Vehicle control method and device, electronic equipment, storage medium and vehicle

Abstract

The application relates to the technical field of vehicles and discloses a vehicle control method and device, an electronic device, a storage medium and a vehicle. The method comprises the following steps: acquiring driving behavior data of a driver, driving situation data of the vehicle and environment interaction data between an external environment and the vehicle during driving of the vehicle; determining a target danger level of the vehicle during driving according to the driving behavior data, the driving situation data and the environment interaction data; and controlling the vehicle to execute a target control strategy according to the target danger level. Through the technical scheme, the problem that malicious driving behavior cannot be accurately identified and prevented in the related art is solved, the identification accuracy of the malicious driving behavior is improved, and the occurrence of the malicious driving behavior is limited.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, specifically to vehicle control methods, devices, electronic equipment, storage media, and vehicles. Background Technology

[0002] In actual vehicle driving, some users engage in malicious driving (including road rage, retaliatory driving, terrorist hijacking, etc.) for various reasons, resulting in multiple injuries and posing a significant threat to society. Accurately identifying and preventing malicious driving behavior is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] This application provides a vehicle control method, device, electronic device, storage medium, and vehicle to solve the problem in the related art of accurately identifying and preventing malicious driving behavior.

[0004] In a first aspect, this application provides a vehicle control method, the method comprising: Acquire driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during vehicle operation; Based on driving behavior data, driving situation data, and environmental interaction data, the target hazard level of the vehicle during the driving process is determined; Based on the target hazard level, control the vehicle to execute the target control strategy.

[0005] This vehicle control method, by simultaneously acquiring and fusing three types of heterogeneous data—driver behavior, vehicle dynamics, and environmental interaction—achieves comprehensive and multi-dimensional perception and assessment of driving scenarios, overcoming the limitations of relying on a single data source and the high false alarm rate. Based on this comprehensive assessment, the target hazard level is determined, and a matching control strategy is executed, forming a closed-loop control system of "perception-decision-execution." This enables more accurate and reliable identification of malicious driving intentions and allows for proactive intervention in a reasonable and effective manner, significantly improving the vehicle's active safety performance when dealing with malicious driving behaviors.

[0006] In one optional implementation, the target hazard level of the vehicle during driving is determined based on driving behavior data, driving situation data, and environmental interaction data, including: Based on driving behavior data, the driver's first dangerous driving score is determined; Based on driving situation data, determine the vehicle's second dangerous driving score; Based on environmental interaction data, determine the vehicle's third-hazard driving score in the environment; The target hazard level of a vehicle during its driving is determined based on the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score.

[0007] This implementation method decomposes the hazard level determination process into three independent dimensions: risk quantification scoring of the driver, the vehicle itself, and the interaction with the environment. A comprehensive decision is then made, achieving a structured and refined risk assessment. This method avoids overall misjudgment caused by anomalies in data from a single dimension. Through cross-validation and complementarity of multi-dimensional evidence, the final determined target hazard level is more comprehensive, objective, and accurate.

[0008] In one optional implementation, determining the target hazard level of the vehicle during driving based on a first dangerous driving score, a second dangerous driving score, and a third dangerous driving score includes: The dangerous driving score is obtained by weighted summing of the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score. The target hazard level is determined based on the hazard driving score.

[0009] This implementation method assigns appropriate weights to risk scores across different dimensions and performs a weighted summation, providing a flexible and quantifiable means of multi-source information fusion. The weights can be adjusted according to the specific scenario, enabling the comprehensive score to more scientifically reflect the overall risk level. This provides a stable and reliable numerical basis for subsequent risk classification, enhancing the adaptability and scientific rigor of the system's decision-making.

[0010] In one alternative implementation, determining the target hazard level based on the dangerous driving score includes: If the dangerous driving score is greater than or equal to the first dangerous threshold, the target dangerous level is set to the first risk level. If the dangerous driving score is less than the first dangerous threshold and greater than or equal to the second dangerous threshold, then the target dangerous level is set to the second risk level. If the dangerous driving score is less than the second danger threshold, the target danger level is set to the third risk level.

[0011] This implementation method maps continuous risk scores to discrete risk levels by setting explicit hazard thresholds, establishing a clear, stable, and easy-to-implement decision rule. This scheme ensures that the system can make completely consistent level judgments for scenarios with similar risk levels, avoiding decision uncertainty and providing precise and unambiguous triggering conditions for subsequently triggering fixed and explicit hierarchical control strategies.

[0012] In one alternative implementation, determining the driver's first dangerous driving score based on driving behavior data includes: Acquire driver physiological parameters, facial expressions, eye status, voice data, steering wheel operation data, and accelerator / brake operation data from driving behavior data; Based on at least one of physiological parameters, facial expressions, eye status, voice data, steering wheel operation data, and accelerator / brake operation data, a first dangerous driving score is determined using a driver dangerous driving identification model. The driver dangerous driving identification model is used to map driving behavior data to the first dangerous driving score.

[0013] This implementation method integrates multimodal physiological and operational data and utilizes a pre-trained dedicated recognition model for calculation, achieving end-to-end automated assessment of driver status from raw signals to risk scores. This method intelligently captures complex feature combinations, offering greater accuracy and efficiency than rule-based judgments, significantly improving the objectivity and reliability of individual driver risk quantification.

[0014] In one alternative implementation, determining a second dangerous driving score for the vehicle based on driving situation data includes: Obtain at least one of the following from the driving situation data: vehicle trajectory, following distance, and acceleration change; Based on at least one of the driving trajectory, following distance, and acceleration changes, a second dangerous driving score is determined using a vehicle dangerous driving identification model. The vehicle dangerous driving identification model is used to map driving situation data to the second dangerous driving score.

[0015] This implementation focuses on the vehicle's core dynamic parameters, such as trajectory, distance, and acceleration, and utilizes a dedicated model for evaluation, achieving an objective and independent quantification of the aggressiveness and abnormality of vehicle behavior. This assessment is independent of the driver's state, providing direct evidence based on the vehicle's objective performance for determining malicious driving, thus strengthening the foundation of comprehensive risk assessment.

[0016] In one alternative implementation, determining the vehicle's third hazard driving score in the environment based on environmental interaction data includes: Acquire surrounding images of the environment and the vehicle's collision intent from environmental interaction data; the images include crowds or vehicles. Based on surrounding images and collision intent, a third dangerous driving score is determined using an environmental hazard intent recognition model. This model maps environmental interaction data to the third dangerous driving score.

[0017] This implementation analyzes the interaction images between the vehicle and its external environment and identifies specific intentions, such as collisions, enabling a specialized assessment of the social harm risks posed by driving behavior in specific environments. This solution expands the perspective of risk assessment from inside the vehicle to the external environment, providing accurate early warnings of intentional malicious acts against pedestrians or other vehicles, filling a gap in traditional systems for assessing the hazards of environmental interactions.

[0018] In one alternative implementation, the vehicle is controlled to execute a target control strategy based on the target hazard level, including: If the target hazard level is Level 1, then at least one of the following should be executed: mandatory takeover, automatic relocation to a safe area, or emergency braking. If the target hazard level is the second risk level, then control the vehicle to perform at least one of the following: speed limit, distance keeping, and lane keeping. If the target hazard level is level three, the driver will be alerted via at least one of voice, text, or vibration. Record sensor data and operation logs of the vehicle executing the target control strategy.

[0019] This implementation method pre-sets a tiered set of control instructions, ranging from alerts and dynamic restrictions to mandatory takeover, for different risk levels, achieving gradient intervention that precisely matches the risk level. This solution ensures non-intrusive guidance for drivers during low-risk situations and decisive measures to ensure safety during high-risk situations. Furthermore, by recording data throughout the intervention process, it provides a complete basis for post-incident traceability analysis and system optimization, constructing a reliable and auditable safety closed loop.

[0020] Secondly, this application provides a vehicle control device, the device comprising: The data acquisition module is used to acquire driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during the vehicle's operation. The hazard level determination module is used to determine the target hazard level of a vehicle during its driving process based on driving behavior data, driving status data, and environmental interaction data. The vehicle control module is used to control the vehicle to execute the target control strategy based on the target hazard level.

[0021] Thirdly, this application provides a vehicle, the vehicle including: a controller, the controller including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method of the first aspect and any corresponding one thereof. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1This is a flowchart of a vehicle control method according to an embodiment of this application; Figure 2 This is a flowchart of another vehicle control method according to an embodiment of this application; Figure 3 This is a flowchart of a vehicle control method for preventing malicious driving according to an embodiment of this application; Figure 4 This is a flowchart of an in-vehicle driving status monitoring according to an embodiment of this application; Figure 5 This is a flowchart of an intervention and control procedure for malicious driving according to an embodiment of this application; Figure 6 This is a structural block diagram of a vehicle control device according to an embodiment of this application; Figure 7 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] It is understood that before using the technical solutions disclosed in the various embodiments of this application, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this application in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0026] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0027] Current traditional vehicle safety mainly focuses on protection after an accident and protection against driver error. In general, after a vehicle is started, the driver controls the speed and direction. However, some people may use vehicles as a tool to vent their anger, attacking pedestrians or other vehicles and causing multiple injuries, leading to irreparable tragedies.

[0028] This application proposes a vehicle control method to prevent malicious driving. During driving, through the coordinated work of onboard sensors and the control system, the method intervenes in stages and intervenes in the control of the vehicle's throttle and brakes in a timely manner through a comprehensive control strategy to prevent further damage.

[0029] According to an embodiment of this application, a vehicle control method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] This embodiment provides a vehicle control method that can be used in the central control domain of a vehicle. Figure 1 This is a flowchart of a vehicle control method according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps: Step S101: Obtain driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during the vehicle's operation.

[0031] Specifically, driving behavior data refers to data used to characterize the driver's physiological state and operational behavior, such as driver's heart rate, facial expressions, steering wheel angle, and accelerator / brake pedal travel collected by in-vehicle sensors. Driving situation data refers to data used to characterize the vehicle's own motion state and relative position, such as vehicle speed, acceleration, and trajectory obtained through vehicle bus, GPS, and inertial measurement unit, as well as following distance measured by radar or cameras. Environmental interaction data refers to data used to characterize external environmental elements and the vehicle's potential interaction intentions with that environment, such as images of surrounding pedestrians and vehicles captured by external cameras, and analysis of these images to determine whether the vehicle intends to collide with or track a specific target.

[0032] By integrating a multi-source sensor network that includes in-vehicle sensors (monitoring the driver), vehicle body sensors (monitoring the vehicle itself), and external sensors (monitoring the environment), three types of key data are collected simultaneously: driving behavior data reflecting the driver's state and operation, such as heart rate, facial expressions, steering wheel and pedal operations; driving status data reflecting the vehicle's movement and relative position, such as vehicle speed, trajectory, and following distance; and environmental interaction data reflecting the external environment and potential threats, such as images of surrounding pedestrians / vehicles and collision intentions analyzed based on them.

[0033] This establishes a complete and multi-dimensional data foundation for subsequent risk assessment. It fundamentally changes the limitations of traditional solutions that rely on a single information source. Through the synchronous acquisition and fusion of multi-modal data, it ensures that the system can make comprehensive judgments based on three related dimensions: driver intent, vehicle dynamics, and external environmental threats. This makes it possible to accurately distinguish between normal aggressive driving and truly malicious driving, effectively reducing the risk of misjudgment. This is a prerequisite for achieving precise and reliable tiered intervention.

[0034] Step S102: Determine the target hazard level of the vehicle during driving based on driving behavior data, driving situation data, and environmental interaction data.

[0035] Specifically, the target hazard level refers to the level at which a driver may cause a dangerous event while driving a vehicle, representing a quantitative classification result that characterizes the overall risk level of the current driving scenario.

[0036] By using the heterogeneous data acquired in the previous step as input, such as driver behavior, vehicle status, and environmental interactions, and conducting comprehensive analysis through established rules or models (such as weighted scoring or machine learning models), a unified and clear target hazard level is ultimately output. This achieves a crucial transformation from "multi-source perception data" to "single decision-making basis." It integrates complex, multi-dimensional safety cues into a clear hazard level signal, thereby providing precise and actionable instructions to the downstream execution layer.

[0037] Step S103: Based on the target hazard level, control the vehicle to execute the target control strategy.

[0038] Specifically, target control strategy refers to a set of vehicle control commands that are pre-matched with and executed according to different target hazard levels.

[0039] The system receives the target hazard level determined in the previous step as an input signal, and automatically triggers and executes a series of specific vehicle control commands corresponding to the level according to the preset mapping rules, thereby directly intervening in the vehicle's longitudinal (throttle / brake) and lateral (steering) dynamics.

[0040] For example, by using a pre-defined mapping table between target hazard levels and target control strategies, the corresponding target control strategies can be determined based on the target hazard level.

[0041] The vehicle control method provided in this embodiment acquires driver behavior data, vehicle driving status data, and environmental interaction data between the vehicle and the external environment during the vehicle's operation. This enables comprehensive and multi-dimensional synchronous perception of driver status, vehicle dynamics, and external environmental threats, providing a complete and reliable data foundation for accurately identifying malicious driving. Based on the driver behavior data, driving status data, and environmental interaction data, the method determines the target hazard level of the vehicle during operation. Through multi-source information fusion and comprehensive evaluation, complex multi-dimensional clues are transformed into a quantifiable and clear risk level, achieving accurate judgment of the degree of danger of driving behavior. According to the target hazard level, the method controls the vehicle to execute the target control strategy, achieving graded proactive intervention from early warning to mandatory takeover, matching the risk level, forming a closed-loop control of "perception-decision-execution". This solves the problems of high false alarm rate in malicious driving identification due to reliance on a single dimension of judgment in existing technologies, and the potential for driver resentment or new safety risks due to the single and lack of gradation in intervention methods. Through multimodal perception, intelligent comprehensive judgment and hierarchical collaborative control, the technology has achieved the ability to more accurately and reliably identify malicious driving intentions and to proactively intervene in vehicles in a more reasonable and effective manner to prevent or mitigate harm, thereby significantly improving the active safety of vehicles in dealing with malicious driving scenarios.

[0042] This embodiment provides a vehicle control method that can be used in the vehicle's central domain controller. Figure 2 This is a flowchart of another vehicle control method according to an embodiment of this application, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during the vehicle's operation.

[0043] Please see details Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0044] Step S202: Determine the target hazard level of the vehicle during driving based on driving behavior data, driving situation data, and environmental interaction data.

[0045] Specifically, step S202 includes: Step S2021: Determine the driver's first dangerous driving score based on driving behavior data.

[0046] Specifically, the First Dangerous Driving Score is a numerical value calculated using a specific algorithm based on the driver's physiological state and operational behavior data, used to quantify the degree of danger of their individual driving state. For example, a high score may indicate that the driver is in a state of anger, tension, or aggressive driving tendencies.

[0047] By constructing and running a driver risk assessment model, such as a driver dangerous driving identification model, this model takes the driving behavior data obtained in step S201 as input, such as heart rate, facial expression, steering wheel angle, and pedal travel. It analyzes and weights these features using a rule engine or machine learning algorithm, ultimately outputting a first dangerous driving score that characterizes the driver's own level of risk. This achieves an independent and quantitative risk assessment of the driver's state and actions.

[0048] Step S2022: Determine the second dangerous driving score of the vehicle based on the driving situation data.

[0049] Specifically, the second dangerous driving score is a numerical value calculated based on the vehicle's own motion state and relative position data, used to quantify the degree of danger of the vehicle's dynamic behavior. For example, a high score may indicate that the vehicle is performing frequent weaving, dangerous following, or unwarranted rapid acceleration / deceleration.

[0050] A vehicle dynamic risk assessment model (such as a vehicle dangerous driving identification model) is constructed and run. This model takes the driving situation data obtained in step S201 as input, such as the vehicle's trajectory, following distance to the vehicle in front, and changes in longitudinal and lateral acceleration. It analyzes the abnormal patterns and degree of danger of these dynamic parameters through predefined rules or algorithms, and finally outputs a second dangerous driving score that characterizes the vehicle's own behavioral risk. This provides an independent and objective risk quantification of the vehicle's dynamic performance, transforming multi-dimensional vehicle kinematic data into a unified risk index, and providing a standard risk assessment for the vehicle in subsequent comprehensive evaluations.

[0051] Step S2023: Based on the environmental interaction data, determine the third dangerous driving score of the vehicle in the environment.

[0052] Specifically, the third hazardous driving score refers to a numerical value calculated through analysis of data on the interaction between the vehicle and its external environment. This value is used to quantify the potential hazard level of vehicle behavior in the current environmental scenario. For example, a high score might indicate that the vehicle is heading towards a crowd, deliberately approaching a specific target, or lingering in an unsafe area.

[0053] By constructing and running an environmental interaction risk assessment model (such as an environmental hazard intent recognition model), this model takes the environmental interaction data obtained in step S201 as input, such as images of surrounding pedestrians and vehicles captured by the vehicle's external camera, and collision or tracking intents identified based on image analysis. Through scene understanding and intent recognition algorithms, it assesses potential collision risks or malicious target orientations in the current environment, ultimately outputting a third-hazard driving score characterizing the hazard of behavior in a specific environmental scenario. This achieves independent risk assessment of the vehicle's external environment and interaction intents, identifying driving intentions and behavioral patterns that may pose a serious threat to third parties (such as pedestrians and other vehicles). This assessment complements the assessment of driver status and vehicle dynamics, enabling the system to detect and warn of the most socially harmful malicious driving scenarios, such as intentional collisions, thus achieving comprehensive safety protection from the internal "vehicle-person" closed loop to the external "vehicle-environment" interaction.

[0054] Step S2024: Determine the target hazard level of the vehicle during the driving process based on the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score.

[0055] Specifically, using the first, second, and third dangerous driving scores obtained in the aforementioned steps as input, the system comprehensively analyzes these three scores through a predetermined algorithm, such as weighted summation, rule-based logical judgment, or using a trained classification model. Based on the comparison of the analysis results with a preset threshold, the system maps the current state to the corresponding target danger level. This overcomes the potential bias and misjudgment risk of relying on any single-dimensional score for decision-making (e.g., judging a vehicle as high-risk solely because the driver is emotionally agitated, ignoring the vehicle's actual safe operating environment). By integrating information from all three sources for comprehensive decision-making, the target danger level derived by the system is more holistic, accurate, and robust, providing a reliable and authoritative decision-making basis for implementing the most suitable and reasonable vehicle control strategy.

[0056] Step S203: Based on the target hazard level, control the vehicle to execute the target control strategy.

[0057] Please see details Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0058] This embodiment provides a vehicle control method. First, it simultaneously collects multi-dimensional data on driver status, vehicle dynamics, and the external environment to achieve comprehensive perception of the driving scenario. Then, based on this data, it quantitatively assesses driver behavior risk, vehicle dynamic risk, and environmental interaction risk, and generates a comprehensive target hazard level through a multi-factor fusion decision-making mechanism. Finally, based on this level, it automatically triggers and executes a matched tiered control strategy, ranging from gentle warnings to mandatory takeover. This method achieves a complete closed loop from multi-source information perception and intelligent fusion assessment to precise gradient execution, thereby enabling more accurate and reliable identification of malicious driving intentions and proactive intervention in the vehicle in a reasonable and effective manner, significantly improving the vehicle's active safety protection capabilities and system reliability in complex scenarios.

[0059] In some optional implementations, step S2024 above includes: Step a1: The first dangerous driving score, the second dangerous driving score, and the third dangerous driving score are weighted and summed to obtain the dangerous driving score.

[0060] Step a2: Determine the target hazard level based on the hazard driving score.

[0061] Specifically, the dangerous driving score refers to a single comprehensive value that represents the overall level of driving danger by weighting and fusing the risk scores of the driver, vehicle, and environment.

[0062] Appropriate weighting coefficients are assigned to the first, second, and third dangerous driving scores, respectively. The weighted scores are then summed to calculate the dangerous driving score. The weighting coefficients can be adjusted according to different scenarios or safety strategies to reflect the emphasis on a particular risk dimension. For example, in densely populated areas, more attention is paid to environmental interaction risks.

[0063] The system presets score threshold ranges corresponding to different hazard levels. It compares the calculated hazardous driving score with these thresholds to classify the vehicle into the corresponding target hazard level.

[0064] This implementation provides a clear, quantifiable, and flexible integrated decision-making path. Through weighted summation, it unifies multi-dimensional risk assessments into a single scalar, achieving information dimensionality reduction and fusion, making the decision-making logic concise and clear. Through threshold-based mapping, it ensures the objectivity and consistency of risk level classification. Overall, this scheme enables the final target hazard level to reflect the combined effects of risks across all dimensions in a balanced and predictable manner, laying a solid foundation for the subsequent implementation of precisely matched hierarchical control strategies.

[0065] In some alternative implementations, step a2 above includes: Step b1: If the dangerous driving score is greater than or equal to the first dangerous threshold, then the target dangerous level is set to the first risk level.

[0066] Step b2: If the dangerous driving score is less than the first dangerous threshold and greater than or equal to the second dangerous threshold, then the target dangerous level is set to the second risk level.

[0067] Step b3: If the dangerous driving score is less than the second danger threshold, then the target danger level is set to the third risk level.

[0068] Specifically, the first and second hazard thresholds are numerical values ​​used to measure the severity of dangerous driving scores. Using two pre-defined thresholds, where the first hazard threshold is greater than the second, consecutive dangerous driving scores are divided into different intervals, each corresponding to a different risk level. The first risk level corresponds to the highest degree of danger, such as high risk. The second risk level corresponds to the moderate degree of danger, such as medium risk. The third risk level corresponds to the lowest degree of danger, such as low risk.

[0069] The dangerous driving score is compared with a first danger threshold and a second danger threshold. Based on the score's numerical range, it is categorized into one of three discrete "target danger levels." By setting fixed thresholds for range division, it clearly maps the comprehensive risk assessment result to three discrete decision levels. This ensures the consistency and predictability of system decisions, ensuring that the same level of danger always triggers the same risk level. This provides reliable and unambiguous triggering conditions for subsequent execution of fixed control strategies that strictly correspond to each level, thereby achieving a precise match between system response and the actual risk level.

[0070] It should be noted that the risk level is updated as the dangerous driving score changes. At different stages of the same behavior, the dangerous driving score may be different, corresponding to different risk levels, and the alerts or interventions triggered will be different.

[0071] In some optional implementations, step S2021 above includes: Step c1: Obtain the driver's physiological parameters, facial expressions, eye status, voice data, steering wheel operation data, and accelerator / brake operation data from the driving behavior data.

[0072] Step c2: Based on at least one of physiological parameters, facial expressions, eye status, voice data, steering wheel operation data, and accelerator / brake operation data, a first dangerous driving score is determined using a driver dangerous driving identification model. The driver dangerous driving identification model is used to map driving behavior data to the first dangerous driving score.

[0073] Specifically, physiological parameters refer to driver physical indicators monitored by in-vehicle biosensors, such as heart rate and respiratory rate. Facial expressions refer to the driver's facial emotional characteristics analyzed by in-vehicle cameras, such as whether they show anger or tension. Eye movement data refers to driver eye movement information obtained through visual tracking technology, such as gaze direction, blink frequency, or pupil changes. Voice data refers to driver voice information collected through in-vehicle microphones, used to analyze tone, volume, and keywords, such as shouting. Steering wheel operation data refers to operation characteristics obtained through steering sensors, such as steering angle, angular velocity, and whether there are abrupt or frequent corrections. Accelerator / brake operation data refers to operation characteristics obtained through pedal sensors, such as pedal opening, pedal speed, and whether there are frequent deep braking. The driver dangerous driving recognition model refers to a trained algorithm or computational model whose function is to receive one or more of the above-mentioned types of driving behavior data as input and output a quantified first dangerous driving score, such as linear regression or a fully connected classifier.

[0074] First, raw data reflecting the driver's physiological state and driving intentions, directly or indirectly, is collected from various types of vehicle sensors. This data is then input into a pre-trained dedicated recognition model. This model incorporates a complex mapping relationship between data features and risk levels, enabling automatic and efficient feature extraction and comprehensive calculation, ultimately outputting a standardized risk score. This achieves end-to-end automated assessment of the driver's state, from raw multimodal signals to a unified risk score. It avoids the difficulties of manually setting complex rules and, through a data-driven approach, can more intelligently and accurately capture subtle and complex combinations of features that foreshadow malicious driving, thus significantly improving the objectivity, accuracy, and efficiency of driver-level risk assessment.

[0075] In some optional implementations, step S2022 above includes: Step d1: Obtain at least one of the following from the driving situation data: vehicle trajectory, following distance, and acceleration change.

[0076] Step d2: Based on at least one of the driving trajectory, following distance, and acceleration change, determine the second dangerous driving score using the vehicle dangerous driving identification model. The vehicle dangerous driving identification model is used to map driving situation data to the second dangerous driving score.

[0077] Specifically, driving trajectory refers to the geometric and dynamic characteristics of the vehicle's actual driving path, such as whether there are frequent, unannounced lateral swaying (wobbling) or malicious tailgating. Following distance refers to the real-time distance between the vehicle and the vehicle in front, used to determine whether there is dangerous close-range tailgating. Acceleration change refers to the degree of drastic change in vehicle speed, including sudden acceleration or deceleration not required for hazard avoidance. The vehicle dangerous driving identification model is a dedicated algorithm model whose function is to analyze one or more of the above driving situation data, identify dangerous patterns, and output a quantified second dangerous driving score.

[0078] Core dynamic parameters that directly characterize the aggressiveness and abnormality of vehicle movement are selectively acquired from the vehicle's bus and sensors. These parameters are then input into a trained, dedicated recognition model. This model incorporates logic for judging dangerous driving patterns (such as weaving, close following, and sudden stops / starts), automatically quantifying the degree of danger of vehicle dynamics. This achieves risk quantification of vehicle behavior based on objective data, independent of the driver's state. It directly assesses the aggressiveness or abnormality exhibited by the vehicle, providing solid evidence for the system's judgment that is independent of the driver's subjective state, effectively supplementing the driver's assessment dimension and making the overall risk assessment more comprehensive and reliable.

[0079] In some optional implementations, step S2023 above includes: Step e1: Obtain surrounding images of the environment and the vehicle's collision intent from the environmental interaction data. The images include crowds or vehicles. Step e2: Based on the surrounding images and the collision intent, the third dangerous driving score is determined using the environmental hazard intent recognition model. The environmental hazard intent recognition model is used to map environmental interaction data to the third dangerous driving score.

[0080] Specifically, the surrounding environment image refers to the real-time image of the vehicle's surrounding environment captured by visual sensors such as external cameras. This image includes information about potentially hazardous targets, such as pedestrians, other vehicles, and obstacles. The vehicle's collision intent refers to the predicted behavioral tendency identified after analyzing surrounding images and vehicle trajectory information, indicating a vehicle's accelerated approach towards a specific target (such as a crowd or another vehicle) or a potential collision. The environmental hazard intent recognition model is a dedicated computer vision and behavior analysis algorithm model. Its function is to comprehensively analyze surrounding images and vehicle status to infer the vehicle's interaction intent with the environment (especially malicious intent) and output a quantified third-risk driving score.

[0081] The system acquires vehicle perception data (images) of the external environment and performs preliminary analysis to determine behavioral intent. This information is then input into a pre-trained dedicated recognition model. This model can deeply understand the context of a scene and assess the potential harm of vehicle behavior in a specific environment. It enables independent quantitative assessment of the social harm risks posed by vehicle behavior in specific environments. It allows the system to not only focus on the state of the driver and the vehicle itself, but also proactively identify threats posed by the vehicle to the external world (especially to vulnerable road users), thereby accurately issuing warnings for malicious driving scenarios with clear intent and extremely high harm (such as intentional collisions).

[0082] In some optional implementations, step S203 above includes: Step f1: If the target hazard level is the first risk level, then execute at least one of the following: forced takeover, automatic driving to a safe area, or emergency braking.

[0083] Step f2: If the target hazard level is the second risk level, then control the vehicle to perform at least one of the following: speed limit, distance maintenance, and lane keeping.

[0084] Step f3: If the target hazard level is the third risk level, then the driver is alerted by at least one of voice, text, or vibration.

[0085] Step f4: Record the sensor data and operation logs of the vehicle executing the target control strategy.

[0086] Specifically, forced takeover refers to the vehicle control system automatically releasing the driver's direct control over the vehicle's lateral (steering) and / or longitudinal (driving and braking) directions under specific conditions (such as detecting high-risk malicious driving). Emergency braking refers to the vehicle control system automatically triggering and applying maximum or near-maximum braking force to rapidly decelerate or stop the vehicle when it judges that an impending collision or other extreme dangerous situation is imminent. Distance keeping refers to the vehicle control system automatically adjusting its own speed (such as controlling the accelerator and brake) to maintain a preset or dynamically calculated safe distance between the vehicle and the vehicle in front. Lane keeping refers to the vehicle control system automatically applying small-amplitude steering torque or adjustments to assist or force the vehicle to maintain its trajectory within the current lane markings. Sensor data refers to the raw or processed monitoring data generated by various sensors inside and outside the vehicle during the execution of the control strategy, such as data collected by cameras, radar, spatial attitude sensors, pedal position sensors, etc. The operation log is a formatted data file that records the system status, decision-making process, executed instructions, and key events (such as intervention triggers and level switching) in chronological order during the execution of the control strategy.

[0087] This implementation details a tiered execution strategy corresponding to different target hazard levels. A tiered set of control instructions, ranging from mild to severe and from warning to mandatory, is preset for the third (low), second (medium), and first (high) risk levels, and is automatically invoked and executed after the risk level is determined. Simultaneously, the system records all data related to the intervention process. This achieves proactive safety intervention that is precisely matched to the risk level, tiered, and traceable. It ensures that the system response avoids excessive interference with the driver when the risk is low, while taking sufficiently forceful measures to mitigate the danger when the risk increases. The data recording function provides objective evidence for post-event analysis, responsibility clarification, and system optimization, thus constructing a complete, reliable, and auditable closed loop for preventing malicious driving.

[0088] Figure 3 This is a flowchart of a vehicle control method for preventing malicious driving according to an embodiment of this application. Figure 3 As shown, external sensors and cameras collect information about the external environment, while internal sensors and cameras acquire information about the driving status inside the vehicle. This information is aggregated and processed by the controller for comprehensive calculation and judgment. Based on the judgment results, the controller initiates a tiered intervention mechanism, including four progressively escalating intervention levels: warnings and reminders, restricting vehicle movement, forced takeover and parking, and remote evidence recording. This diagram clearly illustrates the complete control loop from environmental and status perception to central decision-making and tiered safety execution.

[0089] Figure 4 This is a flowchart of an in-vehicle driving status monitoring method according to an embodiment of this application. Figure 4 As shown, the monitoring system mainly comprises three parallel and complementary monitoring modules: a biometric monitoring module responsible for monitoring the driver's heart rate, respiratory rate, facial expression, eye movement, and voice tone; a driving behavior monitoring module responsible for monitoring steering wheel rotation, accelerator pedal position, and brake pedal movement; and a vehicle dynamics monitoring module responsible for monitoring vehicle trajectory, following distance, and rapid acceleration / deceleration. Notably, the system integrates vehicle dynamics with environmental information, including a specific malicious scenario identification feature: "rushing towards a crowd." This intuitively presents a progressive, comprehensive state perception system, from the driver's physiological state and direct operational behavior to the vehicle's external dynamics, providing structured multi-source input for subsequent comprehensive risk assessment.

[0090] Figure 5 This is a flowchart illustrating an intervention and control process for malicious driving, according to an embodiment of this application. Figure 5As shown, the system executes actions sequentially from low to high hazard level: Level 1, warnings and alerts, including audible and visual alerts, voice reassurance, and tactile feedback; Level 2, restricting vehicle dynamics, including speed limit mode, maintaining distance, and lane keeping; Level 3, forced takeover and stopping, including forced takeover, automatic safe stopping, and emergency braking; and Level 4, remote evidence recording throughout, including data black box recording, cloud alarms, and authorized remote locking. This constructs a complete, multi-layered safety intervention chain from warnings and restrictions to forced takeover and post-incident tracing. It achieves human-machine collaborative intervention that is reasonable in response, reliable in execution, and in line with driving safety expectations. It can guide drivers to correct their behavior autonomously through gentle reminders, gradually restrict aggressive vehicle dynamics when necessary, and ultimately force takeover by the system in extreme situations to ensure safety. Simultaneously, the remote evidence recording function provides complete data support for post-incident tracing and liability determination. Overall, this solution effectively solves the problems of traditional safety systems having single intervention methods, being prone to false triggering, or having insufficient response, significantly improving the practicality, security, and acceptability of malicious driving prevention systems.

[0091] This embodiment also provides a vehicle control device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0092] This embodiment provides a vehicle control device, such as... Figure 6 As shown, it includes: The data acquisition module 601 is used to acquire driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during the vehicle's operation. The hazard level determination module 602 is used to determine the target hazard level of the vehicle during driving based on driving behavior data, driving status data and environmental interaction data; The vehicle control module 603 is used to control the vehicle to execute the target control strategy based on the target hazard level.

[0093] In some alternative implementations, the hazard level determination module 602 includes: The first dangerous driving score determination unit is used to determine the driver's first dangerous driving score based on driving behavior data; The second dangerous driving score determination unit is used to determine the second dangerous driving score of a vehicle based on driving situation data. The third dangerous driving score determination unit is used to determine the third dangerous driving score of a vehicle in the environment based on environmental interaction data. The target hazard level determination unit is used to determine the target hazard level of a vehicle during its driving process based on the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score.

[0094] In some optional implementations, the target hazard level determination unit includes: The dangerous driving score determination subunit is used to weighted sum the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score to obtain the dangerous driving score. The target hazard level determination subunit is used to determine the target hazard level based on the hazardous driving score.

[0095] In some alternative implementations, the target hazard level determination subunit is used for: If the dangerous driving score is greater than or equal to the first dangerous threshold, the target dangerous level is set to the first risk level. If the dangerous driving score is less than the first dangerous threshold and greater than or equal to the second dangerous threshold, then the target dangerous level is set to the second risk level. If the dangerous driving score is less than the second danger threshold, the target danger level is set to the third risk level.

[0096] In some alternative implementations, the first dangerous driving score determination unit includes: The first acquisition subunit is used to acquire the driver's physiological parameters, facial expressions, eye status, voice data, steering wheel operation data, and accelerator / brake operation data from the driving behavior data; The first dangerous driving score determination subunit is used to determine a first dangerous driving score based on at least one of physiological parameters, facial expressions, eye status, voice data, steering wheel operation data, and accelerator / brake operation data, through a driver dangerous driving identification model. The driver dangerous driving identification model is used to map driving behavior data to the first dangerous driving score.

[0097] In some alternative implementations, the second dangerous driving score determination unit includes: The second acquisition subunit is used to acquire at least one of the following in the driving situation data: vehicle trajectory, following distance, and acceleration change. The second dangerous driving score determination subunit is used to determine the second dangerous driving score based on at least one of the driving trajectory, following distance, and acceleration change, through the vehicle dangerous driving identification model. The vehicle dangerous driving identification model is used to map driving situation data to the second dangerous driving score.

[0098] In some optional implementations, the third dangerous driving score determination unit includes: The third acquisition subunit is used to acquire surrounding images of the environment and the vehicle's collision intent from the environmental interaction data. The images include crowds or vehicles. The third dangerous driving score determination subunit is used to determine the third dangerous driving score based on surrounding images and collision intent through an environmental hazard intent recognition model. The environmental hazard intent recognition model is used to map environmental interaction data to the third dangerous driving score.

[0099] In some alternative implementations, the vehicle control module 603 includes: The first risk level unit is used to execute at least one of the following if the target hazard level is the first risk level: forced takeover, automatic driving to a safe area, and emergency braking. The second risk level unit is used to control the vehicle to perform at least one of the following if the target hazard level is the second risk level: speed limit, distance keeping, and lane keeping. The third risk level unit is used to alert the driver via at least one of voice, text, or vibration if the target hazard level is the third risk level. The logging unit is used to record sensor data and operation logs of the vehicle executing the target control strategy.

[0100] The vehicle control device provided in this application can execute the vehicle control method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0101] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0102] The following is a detailed reference. Figure 7 The diagram illustrates a structural schematic suitable for implementing the electronic device described in the embodiments of this application. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 701, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 702 or a program loaded from memory 708 into random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device. The processor 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0103] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 7 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0104] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 709, or installed from memory 708, or installed from ROM 702. When the computer program is executed by processor 701, it performs the functions defined in the vehicle control method of embodiments of this application.

[0105] Figure 7 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0106] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the vehicle control method shown in the above embodiments is implemented.

[0107] A portion of this application can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to this application through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0108] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A vehicle control method, characterized in that, The method includes: Acquire driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during vehicle operation. Based on the driving behavior data, the driving situation data, and the environmental interaction data, the target hazard level of the vehicle during the driving process is determined; Based on the target hazard level, control the vehicle to execute the target control strategy.

2. The method according to claim 1, characterized in that, Determining the target hazard level of the vehicle during its driving process based on the driving behavior data, the driving situation data, and the environmental interaction data includes: Based on the driving behavior data, the driver's first dangerous driving score is determined; Based on the driving situation data, a second dangerous driving score for the vehicle is determined; Based on the environmental interaction data, the third dangerous driving score of the vehicle in the environment is determined; The target hazard level of the vehicle during its driving process is determined based on the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score.

3. The method according to claim 2, characterized in that, Determining the target hazard level of the vehicle during operation based on the first dangerous driving score, the second dangerous driving score, and the third dangerous driving score includes: The first dangerous driving score, the second dangerous driving score, and the third dangerous driving score are weighted and summed to obtain the dangerous driving score. The target hazard level is determined based on the dangerous driving score.

4. The method according to claim 3, characterized in that, Determining the target hazard level based on the dangerous driving score includes: If the dangerous driving score is greater than or equal to the first dangerous threshold, then the target dangerous level is set to the first risk level; If the dangerous driving score is less than the first dangerous threshold and greater than or equal to the second dangerous threshold, then the target dangerous level is set to the second risk level. If the dangerous driving score is less than the second danger threshold, the target danger level is set to the third risk level.

5. The method according to claim 2, characterized in that, The step of determining the driver's first dangerous driving score based on the driving behavior data includes: Acquire the driver's physiological parameters, facial expressions, eye state, voice data, steering wheel operation data, and accelerator / brake operation data from the driving behavior data; Based on at least one of the physiological parameters, facial expressions, eye states, voice data, steering wheel operation data, and accelerator / brake operation data, a first dangerous driving score is determined using a driver dangerous driving identification model. The driver dangerous driving identification model is used to map the driving behavior data to the first dangerous driving score.

6. The method according to claim 2, characterized in that, The step of determining the second dangerous driving score of the vehicle based on the driving situation data includes: Obtain at least one of the following from the driving situation data: the vehicle's driving trajectory, following distance, and acceleration change; Based on at least one of the driving trajectory, the following distance, and the acceleration change, a second dangerous driving score is determined using a vehicle dangerous driving identification model, wherein the vehicle dangerous driving identification model is used to map the driving situation data to the second dangerous driving score.

7. The method according to claim 2, characterized in that, The step of determining the third dangerous driving score of the vehicle in the environment based on the environmental interaction data includes: The surrounding image of the environment and the vehicle's collision intent are obtained from the environmental interaction data, and the image includes crowds or vehicles; Based on the surrounding images and the collision intent, the third dangerous driving score is determined using an environmental hazard intent recognition model, which maps the environmental interaction data to the third dangerous driving score.

8. The method according to claim 4, characterized in that, The step of controlling the vehicle to execute a target control strategy based on the target hazard level includes: If the target hazard level is the first risk level, then at least one of the following is executed: forced takeover, automatic driving to a safe area, or emergency braking; If the target hazard level is the second risk level, then control the vehicle to perform at least one of the following: speed limit, distance maintenance, and lane keeping. If the target hazard level is the third risk level, then the driver will be alerted by at least one of voice, text, and vibration. Record the sensor data and operation logs of the vehicle executing the target control strategy.

9. A vehicle control device, characterized in that, The device includes: The data acquisition module is used to acquire driver behavior data, vehicle driving status data, and environmental interaction data between the external environment and the vehicle during the vehicle's operation. The hazard level determination module is used to determine the target hazard level of the vehicle during driving based on the driving behavior data, the driving situation data, and the environmental interaction data. The vehicle control module is used to control the vehicle to execute the target control strategy based on the target hazard level.

10. A vehicle, characterized in that, The vehicle includes a controller, which includes a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method of any one of claims 1 to 8.

Images