Vehicle control method, device and vehicle

Abstract

The application discloses a vehicle control method, device and vehicle. The method comprises the following steps: acquiring system state data of a driving system in the vehicle, environment perception data of an environment in which the vehicle is located, and object state data of a target object in the vehicle, wherein the object state data is used for describing the input degree of the target object to a driving task; determining a target control right allocation coefficient of the vehicle based on the system state data, the environment perception data and the object state data; determining a target control parameter from a preset scenario continuous spectrum model based on the target control right allocation coefficient; and controlling the vehicle based on the system state data, the target control parameter and a target constraint condition. The application solves the technical problem of low passenger comfort in vehicle control in the related art.

Description

Technical Field

[0001] This application relates to the field of vehicles, and more specifically, to a vehicle control method, device, and vehicle. Background Technology

[0002] Current intelligent vehicle human-machine interaction and context perception systems mostly adopt a rule-based architecture driven by discrete events. They switch modes only based on binary signals such as whether the autonomous driving function is activated or whether an alarm is triggered. This causes passengers to frequently encounter discontinuous sensory stimuli and cognitive load in different driving situations, resulting in low passenger comfort when controlling the vehicle using related technologies.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a vehicle control method, device, and vehicle to at least solve the technical problem of low occupant comfort when controlling a vehicle in related technologies.

[0005] According to one aspect of the embodiments of this application, a vehicle control method is provided, comprising: acquiring system state data of a driving system in a vehicle, environmental perception data of the environment in which the vehicle is located, and object state data of a target object in the vehicle, wherein the object state data is used to describe the degree of engagement of the target object with the driving task; determining a target control right allocation coefficient of the vehicle based on the system state data, environmental perception data, and object state data, wherein the target control right allocation coefficient is used to quantify the control right allocation ratio between the driving system and the target object at the current moment; determining target control parameters from a preset scenario continuous spectrum model based on the target control right allocation coefficient, wherein the preset scenario continuous spectrum model is used to describe the continuous mapping relationship between the control right allocation coefficient and the control parameters; and controlling the vehicle based on the system state data, the target control parameters, and target constraints, wherein the target constraints are used to describe the constraints that need to be satisfied during the control process of the vehicle.

[0006] Furthermore, based on system state data, environmental perception data, and object state data, the target control right allocation coefficient for the vehicle is determined, including: normalizing the system state data, environmental perception data, and object state data to obtain multiple influencing factors corresponding to the system state data, environmental perception data, and object state data respectively; determining the weights corresponding to the multiple influencing factors respectively; and using a weighted fusion model to fuse the multiple influencing factors and their corresponding weights to obtain the target control right allocation coefficient.

[0007] Furthermore, the multiple influencing factors include: system state factors, environmental perception factors, and object state factors; the weights corresponding to these multiple influencing factors include: system state weights, environmental perception weights, and object state weights; a weighted fusion model is used to fuse these multiple influencing factors and their corresponding weights to obtain the target control allocation coefficients, including: determining the system control quantity based on the system state factors and system state weights, where the system control quantity is directly proportional to the system state factors, and the system control quantity represents the driving system's ability to undertake driving tasks; determining the environmental control quantity based on the environmental perception factors and environmental perception weights, where the environmental control quantity is inversely proportional to the environmental perception factors, and the environmental control quantity represents the urgency of the driving system taking over driving tasks under the current environmental risk level; determining the object control quantity based on the object state factors and object state weights, where the object control quantity is inversely proportional to the object state factors, and the object control quantity represents the urgency of the driving system taking over driving tasks under the current level of engagement of the target object with the driving task; and determining the target control allocation coefficients based on the system control quantity, environmental control quantity, and object control quantity.

[0008] Furthermore, the vehicle is controlled based on system state data, target control parameters, and target constraints, including: determining the current control parameters of the vehicle based on system state data; determining the control duration for completing vehicle control based on target constraints, current control parameters, and target control parameters; and controlling the vehicle based on the control duration, target constraints, and target control parameters.

[0009] Furthermore, based on the target constraints, current control parameters, and target control parameters, the control duration for completing vehicle control is determined, including: determining the control parameter difference based on the current control parameters and the target control parameters; and determining the control duration based on the control parameter difference and the target constraints.

[0010] Furthermore, the method also includes: if the environmental perception data exceeds a preset environmental risk threshold within a preset time period, executing the methods in the various embodiments of this application.

[0011] Furthermore, based on environmental perception factors and environmental perception weights, environmental control quantities are determined; based on object state factors and object state weights, object control quantities are determined, including: determining the difference between a preset value and an environmental perception factor as a first difference; determining the difference between a preset value and an object state factor as a second difference; determining the environmental control quantity based on the first difference and the environmental perception weights; and determining the object control quantity based on the second difference and the object state weights.

[0012] According to another aspect of the embodiments of this application, a vehicle control device is also provided, comprising: an acquisition module, configured to acquire system state data of a driving system in a vehicle, environmental perception data of the environment in which the vehicle is located, and object state data of a target object in the vehicle, wherein the object state data is used to describe the degree of engagement of the target object with the driving task; a first determination module, configured to determine a target control right allocation coefficient of the vehicle based on the system state data, environmental perception data, and object state data, wherein the target control right allocation coefficient is used to quantify the control right allocation ratio between the driving system and the target object at the current moment; a second determination module, configured to determine target control parameters from a preset scenario continuous spectrum model based on the target control right allocation coefficient, wherein the preset scenario continuous spectrum model is used to describe the continuous mapping relationship between the control right allocation coefficient and the control parameters; and a control module, configured to control the vehicle based on the system state data, the target control parameters, and target constraints, wherein the target constraints are used to describe the constraints that need to be satisfied during the control process of the vehicle.

[0013] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0014] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0017] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0018] In this application embodiment, a vehicle control method is proposed. First, the system state data of the driving system in the vehicle, the environmental perception data of the environment in which the vehicle is located, and the object state data of the target object in the vehicle are acquired. Next, based on the system state data, environmental perception data, and object state data, the target control right allocation coefficient of the vehicle is determined. Then, based on the target control right allocation coefficient, the target control parameters are determined from a preset scenario continuous spectrum model. Finally, the vehicle is controlled based on the system state data, the target control parameters, and the target constraints. This application transforms the human-machine co-driving scenario into continuously changing target control allocation coefficients, achieving a dynamic correlation between perceptual input and environmental output. Simultaneously, it utilizes a pre-defined scenario continuous spectrum model to achieve a smooth, stepless mapping of target control parameters with the target control allocation coefficients. This avoids abrupt changes in environmental parameters and sensory discontinuities caused by discrete mode switching in related technologies. It ensures that vehicle environmental adjustments are always dynamically synchronized with the actual engagement of the target and the driving system's capabilities, thereby eliminating abrupt interventions and cognitive loads caused by "trigger-based responses." This achieves precise, gradual, and coordinated adaptation between environmental adjustments and the driving scenario, resulting in a systematic improvement in the continuity of occupant perception and physiological comfort at different driving stages. Furthermore, it solves the technical problem of low occupant comfort during vehicle control in related technologies. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is an embodiment of a vehicle control method according to an embodiment of this application;

[0021] Figure 2 This is a flowchart of an optional vehicle control method according to an embodiment of this application;

[0022] Figure 3 This is a schematic diagram of a system hardware architecture and data flow according to an embodiment of this application;

[0023] Figure 4 This is a schematic diagram of a vehicle scenario continuous spectrum model according to an embodiment of this application;

[0024] Figure 5 This is a timing control comparison diagram according to an embodiment of this application;

[0025] Figure 6 This is a schematic diagram of a layered wake-up according to an embodiment of this application;

[0026] Figure 7This is a schematic diagram of a vehicle control device according to an embodiment of this application. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] According to an embodiment of this application, an embodiment of a vehicle control method 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] Figure 1 This is an embodiment of a vehicle control method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0031] Step S102: Obtain system status data of the driving system in the vehicle, environmental perception data of the environment in which the vehicle is located, and object status data of the target object in the vehicle. The object status data is used to describe the degree of engagement of the target object with the driving task.

[0032] The aforementioned vehicles can refer to intelligent mobile vehicles equipped with autonomous driving capabilities. These vehicles may include, but are not limited to, passenger cars, commercial vehicles, and specialized autonomous driving platforms; the specific vehicle must be determined based on the actual control objectives. The vehicle can serve as the physical carrier of the control method proposed in this application, carrying all sensing sensors, computing units, and environmental actuators, and acting as a platform for acquiring and responding to system status, environmental data, and object status.

[0033] The aforementioned driving system refers to the automated hardware and software systems within a vehicle used to perform some or all of the driving tasks. The driving system may include, but is not limited to, perception modules, decision-making and planning modules, execution and control modules, and their integrated autonomous driving domain controllers; the specific driving system needs to be determined based on the vehicle type. The driving system can be responsible for perceiving the environment, assessing its own capabilities, generating control commands, and collaborating with the target object to complete the driving task in a human-machine co-driving mode.

[0034] The aforementioned system status data refers to a quantitative set of data representing the current operational capability, stability, and task execution reliability of the driving system. System status data may include, but is not limited to, system functional availability, system confidence, positioning accuracy, execution response latency, and function activation status; the specific system status data needs to be determined based on actual requirements. System status data can be used to represent the current state of the driving system.

[0035] The aforementioned environmental perception data refers to real-time data characterizing the risk level of the external environment in which the vehicle is located. Environmental perception data may include, but is not limited to, obstacle distance, traffic flow density, road type, and weather conditions; the specific environmental perception data needs to be determined based on the vehicle's environment. Environmental perception data can be used to quantify the complexity and danger of the current driving scenario.

[0036] The aforementioned target object can refer to the primary driving participant currently inside the vehicle. The behavior and physiological state of the target object directly affect the responsibility boundary of human-machine collaboration, and the target object can be used as the core of human factor feedback for the allocation of vehicle control.

[0037] The aforementioned object state data refers to multimodal physiological and behavioral data used to quantify the target object's level of attention and intensity of behavioral participation in the current driving task. Object state data may include, but is not limited to, hand operations (whether steering wheel grip strength is below a threshold, whether there are non-driving operations), object gaze (percentage of time the gaze deviates from the center of the road, blinking frequency), object physiological signals (heart rate variability, skin conductance response, used to determine fatigue or distraction), and voice interaction behavior (whether there is prolonged lack of response to system prompts), etc. Specific object state data needs to be determined based on actual needs. Object state data can be used to reflect the current state of the target object, determine the target object's level of engagement with the driving task at the current moment. The lower the object's engagement, the higher the urgency for system takeover, thereby achieving a "system in place when the person is not present" safety fallback mechanism.

[0038] In one optional embodiment, firstly, system status data of the vehicle driving system is acquired in real time to characterize the operational capability and reliability of the autonomous driving function; simultaneously, environmental perception data of the vehicle's environment is collected to quantify external risk conditions such as roads, traffic, and weather; and object status data of the target object is collected to monitor its gaze direction, hand operations, physiological signals, and other behavioral and physiological indicators, thereby objectively reflecting its level of attention to the driving task. These three types of data serve as the basic input for human-machine co-driving scenario assessment, collectively constituting the perception layer basis for dynamic allocation of control rights. This achieves multi-dimensional, synchronous, and objective quantitative collection of driving system capabilities, external environmental risks, and driver participation status, providing a real, reliable, and computable input foundation for subsequent adaptive allocation of control rights based on a continuous spectrum, effectively supporting the dynamic identification and smooth transition of human-machine responsibility boundaries.

[0039] For example, three types of data can be collected synchronously through the vehicle domain controller: System status data is output in real time by the functional modules within the autonomous driving domain controller, including internal operating indicators such as perception confidence, positioning accuracy, actuator response status, and auxiliary function activation status. Environmental perception data is acquired in real time by external perception devices such as surround-view cameras, millimeter-wave radar, lidar, and weather sensors, covering environmental risk parameters such as obstacle distribution, road curvature, weather conditions, and traffic flow density. Object status data is collected through in-vehicle biosensors (such as driver detection cameras, steering wheel grip sensors, and heart rate / conductivity of skin monitoring modules) to collect the driver's gaze focus, head posture, hand contact behavior, and physiological stress signals, in order to quantify the driver's level of attention to the driving task. After time alignment and preprocessing, the three types of data are uniformly input into the situation assessment module as the real-time perception basis for continuous control allocation.

[0040] Step S104: Based on system status data, environmental perception data, and object status data, determine the target control right allocation coefficient of the vehicle. The target control right allocation coefficient is used to quantify the control right allocation ratio between the driving system and the target object at the current moment.

[0041] The aforementioned target control allocation coefficient can be a continuous value between 0 and 1, used to quantify the dynamic allocation ratio of control responsibility for the vehicle driving task between the driving system and the target object at the current moment. This target control allocation coefficient is a unified semantic variable for the system's adaptive decision-making; the closer its value is to 1, the greater the control responsibility undertaken by the driving system; the closer it is to 0, the more dominant the target object is in the driving task.

[0042] The target control allocation coefficient can be used as the core quantitative representation of the human-machine co-driving scenario. The target control allocation coefficient integrates multi-source heterogeneous system state, environmental perception and object state data into a unified, continuous and computable decision input, which is used to drive the smooth adaptive adjustment of subsequent vehicle environmental parameters (such as light, sound, seat and other parameters), and provides a continuous and non-abrupt trigger basis for the takeover wake-up process, thereby eliminating the interaction fragmentation and psychological load caused by discrete mode switching.

[0043] The methods for determining the aforementioned target control allocation coefficients may include, but are not limited to, the following methods:

[0044] The first method, the dynamic weighted linear fusion method, involves normalizing the three types of input data—system state data, environmental perception data, and object state data—using a weighted summation formula and then linearly combining them to obtain the target control weight allocation coefficients. This method is the preferred implementation of this application and has the advantages of being calculable, adjustable, interpretable, and easy to implement in hardware.

[0045] The second method, based on fuzzy logic reasoning and membership mapping, constructs fuzzy membership functions (e.g., "high / medium / low system capability", "low / medium / high environmental risk", "strong / medium / weak driver engagement") for three types of input data: system state data, environmental perception data, and object state data. Fuzzy reasoning is then performed using a pre-set fuzzy rule base (e.g., "if system capability is high, environmental risk is low, and driver engagement is weak, then control is biased towards the system"). Finally, the centroid method is used to defuzzify and output continuous a(t) values, which are the target control allocation coefficients. This method is suitable for scenarios with high sensor noise and high data uncertainty.

[0046] The third method employs a pre-trained regression model (such as random forest regression or lightweight neural networks) to learn from historical driving data. The inputs are normalized system state data, environmental perception data, and object state data, with the output being a continuous a(t) value, i.e., the target control allocation coefficient. Model training is based on expert-annotated "ideal control allocation" samples (such as jointly annotated data from human-machine cooperative driving videos and physiological feedback) to achieve nonlinear mapping. This method is suitable for scenarios with abundant data and iterative system upgrades.

[0047] The above method for determining the target control allocation coefficient is only an example. The specific method should be determined according to actual needs and application scenarios, and is not limited here.

[0048] In one optional embodiment, a continuously changing target control allocation coefficient is generated by co-calculating three types of quantified data: system state data, environmental perception data, and object state data. This coefficient, with a value in the range [0, 1], represents the dynamic allocation ratio of driving control responsibility between the autonomous driving system and the driver at the current moment. A coefficient approaching 1 indicates system dominance, and approaching 0 indicates driver dominance. Furthermore, this coefficient does not depend on a preset mode threshold but evolves continuously in real time with the input data. This achieves precise, continuous, and stepless quantization of human-machine control allocation, avoiding the abrupt changes in interaction caused by traditional discrete mode switching. It provides a unified, stable, and computable decision-making basis for subsequent continuous spectrum-based environmental adaptive adjustment and hierarchical wake-up mechanisms, thereby improving the smoothness and safety of human-machine collaboration.

[0049] Step S106: Determine the target control parameters from the preset scenario continuous spectrum model based on the target control weight allocation coefficients, wherein the preset scenario continuous spectrum model is used to describe the continuous mapping relationship between the control weight allocation coefficients and the control parameters.

[0050] The aforementioned pre-defined scenario continuous spectrum model can refer to a pre-defined, non-discrete, interpolable multivariate mapping table or mathematical function curve. This model can be used to translate the abstract, continuous control allocation coefficient a(t) losslessly, smoothly, and predictably into specific vehicle environmental parameter target values, achieving a precise mapping from "human-machine responsibility ratio" to "occupant perception experience." Its core value lies in breaking away from traditional "switch-type" environmental control logic, enabling parameters such as headlight color temperature, ambient brightness, air conditioning fan speed / direction, seat posture, human-machine interface (HMI) information density, audio sound field, and tactile feedback intensity to adjust synchronously, progressively, and collaboratively with the continuous change of a(t), thereby constructing a "seamless transition" human-machine co-driving experience.

[0051] The aforementioned target control parameters refer to a set of continuous target values ​​output by the preset scenario continuous spectrum model based on the current target control weight allocation coefficient a(t), used to drive the vehicle's environmental actuators. These parameters encompass environmental configuration instructions across five perception dimensions: light, sound, temperature, touch, and information. Target control parameters may include, but are not limited to, light control parameters, sound control parameters, temperature control parameters, touch control parameters, information control parameters, and driving control parameters. Specific target control parameters need to be determined based on the target control weight allocation coefficient. Target control parameters can be used to transform abstract control weight allocation results into specific physical instructions that can be responded to by the actuators, enabling synchronous, coordinated, and gradual adjustment of the vehicle system.

[0052] In one optional embodiment, based on the target control allocation coefficient, the corresponding multi-dimensional environmental target configuration, i.e., the target control parameters, is retrieved or calculated from a preset scenario continuous spectrum model. This preset scenario continuous spectrum model uses the target control allocation coefficient a(t) as the sole input variable. Through a pre-calibrated continuous function, it seamlessly maps the control allocation ratio to target values ​​of control parameters in five dimensions: light, sound, temperature, touch, and information, such as ambient lighting color temperature, seat tilt angle, HMI information density, and driving control parameters. Each parameter evolves synchronously and smoothly with the continuous change of the target control allocation coefficient, without relying on any discrete mode switching logic. It generates the corresponding target control parameters in real time based solely on the current target control allocation coefficient, achieving a precise, continuous, and coordinated correlation between the control state and the environmental response.

[0053] Step S108: Based on system state data, target control parameters, and target constraints, the vehicle is controlled. The target constraints describe the constraints that need to be met during the vehicle control process.

[0054] The aforementioned target constraints can refer to a set of safety boundaries, physical limitations, and human safety thresholds that must be satisfied during vehicle control. Target constraints may include, but are not limited to, physical execution limit constraints, human safety threshold constraints, system stability and synchronization constraints, etc., and the specific target constraints need to be determined based on the actual situation. Target constraints can be used to prevent runaway output of target control parameters due to abnormal inputs, calibration errors, or extreme operating conditions. This ensures that environmental adjustment behavior remains within safe, comfortable, and acceptable physical and physiological boundaries without interfering with continuous adaptive logic, avoiding driver discomfort, distraction, or physiological stress caused by excessive adjustment (such as excessively strong microcurrents or excessively bright lights), thus supporting the core design goal of "unobtrusive yet safe."

[0055] In one optional embodiment, based on system state data, target control parameters, and target constraints introduced by the system in the environmental coupling controller, the commands of each environmental actuator are subject to real-time amplitude and rate constraints. This ensures that the adjustment of parameters such as ambient light brightness, color temperature change rate, seat tilt angle, micro-current intensity, and augmented reality head-up display (AR-HUD) information density never exceeds the physical execution limits, human safety thresholds, and dynamic response rate limits of each actuator. This achieves "seamless but reliable" environmental collaborative control, providing verifiable and mass-producible safety execution assurance for the vehicle situation adaptive system.

[0056] In this application embodiment, a vehicle control method is proposed. First, the system state data of the driving system in the vehicle, the environmental perception data of the environment in which the vehicle is located, and the object state data of the target object in the vehicle are acquired. Next, based on the system state data, environmental perception data, and object state data, the target control right allocation coefficient of the vehicle is determined. Then, based on the target control right allocation coefficient, the target control parameters are determined from a preset scenario continuous spectrum model. Finally, the vehicle is controlled based on the system state data, the target control parameters, and the target constraints. This application transforms the human-machine co-driving scenario into continuously changing target control allocation coefficients, achieving a dynamic correlation between perceptual input and environmental output. Simultaneously, it utilizes a pre-defined scenario continuous spectrum model to achieve a smooth, stepless mapping of target control parameters with the target control allocation coefficients. This avoids abrupt changes in environmental parameters and sensory discontinuities caused by discrete mode switching in related technologies. It ensures that vehicle environmental adjustments are always dynamically synchronized with the actual engagement of the target and the driving system's capabilities, thereby eliminating abrupt interventions and cognitive loads caused by "trigger-based responses." This achieves precise, gradual, and coordinated adaptation between environmental adjustments and the driving scenario, resulting in a systematic improvement in the continuity of occupant perception and physiological comfort at different driving stages. Furthermore, it solves the technical problem of low occupant comfort during vehicle control in related technologies.

[0057] Optionally, based on system state data, environmental perception data, and object state data, the target control right allocation coefficient of the vehicle is determined, including: normalizing the system state data, environmental perception data, and object state data to obtain multiple influencing factors corresponding to the system state data, environmental perception data, and object state data respectively; determining the weights corresponding to the multiple influencing factors respectively; and using a weighted fusion model to fuse the multiple influencing factors and their corresponding weights to obtain the target control right allocation coefficient.

[0058] The aforementioned normalization process refers to the standardization process of mapping system state data, environmental perception data, and object state data from their original dimensions and numerical ranges to the interval [0, 1]. Types of normalization include, but are not limited to, min-max normalization, Z-score normalization, and piecewise linear mapping. The specific normalization method needs to be determined based on the data type and actual requirements. Normalization can eliminate numerical bias caused by differences in sensor sampling range, units, or uneven data distribution, ensuring that the contribution weight of each influencing factor in subsequent weighted fusion is determined solely by its intrinsic importance.

[0059] The aforementioned multiple influencing factors can refer to independent input variables that quantify and characterize the overall vehicle situation, formed by feature extraction from normalized system state data, environmental perception data, and object state data. These multiple influencing factors may include, but are not limited to, system state factors, environmental perception factors, and object state factors; the specific influencing factors need to be determined based on the actual situation. These multiple influencing factors can be used to decouple complex and fuzzy driving situations into several calculable, monitorable, and calibrable numerical variables, providing structured input for the calculation of continuous control right allocation coefficients, and realizing semantic abstraction from "perceived data" to "situational semantics."

[0060] The weights corresponding to the aforementioned multiple influencing factors can refer to the dynamic weighting coefficients assigned to each influencing factor. These weights may include, but are not limited to, system state weights, environmental perception weights, and object state weights; the specific weights need to be determined based on the influencing factors. The weight corresponding to each influencing factor characterizes its relative importance to the control allocation decision in the current driving situation, and its value ranges from [0, 1], with the sum of all weights always equal to 1. The weights corresponding to multiple influencing factors can be used to achieve dynamic priority adjustment of situational awareness, enabling the system to automatically strengthen or weaken the influence of a certain dimension based on real-time risks and task requirements (e.g., increasing the environmental risk entropy weight when there is high risk, and decreasing the system capability weight when the driver is focused), thereby improving the responsiveness and situational fit of control allocation.

[0061] The aforementioned weighted fusion model can refer to a fusion mechanism based on linear or nonlinear mathematical operations, which is used to combine multiple normalized influence factors with their respective dynamic weights to calculate and output a continuous, unitless target control weight allocation coefficient a(t), as a unified quantitative expression of the whole vehicle scenario.

[0062] In one optional embodiment, the system state data, environmental perception data, and object state data are first normalized to eliminate differences in their dimensions and numerical ranges, resulting in multiple independent, dimensionless influence factors corresponding to the system state data, environmental perception data, and object state data, respectively. Then, the weights corresponding to each influence factor are dynamically assigned based on the current driving task and risk state, ensuring that the sum of the weights is always 1. Finally, a weighted linear fusion model is used to linearly combine all influence factors with their corresponding weights to generate target control allocation coefficients that continuously take values ​​in the interval [0, 1]. This process ensures the comparability of multi-source heterogeneous data through normalization, achieves adaptive priority adjustment of situational elements through dynamic weight allocation, and transforms dispersed perception information into continuous, calculable control allocation coefficients through a weighted fusion model. This provides accurate, stable, and interpretable input for subsequent adaptive environmental adjustment, improving the system's quantitative perception capability and response consistency in complex driving situations.

[0063] Optionally, the multiple influencing factors include: system state factors, environmental perception factors, and object state factors; the weights corresponding to the multiple influencing factors include: system state weights, environmental perception weights, and object state weights; a weighted fusion model is used to fuse the multiple influencing factors and their corresponding weights to obtain the target control allocation coefficient, including: determining the system control quantity based on the system state factors and system state weights, where the system control quantity is directly proportional to the system state factors, and the system control quantity is used to represent the driving system's ability to undertake the driving task; determining the environmental control quantity based on the environmental perception factors and environmental perception weights, where the environmental control quantity is inversely proportional to the environmental perception factors, and the environmental control quantity is used to represent the urgency of the driving system taking over the driving task under the current environmental risk level; determining the object control quantity based on the object state factors and object state weights, where the object control quantity is inversely proportional to the object state factors, and the object control quantity is used to represent the urgency of the driving system taking over the driving task under the current level of the target object's engagement in the driving task; and determining the target control allocation coefficient based on the system control quantity, environmental control quantity, and object control quantity.

[0064] The aforementioned system state factor can refer to a scalar value, after normalization, characterizing the current functional availability, operational confidence, and task execution capability of the driving system, with a value range of [0, 1]. In this application, the system state factor can be the system capability score Cs(t). The system state factor can be used to quantify what the driving system "can do" at the current moment, serving as a direct reflection of "system capability" in the allocation of control, and is the inherent basis for the system to claim control.

[0065] The aforementioned environmental perception factor can refer to a scalar value, after normalization, characterizing the comprehensive risk level of the external environment in which the vehicle is located, with a value range of [0, 1]. In this application, the environmental perception factor can be the environmental normalized risk entropy Ne(t). The environmental perception factor can be used to quantify "whether the environment is suitable for the system's autonomous driving," reflecting the degree of constraint of the external environment on the system's autonomous capabilities, and used to assess whether the system should proactively reduce control or prepare to take over.

[0066] The aforementioned object state factor can refer to a scalar value, after normalization, representing the driver's current level of engagement with the driving task, with a value range of [0, 1]. In this application, the object state factor can be the driver engagement index Ed(t). The object state factor can be used to quantify whether the driver is in an effective monitoring state, serving as a key basis for determining whether the system needs to actively intervene or prepare for takeover.

[0067] The aforementioned system state weights can refer to dynamic coefficients, i.e., Ws(t), used to adjust the contribution intensity of system state factors in the allocation of target control rights. System state weights can be used to control the influence of "system capabilities" in decision-making. When the system functions fully, its weight is increased; when the system is limited or the task is unclear, its weight is decreased, avoiding "capable but irresponsible" or "incompetent but stubborn."

[0068] The aforementioned environmental perception weight can refer to a dynamic coefficient, We(t), used to adjust the contribution intensity of environmental perception factors in the allocation of target control rights. Environmental perception weight can be used to enhance the dominance of "high-risk environments" in the allocation of control rights, ensuring that the system automatically reduces a(t) when the risk increases, thus realizing the "safety first" principle.

[0069] The aforementioned object state weights can refer to dynamic coefficients, i.e., Wd(t), used to adjust the contribution of object state factors in the allocation of target control. Object state weights can increase the urgency of system takeover when the driver's attention is distracted or they are off-task.

[0070] The aforementioned system control quantity can refer to the positive output value that characterizes the system's assertion of control, calculated based on system state factors and system state weights. The system control quantity can be used to represent "how much driving responsibility the system believes it can assume," and is the "system assertive force" in the allocation of control, used to counteract and balance "environmental urgency" and "human-caused risks."

[0071] The aforementioned environmental control quantity can refer to a negative output value, calculated based on environmental perception factors and environmental perception weights, that characterizes the degree to which environmental risk inhibits system control. The environmental control quantity can be used to reflect "the urgency of the system actively relinquishing control under the current environment," that is, the more dangerous the environment, the smaller the environmental control quantity, and the more the system should yield. As a "negative inhibition factor," the environmental control quantity is used to reduce overall control.

[0072] For example, the environmental control quantity can be We×(1-Ne(t)), where We(t) is the environmental perception weight and Ne(t) is the environmental perception factor. When Ne(t) approaches 0.1, the environmental control quantity increases to 0.9, indicating that the environment is safe and the driving system can assume more responsibility. When Ne(t) approaches 0.9, the environmental control quantity decreases to 0.1, indicating that the environment is high-risk and the driving system needs to yield. The above design reflects the technical logic of prioritizing safety. The above values ​​are only examples; specific values ​​need to be determined according to the actual situation and are not limited here.

[0073] The aforementioned object control quantity can be considered a negative output value, calculated based on object state factors and object state weights, characterizing the degree to which the risk of driver disengagement inhibits system control. The object control quantity reflects the urgency of the system taking over due to driver inattention; the lower the driver's participation / attention, the smaller the object control quantity, and the higher the urgency of system takeover. The object control quantity and environmental control quantity together constitute the "human-caused risk inhibition mechanism."

[0074] In one optional embodiment, the system control quantity, environmental control quantity, and object control quantity are calculated by linearly combining the system state factor, environmental perception factor, and object state factor with their corresponding system state weights, environmental perception weights, and object state weights, respectively. The system control quantity is the product of the system state factor and the system state weight, representing the driving system's proactive task-taking capability. The environmental control quantity is the product of the environmental perception weight and (1 minus the environmental perception factor), representing the urgency of system takeover due to increased environmental risk. The object control quantity is the product of the object state weight and (1 minus the object state factor), representing the urgency of system takeover due to decreased driver engagement. Finally, the target control allocation coefficient is determined by comprehensively considering the system control quantity, environmental control quantity, and object control quantity through a preset product or weighted combination relationship.

[0075] The above process, by transforming the three types of perception factors and dynamic weights into control quantities with clear physical meanings, constructs a triple-logic control allocation mechanism, realizing a detailed, structured, interpretable, and quantifiable understanding of human-machine co-driving scenarios. This enables the control allocation coefficient to accurately respond to the coordinated changes in system capabilities, environmental risks, and driver status, effectively avoiding abrupt interactions triggered by a single threshold, and improving the smoothness, safety, and human-factor adaptability of the system's control transition in complex scenarios.

[0076] Optionally, the vehicle is controlled based on system state data, target control parameters, and target constraints, including: determining the current control parameters of the vehicle based on system state data; determining the control duration for completing vehicle control based on target constraints, current control parameters, and target control parameters; and controlling the vehicle based on the control duration, target constraints, and target control parameters.

[0077] The aforementioned current control parameters refer to quantified values ​​that characterize the actual control state of the vehicle at the current moment, calculated or read in real time based on system state data. Current control parameters may include, but are not limited to, dynamic control parameters (such as longitudinal acceleration, steering angular velocity, braking pressure, etc.), function execution control parameters (such as air conditioning set temperature, ambient lighting brightness, seat posture angle, voice prompt volume, etc.), and system state control parameters (such as the current activation level of the autonomous driving function, system confidence output value, etc.). Specific current control parameters need to be determined based on the actual control objectives and system state data. Current control parameters can be used as a starting point benchmark for the control process, dynamically compared with target control parameters, quantifying the gap between the current state and the desired state. They serve as inputs for calculating control duration, evaluating system response capabilities and constraint satisfaction, ensuring that control actions have a traceable basis for "where to begin."

[0078] The aforementioned control duration refers to the time required for a smooth transition from the current control state to the target control state, derived from a pre-defined calculation model based on the dynamic relationship between the target constraints, current control parameters, and target control parameters. Control duration can be used to transform the "control objective" into a "gradual process over time," avoiding driving interference or system overload caused by abrupt commands. By controlling duration, the "rate controllability" of control actions is achieved, ensuring the smoothest, most comfortable, and safest state transition while meeting target constraints (such as maximum acceleration, rate of temperature change, and brightness gradient).

[0079] In one optional embodiment, the vehicle's current control parameters are first obtained based on system state data, serving as the initial state for the control action. Then, the current control parameters are combined with target control parameters and target constraints (such as rate of change, physical limits, and comfort thresholds) to calculate the minimum safe control duration required for the state transition, i.e., the control duration, using a linear or constraint model. Finally, based on the control duration, target constraints, and target control parameters, a closed-loop control strategy with rate limiting is employed to drive the actuator for gradual adjustment. The entire process revolves around "time smoothing," achieving a seamless, controllable, and compliant transition from the "current state" to the "target state."

[0080] Optionally, the control duration for completing vehicle control is determined based on the target constraints, current control parameters, and target control parameters, including: determining the control parameter difference based on the current control parameters and the target control parameters; and determining the control duration based on the control parameter difference and the target constraints.

[0081] The aforementioned control parameter difference refers to the absolute or relative deviation between the current control parameters and the target control parameters, used to quantify the required state transition range of the vehicle. The control parameter difference serves as a core bridge connecting the "current state" and the "target state," and is a prerequisite input for calculating the control duration. Its value directly determines how long the system needs to complete the transition while meeting the target constraints. Accurately quantifying the difference avoids interference caused by "too fast" control actions or response lag caused by "too slow" actions, achieving precise timing planning and efficient resource utilization in the control process.

[0082] In one optional embodiment, the control parameter difference is first calculated by comparing the target control parameter with the current control parameter. This difference can be the absolute deviation or normalized deviation between the target and current control parameters, reflecting the magnitude of the state transition. Subsequently, the control parameter difference is combined with the target constraint, and the minimum control time required to complete the state transition is determined through linear division. This mechanism achieves a clear mathematical relationship to map the "magnitude-time" of control actions, ensuring that all environmental adjustment behaviors are smoothly executed under safe, comfortable, and controllable constraints.

[0083] Optionally, the method further includes: if the environmental perception data exceeds a preset environmental risk threshold within a preset time period, executing the methods in the various embodiments of this application.

[0084] The aforementioned preset time refers to the minimum time window required to maintain environmental perception data in an abnormal state (i.e., exceeding a preset environmental risk threshold). This preset time prevents the system from mistakenly triggering control switching or environmental adjustments due to perceived noise, momentary sensor jitter, or brief obstructions (such as a vehicle changing lanes ahead or tree shadows obscuring the radar). By introducing a time verification mechanism, the system's response reliability to real, continuous high-risk scenarios is improved, ensuring that the system intervenes only when necessary, achieving a safety logic of "steady-state risk triggering and dynamic response."

[0085] The aforementioned preset environmental risk threshold can refer to a critical risk level set for environmental perception factors (such as Ne(t)) to determine whether the current driving environment constitutes a "critical risk level that requires triggering the adaptive control mechanism". The preset environmental risk threshold can serve as a "decision triggering threshold" for environmental risk, used to determine whether the precondition for "executing the method of this application" is met. The system only initiates the recalculation of the continuous control allocation coefficient a(t) and the environmental coupling adjustment process if and only if the environmental perception data remains above this threshold for a preset time period, thereby avoiding invalid calculations and resource waste in low-risk scenarios and improving system energy efficiency and response accuracy.

[0086] In one optional embodiment, the system only recognizes a real and continuous high-risk situation when environmental sensing data (such as Ne(t)) continuously exceeds a preset environmental risk threshold and the duration of this over-limit state reaches a preset time. This triggers the continuous situation assessment, control allocation, and seamless environmental coupling adjustment process described in this application. This mechanism, through dual verification using a threshold and time window, effectively distinguishes between transient interference and real risks, ensuring that the system intervenes only at critical times, thus improving the accuracy, security, and resource efficiency of the response.

[0087] Optionally, an environmental control quantity is determined based on an environmental perception factor and an environmental perception weight; an object control quantity is determined based on an object state factor and an object state weight, including: determining the difference between a preset value and an environmental perception factor as a first difference; determining the difference between a preset value and an object state factor as a second difference; determining the environmental control quantity based on the first difference and the environmental perception weight; and determining the object control quantity based on the second difference and the object state weight.

[0088] The aforementioned preset value can refer to 1. The preset value can be used to make the environmental perception factor inversely proportional to the environmental control quantity, and the object state factor inversely proportional to the object control quantity. By subtracting from the preset value, a mathematical mapping is achieved to convert the "high-risk / low-input" state into a "positive inhibition signal," ensuring that the calculation logic of the environmental control quantity and the object control quantity possesses physical interpretability and safety symmetry.

[0089] In one optional embodiment, firstly, a uniform preset value (1) is set as a reference benchmark for the environmental perception factor and the object state factor; then, the differences between the preset value and the environmental perception factor and the object state factor are calculated respectively to form a first difference (1-Ne(t)) and a second difference (1-Ed(t)), both of which represent the degree of "deviation from the ideal safe state"; subsequently, the first difference is multiplied by the environmental perception weight to obtain the environmental control quantity, and the second difference is multiplied by the object state weight to obtain the object control quantity, thereby realizing the negative feedback control logic of "the higher the environmental risk, the more the system yields; the lower the human factor input, the more the system intervenes". This mechanism has a symmetrical structure, deterministic calculation, and provable logic, providing a stable and interpretable input source for the stationary calculation of the continuous target control weight allocation coefficient a(t), and improving the refinement and safety of human-machine co-driving situation judgment.

[0090] In an optional embodiment, the formula for calculating the target control allocation coefficient a(t) is as follows:

[0091] ;

[0092] In the formula, The system capability score is calculated based on the autonomous driving function status and system confidence level, and ranges from [0, 1]. Used to represent the environmental normalized risk entropy: a quantitative risk that integrates obstacles, traffic flow, roads, and weather, ranging from [0, 1]. The driver engagement index is defined by integrating operational behavior, line of sight, and physiological signals, ranging from [0, 1]. Ws(t), We(t), and Wd(t) represent the dynamic weights corresponding to the three types of sensory data, respectively, and these dynamic weights can be adjusted in real time according to the task and risk, with a total sum of 1. When the environmental risk Ne(t) increases sharply, the value of the (1-Ne(t)) term in the formula decreases, actively lowering the threshold. The value drives the system back towards human supervision, which reflects the core principle of safety first.

[0093] Figure 2 This is a flowchart of an optional vehicle control method according to an embodiment of this application, such as... Figure 2As shown, this method begins with real-time acquisition and quantization of multi-source data Cs(t), Ne(t), Ed(t), and weight W(t); then, a dynamic weighted fusion model is executed to calculate the continuous control weight coefficient a(t); next, it is determined whether the continuous control weight coefficient a(t) has drifted significantly. If no significant drift has occurred, the instruction is sent to each environmental actuator; if a significant drift has occurred, the "contextual continuity spectrum" mapping relationship is queried to determine the target environmental parameter set; then, the "seamless coupling scheduling" algorithm is executed to generate instructions with dynamic rate constraints; then, the instructions are sent to each environmental actuator; finally, the adjusted occupant physiological feedback data is collected to continue the vehicle control process.

[0094] Specifically, firstly, the system collects and quantifies three types of perception data in real time: system capability score Cs(t), environmental normalized risk entropy Ne(t), and driver engagement index Ed(t), and loads preset dynamic weights Ws(t), We(t), and Wd(t); then, it calculates the continuous control weight coefficient a(t) through a dynamic weighted fusion model; then, it determines whether a(t) has drifted significantly compared to the previous moment. If there is no drift, it directly uses the current environmental parameters to send commands to actuators such as ambient lighting, air conditioning, seats, AR-HUD, and microcurrent arrays. If a significant drift occurs, the preset "contextual continuity spectrum" mapping table is queried to determine the offline-calibrated target environmental parameter set corresponding to the current a(t) value. The "seamless coupling scheduling" algorithm is then invoked to generate a gradual control command with dynamic rate constraints based on the maximum change rate set for each environmental parameter type (e.g., lights ≤ 5% / s, seats ≤ 2° / s). This command is then sent to the actuator to achieve smooth adjustment. Finally, physiological feedback data such as the driver's heart rate variability and skin conductance response after adjustment are collected as offline inputs for subsequent weight adjustments, and the cycle continues into the next control cycle.

[0095] This process, through a mechanism of "drift-triggered mapping + rate constraint control," avoids frequent environmental disturbances caused by data jitter. While ensuring timely situational response, it achieves low-disturbance, high-coordination environmental regulation and improved physiological comfort, reducing driver cognitive load and ensuring the determinism of system control logic and the verifiability of functional safety. The values ​​in the above process are for illustrative purposes only; specific values ​​need to be determined based on actual needs and are not limited here.

[0096] Figure 3 This is a schematic diagram of a system hardware architecture and data flow according to an embodiment of this application, such as... Figure 3As shown, the system's hardware architecture mainly consists of a multi-source perception layer (input), a core processing layer, and a vehicle environment actuator network (output). The multi-source perception layer includes: intelligent driving domain state (functional availability / system confidence), occupant state perception (driver behavior / physiological indicators), and environmental perception (risk entropy / road type / weather). The core processing layer mainly includes: a continuous context assessment algorithm (dynamic weighted fusion model), an environmental coupling controller, and a continuous spectrum mapping and environmental parameter decision module. The vehicle environment actuator network includes: a visual adjustment unit (ambient lighting / AR-HUD / switching glass, etc.), an auditory adjustment unit (3D audio / active noise cancellation, etc.), and a thermo-tactile adjustment unit (multi-zone climate control / 4D haptic seats, etc.).

[0097] After the multi-dimensional heterogeneous data, including intelligent driving domain state, occupant state perception, and environmental perception, is input into the multi-source perception layer, the continuous context evaluation algorithm (dynamic weighted fusion model) of the core processing layer processes the multi-dimensional heterogeneous data to obtain the continuous control weight coefficient a(t). Then, a(t) is processed by the continuous spectrum mapping and environmental parameter decision module to obtain the collaborative scheduling instruction. Finally, the collaborative scheduling instruction is sent to the vehicle environmental actuator network to control the visual adjustment unit, auditory adjustment unit, and thermo-tactile adjustment unit to perform corresponding adjustments.

[0098] This application constructs a three-tier architecture of "multi-source perception layer - core processing layer - environmental actuator network," which uses continuous control weight coefficients a(t) to uniformly represent the human-machine co-driving situation. This integrates the three previously isolated inputs—intelligent driving state, driver state, and environmental risk—into a continuous and computable global situational signal. This, in turn, drives the coordinated adjustment of the three environmental dimensions of vision, hearing, and tactile sensation, overcoming the abrupt interference of traditional discrete mode switching and achieving smooth, synchronous, and low-disturbance gradual change of environmental parameters with a(t). At the same time, by embedding continuous spectrum mapping and dynamic rate constraints into the core processing layer, it ensures that environmental adjustment accurately matches situational changes while conforming to human perception thresholds and physiological comfort, reducing the driver's cognitive load and psychological tension, and improving the safety and continuity of human-machine collaboration.

[0099] Figure 4 This is a schematic diagram of a vehicle scenario continuous spectrum model according to an embodiment of this application, such as... Figure 4 As shown, Figure 4 The two-dimensional mapping relationship of the whole vehicle scenario continuous spectrum model is shown in the figure. The horizontal axis is the continuous control right allocation coefficient a(t), which ranges from 0 (manual driving) to 1 (fully automated driving). The vertical axis is the environmental support tendency (tendency to relax), that is, the target occupant state supported by the whole vehicle environment configuration, including light environment (brightness, color temperature), sound environment (volume, music type), thermal environment (temperature, wind speed), tactile environment (seat posture, vibration intensity) and information environment (HUD information density). Figure 4A smooth, continuous spectrum curve is plotted, connecting five reference mode nodes: the left end is "manual driving mode", the right end is "fully automated driving mode", and the middle ends are "cooperative assistance mode", "supervisory escort mode" and "safe management mode" in sequence. The curve represents the mapping trajectory of each environmental parameter as the value of a(t) changes continuously. Each reference mode is the coordinate point of a specific value of a(t) on the curve, which is a non-discrete switch mode. The arrows indicate that the real-time value of a(t) of the system moves smoothly along the curve, driving the environmental parameters to change synchronously and gradually.

[0100] This application constructs a vehicle context continuous spectrum model with the continuous control power coefficient a(t) as the horizontal axis and the environmental support tendency as the vertical axis. This transforms the originally discrete driving mode into a smooth and continuous multi-dimensional environmental mapping trajectory, allowing the five major environmental parameters of light, sound, heat, touch, and information to change steplessly with the value of a(t). This solves the sensory abruptness and psychological discontinuity caused by the traditional "switch-type" mode switching, and achieves a natural, seamless, and collaborative matching between the occupant's state and the vehicle context, thereby improving the comfort, immersion, and cognitive continuity during human-machine co-driving.

[0101] Figure 5 This is a timing control comparison diagram according to an embodiment of this application, such as... Figure 5 As shown, the left side is the mode transition timing diagram of the traditional solution. It can be seen that the traditional solution suddenly triggers a high-intensity alarm at time T0, abruptly interrupting the driver's state and causing a stress response. The right side is the mode transition timing diagram of the proposed solution. It can be seen that the proposed solution begins before T0, achieving a seamless pre-adjustment stage driven by the continuous change of the a(t) value, gradual change of headlight color temperature, gentle adjustment of seat posture, change of air conditioning fan speed, and micro-current background connection. Until time T0, a gentle explicit prompt is added, and the driver's state is guided by the environment to achieve a seamless and smooth transition.

[0102] This shows that traditional solutions deliver sudden, highly disruptive outputs at time T0. In contrast, this application, based on the continuous change of a(t) value before T0, drives the environment to perform early, gentle, and multi-sensory pre-adjustments, allowing the driver to unconsciously complete state preparation and improving the driver experience.

[0103] Figure 6 This is a schematic diagram of a layered wake-up according to an embodiment of this application. Figure 6 The diagram illustrates the intelligent hierarchical wake-up process executed by the system when a sudden increase in environmental risk Ne(t) leads to a continuous decrease in a(t). The process mainly includes: the first stage, risk perception and subconsciousness: the value of a(t) begins to decrease; the second stage, physiological arousal and status: the value of a(t) enters the low-risk zone; and the third stage: explicit takeover request: the value of a(t) approaches the manual driving range.

[0104] Specifically, the process begins in the first stage, risk perception and subconscious awareness: the a(t) value begins to decrease. In this stage, the risk Ne(t) increases, the dynamic weight We(t) increases, and the a(t) value continues to decrease, at which point a subconscious warning is activated. As a(t) continues to decrease, it crosses the "safe management boundary" and enters the second stage, physiological arousal and situational awareness: the a(t) value enters the low-risk zone: environmental pre-adjustment reverts, lights become colder / seat tightens, and the AR-HUD prompts an escalation of the situation. When a(t) continues to decrease, it crosses the "supervision and escort boundary" and enters the third stage: explicit takeover request: the a(t) value approaches the manual driving range. In this stage, an explicit multimodal takeover request is triggered, which is managed by the driver or the system to execute driving behavior requests. Figure 6 Each level of wake-up action corresponds one-to-one with the descent interval of a(t). The arrows indicate the hierarchical progression driven by the continuous change of a(t). There are no independent event trigger points. This wake-up process is a continuous, progressive, multi-sensory collaborative adaptive wake-up mechanism.

[0105] According to an embodiment of this application, a vehicle control device is provided. It should be noted that this device can be used to execute the vehicle control method described above. Specific embodiments and preferred application scenarios are the same as those of the vehicle control method described above, and will not be repeated here.

[0106] Figure 7 This is a schematic diagram of a vehicle control device according to an embodiment of this application, such as... Figure 7 As shown, the device includes: an acquisition module 702, a first determination module 704, a second determination module 706, and a control module 708.

[0107] The acquisition module 702 is used to acquire system state data of the driving system in the vehicle, environmental perception data of the environment in which the vehicle is located, and object state data of the target object in the vehicle. The object state data is used to describe the degree of engagement of the target object with the driving task. The first determination module 704 is used to determine the target control right allocation coefficient of the vehicle based on the system state data, environmental perception data, and object state data. The target control right allocation coefficient is used to quantify the control right allocation ratio between the driving system and the target object at the current moment. The second determination module 706 is used to determine the target control parameters from a preset scenario continuous spectrum model based on the target control right allocation coefficient. The preset scenario continuous spectrum model is used to describe the continuous mapping relationship between the control right allocation coefficient and the control parameters. The control module 708 is used to control the vehicle based on the system state data, the target control parameters, and the target constraints. The target constraints are used to describe the constraints that need to be met during the control process of the vehicle.

[0108] Optionally, the first determining module is used to normalize the system state data, environmental perception data, and object state data to obtain multiple influencing factors corresponding to the system state data, environmental perception data, and object state data, respectively; determine the weights corresponding to the multiple influencing factors; and use a weighted fusion model to fuse the multiple influencing factors and their corresponding weights to obtain the target control right allocation coefficient.

[0109] Optionally, the multiple influencing factors include: system state factors, environmental perception factors, and object state factors; the weights corresponding to the multiple influencing factors include: system state weights, environmental perception weights, and object state weights; the first determining module is further used to determine the system control quantity based on the system state factors and system state weights, wherein the system control quantity is directly proportional to the system state factors, and the system control quantity is used to represent the driving system's ability to undertake the driving task; to determine the environmental control quantity based on the environmental perception factors and environmental perception weights, wherein the environmental control quantity is inversely proportional to the environmental perception factors, and the environmental control quantity is used to represent the urgency of the driving system taking over the driving task under the current environmental risk level; to determine the object control quantity based on the object state factors and object state weights, wherein the object control quantity is inversely proportional to the object state factors, and the object control quantity is used to represent the urgency of the driving system taking over the driving task under the current level of investment of the target object in the driving task; and to determine the target control right allocation coefficient based on the system control quantity, environmental control quantity, and object control quantity.

[0110] Optionally, the control module is used to determine the current control parameters of the vehicle based on system status data; determine the control duration for completing vehicle control based on target constraints, current control parameters, and target control parameters; and control the vehicle based on the control duration, target constraints, and target control parameters.

[0111] Optionally, the control module is also used to determine the degree of difference between the current control parameters and the target control parameters; and to determine the control duration based on the degree of difference between the control parameters and the target constraints.

[0112] Optionally, the device is also used to execute the methods in various embodiments of this application if the environmental perception data exceeds a preset environmental risk threshold within a preset time period.

[0113] Optionally, the first determining module is further configured to determine the difference between the preset value and the environmental perception factor as a first difference; determine the difference between the preset value and the object state factor as a second difference; determine the environmental control quantity based on the first difference and the environmental perception weight; and determine the object control quantity based on the second difference and the object state weight.

[0114] Embodiments of this application also provide an electronic device, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0115] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0116] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0117] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0118] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0119] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0121] 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0122] Furthermore, the functional units in the various embodiments of this application 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0123] If the integrated unit 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 application, in essence, or the part that contributes to the prior art, or all or 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 described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0124] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A vehicle control method, characterized in that, include: The system status data of the driving system in the vehicle, the environmental perception data of the environment in which the vehicle is located, and the object status data of the target object in the vehicle are acquired, wherein the object status data is used to describe the degree of engagement of the target object with the driving task. Based on the system status data, the environmental perception data, and the object status data, the target control allocation coefficient of the vehicle is determined, wherein the target control allocation coefficient is used to quantify the control allocation ratio between the driving system and the target object at the current moment; Based on the target control power allocation coefficients, target control parameters are determined from a preset scenario continuous spectrum model, wherein the preset scenario continuous spectrum model is used to describe the continuous mapping relationship between the control power allocation coefficients and the control parameters; The vehicle is controlled based on the system state data, the target control parameters, and the target constraints, wherein the target constraints describe the constraints that need to be met during the control process of the vehicle.

2. The method according to claim 1, characterized in that, Based on the system status data, the environmental perception data, and the object status data, the target control rights allocation coefficient of the vehicle is determined; The system state data, the environmental perception data, and the object state data are normalized to obtain multiple influencing factors corresponding to the system state data, the environmental perception data, and the object state data, respectively. Determine the weights corresponding to the plurality of influencing factors respectively; The weighted fusion model is used to fuse the multiple influencing factors and their corresponding weights to obtain the target control allocation coefficient.

3. The method according to claim 2, characterized in that, The multiple influencing factors include: system state factors, environmental perception factors, and object state factors; the weights corresponding to the multiple influencing factors include: system state weights, environmental perception weights, and object state weights; a weighted fusion model is used to fuse the multiple influencing factors and their corresponding weights to obtain the target control right allocation coefficients, including: Based on the system state factor and the system state weight, a system control quantity is determined, wherein the system control quantity is directly proportional to the system state factor, and the system control quantity is used to represent the driving system's ability to undertake the driving task; Based on the environmental perception factor and the environmental perception weight, an environmental control quantity is determined, wherein the environmental control quantity is inversely proportional to the environmental perception factor, and the environmental control quantity is used to represent the urgency of the driving system taking over the driving task under the current environmental risk level; Based on the object state factor and the object state weight, an object control quantity is determined, wherein the object control quantity and the object state factor are inversely proportional, and the object control quantity is used to represent the urgency of the driving system taking over the driving task given the current level of engagement of the target object with the driving task. Based on the system control quantity, the environmental control quantity, and the object control quantity, the target control right allocation coefficient is determined.

4. The method according to claim 1, characterized in that, Based on the system state data, the target control parameters, and the target constraints, the vehicle is controlled, including: Based on the system status data, the current control parameters of the vehicle are determined; Based on the target constraints, the current control parameters, and the target control parameters, determine the control duration for completing vehicle control; The vehicle is controlled based on the control duration, the target constraints, and the target control parameters.

5. The method according to claim 4, characterized in that, Based on the target constraints, the current control parameters, and the target control parameters, the control duration for completing vehicle control is determined, including: Based on the current control parameters and the target control parameters, determine the degree of difference in control parameters; The control duration is determined based on the difference in control parameters and the target constraints.

6. The method according to claim 1, characterized in that, The method further includes: If the environmental perception data exceeds a preset environmental risk threshold within a preset time period, the method described in any one of claims 1 to 5 shall be executed.

7. A vehicle control device, characterized in that, include: The acquisition module is used to acquire system status data of the driving system in the vehicle, environmental perception data of the environment in which the vehicle is located, and object status data of the target object in the vehicle, wherein the object status data is used to describe the degree of engagement of the target object with the driving task. The first determining module is used to determine the target control allocation coefficient of the vehicle based on the system state data, the environmental perception data, and the object state data, wherein the target control allocation coefficient is used to quantify the control allocation ratio between the driving system and the target object at the current moment; The second determining module is used to determine the target control parameters from the preset scenario continuous spectrum model based on the target control power allocation coefficients, wherein the preset scenario continuous spectrum model is used to describe the continuous mapping relationship between the control power allocation coefficients and the control parameters; The control module is used to control the vehicle based on the system status data, the target control parameters, and the target constraints, wherein the target constraints describe the constraints that need to be met during the control process of the vehicle.

8. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images