Intelligent cockpit control method and device of vehicle, vehicle and readable storage medium

Abstract

The embodiment of the application provides a kind of intelligent cabin control method, device, vehicle and readable storage medium of vehicle, the method comprises: based on the physiological state perception signal of the driver in the intelligent cabin, the physiological steady state index of driver is determined, wherein, physiological steady state index is used to characterize the physiological state of driver;In response to physiological steady state index greater than index threshold, the operating state parameter of vehicle is acquired;In response to operating state parameter characterizes that vehicle is in safe parking state, control strategy matched with physiological steady state index is determined, wherein, intelligent cabin includes multi-modal environment executor, control strategy is used to represent the rule that multi-modal environment executor cooperatively constructs intelligent cabin environment matched with physiological state;According to control strategy, multi-modal environment executor cooperatively constructs intelligent cabin environment.The technical problem that the intelligent cabin environment of vehicle cannot match the physiological state of user is solved.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a smart cockpit control method, device, vehicle, and readable storage medium for a vehicle. Background Technology

[0002] With the rapid development of intelligent connected vehicle technology, the functional positioning of vehicle cabins is evolving from a simple means of transportation to a "third living space," and users' demand for physical and mental health management and emotional regulation in the cabin is increasing day by day.

[0003] In related technologies, the main reliance is on users manually triggering preset meditation audio or scene modes through touch screen or voice commands on the central control screen to assist users in adjusting their physical and mental state and relieving stress. However, this method has significant passive and static defects. Once a scene is selected, it cannot be adjusted adaptively, resulting in a rigid experience for users' physical and mental adjustment, making it difficult to achieve the expected effect. There is a technical problem that the vehicle's intelligent cockpit environment cannot match the user's physiological state, thus failing to provide users with an immersive relaxation experience.

[0004] There is currently no good solution to the above-mentioned technical problems. Summary of the Invention

[0005] This application provides a smart cockpit control method, device, vehicle, and readable storage medium for a vehicle, so as to at least solve the technical problem that the adjustment mode of the smart cockpit environment of the vehicle cannot match the user's physiological state.

[0006] According to one aspect of the embodiments of this application, a smart cockpit control method for a vehicle is provided. The method includes: determining a physiological homeostasis index of the driver / passenger based on a physiological state perception signal of the driver / passenger within the smart cockpit, wherein the physiological homeostasis index is used to characterize the physiological state of the driver / passenger; acquiring vehicle operating state parameters in response to the physiological homeostasis index being greater than an index threshold; determining a control strategy matching the physiological homeostasis index in response to the operating state parameters indicating that the vehicle is in a safe parking state, wherein the smart cockpit includes multimodal environmental actuators, and the control strategy is used to characterize the rules for the multimodal environmental actuators to collaboratively construct a smart cockpit environment matching the physiological state; and controlling the multimodal environmental actuators to collaboratively construct the smart cockpit environment according to the control strategy.

[0007] Furthermore, based on the physiological state perception signals of the driver and passengers in the intelligent cockpit, the physiological homeostasis index of the driver and passengers is determined, including: extracting multimodal physiological features from the physiological state perception signals, wherein the multimodal physiological features are used to characterize the autonomic nervous regulation state of the driver and passengers from multiple dimensions respectively; and determining the physiological homeostasis index of the driver and passengers based on the multimodal physiological features.

[0008] Furthermore, based on multimodal physiological characteristics, the physiological homeostasis index of the driver and passenger is determined, including: inputting the multimodal physiological characteristics into a physiological state assessment model for mapping to obtain the physiological homeostasis index of the driver and passenger, wherein the physiological state assessment model is used to characterize the mapping relationship between multimodal physiological characteristics and physiological homeostasis index.

[0009] Further, determining the control strategy that matches the physiological homeostasis index includes: determining the index interval to which the physiological homeostasis index belongs; retrieving the control strategy that matches the index interval from the control strategy library, wherein the control strategy library stores different control strategies corresponding to different index intervals.

[0010] Furthermore, the control strategy includes environmental parameters of the intelligent cockpit. According to the control strategy, the multimodal environmental actuators are controlled to collaboratively construct the intelligent cockpit environment, including: generating control commands according to the environmental parameters; and responding to the control commands to control the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment.

[0011] Furthermore, after controlling the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment, the method further includes: determining the current physiological homeostasis index of the driver / passenger based on the real-time physiological state perception signal of the driver / passenger, wherein the current physiological homeostasis index is used to characterize the physiological state of the driver / passenger at the current moment; adjusting the control strategy based on the current physiological homeostasis index and the physiological change curve in the control strategy to obtain the adjusted control strategy, wherein the physiological change curve is used to characterize the ideal change trend of the driver / passenger's physiological state within a preset period; and controlling the multimodal environmental actuators to collaboratively adjust the intelligent cockpit environment according to the adjusted control strategy, wherein the adjusted intelligent cockpit environment matches the current physiological homeostasis index.

[0012] Furthermore, based on the current physiological homeostasis index and the physiological change curve in the control strategy, the control strategy is adjusted to obtain the adjusted control strategy, including: determining a preset physiological homeostasis index corresponding to the current physiological homeostasis index from the physiological change curve, wherein the preset physiological homeostasis index is used to characterize the ideal physiological state of the driver / passenger at the current moment; determining the deviation value between the current physiological homeostasis index and the preset physiological homeostasis index, wherein the deviation value is used to characterize the degree of deviation between the actual physiological state and the ideal physiological state of the driver / passenger; and adjusting the control strategy based on the deviation value to obtain the adjusted control strategy.

[0013] Furthermore, the method also includes: in response to the current physiological homeostasis index being less than the index threshold, controlling the multimodal environment actuator to switch the intelligent cockpit environment to a safety baseline state and generating an evaluation report, wherein the safety baseline state is used to characterize that the intelligent cockpit environment meets the vehicle's safe driving specifications, and the evaluation report is used to characterize the correlation between the regulatory behavior of the intelligent cockpit environment and the physiological state change trend of the driver and passengers.

[0014] Furthermore, the operating status parameters include at least the gear position, charging connection status, and door status. After obtaining the vehicle's operating status parameters, the method further includes: determining that the vehicle is in a safe parking state in response to the gear position being in the parking gear and the door status being locked, or determining that the vehicle is in a safe parking state in response to the charging connection status being connected and the door status being locked.

[0015] According to another aspect of the embodiments of this application, a smart cockpit control device for a vehicle is also provided. The device includes: a first determining unit, configured to determine a physiological homeostasis index of a driver / passenger based on a physiological state perception signal of the driver / passenger within the smart cockpit, wherein the physiological homeostasis index is used to characterize the physiological state of the driver / passenger; an acquiring unit, configured to acquire vehicle operating state parameters in response to the physiological homeostasis index being greater than an index threshold; a second determining unit, configured to determine a control strategy matching the physiological homeostasis index in response to the operating state parameters indicating that the vehicle is in a safe parking state, wherein the smart cockpit includes multimodal environmental actuators, and the control strategy is used to characterize the rules for the multimodal environmental actuators to collaboratively construct a smart cockpit environment matching the physiological state; and a control unit, configured to control the multimodal environmental actuators to collaboratively construct the smart cockpit environment according to the control strategy.

[0016] 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.

[0017] According to another aspect of the embodiments of this application, an electronic device 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.

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] In this embodiment, based on the physiological state perception signal of the driver and passenger in the smart cockpit, a physiological homeostasis index of the driver and passenger is determined, wherein the physiological homeostasis index is used to characterize the physiological state of the driver and passenger; in response to the physiological homeostasis index being greater than an index threshold, vehicle operating state parameters are acquired; in response to the operating state parameters indicating that the vehicle is in a safe parking state, a control strategy matching the physiological homeostasis index is determined, wherein the smart cockpit includes multimodal environmental actuators, and the control strategy is used to characterize the rules for the multimodal environmental actuators to collaboratively construct a smart cockpit environment matching the physiological state; according to the control strategy, the multimodal environmental actuators are controlled to collaboratively construct the smart cockpit environment. In other words, in this embodiment, the physiological homeostasis index, which characterizes the physiological state of the driver and passenger, is determined by the physiological state perception signal. This achieves a shift from passive response to active assessment based on objective physiological data, improving the accuracy and objectivity of the driver and passenger's physiological state identification. Furthermore, when the driver and passenger's physiological homeostasis index is greater than the index threshold, it is determined whether the vehicle is in a safe parking state. The control strategy for the intelligent cockpit environment is determined only when the vehicle is in a safe parking state, achieving the goal of deeply coupling physiological state assessment with vehicle safety scenarios. This effectively avoids potential interference to driving safety caused by environmental adjustments during driving. Finally, based on the control strategy matched with the physiological homeostasis index, a multimodal environmental actuator is collaboratively controlled to construct the intelligent cockpit environment. This achieves dynamic and accurate matching between environmental parameters and the driver and passenger's physiological state. Through the collaborative action of the multimodal environmental actuator, an immersive intelligent cockpit environment is created, significantly improving the accuracy, safety, and user experience of intelligent cockpit health interaction. This solves the technical problem that the adjustment method of the vehicle's intelligent cockpit environment cannot match the user's physiological state. Attached Figure Description

[0023] 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:

[0024] Figure 1 This is a flowchart of a smart cockpit control method for a vehicle according to an embodiment of this application;

[0025] Figure 2 This is a schematic diagram of a vehicle intelligent cockpit control system according to an embodiment of this application;

[0026] Figure 3 This is a flowchart of another intelligent cockpit control method for a vehicle according to an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of a multimodal data fusion and meditation depth assessment model according to an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of a biofeedback closed-loop regulation principle according to an embodiment of this application;

[0029] Figure 6 This is a schematic diagram of a vehicle intelligent cockpit control device according to an embodiment of this application. Detailed Implementation

[0030] 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.

[0031] 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.

[0032] According to an embodiment of this application, an embodiment of a smart cockpit control method for a vehicle 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.

[0033] This embodiment provides a method for controlling a vehicle's intelligent cockpit. Figure 1 This is a flowchart of a vehicle intelligent cockpit control method according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps:

[0034] Step S101: Based on the physiological state perception signal of the driver and passenger in the intelligent cockpit, determine the physiological homeostasis index of the driver and passenger, wherein the physiological homeostasis index is used to characterize the physiological state of the driver and passenger.

[0035] In the technical solution provided in step S101 of this application, the aforementioned physiological state perception signal refers to the raw data stream reflecting changes in the physiological functions of the driver and passenger, collected by non-contact sensors deployed in the vehicle's intelligent cockpit. This physiological state perception signal mainly includes visual signals, radar signals, and acoustic signals. Visual signals primarily include the driver's facial expressions, eye opening and closing, and head posture. Radar signals primarily include the respiratory waveform caused by chest micro-movements and heart rate variability (HRV) signals. Acoustic signals primarily include the airflow spectrum generated by deep breathing. The aforementioned physiological homeostasis index is a quantitative value (e.g., a score of 0-100) generated after algorithmic processing, used to objectively characterize the degree to which the driver's autonomic nervous system converges to the resting baseline, i.e., the comprehensive level of the driver's physiological relaxation state or meditation depth. The higher the physiological homeostasis index, the more deeply relaxed or focused the driver is in a meditative state.

[0036] In this embodiment, a multimodal sensor array integrated in the smart cockpit can be used to collect physiological state perception signals of the driver and passengers. This signal acquisition process is non-contact acquisition, that is, the user does not need to wear any contact sensors or special medical / monitoring equipment, but the physiological state perception signals are automatically acquired through non-contact means (such as vision, radar, sound waves, etc.) without the user's awareness.

[0037] For example, a Driver Monitoring System (DMS) deployed behind the steering wheel or on the roof of the smart cockpit can capture key facial features, eye closure duration, and head movement energy of the driver and passengers in real time. A millimeter-wave radar deployed on the center console can non-contactly monitor chest movements and extract high-precision data on respiratory rate, respiratory depth, and heart rate variability. Simultaneously, a microphone array integrated into the smart cockpit can capture the specific airflow acoustic spectrum characteristics generated by the driver and passengers' natural breathing. Under the coordinated control of the smart cockpit domain controller, these sensors synchronously collect multi-source heterogeneous data from the driver and passengers, forming a continuous sequence of physiological state perception signals.

[0038] Optionally, after acquiring the physiological state perception signal of the driver / passenger, a physiological homeostasis index of the driver / passenger can be determined based on the acquired physiological state perception signal, wherein the physiological homeostasis index is used to characterize the physiological state of the driver / passenger.

[0039] For example, the determination process of this physiological homeostasis index mainly includes three steps: feature extraction, multimodal fusion, and model inference. Firstly, the physiological state perception signals of the driver / passenger can be preprocessed and features extracted: for visual signals, the eye closure index and head micromotion energy are calculated; for radar signals, the HRV time-domain index and respiratory cycle envelope variance are extracted using Fast Fourier Transform (FFT) and peak detection algorithms. The HRV time-domain index can be the root mean square of successive differences (RMSSD), one of the most commonly used time-domain indicators for assessing heart rate variability in drivers / passengers, primarily reflecting the activity intensity of the parasympathetic nervous system (responsible for rest, relaxation, and recovery). For acoustic signals, the power spectral density characteristics of the respiratory sound spectrum can be extracted. Next, the extracted multimodal physiological features are normalized, and the normalized multimodal physiological features are then concatenated into vectors and input into a multi-head Long Short-Term Memory (LSTM) network model based on an attention mechanism to obtain the physiological homeostasis index of the driver / passenger. This LSTM model is trained on a large amount of historical data annotated by professional meditators and is capable of learning the complex mapping relationship between multimodal physiological features and the physiological homeostasis index.

[0040] Optionally, the physiological homeostasis index output by the LSTM model can be a value between 0 and 100. This value not only reflects the user's current level of physiological relaxation, but also dynamically weighs the importance of different modalities (such as HRV and eye state) in judging the physiological relaxation state of the driver and passengers through an attention weighting mechanism, thereby ensuring the accuracy and robustness of the assessment of the physiological state of the driver and passengers.

[0041] In this step, physiological state perception signals are automatically collected without the driver or passenger's awareness through non-contact means, thereby accurately quantifying and evaluating the physiological homeostasis index. This breaks through the limitations of traditional passive response, establishes an active evaluation mechanism based on objective physiological perception data, effectively solves the problem of subjective judgment bias, and significantly improves the accuracy and objectivity of physiological state recognition of the driver or passenger.

[0042] Step S102: In response to the physiological homeostasis index being greater than the index threshold, the vehicle's operating status parameters are obtained.

[0043] In the technical solution provided by step S102 of this application, after determining the physiological homeostasis index of the driver / passenger through step S101, the physiological homeostasis index can be compared with an index threshold to further determine the user's intention to relax. The index threshold is usually set based on the driver / passenger's historical physiological baseline data or standard meditation depth requirements (for example, set to 60 points, indicating that the driver / passenger has entered a preliminary state of relaxation or meditation preparation). When the physiological homeostasis index exceeds the index threshold, it indicates that the driver / passenger has a clear intention to relax or meditate.

[0044] Optionally, after determining that the driver or passenger has a clear intention to relax or meditate, in order to ensure that the adjustment of the smart cockpit environment is carried out in a safe state, the vehicle's motion state parameters can be used to further assess whether the vehicle is in a safe parking state, so as to couple the safety boundary of the smart cockpit's health interaction function with the vehicle's real-time operating status.

[0045] In this embodiment, when the physiological homeostasis index is determined to be greater than the index threshold, the vehicle's operating status parameters can be actively acquired. These operating status parameters are mainly read in real time through the vehicle controller area network (CAN) bus or the vehicle Ethernet interface, and specifically include key data such as vehicle speed, gear (e.g., P, D, R), driving mode (e.g., sport mode, eco mode), door lock status, charging gun connection status, and estimated time of arrival (ETA) of the navigation system.

[0046] Optionally, the purpose of obtaining the vehicle's operating status parameters is to determine the vehicle's safety scenario, that is, to determine whether the vehicle is currently in a safe parking state. If the vehicle speed is 0, the gear is in P, or the vehicle is in DC fast charging mode and the doors are locked, then the vehicle is determined to be in a safe parking state; if the vehicle speed is greater than a preset threshold (e.g., 5 km / h), the gear is in D, N, or R, and the doors are unlocked, then the vehicle is determined to be in a non-parked state.

[0047] In this step, when the physiological homeostasis index of the driver or passenger exceeds a preset threshold, the linkage adjustment of the cabin environment is not directly triggered. Instead, the vehicle's operating status parameters are obtained, and the vehicle is assessed to determine whether it is in a safe parking state. This effectively avoids misoperation in unsafe scenarios and ensures that subsequent intelligent cabin environment adjustments are only performed when the vehicle is stationary and safe.

[0048] Step S103: In response to the operating state parameters indicating that the vehicle is in a safe parking state, a control strategy matching the physiological homeostasis index is determined. The intelligent cockpit includes a multimodal environmental actuator, and the control strategy is used to characterize the rules by which the multimodal environmental actuator collaboratively constructs an intelligent cockpit environment that matches the physiological state.

[0049] In the technical solution provided in step S103 of this application, the aforementioned safe parking state refers to a safe scenario where the vehicle is stationary and no complex operation by the driver is required. When the vehicle is in a safe parking state, it is confirmed that the current environmental adjustments will not distract the driver or affect driving safety, thereby unlocking the deep adjustment permissions of the multimodal environmental actuator.

[0050] In this embodiment, after confirming that the vehicle is in a safe parked state, a control strategy matching the physiological homeostasis index of the driver and passengers can be further determined. This control strategy includes multimodal environmental actuators within the smart cockpit, collaboratively constructing rules for a smart cockpit environment that matches the physiological state of the driver and passengers.

[0051] For example, control strategies can be retrieved from a pre-set control strategy library or dynamically generated based on the physiological homeostasis index of the driver / passenger. Multimodal environmental actuators within the smart cockpit include ambient lighting systems, intelligent fragrance systems, air conditioning systems, seat systems (including tilt, massage, and ventilation), active noise cancellation systems, and multimedia systems. Control strategies include the coordination rules between these multimodal environmental actuators and the corresponding environmental adjustment parameters. For instance, when the physiological homeostasis index indicates that the driver / passenger is in a slightly relaxed state, the control strategy may include adjusting the ambient lighting to a soft, warm tone, releasing a trace amount of cedarwood fragrance, and maintaining the seat at its current angle; while when the physiological homeostasis index indicates that the user is in a deeply relaxed state, the control strategy may include adjusting the seat to zero-gravity mode, switching the fragrance to a high-concentration lavender, operating the active noise cancellation at full power to isolate external noise, and generating ambient lighting effects synchronized with the driver / passenger's breathing frequency. In other words, different physiological homeostasis indices correspond to different control strategies; this is merely an example and does not limit the specific content of the control strategies.

[0052] Optionally, the aforementioned control strategy is not statically fixed but dynamically adaptable. That is, the control strategy can be dynamically adjusted based on the real-time physiological state perception signals of the driver and passengers. For example, non-contact sensors can continuously collect the physiological state perception signals of the driver and passengers, calculate the physiological homeostasis index of the driver and passengers in real time, and adjust the control strategy in real time based on the physiological homeostasis index to match the physiological state of the driver and passengers, creating an immersive and relaxing environment for them.

[0053] In this step, when the vehicle is safely parked, determining a control strategy that matches the physiological homeostasis index of the driver and passengers can provide data adjustment basis for the subsequent multimodal environmental actuators in the smart cockpit to collaboratively construct a smart cockpit environment that matches the physiological state of the driver and passengers. Moreover, this control strategy can be dynamically adjusted according to the real-time physiological state perception signals of the driver and passengers, ensuring that the smart cockpit environment can accurately match the real-time physiological relaxation needs of the driver and passengers, thereby constructing an immersive and personalized smart cockpit environment.

[0054] Step S104: According to the control strategy, control the multimodal environment actuators to collaboratively construct the intelligent cockpit environment.

[0055] In the technical solution provided in step S104 of this application, after the control strategy is determined, the multimodal environment actuators can be controlled to collaboratively construct the intelligent cockpit environment.

[0056] In this embodiment, after obtaining the control strategy, the control strategy can be decomposed into specific control instructions for each independent environmental actuator. For example, for an ambient lighting system, control commands can be generated based on a control strategy to adjust the color, brightness, and dynamic rhythm of the ambient light, as well as generate pulsating light effects synchronized with the user's breathing frequency, or a fading effect that gradually dims as the depth of meditation increases; for an intelligent fragrance system, control commands can be generated based on a control strategy, and then, based on these commands, micro-pumps can be controlled to precisely release specific concentrations and types of fragrances (such as lavender or cedar) to assist in olfactory relaxation guidance; for a seating system, control commands can be generated based on a control strategy to drive the motor to adjust the seat tilt angle (e.g., switch to zero gravity mode), activate the airbag massage program, or adjust ventilation / heating parameters to provide tactile comfort; for an acoustic system, control commands can be generated based on the control commands to actively reduce external road noise and play white noise or binaural beats that match the ambient atmosphere, while the speaking speed and volume of the guiding voice are also dynamically adjusted according to the control strategy.

[0057] Optionally, the above adjustments are not one-time adjustments, but a continuous adjustment process. For example, the multimodal environmental actuators can be continuously adjusted based on the real-time physiological state perception signals of the driver and passengers, i.e., real-time physiological feedback signals, to enhance the user's relaxation experience in the intelligent cockpit environment. For instance, if the real-time physiological state perception signals indicate that the driver and passengers have entered a deep relaxation state, the ambient lighting will simultaneously undergo a slow, gradual change in brightness, the fragrance will release a subtle, refreshing scent, and the seat massage waveform will be adjusted to a soothing low-frequency mode, thereby creating multi-dimensional synergistic stimulation in terms of vision, smell, touch, and hearing, enhancing the user's relaxation experience.

[0058] In this step, through the collaboration of multimodal environmental actuators in the intelligent cockpit environment, the internal physiological state represented by the physiological homeostasis index of the driver and passengers can be successfully externalized into external, perceptible changes in the cockpit environment, realizing closed-loop control from "physiological perception" to "environmental intervention", and ultimately achieving the health interaction goal of relieving stress, improving focus or aiding sleep.

[0059] In steps S101 to S104 of this application, the physiological homeostasis index, which characterizes the physiological state of the driver / passenger, is determined through physiological state perception signals. This achieves a shift from passive response to active assessment based on objective physiological data, improving the accuracy and objectivity of physiological state identification. Furthermore, when the physiological homeostasis index of the driver / passenger is greater than the index threshold, it is determined whether the vehicle is in a safe parking state. The control strategy for the intelligent cockpit environment is determined only when the vehicle is in a safe parking state, achieving deep coupling between physiological state assessment and vehicle safety scenarios. This effectively avoids potential interference to driving safety caused by environmental adjustments during driving. Finally, based on the control strategy matched with the physiological homeostasis index, a multimodal environmental actuator is collaboratively controlled to construct the intelligent cockpit environment. This achieves dynamic and accurate matching between environmental parameters and the physiological state of the driver / passenger. Through the collaborative action of the multimodal environmental actuator, an immersive intelligent cockpit environment is created, significantly improving the accuracy, safety, and user experience of intelligent cockpit health interaction. This solves the technical problem that the adjustment method of the vehicle's intelligent cockpit environment cannot match the user's physiological state.

[0060] The intelligent cockpit control method for the vehicle in this application will be further described below.

[0061] As an optional implementation, step S101, based on the physiological state perception signal of the driver and passenger in the intelligent cockpit, determines the physiological homeostasis index of the driver and passenger, including: extracting multimodal physiological features from the physiological state perception signal, wherein the multimodal physiological features are used to characterize the autonomic nervous regulation state of the driver and passenger from multiple dimensions respectively; and determining the physiological homeostasis index of the driver and passenger based on the multimodal physiological features.

[0062] In this embodiment, the aforementioned multimodal physiological features refer to quantitative indicators extracted from the original physiological state perception signals, used to reflect the autonomic nervous system regulation state of the driver / passenger from multiple dimensions. For example, these multimodal physiological features include: HRV time-domain indicators (e.g., RMSSD) from millimeter-wave radar perception signals, which mainly reflect the parasympathetic activity of the driver / passenger; a higher value indicates a better degree of relaxation; eye closure index and head motion energy from DMS camera perception signals, the former representing the degree of closure of visual input and the latter reflecting the degree of body stillness, both jointly characterizing the relaxed state of external behavior; and respiratory synchronicity coefficient from microphone array perception signals, reflecting the stability of the respiratory rhythm. These features characterize the autonomic nervous system regulation state of the driver / passenger from physiological to behavioral dimensions, including cardiovascular regulation, behavioral posture, and respiratory rhythm.

[0063] Optionally, the process of determining the physiological homeostasis index of a driver or passenger based on the physiological state perception signals within the smart cockpit can be divided into a feature extraction stage and an index determination stage. In the feature extraction stage, the collected physiological state perception signals can be preprocessed and cleaned. For example, for millimeter-wave radar perception signals, respiratory waveforms and heartbeat signals can be separated from chest cavity micro-motion signals using Fast Fourier Transform (FFT) and peak detection algorithms, and then the HRV time-domain index can be calculated; for camera perception signals, computer vision algorithms can be used to identify facial key points and calculate eye closure and head motion energy; for microphone array perception signals, the power spectral density of the respiratory sound spectrum can be extracted to analyze the stability of the respiratory frequency. All extracted feature data are then normalized to eliminate individual differences and dimensional influences, forming unified multimodal physiological characteristics.

[0064] Optionally, during the index determination stage, the physiological homeostasis index of the driver / passenger can be further determined based on multimodal physiological characteristics.

[0065] The following section will provide further details on the determination phase of this index.

[0066] As an optional implementation method, the physiological homeostasis index of the driver / passenger is determined based on multimodal physiological characteristics, including: inputting the multimodal physiological characteristics into a physiological state assessment model for mapping to obtain the physiological homeostasis index of the driver / passenger, wherein the physiological state assessment model is used to characterize the mapping relationship between multimodal physiological characteristics and physiological homeostasis index.

[0067] In this embodiment, a pre-established physiological state assessment model is constructed. This model is a pre-trained deep learning model, such as a multi-head LSTM neural network model based on an attention mechanism. Trained on extensive historical data annotated by professional meditators, this model has learned and solidified the mapping relationship between "multimodal physiological characteristics" (e.g., HRV index, eye closure, respiratory synchrony, etc.) and "physiological homeostasis indices" (e.g., a quantitative score of 0-100). This model is not a simple linear weighted calculation but rather captures the dynamic correlations between different physiological characteristics and their changes over time, thus more accurately reflecting the true regulatory state of the autonomic nervous system.

[0068] Optionally, when determining the physiological homeostasis index of a driver / passenger based on multimodal physiological characteristics, the determined multimodal physiological characteristics can be input into the physiological state assessment model. This physiological state assessment model, through its gating mechanism, can effectively capture the dependencies and dynamic trends among multimodal physiological characteristics, and then output a physiological homeostasis index matching those multimodal physiological characteristics. This index can be a state score of 1-100, objectively representing the driver / passenger's current level of physiological relaxation or depth of meditation, providing accurate data support for subsequent judgments on whether trigger thresholds (e.g., exceeding the index threshold) are met and for formulating corresponding environmental control strategies.

[0069] In the above steps, by extracting multimodal physiological features from multimodal physiological perception signals, the system can comprehensively consider the physiological state of the driver / passenger from different dimensions of autonomic nervous regulation (such as stress level represented by heart rate variability, relaxation level represented by respiratory rhythm, and attention state represented by visual features). This multi-source information fusion mechanism effectively overcomes the shortcomings of single physiological indicators being susceptible to noise interference or having limited representational capabilities, and significantly improves the robustness, comprehensiveness, and accuracy of physiological homeostasis index calculation.

[0070] As an optional implementation, step S103, determining a control strategy that matches the physiological homeostasis index, includes: determining the index interval to which the physiological homeostasis index belongs; retrieving a control strategy that matches the index interval from the control strategy library, wherein the control strategy library stores different control strategies corresponding to different index intervals.

[0071] In this embodiment, when determining a control strategy that matches the physiological homeostasis index, a matching control strategy can be quickly retrieved from a preset control strategy library based on the magnitude of the physiological homeostasis index through discretized interval mapping.

[0072] Optionally, the physiological homeostasis index (0-100) is divided into multiple index ranges with different semantics. For example, a low index range (e.g., 0-40 points) may correspond to "routine rest" or "not in a meditative state," where complex environmental adjustments are not initiated or only basic comfort is maintained; a medium index range (e.g., 41-80 points) may correspond to "mild relaxation" or "stress reduction preparation," where basic ambient lighting and gentle fragrance release can be initiated; and a high index range (e.g., 81-100 points) may correspond to "deep meditation" or "deep recovery," where the highest level of immersive experience is unlocked. Based on this, after calculating the physiological homeostasis index of the driver / passenger, the index range to which the physiological homeostasis index belongs can be further determined to ascertain the current physiological state of the driver / passenger.

[0073] Optionally, after determining the index range to which the physiological homeostasis index of the passenger belongs, a control strategy matching the index range can be retrieved from the control strategy library. This control strategy library is a data structure pre-stored in the intelligent cockpit domain controller, where a complete control strategy (i.e., multimodal environmental actuator coordination rules) is bound to each preset index range. This control strategy defines in detail how the various environmental actuators within the intelligent cockpit should cooperate when the passenger is in that physiological state to construct an intelligent cockpit environment that matches the passenger's physiological state.

[0074] For example, if the physiological homeostasis index belongs to the high index range, the control strategy corresponding to the high index range may include specific adjustment rules such as: "adjusting the seat to zero gravity mode (angle X degrees, lumbar support inflation Y%), releasing high concentration of lavender from the fragrance system (concentration Z%), operating the active noise cancellation at full power, gradually darkening the ambient light with a low-frequency rhythm synchronized with the breathing frequency, and reducing the guided speech speed by 10%". This is only an example and does not limit the specific adjustment rules.

[0075] In the above steps, the one-to-one mapping mechanism of "exponential interval - control strategy" can avoid complex real-time calculations and ensure that when the vehicle is in a safe parking state and the user's physiological state meets the requirements, the corresponding immersive environment configuration can be loaded quickly and accurately, so as to achieve a seamless connection from physiological perception to environmental execution.

[0076] As an optional implementation, step S104, the control strategy includes the environmental parameters of the intelligent cockpit, and according to the control strategy, controlling the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment includes: generating control commands according to the environmental parameters; and responding to the control commands, controlling the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment.

[0077] In this embodiment, after the control strategy is determined, the abstract control strategy can be transformed into the execution stage of specific physical actions. Its core lies in generating precise control commands to drive various hardware actuators in the smart cockpit to work together, thereby creating an immersive environment that matches the physiological state of the driver and passengers.

[0078] Optionally, the control strategy includes environmental parameters of the smart cockpit, which are used to guide the actions to be performed by the multimodal environmental actuators in the smart cockpit. For example, the environmental parameters may include, but are not limited to, the color temperature, brightness and pulsation frequency of the ambient light, the release concentration and fragrance combination of the fragrance system, the air outlet temperature and wind speed of the air conditioning system, the tilt angle and massage waveform of the seat, and the gain value of the active noise cancellation system, etc., which are not specifically limited here.

[0079] Optionally, during the generation of control commands, the aforementioned environmental parameters can be parsed and converted into digital signals or communication protocol commands recognizable by each environmental actuator. For example, for an intelligent fragrance system, commands for micropump activation pulse width and flow rate can be generated to precisely control the release of a specific fragrance (such as cedar or lavender); for an ambient lighting system, commands for pulse width modulation duty cycle can be generated to achieve accurate color reproduction of the light and a gradual change in brightness synchronized with the user's breathing frequency; for a seat system, motor drive signals can be generated to control the seat back angle to adjust to a zero-gravity posture and activate the airbag massage program. These are merely illustrative examples.

[0080] Optionally, after generating control commands, these commands can be synchronously sent to each independent actuator via the vehicle network to drive the various controllers to collaboratively construct an intelligent cockpit environment. This process emphasizes synergy, such as visual synergy, olfactory synergy, and tactile and auditory synergy. Visual synergy can include ambient lighting that can pulsate at a specific rhythm based on control commands, with its frequency possibly synchronized with the beat of the guiding voice or the user's real-time breathing frequency, creating a visual sense of guidance. Olfactory synergy can include a fragrance generator that releases a trace of fragrance as the lights dim or a voice prompt urges the user to take a deep breath, assisting in the activation of the parasympathetic nervous system through the olfactory pathway. Tactile and auditory synergy can include seats providing physical support by adjusting the tilt angle and massage intensity, while an active noise cancellation system suppresses external noise, combined with ambient soundscapes played by the multimedia system (such as white noise or binaural beats), forming a closed and immersive sound field.

[0081] In the above steps, through the synchronous response and collaborative operation of multimodal actuators, the intelligent cockpit can be transformed from an ordinary traffic space into a dynamically adjusted health intervention space. Its environmental conditions (such as light, air, temperature, sound, and touch) follow and respond to the physiological homeostasis index of the driver and passengers in real time, thereby achieving the meditation-assisted effect of deep relaxation or enhanced concentration.

[0082] Optionally, after controlling the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment, the intelligent cockpit environment can also be dynamically adjusted based on the real-time physiological feedback signals of the driver and passengers. This ensures that the intelligent cockpit environment (such as lighting rhythm, sound intensity, and fragrance concentration) maintains a high degree of synchronization and adaptive matching with the real-time physiological state of the driver and passengers, thereby enhancing their immersive experience. The process of dynamically adjusting the intelligent cockpit environment will be further described below.

[0083] As an optional implementation, after controlling the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment, the method further includes: determining the current physiological homeostasis index of the driver / passenger based on the real-time physiological state perception signal of the driver / passenger, wherein the current physiological homeostasis index is used to characterize the physiological state of the driver / passenger at the current moment; adjusting the control strategy based on the current physiological homeostasis index and the physiological change curve in the control strategy to obtain the adjusted control strategy, wherein the physiological change curve is used to characterize the ideal change trend of the driver / passenger's physiological state within a preset period; and controlling the multimodal environmental actuators to collaboratively adjust the intelligent cockpit environment according to the adjusted control strategy, wherein the adjusted intelligent cockpit environment matches the current physiological homeostasis index.

[0084] In this embodiment, the aforementioned real-time physiological state perception signal refers to the dynamic physiological data stream of the driver / passenger continuously collected through non-contact sensors (e.g., millimeter-wave radar, DMS camera, microphone) during the driver / passenger's relaxation process. These real-time physiological state perception signals include the driver / passenger's real-time heart rate variability (HRV), respiratory rate, respiratory depth, eye closure degree, and head posture, reflecting the driver / passenger's immediate physiological fluctuations during relaxation. The aforementioned current physiological homeostasis index refers to a quantitative value (e.g., 0-100) calculated in real-time based on the aforementioned real-time physiological state perception signal using a physiological state assessment model (e.g., an LSTM network based on an attention mechanism). This current physiological homeostasis index characterizes the degree of physiological relaxation or depth of meditation experienced by the driver / passenger at the current moment and serves as a direct input basis for dynamically adjusting the intelligent cockpit environment.

[0085] The aforementioned physiological change curve refers to the trajectory of the target physiological parameters of the driver / passenger in the preset control strategy. This physiological change curve depicts the evolution trend of the physiological indicators (such as HRV, respiratory synchrony) of the driver / passenger under ideal conditions within a preset period (e.g., 10 minutes).

[0086] Optionally, after controlling the multimodal environmental actuators to construct the intelligent cockpit environment according to the initially determined control strategy, the real-time monitoring and feedback phase can begin. For example, this involves continuously collecting real-time physiological state perception signals from the driver and passengers, and recalculating the current physiological homeostasis index at fixed time intervals (e.g., every second or every few seconds). Following this, the control strategy is dynamically adjusted. For example, the current physiological homeostasis index is compared with a physiological change curve. The purpose of this comparison is to assess whether the driver's / passengers' current actual physiological state deviates from the ideal trajectory, and the direction and extent of this deviation.

[0087] Optionally, if the current physiological homeostasis index of the driver / passenger is lower than the corresponding preset physiological homeostasis index in the physiological change curve, it indicates that the driver / passenger has not entered the expected relaxation depth; if the current physiological homeostasis index of the driver / passenger is lower than or higher than the corresponding preset physiological homeostasis index in the physiological change curve, it indicates that the driver / passenger has adapted well. Subsequently, based on this comparison result, the control strategy can be adjusted to obtain the adjusted control strategy. Following the adjusted control strategy, the multimodal environment actuators are controlled to collaboratively adjust the intelligent cockpit environment, wherein the adjusted intelligent cockpit environment matches the current physiological homeostasis index.

[0088] For example, the logic behind adjusting the control strategy is to narrow the gap between the driver's / passenger's actual physiological state and their ideal physiological state. For instance, if a discrepancy is detected between the driver / passenger's breathing rhythm and their physiological change curve, the speaking speed of the guiding voice or the rhythm of the background sound can be fine-tuned to synchronize the driver / passenger. If a slow increase in the driver / passenger's HRV is detected, the frequency of the ambient lighting's soothing pulses or the concentration of fragrance can be increased to aid relaxation through multi-sensory stimulation. If the driver / passenger has entered a state of deep relaxation, the volume of the guiding voice can be gradually reduced, or even only ambient sound can be retained to avoid excessive interference. This is merely an illustrative example.

[0089] In the above steps, the physiological state signals of the driver and passengers are monitored in real time, and their current physiological homeostasis index is calculated. This index is then used to adjust the control strategy, constructing a dynamically evolving intelligent cockpit environment that matches the current physiological homeostasis index. This process forms a real-time closed loop of "perception-evaluation-adjustment-re-perception," ensuring that the intelligent cockpit environment always meets the physiological needs of the driver and passengers, thereby maximizing their relaxation effect.

[0090] As an optional implementation, the control strategy is adjusted based on the current physiological homeostasis index and the physiological change curve in the control strategy to obtain an adjusted control strategy. This includes: determining a preset physiological homeostasis index corresponding to the current physiological homeostasis index from the physiological change curve, wherein the preset physiological homeostasis index is used to characterize the ideal physiological state of the driver / passenger at the current moment; determining the deviation value between the current physiological homeostasis index and the preset physiological homeostasis index, wherein the deviation value is used to characterize the degree of deviation between the actual physiological state and the ideal physiological state of the driver / passenger; and adjusting the control strategy based on the deviation value to obtain the adjusted control strategy.

[0091] In this embodiment, the control strategy is adjusted based on the current physiological homeostasis index and the physiological change curve in the control strategy. To obtain the adjusted control strategy, an ideal reference point corresponding to the current physiological homeostasis index can first be determined from the physiological change curve. For example, the current moment can be used as an index to find the corresponding time node from a preset physiological change curve, thereby obtaining the preset physiological homeostasis index corresponding to that moment. This preset physiological homeostasis index represents the expected level of physiological relaxation that the driver / passenger should achieve at the current moment.

[0092] Optionally, after determining the preset physiological homeostasis index, the current physiological homeostasis index (i.e., the actual level of relaxation achieved by the user) calculated in real time can be compared with the preset physiological homeostasis index (i.e., the ideal level of relaxation expected by the user) obtained above, and the difference between the two, i.e., the deviation value, can be calculated. This deviation value has both directional and magnitude meaning. For example, if the current physiological homeostasis index < the preset physiological homeostasis index, the deviation value is negative, indicating that the actual level of relaxation of the driver / passenger is lower than expected, and the driver / passenger may be tense, distracted, or not yet in a state of relaxation; if the current physiological homeostasis index > the preset physiological homeostasis index, the deviation value is positive, indicating that the actual level of relaxation of the driver / passenger is higher than expected, and the driver / passenger may have entered a state of deep relaxation or meditation; if the current physiological homeostasis index ≈ the preset physiological homeostasis index, the deviation value is close to zero, indicating that the actual level of physiological relaxation of the driver / passenger matches the ideal level of relaxation.

[0093] Optionally, after identifying the deviation, the control strategy can be adjusted based on the deviation value. For example, based on the magnitude and direction of the deviation, a reinforcement learning model or a preset rule base can be invoked to fine-tune the original control strategy, generating an adjusted control strategy. The specific logic of the adjustment follows the principle of correction. For example, when the deviation value is negative (the driver / passenger's relaxation state is not up to standard), the guidance intensity tends to be increased. For example, the guidance speech speed may be increased to prompt deep breathing, the pulsation of ambient lighting may be increased to attract attention, or the fragrance concentration may be adjusted to assist in emotional soothing, aiming to push the user's physiological state closer to the ideal curve. When the deviation value is positive (the driver / passenger's relaxation state is good), interference tends to be reduced to protect the deep relaxation state. For example, the volume of the guidance voice may be reduced, the ambient lighting effects may be made smoother and softer, or some strong sensory stimuli may be paused to avoid interrupting the user's deep relaxation experience. When the deviation value is within the allowable error range, the current control strategy is maintained or a very small smooth correction is made.

[0094] In the above steps, by dynamically adjusting the control strategy based on the deviation between the current physiological homeostasis index and the preset physiological change curve, the intelligent cockpit environment achieves real-time adaptive matching to the user's physiological state. This mechanism constructs an efficient biofeedback closed loop, which can guide the physiological indicators of the driver and passengers to converge quickly to the target relaxed state, effectively solving the technical problems of rigid control strategies, lack of personalized adaptation, and lag in response of existing in-vehicle meditation systems.

[0095] As an optional implementation, the intelligent cockpit control method of the vehicle further includes: in response to the current physiological homeostasis index being less than an index threshold, controlling the multimodal environment actuator to switch the intelligent cockpit environment to a safety baseline state and generating an evaluation report, wherein the safety baseline state is used to characterize that the intelligent cockpit environment meets the vehicle's safe driving specifications, and the evaluation report is used to characterize the correlation between the regulatory behavior of the intelligent cockpit environment and the physiological state change trend of the driver and passengers.

[0096] In this embodiment, as described above, the aforementioned index threshold is a preset critical value (e.g., 60 points) used to distinguish whether the driver or passenger has a need for relaxation. If the current physiological homeostasis index is lower than this index threshold, it means that the driver or passenger has not reached the system's expected level of relaxation or focus; that is, the driver or passenger currently has no need for relaxation. In this case, continuing to maintain a high-intensity immersive smart cockpit environment may not only be ineffective but may also affect the driver or passenger's subsequent activities due to excessive stimulation or sensory interference.

[0097] The aforementioned safety baseline state represents a default or basic environmental configuration mode for the intelligent cockpit. In this state, strong sensory enhancements used to create an immersive, meditative atmosphere (such as dramatic lighting changes, complex fragrance releases, deep seat massages, and strong active noise cancellation) are downgraded or disabled, retaining only basic functions that ensure fundamental comfort and safety (such as moderate air conditioning, normal interior lighting, and minimal background noise suppression). Its core purpose is to ensure the intelligent cockpit environment complies with the vehicle's safe driving standards, avoiding any factors that might distract the driver or cause sensory confusion.

[0098] The aforementioned assessment report is a data-driven summary document that records the entire process of the immersive relaxation session for drivers and passengers. It not only shows the final changes in physiological indicators but also focuses on analyzing the correlation between regulatory behaviors (i.e., which adjustments to lighting, fragrance, sound, etc., were made by the intelligent cockpit system) and trends in physiological state changes (i.e., the changes in the drivers' and passengers' heart rate, HRV, respiration, etc., over time). For example, the assessment report might state: "At time T1, when the intensity of the ambient red light was increased, the drivers' and passengers' HRV increased by 5% within the following minute, indicating that the visual stimulus had a significant relaxing effect on them."

[0099] Optionally, the process can be further described as follows: when the current physiological homeostasis index of the driver or passenger is continuously lower than the index threshold, it is determined that the driver or passenger has failed to effectively enter a relaxed state or that the current physiological state of the driver or passenger is not suitable to continue in the smart cockpit environment. In this case, the safe exit and reset procedure can be triggered immediately.

[0100] For example, multimodal environmental actuators in the smart cockpit can be controlled to quickly and smoothly switch the smart cockpit environment from the current immersive meditation mode back to a safe baseline state. For instance, dynamically changing ambient lighting can be gradually dimmed or turned off, restoring constant, soft basic lighting; fragrance release can be stopped or switched to the default basic fresh mode; seat massage can be stopped or adjusted to simple static support; active noise cancellation and complex background soundscapes can be reduced or turned off, restoring a normal in-vehicle communication sound field; the volume of guiding voice commands can be stopped or reduced until it stops completely to minimize sensory interference for the user. These are merely illustrative examples.

[0101] Optionally, the above switching process ensures the safety of the intelligent cockpit environment, prevents sensory overload or potential distraction risks caused by the driver or passengers not being in the right state, and complies with the vehicle's safe driving standards.

[0102] Optionally, during or after the intelligent cockpit environment switch, an evaluation report can be generated based on all data collected during the session. This report can then be used to perform a correlation analysis between the adjustment commands at each moment (e.g., increased light brightness, slower speech) and the physiological data changes of the driver / passenger at the same moment (e.g., decreased respiratory rate, increased HRV). The final report not only informs the driver / passenger about the effectiveness of the relaxation session but also provides personalized insights (e.g., the user responded well to auditory guidance but weakly to visual guidance), providing data support for future preference settings and feedback data for the system's own model optimization (e.g., reward function adjustment in reinforcement learning).

[0103] As an optional implementation, the operating status parameters include at least gear position, charging connection status, and door status. After obtaining the vehicle's operating status parameters, the vehicle's smart cockpit control method further includes: determining that the vehicle is in a safe parking state in response to the gear position being in parking gear and the door status being locked; or determining that the vehicle is in a safe parking state in response to the charging connection status being connected and the door status being locked.

[0104] In this embodiment, when determining whether a vehicle is in a safe parking state based on its operating status parameters, the acquired operating status parameters can be analyzed to determine whether the vehicle is in a safe parking state.

[0105] Optionally, since the gear position reflects the status of the vehicle's transmission system: the parking gear (usually P gear) indicates that the vehicle's drive wheels are mechanically locked, the vehicle is stationary and cannot be moved by the accelerator; the charging connection status reflects the vehicle's external energy interaction status, where the connected status indicates that the charging gun is plugged into the vehicle's charging port, usually meaning that the vehicle is undergoing DC fast charging or AC slow charging, and the vehicle is in a long-term stationary charging operation; the door status reflects the physical enclosure of the smart cockpit, where the locked status indicates that all doors (including the trunk) are closed and locked, indicating that passengers have entered the cabin and are in a relatively closed, private and stable environment, eliminating the risk of accidental falls or external interference caused by opening the doors.

[0106] Optionally, when the vehicle is in the parking gear and the doors are locked, the vehicle is determined to be in a safe parking state; or, when the vehicle is connected to the charging station and the doors are locked, the vehicle is determined to be in a safe parking state.

[0107] Optionally, when it is confirmed that the vehicle is in a safe parking state, it means that the smart cockpit allows the driver and passengers to have a relaxing experience. Then, the vehicle's control strategy can be further determined, and according to the control strategy, the multimodal environmental actuators in the smart cockpit can be controlled to collaboratively construct the smart cockpit environment to create an immersive and relaxing atmosphere for the driver and passengers.

[0108] Optionally, if the vehicle speed exceeds a speed threshold (e.g., 5 km / h), the smart cockpit will not allow passengers to relax in order to ensure driving safety and prevent any potential interference with the driver's sensory adjustment. Furthermore, if the vehicle speed exceeds the speed threshold (e.g., 5 km / h) and the multimodal environmental actuators in the smart cockpit are operating certain functions to assist the driver in relaxation, these functions must be turned off to ensure driving safety. For example, all environmental adjustments such as ambient lighting, fragrance, and seat massage will be gradually dimmed and turned off within 10 seconds; only the lowest volume (e.g., 40 dB) of voice guidance will be maintained until the user actively ends the session or the vehicle stops.

[0109] Optionally, if the vehicle's navigation system anticipates reaching the destination within a preset timeframe, the smart cockpit can proactively suggest a "refreshing short meditation." For example, it can gently ask via the central control screen or voice assistant, "We're almost there. Would you like a 5-minute refreshing meditation?" to help passengers regain their alertness before the journey ends.

[0110] In the above steps, by constructing a dynamic safety boundary management strategy based on vehicle operating status (such as vehicle speed, gear, and charging status) and travel scenarios (such as navigation ETA), the intelligent cockpit health function achieves unity in safety, initiative, and scenario adaptability: On the one hand, it strictly defines "safe parking" as the entry condition for a relaxing experience, and immediately interrupts unnecessary sensory adjustments and downgrades to retain the least disruptive voice guidance when a driving status is detected (such as vehicle speed > 5km / h), fundamentally eliminating safety hazards during driving; on the other hand, by combining charging scenarios and trip destination prediction, it intelligently triggers personalized meditation suggestions of different durations (such as deep recovery or short meditations for rejuvenation), achieving a seamless connection from passive response to proactive anticipation and from single scenario to full trip coverage under the premise of ensuring absolute driving safety, significantly improving the consistency, comfort, and effectiveness of user experience and health management.

[0111] The above technical solutions of the embodiments of this application will be further illustrated below with reference to preferred embodiments.

[0112] Figure 2 This is a schematic diagram of a vehicle intelligent cockpit control system according to an embodiment of this application, as shown below. Figure 2 As shown, the intelligent cockpit control system 200 of the vehicle includes: a multimodal intent and state perception module 201, a scene fusion and safety boundary management module 202, a personalized meditation guidance content library 203, a multi-dimensional cockpit environment collaborative execution module 204, a biofeedback and dynamic adjustment engine 205, and a seamless interaction and effect reporting module 206.

[0113] The multimodal intent and state awareness module 201 is used to collect the user's facial expressions, eye state, head posture, heart rate, respiratory rate and heart rate variability data through non-contact sensors (such as DMS / OMS cameras, millimeter-wave radar), and combine them with deep breathing feature signals collected by in-cabin sensors to comprehensively determine whether the user has entered a meditation preparation state and the current meditation depth.

[0114] The Scene Fusion and Safety Boundary Management Module 202 is used to acquire vehicle status data (such as vehicle speed, gear, driving mode), environmental data (such as time, geographical location), and user preset schedules in real time, and dynamically define safe scenarios where meditation assistance can be enabled (such as when the vehicle is stationary or charging) and the allowed adjustment intensity.

[0115] The Personalized Meditation Guided Content Library 203 is used to pre-set meditation guided programs with various goals (such as stress reduction, focus, and sleep aid), durations, and styles, including guided voice, background soundscape, visual cue sequences, and target physiological parameter curves.

[0116] The multi-dimensional cockpit environment collaborative execution module 204 is used to control the ambient lighting, fragrance generator, zoned air conditioning, seats (e.g., tilt, massage, ventilation), active noise cancellation system and multimedia system in the cockpit, and collaboratively create an immersive intelligent cockpit environment according to the guided program.

[0117] The Biofeedback and Dynamic Regulation Engine 205 is used to compare the user's real-time physiological data (such as HRV and respiratory synchrony) with the target curve in the guidance program in real time. Through reinforcement learning models, it dynamically adjusts parameters such as guidance speech rate, sound intensity, light rhythm, and fragrance concentration to form a real-time closed loop of "perception-guidance-re-perception".

[0118] The Seamless Interaction and Effect Report Module 206 supports seamless interaction methods such as gestures, eye movements, and whispered voice, and generates a visual report after each session to show the depth of meditation, the trend of physiological indicators, and personalized suggestions.

[0119] The following section will further describe the control method of the vehicle's intelligent cockpit in this application embodiment, using the user's meditation process as an example. Figure 3 This is a flowchart of another intelligent cockpit control method for a vehicle according to an embodiment of this application, such as... Figure 3 As shown, the method includes the following steps.

[0120] Step S301: Multimodal sensing data acquisition.

[0121] In this embodiment, user data is continuously collected through a non-contact sensor array deployed within the cockpit. Specifically, the DMS / OMS camera captures the user's facial expressions, eye status (e.g., duration of eye closure, eye movements), and head posture; millimeter-wave radar collects chest cavity micro-movement signals, extracting respiratory rate, respiratory depth, and heart rate variability (HRV); and an array microphone captures the airflow spectrum of the user's deep breaths. Simultaneously, the system acquires vehicle status data (e.g., vehicle speed, gear position, charging status, etc.) in real time via the CAN bus.

[0122] Step S302: Identify whether the user has entered a meditation preparation state.

[0123] In this embodiment, the collected multimodal data is input into a multi-head LSTM network model based on an attention mechanism. The model performs a fusion analysis of visual features (e.g., facial relaxation, closed eyes) and physiological features (e.g., slower and deeper breathing, HRV changes) to calculate a "meditation intention score". If the score exceeds a preset threshold, the user is determined to have entered a meditation preparation state, and step S303 is executed; otherwise, the process returns to step S301.

[0124] Step S303: Determine that the vehicle's current speed is ≤5 km / h.

[0125] In this embodiment, the safety boundary determination logic is entered. First, the vehicle status is checked. If the detected vehicle speed is greater than 5km / h (or the vehicle is in a non-parked / non-charging state), it is considered an unsafe scenario, and the process jumps to step S304. If the vehicle speed is ≤5km / h (usually corresponding to parking in P gear or crawling at very low speed, and combined with conditions such as door lock and charging connection, it is determined to be a safe parking), then the current situation is determined to be a safe scenario, allowing the entry into the meditation assistance process, and the process jumps to step S305.

[0126] Step S304: Stay in standby / exit safely.

[0127] In this embodiment, if the vehicle is in motion or the safe start conditions are not met, the system remains in standby mode or executes a safe exit procedure. If the meditation function was previously running, the system will gradually turn off all environmental adjustments such as ambient lighting, fragrance, and seat massage within 10 seconds, retaining only the lowest volume (e.g., 40dB) of voice guidance until the user actively ends the function or the vehicle comes to a complete stop, ensuring that driving safety is not disturbed.

[0128] Step S305: Bootloader matching and initialization.

[0129] In this embodiment, based on the user's historical preferences, the current time period, and the identified meditation intentions (e.g., stress reduction, sleep aid, focus), the most suitable guided program is matched from a personalized meditation guidance content library. The program's parameters are initialized, including the guiding voice, background soundscape, visual cue sequence, and a preset "target physiological parameter curve." Simultaneously, user confirmation is obtained through seamless interaction (e.g., screen prompts or voice queries).

[0130] Step S306: Multiple actuators work together to construct the lighting / music / fragrance / seats in the intelligent cockpit environment.

[0131] In this embodiment, instructions are sent to the multi-dimensional cabin environment collaborative execution module. The ambient lighting system sets the initial color temperature and brightness according to the guidance program; the fragrance generator releases the corresponding initial fragrance concentration; the zoned air conditioning adjusts to a comfortable temperature and gentle airflow; the seat adjusts to a preset relaxation posture (e.g., a slight recline in P gear or a zero-gravity posture in charging mode); and the active noise cancellation system activates to block out external noise. All actuators work together to initially create an immersive and relaxing atmosphere.

[0132] Step S307, Real-time biological data acquisition.

[0133] In this embodiment, during meditation, the system continuously and frequently collects the user's real-time physiological data. Millimeter-wave radar outputs the latest heart rate and HRV time-domain metrics (e.g., RMSSD) in real time, visual sensors monitor eye closure and head micro-movements in real time, and acoustic sensors monitor respiratory rhythm. All data is transmitted in real time to the biofeedback and dynamic regulation engine.

[0134] Step S308: Calculate the real-time meditation depth index.

[0135] In this embodiment, the biofeedback and dynamic regulation engine uses a preset meditation state quantification model to input the real-time physiological data collected in step S307 (such as normalized HRV, respiratory wave synchronicity coefficient, eye closure degree, etc.) into the model to calculate the current "meditation depth index" (0-100 points), which is used to quantify the user's current relaxation level.

[0136] Step S309: Adjust the cabin environment parameters.

[0137] In this embodiment, the "real-time meditation depth index" is compared with the "target physiological parameter curve" in the guidance program to calculate the deviation value. Based on a policy gradient reinforcement learning algorithm, the cabin environment parameters are adjusted. For example, if the user's relaxation level is lower than expected, the guiding speech speed may be increased or the light pulsation may be enhanced to guide breathing; if the user is already deeply relaxed, the sound and light stimulation is reduced to avoid interference. Subsequently, the instructions are sent to the actuator module to adjust the light rhythm, sound intensity, fragrance concentration, and guiding speech speed in real time.

[0138] Step S310: Determine whether the session has ended.

[0139] In this embodiment, the conditions for the end of the session are monitored. The end conditions include: the user actively issues an end command through gesture, voice, or eye movement; the guide program finishes playing naturally; or the vehicle status changes (e.g., the vehicle speed exceeds 5 km / h again, triggering a safe exit). If the session has not ended, the process jumps to step S311; if the session is determined to have ended, the process proceeds to step S313.

[0140] Step S311: Adjust the cabin environment parameters.

[0141] In this embodiment, the cabin environment parameters continue to be adjusted based on the user's biofeedback data before the session ends, ensuring a consistent user experience.

[0142] Step S312: Obtain user experience.

[0143] In this embodiment, after adjusting the cabin environment parameters, subjective ratings from users regarding the adjustment can be collected to determine the user experience.

[0144] Step S313: Generate a meditation report.

[0145] In this embodiment, after the session ends and the smart cockpit environment gradually recovers, a visual meditation report can be generated based on all the data collected during the session. The report includes meditation depth score, physiological indicator trends (e.g., heart rate decrease, HRV increase curve), attention duration, etc., and is displayed on the central control screen. Afterwards, steps S307 to S313 can be repeated in a loop.

[0146] In steps S301 to S313 above, through multimodal non-contact perception and reinforcement learning-driven dynamic closed-loop control, a technological leap from passive response to proactive prediction and from single adjustment to multidimensional collaboration is achieved. This not only significantly improves the accuracy of meditation intention recognition and the precision of state assessment, but also creates an immersive environment by linking multidimensional actuators such as lighting, fragrance, seats, and sound fields. Combined with stringent vehicle scenario safety boundary management, it effectively reduces users' stress hormone levels and optimizes the relaxation experience while ensuring driving safety. At the same time, it implements native privacy design and realizes personalized, intelligent, safe and reliable cabin health interaction services.

[0147] Figure 4 This is a schematic diagram of a multimodal data fusion and meditation depth assessment model according to an embodiment of this application, as shown below. Figure 4 As shown, the multimodal data fusion and meditation depth evaluation model includes a multimodal raw data input layer, a feature extraction layer, a temporal-spatial fusion neural network layer, and a quantitative index output layer.

[0148] The multimodal raw data input layer is used to receive raw, multi-source heterogeneous data from various non-contact sensors within the cockpit. This multi-source heterogeneous data includes visual signals (e.g., high-resolution facial video streams and infrared image sequences acquired by DMS / OMS cameras), radar signals (e.g., raw intermediate frequency signals of chest micro-movements acquired by millimeter-wave radar), and audio signals (e.g., ambient audio streams within the cockpit acquired by ceiling array microphones, with particular attention to the respiratory airflow spectrum).

[0149] The feature extraction layer is used to preprocess and decouple the raw data from key features, transforming it into a high-dimensional feature vector. This high-dimensional feature vector may include visual features (e.g., facial movements, eye movements, and head posture), radar features (e.g., respiratory rate, respiratory depth, and HRV), and audio features (e.g., respiratory rhythm).

[0150] The temporal-spatial fusion neural network layer adopts a dual-stream network architecture to simultaneously capture the spatiotemporal dependencies of data and achieves deep fusion through an attention mechanism. This layer includes a temporal branch (using a Long Short-Term Memory (LSTM) network or a Gated Recurrent Unit (GRU) network to specifically process physiological signal sequences with temporal continuity in order to capture the dynamic evolution of the user's physiological state), a spatial branch (using a Convolutional Neural Network (CNN) combined with an Attention module to specifically process visual features with spatial structure (e.g., facial key point distribution), automatically weight the importance of key regions (e.g., eyes and mouth), filter background noise), and an attention fusion layer (introducing a cross-modal attention mechanism to dynamically calculate the correlation weights between visual, radar, and audio features, achieving adaptive alignment and deep fusion of multimodal features, overcoming the limitations of a single modality under changes in illumination, slight shaking, or background noise).

[0151] The quantitative indicator output layer is used to output multi-dimensional psychophysiological quantitative indicators based on the fused multimodal deep features through regression or classification heads. These indicators include meditation readiness, instantaneous focus, relaxation index, and overall meditation depth. Meditation readiness is used to predict the probability of a user entering a meditative state and is used for proactive service triggering. Instantaneous focus reflects the user's current level of concentration. The relaxation index quantifies the user's autonomic nervous relaxation level based on indicators such as HRV and respiratory synchronicity. Overall meditation depth is used as the core output indicator (0-100 points). By combining the above multi-dimensional features, the system can assess the user's current meditative immersion level in real time, providing accurate state feedback for the biofeedback dynamic adjustment engine.

[0152] Figure 5 This is a schematic diagram of a biofeedback closed-loop regulation principle according to an embodiment of this application, as shown below. Figure 5As shown, the target reference curve (i.e., the preset ideal meditation depth or physiological parameter trajectory) and the measured physiological curve (i.e., the user's current actual meditation depth or physiological index, collected and calculated in real time by sensors) are simultaneously input to the comparator. The comparator compares the two in real time, calculates and outputs the control error. This control error quantifies the degree of deviation between the user's current physiological state and the ideal target state, and is the key basis for the system to determine the direction and intensity of adjustment. Subsequently, this control error is input to the intelligent regulator (e.g., an algorithm module based on reinforcement learning or adaptive control). The intelligent regulator generates control commands based on the error signal, combined with the current session state and historical strategies. These control commands then act on the controlled object, namely, to coordinate and fine-tune the multi-dimensional environmental parameters in the cabin (including the rhythm and color temperature of ambient lighting, the concentration of fragrance release, the seat tilt and massage mode, and the fan speed and temperature of the zoned air conditioning) and the guidance content (e.g., the speech rate of voice guidance, the volume and frequency of background soundscape). After the environmental parameters are adjusted, the user's physiology and behavior respond, thereby changing their physical and mental state. The system then collects this new physiological response again through sensors, updates the measured physiological curve, and re-inputs it into the comparator, thus forming a continuous iterative closed loop of "error detection—intelligent decision-making—environmental adjustment—physiological response—real-time feedback." Through this high-frequency real-time adjustment, the system can guide the user's physiological state to gradually converge to the target reference curve, achieving a personalized immersive meditation experience.

[0153] Optionally, the aforementioned biofeedback closed-loop adjustment process can be realized through a biofeedback dynamic adjustment engine in the smart cockpit. This biofeedback dynamic adjustment engine constructs a dynamic closed loop based on a policy gradient reinforcement learning algorithm. It takes the current meditation depth index, the guidance program stage, and the preset target curve as state inputs. Through online learning, it optimizes the strategy to output fine-tuning instructions for the guidance speech rate, background sound volume, and ambient light pulse speed. Its reward function comprehensively considers the current depth index increase, the synchronization of breathing and guidance rhythm, and the user's active interruption penalty, thereby achieving real-time adaptive matching between environmental parameters and the user's physiological state.

[0154] As smart cockpits evolve into "third living spaces," they collect a large amount of biometric data, including facial images, respiratory waveforms, and heart rate variability (HRV), through sensors such as millimeter-wave radar and DMS cameras. This data is sensitive and constitutes personal privacy; improper handling can easily trigger user anxiety about privacy breaches. Existing in-vehicle systems often lack sophisticated privacy protection mechanisms for biometric data, leading to low user trust and hindering the widespread adoption of health-related functions. Therefore, this application establishes a privacy protection framework covering the entire data lifecycle to resolve the conflict between data collection and user privacy rights, ensuring that smart cockpit systems provide personalized services while remaining compliant.

[0155] The following section provides a further introduction to the scenario-based rule base in the embodiments of this application.

[0156] This scenario-based rule base aims to establish a rigorous logical judgment mechanism based on vehicle dynamics and environmental conditions to ensure a balance between vehicle safety and user experience for the meditation assistance function. The rule base primarily comprises three levels of core rules: safety-first rules, in-depth experience rules, and proactive service rules.

[0157] Table 1 is a scenario-based security rule table according to an embodiment of this application.

[0158]

[0159] Optionally, as shown in Table 1, the above scenario rules divide the meditation function into four states: "prohibited / restricted", "restricted", "fully open" and "proactive suggestion" through three key dimensions: vehicle speed threshold, vehicle gear / charging status and navigation ETA. This not only eliminates safety hazards while driving, but also maximizes health benefits in parking scenarios, achieving an organic unity between functional usability and driving safety.

[0160] The privacy protection and data processing mechanisms in the embodiments of this application will be further described below.

[0161] In this embodiment, following the principles of data minimization, localized processing, and end-to-end encryption and anonymity, an implementation path combining hierarchical processing and privacy enhancement technologies is constructed. First, at the data hierarchical and local processing level, the acquired user physiological state perception signals are strictly hierarchically classified. For L1 level raw physiological state perception signals (e.g., raw radar intermediate frequency signals, raw camera video streams), real-time processing is performed only within the secure isolation zone of the vehicle domain controller. After processing, the signals are immediately destroyed in memory and never left the vehicle or are persistently stored. For L2 level desensitized feature vectors (e.g., extracted HRV index, respiratory envelope, eye closure index) and L3 level interaction logs, they are generated only locally. The following table illustrates the mapping relationship between data types, processing methods, and storage strategies for different levels of data.

[0162] Table 2 is a data classification and localization processing strategy mapping table according to an embodiment of this application.

[0163]

[0164] Optionally, at the data anonymization and desensitization level, strict anonymization is performed before data upload. Feature extraction converts the original signal into irreversible feature values. For example, millimeter-wave radar signals are processed using Fast Fourier Transform (FFT) and peak detection algorithms to directly output heart rate values ​​(e.g., "72 bpm"). This process is irreversible, making it impossible to deduce the original radar signal from the heart rate value. Simultaneously, before uploading L2 and L3 data, direct identifiers are stripped to generate a randomized session ID unrelated to the vehicle identification number (VIN) or user account. Precise time and geographic location are also obfuscated; for example, precise time is converted to intervals such as "early morning" or "commuting hours," and geographic location is converted to city-level area codes.

[0165] Optionally, regarding secure transmission and cloud processing, all data to be uploaded is encrypted during transmission to ensure confidentiality and integrity. In the cloud, differential privacy technology (with a privacy budget of ε=0.5) is used to inject Laplace noise into aggregated statistical data to prevent the inference of individual information from statistical results. Simultaneously, a federated learning paradigm is employed, encrypting only the model parameters rather than the raw data, and updating the local model with the global model to ensure that user data is always stored on the local device.

[0166] Finally, in terms of data lifecycle management, vehicle-side storage employs hardware-level encryption, implements strict role-based access control, and grants users the right to view and manage preferences at any time, as well as to delete data completely with one click.

[0167] Through the aforementioned privacy protection and data processing mechanisms, a "privacy-native" intelligent cockpit health system is constructed. Its core objective is to eliminate user concerns about privacy leaks and build user trust in the intelligent cockpit health functions while ensuring the security of user biometric data. Through localized processing and privacy enhancement technologies, it achieves both accurate analysis of multimodal physiological data to provide a personalized meditation experience and ensures minimal exposure and controllability of data in all stages of collection, transmission, storage, and use. This achieves a balance between technological innovation and privacy protection, promoting the sustainable commercial application of intelligent cockpit health services.

[0168] The following section will use two specific application scenarios as examples to further introduce the control method of the intelligent cockpit of the vehicle in this application embodiment.

[0169] For quick stress relief during commutes and charging breaks, when the vehicle is connected to a charging station and the DMS and millimeter-wave radar detect relaxation features such as the driver closing their eyes and taking deep breaths, the system proactively inquires and initiates a 10-minute "Morning Focus" meditation. It creates an initial atmosphere by coordinating seat recline, gentle air conditioning, cedar scent, and enhanced active noise cancellation. During the guidance process, it dynamically adjusts the guiding speech speed and the ambient lighting rhythm synchronized with breathing based on the slow increase in real-time HRV. Finally, it generates an effect report that includes an 85% increase in focus and a 12% decrease in average heart rate.

[0170] For nighttime rest scenarios after long-distance driving, when the vehicle is parked in P gear and the system detects that the user has not gotten out of the car for a long time and is in a relaxed posture, it automatically activates "deep sleep guidance", adjusts the seat to zero gravity mode and switches to lavender fragrance, and plays binaural beat sounds of a specific frequency; the system monitors physiological state in real time, and once it detects that the user's HRV enters the high-frequency range, indicating that they have entered deep relaxation, it gradually turns off the guidance voice and keeps only the ambient sound, and gradually dims the lights until they are off, achieving a seamless deep rest experience.

[0171] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0172] According to an embodiment of this application, an embodiment of a smart cockpit control device for a vehicle is provided. It should be noted that the device can be used to execute the aforementioned smart cockpit control method for a vehicle.

[0173] Figure 6 This is a schematic diagram of a vehicle intelligent cockpit control device according to an embodiment of this application, such as... Figure 6 As shown, the intelligent cockpit control device 600 of the vehicle includes: a first determining unit 601, an acquiring unit 602, a second determining unit 603, and a control unit 604.

[0174] The first determining unit 601 is used to determine the physiological homeostasis index of the driver and passenger based on the physiological state perception signal of the driver and passenger in the intelligent cockpit, wherein the physiological homeostasis index is used to characterize the physiological state of the driver and passenger.

[0175] The acquisition unit 602 is used to acquire the vehicle's operating status parameters in response to the physiological homeostasis index being greater than the exponential threshold.

[0176] The second determining unit 603 is used to determine a control strategy that matches the physiological homeostasis index in response to the operating state parameters indicating that the vehicle is in a safe parking state. The intelligent cockpit includes a multimodal environment actuator, and the control strategy is used to characterize the rules by which the multimodal environment actuator collaboratively constructs an intelligent cockpit environment that matches the physiological state.

[0177] The control unit 604 is used to control the multimodal environment actuators to collaboratively construct the intelligent cockpit environment according to the control strategy.

[0178] Optionally, the first determining unit 601 is further configured to: extract multimodal physiological features from the physiological state perception signal, wherein the multimodal physiological features are used to characterize the autonomic nervous regulation state of the driver / passenger object from multiple dimensions respectively; and determine the physiological homeostasis index of the driver / passenger object based on the multimodal physiological features.

[0179] Optionally, the first determining unit 601 is further configured to: input multimodal physiological characteristics into a physiological state assessment model for mapping to obtain the physiological homeostasis index of the driver / passenger, wherein the physiological state assessment model is used to characterize the mapping relationship between multimodal physiological characteristics and physiological homeostasis index.

[0180] Optionally, the second determining unit 603 is further configured to: determine the index interval to which the physiological homeostasis index belongs; and retrieve a control strategy matching the index interval from the control strategy library, wherein the control strategy library stores different control strategies corresponding to different index intervals.

[0181] Optionally, the control unit 604 is also configured to: generate control commands according to environmental parameters; and, in response to the control commands, control the multimodal environmental actuators to collaboratively construct an intelligent cockpit environment.

[0182] Optionally, the device 600 is further configured to: determine the current physiological homeostasis index of the driver / passenger based on the real-time physiological state perception signal of the driver / passenger, wherein the current physiological homeostasis index is used to characterize the physiological state of the driver / passenger at the current moment; adjust the control strategy based on the current physiological homeostasis index and the physiological change curve in the control strategy to obtain the adjusted control strategy, wherein the physiological change curve is used to characterize the ideal change trend of the driver / passenger's physiological state within a preset period; and control the multimodal environment actuator to coordinately adjust the intelligent cockpit environment according to the adjusted control strategy, wherein the adjusted intelligent cockpit environment matches the current physiological homeostasis index.

[0183] Optionally, the device 600 is further configured to: determine a preset physiological homeostasis index corresponding to the current physiological homeostasis index from the physiological change curve, wherein the preset physiological homeostasis index is used to characterize the ideal physiological state of the driver / passenger at the current moment; determine the deviation value between the current physiological homeostasis index and the preset physiological homeostasis index, wherein the deviation value is used to characterize the degree of deviation between the actual physiological state and the ideal physiological state of the driver / passenger; and adjust the control strategy based on the deviation value to obtain the adjusted control strategy.

[0184] Optionally, the device 600 is further configured to: in response to the current physiological homeostasis index being less than an index threshold, control the multimodal environment actuator to switch the smart cockpit environment to a safety baseline state and generate an evaluation report, wherein the safety baseline state is used to characterize that the smart cockpit environment meets the vehicle's safe driving specifications, and the evaluation report is used to characterize the correlation between the regulatory behavior of the smart cockpit environment and the physiological state change trend of the driver and passengers.

[0185] Optionally, the device 600 is also used to: determine that the vehicle is in a safe parking state in response to the gear being in the parking position, the charging connection being connected, and the door being locked.

[0186] In the intelligent cockpit control device of the vehicle described in this application, a physiological homeostasis index representing the physiological state of the driver and passengers is determined through physiological state perception signals. This realizes a shift from passive response to active assessment based on objective physiological data, improving the accuracy and objectivity of physiological state recognition. Furthermore, when the physiological homeostasis index of the driver and passengers is greater than the index threshold, it determines whether the vehicle is in a safe parking state. The control strategy for the intelligent cockpit environment is determined only when the vehicle is in a safe parking state, achieving deep coupling between physiological state assessment and vehicle safety scenarios. This effectively avoids potential interference to driving safety caused by environmental adjustments during driving. Finally, based on the control strategy matched with the physiological homeostasis index, a multimodal environmental actuator is collaboratively controlled to construct the intelligent cockpit environment. This achieves dynamic and accurate matching between environmental parameters and the physiological state of the driver and passengers. Through the collaborative action of the multimodal environmental actuator, an immersive intelligent cockpit environment is created, significantly improving the accuracy, safety, and user experience of intelligent cockpit health interaction. This solves the technical problem that the adjustment method of the vehicle's intelligent cockpit environment cannot match the user's physiological state.

[0187] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0188] 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.

[0189] 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.

[0190] 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.

[0191] 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.

[0192] 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.

[0193] 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.

[0194] 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.

[0195] 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.

[0196] 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.

[0197] 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 method for controlling an intelligent cockpit in a vehicle, characterized in that, include: Based on the physiological state perception signals of the driver and passenger in the intelligent cockpit, the physiological homeostasis index of the driver and passenger is determined, wherein the physiological homeostasis index is used to characterize the physiological state of the driver and passenger. In response to the physiological homeostasis index being greater than the exponential threshold, the operating status parameters of the vehicle are obtained; In response to the operating state parameters indicating that the vehicle is in a safe parking state, a control strategy matching the physiological homeostasis index is determined, wherein the intelligent cockpit includes a multimodal environmental actuator, and the control strategy is used to characterize the rules by which the multimodal environmental actuator collaboratively constructs an intelligent cockpit environment matching the physiological state; According to the control strategy, the multimodal environmental actuators are controlled to collaboratively construct the intelligent cockpit environment.

2. The method according to claim 1, characterized in that, Based on the physiological state perception signals of the passengers in the intelligent cockpit, the physiological homeostasis index of the passengers is determined, including: Multimodal physiological features are extracted from the physiological state perception signals, wherein the multimodal physiological features are used to characterize the autonomic nervous regulation state of the driver / passenger from multiple dimensions respectively; Based on the multimodal physiological characteristics, the physiological homeostasis index of the driver / passenger is determined.

3. The method according to claim 2, characterized in that, Based on the aforementioned multimodal physiological characteristics, the physiological homeostasis index of the driver / passenger is determined, including: The multimodal physiological features are input into a physiological state assessment model for mapping to obtain the physiological homeostasis index of the driver / passenger. The physiological state assessment model is used to characterize the mapping relationship between the multimodal physiological features and the physiological homeostasis index.

4. The method according to claim 1, characterized in that, Determining a control strategy that matches the physiological homeostasis index includes: Determine the index range to which the physiological homeostasis index belongs; Retrieve a control strategy that matches the index interval from the control strategy library, wherein the control strategy library stores different control strategies corresponding to different index intervals.

5. The method according to claim 1, characterized in that, The control strategy includes the environmental parameters of the intelligent cockpit. According to the control strategy, the multimodal environmental actuators are controlled to collaboratively construct the intelligent cockpit environment, including: Generate control commands based on the environmental parameters; In response to the control command, the multimodal environmental actuators are controlled to collaboratively construct the intelligent cockpit environment.

6. The method according to claim 5, characterized in that, After controlling the multimodal environmental actuators to collaboratively construct the intelligent cockpit environment, the method further includes: Based on the real-time physiological state perception signal of the driver and passenger, the current physiological homeostasis index of the driver and passenger is determined, wherein the current physiological homeostasis index is used to characterize the physiological state of the driver and passenger at the current moment; Based on the current physiological homeostasis index and the physiological change curve in the control strategy, the control strategy is adjusted to obtain the adjusted control strategy, wherein the physiological change curve is used to characterize the ideal change trend of the physiological state of the driver / passenger object within a preset period. According to the adjusted control strategy, the multimodal environment actuator is controlled to coordinately adjust the intelligent cockpit environment, wherein the adjusted intelligent cockpit environment matches the current physiological homeostasis index.

7. The method according to claim 6, characterized in that, Based on the current physiological homeostasis index and the physiological change curve in the control strategy, the control strategy is adjusted to obtain the adjusted control strategy, including: A preset physiological homeostasis index corresponding to the current physiological homeostasis index is determined from the physiological change curve, wherein the preset physiological homeostasis index is used to characterize the ideal physiological state of the driver / passenger at the current moment; Determine the deviation value between the current physiological homeostasis index and the preset physiological homeostasis index, wherein the deviation value is used to characterize the degree of deviation between the actual physiological state and the ideal physiological state of the driver / passenger; Based on the deviation value, the control strategy is adjusted to obtain the adjusted control strategy.

8. The method according to claim 7, characterized in that, The method further includes: In response to the current physiological homeostasis index being less than the index threshold, the multimodal environment actuator is controlled to switch the smart cockpit environment to a safety baseline state and generate an evaluation report. The safety baseline state is used to characterize that the smart cockpit environment meets the vehicle's safe driving specifications, and the evaluation report is used to characterize the correlation between the regulatory behavior of the smart cockpit environment and the physiological state change trend of the driver and passengers.

9. The method according to claim 1, characterized in that, The operating status parameters include at least gear position, charging connection status, and door status. After obtaining the vehicle's operating status parameters, the method further includes: In response to the gear being in the parking position and the door being locked, the vehicle is determined to be in the safe parking state; or, in response to the charging connection being connected and the door being locked, the vehicle is determined to be in the safe parking state.

10. A smart cockpit control device for a vehicle, characterized in that, include: The first determining unit is used to determine the physiological homeostasis index of the driver / passenger based on the physiological state perception signal of the driver / passenger in the intelligent cockpit, wherein the physiological homeostasis index is used to characterize the physiological state of the driver / passenger. The acquisition unit is used to acquire the vehicle's operating status parameters in response to the physiological homeostasis index being greater than the index threshold. The second determining unit is used to determine a control strategy that matches the physiological homeostasis index in response to the operating state parameters indicating that the vehicle is in a safe parking state. The intelligent cockpit includes a multimodal environment actuator, and the control strategy is used to characterize the rules by which the multimodal environment actuator collaboratively constructs an intelligent cockpit environment that matches the physiological state. The control unit is used to control the multimodal environment actuators to collaboratively construct the intelligent cockpit environment according to the control strategy.

11. 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 9.

12. 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 9.

Images