained based on traffic flow status information, suggested style coefficient, and vehicle status.

[0299] In one possible implementation, the acquisition module 501 can also be specifically used for:

[0300] The corresponding scene processing model is determined based on the scene tags sent from the cloud; the scene tags are determined by the cloud based on traffic flow status information.

[0301] Traffic flow status information, suggested style coefficients, and vehicle status are input into the scene processing model to obtain suggested style coefficients.

[0302] In one possible implementation, the determining module 502 can also be specifically used for:

[0303] Based on the recommended style coefficient and the baseline style parameters, the user's adjustment intention is obtained;

[0304] If the intention to adjust indicates that the user agrees to the adjustment, the execution style parameters are determined based on the recommended style coefficient and the user's baseline style parameters.

[0305] If the user refuses to adjust the intended style, the baseline style parameter will be used as the executed style parameter.

[0306] The vehicle's driving status is controlled based on the execution style parameters.

[0307] In one possible implementation, the determining module 502 can also be specifically used for:

[0308] If the difference between the recommended style coefficient and the baseline style parameter is greater than the preset difference value, a query message will be output to the user.

[0309] By responding to user responses to inquiries, we can understand the user's intention to make adjustments.

[0310] In one possible implementation, the determining module 502 can also be specifically used for:

[0311] Dynamic weights are determined based on the confidence level of traffic flow status information and the urgency of traffic scenarios;

[0312] The execution style parameters are determined based on dynamic weights, recommended style coefficients, and baseline style parameters.

[0313] In one possible implementation, the determining module 502 can also be specifically used for:

[0314] The theoretical style correction coefficient is obtained based on the dynamic weights and the recommended style coefficient.

[0315] Based on the dynamic weights and the baseline style parameters, the baseline style correction parameters are obtained;

[0316] The execution style parameters are obtained based on the theoretical style correction coefficient and the baseline style correction parameter.

[0317] In one possible implementation, the control module 503 can also be specifically used for:

[0318] The driving parameters are determined based on the execution style parameters; the type of driving parameters is determined based on the baseline style parameters.

[0319] The vehicle's driving status is smoothly controlled based on driving parameters.

[0320] In one possible implementation, the control module 503 can also be specifically used for:

[0321] When the driving parameter is represented as the expected following distance, the driving parameter is determined based on the expected following distance and the execution style parameter;

[0322] And / or,

[0323] When the driving parameters are represented as acceleration limits, the driving parameters are determined based on the preset boundary of acceleration and the adjustment function of the execution style parameters;

[0324] And / or,

[0325] When the driving parameter is represented as energy recovery torque, the driving parameter is determined based on the maximum recovery torque and the execution style parameter;

[0326] And / or,

[0327] When the driving parameters are represented as lateral decision thresholds, the driving parameters are determined based on theoretical safety gaps, scenario coefficients, and execution style parameters.

[0328] The vehicle driving style adjustment device provided in this embodiment can perform the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0329] Figure 6 This is a schematic diagram of the controller provided in this application. Figure 6 As shown, the controller 60 provided in this embodiment includes at least one processor 601 and a memory 602. Optionally, the controller 60 further includes a communication component 603. The processor 601, memory 602, and communication component 603 are connected via a bus 604.

[0330] In a specific implementation, at least one processor 601 executes computer execution instructions stored in memory 602, causing at least one processor 601 to perform the above-described method.

[0331] The specific implementation process of processor 601 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0332] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0333] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0334] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0335] This application also provides a vehicle, including: a vehicle body and a controller disposed in the vehicle body.

[0336] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0337] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0338] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0339] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0340] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0341] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0342] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0343] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0344] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0345] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for adjusting vehicle driving style, characterized in that, Applications in vehicles include: Obtain traffic flow status information broadcast by the vehicle-road cooperative communication network, and a recommendation style coefficient determined based on the traffic flow status information; The execution style parameters are determined based on the recommended style coefficients and the user's baseline style parameters; The driving state of the vehicle is controlled according to the execution style parameters.

2. The method according to claim 1, characterized in that, The acquisition of traffic flow state information broadcast by the vehicle-road cooperative communication network and the recommendation style coefficient determined based on the traffic flow state information include: The system acquires traffic flow status information broadcast by the vehicle-road cooperative communication network and suggested style coefficients sent from the cloud; the suggested style coefficients are determined by the cloud based on the traffic flow status information. The recommended style coefficient is obtained based on the traffic flow status information, the suggested style coefficient, and the vehicle's own status.

3. The method according to claim 2, characterized in that, The step of obtaining the recommended style coefficient based on the traffic flow state information, the suggested style coefficient, and the vehicle's own state includes: The corresponding scene processing model is determined based on the scene tags sent from the cloud; the scene tags are determined by the cloud based on the traffic flow status information. The traffic flow state information, the suggested style coefficient, and the vehicle's own state are input into the scene processing model to obtain the suggested style coefficient.

4. The method according to claim 1, characterized in that, The step of determining the execution style parameters based on the recommended style coefficient and the user's baseline style parameters includes: Based on the recommended style coefficient and the baseline style parameter, the user's adjustment intention is obtained; If the adjustment intention indicates that the user agrees to the adjustment, the execution style parameters are determined based on the recommended style coefficient and the user's baseline style parameters; If the adjustment intention indicates that the user refuses the adjustment, the baseline style parameter will be used as the execution style parameter; The driving state of the vehicle is controlled according to the execution style parameters.

5. The method according to claim 4, characterized in that, The step of obtaining the user's adjustment intention based on the recommended style coefficient and the baseline style parameter includes: If the difference between the recommended style coefficient and the baseline style parameter is greater than a preset difference value, then a query message is output to the user. In response to the user's reply to the query, the user's intention to make adjustments is obtained.

6. The method according to any one of claims 1 or 4, characterized in that, The step of determining the execution style parameters based on the recommended style coefficient and the user's baseline style parameters includes: Dynamic weights are determined based on the confidence level of traffic flow status information and the urgency of traffic scenarios; The execution style parameters are determined based on the dynamic weights, the recommended style coefficients, and the baseline style parameters.

7. The method according to claim 6, characterized in that, The step of determining the execution style parameters based on the dynamic weights, the recommended style coefficients, and the baseline style parameters includes: Based on the dynamic weights and the recommended style coefficients, the theoretical style correction coefficients are obtained; Based on the dynamic weights and the baseline style parameters, the baseline style correction parameters are obtained; The execution style parameters are obtained based on the theoretical style correction coefficients and the baseline style correction parameters.

8. The method according to claim 1, characterized in that, The step of controlling the vehicle's driving state according to the execution style parameters includes: The driving parameters are determined based on the execution style parameters; the type of the driving parameters is determined based on the baseline style parameters. The driving state of the vehicle is smoothly controlled based on the driving parameters.

9. The method according to claim 8, characterized in that, When the driving parameter is characterized as the desired following distance, the driving parameter is determined based on the desired following distance and the execution style parameter; And / or, When the driving parameter is characterized as an acceleration limit, the driving parameter is determined according to the preset boundary of acceleration and the adjustment function of the execution style parameter; And / or, When the driving parameter is characterized as energy recovery torque, the driving parameter is determined based on the maximum recovery torque and the execution style parameter; And / or, When the driving parameter is represented as a lateral decision threshold, the driving parameter is determined based on the theoretical safety gap, scenario coefficient, and execution style parameter.

10. A vehicle driving style adjustment device, characterized in that, include: The acquisition module is used to acquire traffic flow status information broadcast by the vehicle-road cooperative communication network and recommendation style coefficients determined based on the traffic flow status information; The determination module is used to determine the execution style parameters based on the recommended style coefficients and the user's baseline style parameters; The control module is used to control the driving state of the vehicle according to the execution style parameters.

Images