A method and device for personalizing optimization of a passenger vehicle environment for multiple passengers

Abstract

This application provides a method and apparatus for personalized optimization of passenger vehicle environment for multiple passengers. The method includes: acquiring the perception data of each passenger in real time through a multimodal perception device, obtaining the passenger's historical comfort preference model, calculating the passenger's predicted comfort based on the passenger's perception data, determining the optimal execution instruction set that satisfies the passenger's historical comfort preference model based on the predicted comfort, inputting the current number of passengers in the vehicle, target comfort, optimal execution instruction set, predicted comfort, comfort weight, energy efficiency weight, and total power consumption of the vehicle into a lightweight multi-objective genetic solver with target population size and target iteration number for multi-objective optimization, obtaining Pareto control strategy candidate solutions, weighting and scoring the Pareto control strategy candidate solutions based on energy efficiency weight, selecting the target solution of the control strategy, and generating a corresponding control instruction package based on the target solution of the control strategy and sending it to a distributed actuator for execution.

Description

Technical Field

[0001] This application relates to the field of vehicle environmental control technology, and more specifically, to a method and apparatus for personalized optimization of the environment of a multi-passenger passenger vehicle. Background Technology

[0002] The environmental control system for passenger vehicles has evolved from basic temperature regulation to a comprehensive intelligent cockpit management system that integrates temperature and humidity management, air quality optimization, sound comfort, and personalized experience. Its core objective is to create a comfortable, healthy, and energy-efficient driving and riding space that meets the needs of "people."

[0003] Currently, the environmental control systems of mainstream passenger vehicles mainly include air conditioning systems and audio / noise control systems, which are functionally independent. The air conditioning system adopts a centralized air supply structure to control the average temperature and humidity in the vehicle cabin by adjusting parameters such as compressor power, damper opening, and blower speed. In the acoustic system, the in-vehicle audio system mainly plays throughout the cabin, and active noise cancellation (ANC) technology is mostly used to suppress low-frequency global noise such as engine or road noise.

[0004] However, current mainstream passenger vehicle environmental control systems lack cross-domain coordination between thermal and acoustic management, and zonal control is still based on individual zones, resulting in insufficient spatial resolution. Furthermore, most systems rely on manual user settings, leading to weak feedback mechanisms. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a method and apparatus for personalized optimization of passenger vehicle environments for multiple passengers. By acquiring the thermal and acoustic perception data of each passenger and calculating the passenger's predicted comfort, multi-dimensional information is combined to perform multi-objective optimization to obtain candidate solutions for control strategies. Based on the objective solution of the control strategy, corresponding control command packages are generated and sent to distributed actuators for execution. This achieves cross-domain collaboration between thermal management and acoustic management. Zonal control is no longer based on individual areas; it can provide independent microenvironments for passengers of different body types or preferences on the same seat, improving spatial resolution. Furthermore, adaptive learning based on physiological signals or behavioral feedback can enhance the feedback mechanism.

[0006] In a first aspect, embodiments of this application provide a method for personalized optimization of a multi-passenger vehicle environment, the method comprising: The sensory data of each passenger is acquired in real time through a multimodal sensing device, and the passengers are identified to obtain their historical comfort preference model; wherein the sensory data represents thermal and acoustic sensory data. The passenger's predicted comfort level is calculated based on the passenger's perceived data; wherein the range of the predicted comfort level includes a first threshold and a second threshold, each threshold representing the passenger's predicted comfort level; Based on the predicted comfort level, determine the optimal set of execution instructions that satisfies the passenger's historical comfort preference model, and obtain the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located; Based on the current number of passengers in the vehicle, the target comfort level, the optimal execution instruction set, the predicted comfort level, the comfort level weight, the energy efficiency weight, and the total power consumption of the vehicle, a lightweight multi-objective genetic solver with a target population size and a target iteration number is used for multi-objective optimization to obtain candidate solutions for the Pareto control strategy. The candidate solutions of the Pareto control strategy are weighted and scored based on the energy efficiency weights, the target solution of the control strategy is selected, and the corresponding control instruction package is generated based on the target solution of the control strategy and sent to the distributed actuator for execution.

[0007] In one possible implementation, the multimodal sensing device includes at least a non-contact infrared thermal imager, millimeter-wave radar, optional wearable devices, a miniature wind speed sensor, a near-ear microphone, and a light sensor; the real-time acquisition of sensing data for each passenger includes: The target area of ​​the passenger is measured by a non-contact infrared thermal imager with target parameters to obtain the passenger's infrared thermal imaging information, and the passenger's body surface temperature distribution map is output based on the infrared thermal imaging information; wherein, the non-contact infrared thermal imager is installed in the ceiling of the vehicle, facing the head and upper torso area of ​​each seat, and the target area includes the forehead and / or neck; the target parameters include target pixels and target frame rate; The passenger's chest and abdominal movements are detected by a millimeter-wave radar operating at the target frequency to calculate indirect indicators of heat stress; wherein, the indirect indicators of heat stress include at least respiratory rate; the millimeter-wave radar is installed on the seat back or the inside of the B-pillar. The system receives electrocardiogram (ECG) information from optional wearable devices via a preset communication method. The communication method includes at least Bluetooth and UWB, and the optional wearable devices include at least smartwatches and smart bracelets. The ECG information includes at least heart rate and skin conductance. The local airflow velocity of the passenger is collected by the miniature wind speed sensor, the local sound pressure level of the passenger is collected by the near-ear microphone, and the solar radiation intensity of the passenger is collected by the light sensor located on the roof.

[0008] In one possible implementation, identifying the passenger and obtaining the passenger's historical comfort preference model includes: Facial recognition is performed on the passenger to obtain facial recognition information, and the passenger's user identifier is determined based on the facial recognition information; The user identifier is used to match the passenger's preference profile stored in the cloud or locally; wherein each registered passenger has a unique user identifier; the preference profile includes at least temperature range, acoustic preferences, and physiological sensitivity tags.

[0009] In one possible implementation, calculating the passenger's predicted comfort based on the passenger's perceived data includes: For the passenger, an environmental state vector is determined based on the skin temperature, respiratory rate, heart rate, skin conductance, local airflow velocity, local sound pressure level, and solar radiation intensity. The passenger's environmental state vector is input into a preset Gaussian process regression model, and the predicted comfort level of the passenger is output.

[0010] In one possible implementation, obtaining the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle the passenger is in includes: The comfort level selected by the passenger is determined, and the corresponding target comfort level is determined based on the comfort level; wherein, different comfort levels map to different values; The passenger's comfort weight is determined based on the passenger's type, the driver's driving status, or the seat priority mode selected by the passenger. The energy mode of the vehicle is determined, and the corresponding energy efficiency weight is determined based on the energy mode; wherein the energy mode includes at least a normal mode, an energy-saving mode, and a performance mode.

[0011] In one possible implementation, the method further includes: In response to the passenger exhibiting implicit abnormal behavior, the implicit abnormal data of the passenger is obtained; Obtain the explicit feedback data of the passengers, and incrementally update the Gaussian process regression model online based on the explicit feedback data and implicit anomaly data of the passengers; The acquisition of explicit feedback data from the passenger includes: Determine the comfort level selected by the passenger, and determine the corresponding explicit feedback data based on the comfort level.

[0012] In one possible implementation, the distributed actuator includes a distributed microclimate execution array characterizing the thermal management execution layer and a distributed acoustic execution array characterizing the acoustic management execution layer; The distributed microclimate execution array is used to provide each passenger with independently controllable local temperature, humidity and airflow; the distributed microclimate execution array corresponds to the first target installation location, the first target technical specifications and the first target control method, and the distributed microclimate execution array includes at least a directional adjustable air supply nozzle, a micro blower module and a central temperature and humidity control unit; The distributed acoustic execution array is used to simultaneously achieve local active noise cancellation, directional audio playback, and background sound masking within the same carriage; the distributed acoustic execution array corresponds to the installation location of the second target, the technical specifications of the second target, and the control method of the second target; the distributed microclimate execution array includes at least a directional speaker, an active noise cancellation microphone, and a multi-channel audio processing unit.

[0013] Secondly, embodiments of this application also provide a personalized optimization device for a multi-passenger vehicle environment, the device comprising: The first acquisition module is used to acquire the perception data of each passenger in real time through a multimodal perception device, identify the passenger, and obtain the passenger's historical comfort preference model; wherein, the perception data represents thermal and acoustic perception data; A calculation module is used to calculate the passenger's predicted comfort level based on the passenger's perception data; wherein the range of the predicted comfort level includes a first threshold and a second threshold, each threshold representing the passenger's predicted comfort level; The second acquisition module is used to determine the optimal set of execution instructions that satisfies the passenger's historical comfort preference model based on the predicted comfort level, and to acquire the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located. The third acquisition module is used to perform multi-objective optimization by inputting the current number of passengers in the vehicle, the target comfort level, the optimal execution instruction set, the predicted comfort level, the comfort level weight, the energy efficiency weight, and the total power consumption of the vehicle into a lightweight multi-objective genetic solver with a target population size and a target iteration number, so as to obtain candidate solutions for the Pareto control strategy. The execution module is used to perform weighted scoring on the candidate solutions of the Pareto control strategy based on the energy efficiency weight, select the target solution of the control strategy, and generate a corresponding control instruction package based on the target solution of the control strategy and send it to the distributed actuator for execution.

[0014] In one possible implementation, the multimodal sensing device includes at least a non-contact infrared thermal imager, millimeter-wave radar, optional wearable device, miniature wind speed sensor, near-ear microphone, and light sensor; the first acquisition module is specifically used for: The target area of ​​the passenger is measured by a non-contact infrared thermal imager with target parameters to obtain the passenger's infrared thermal imaging information, and the passenger's body surface temperature distribution map is output based on the infrared thermal imaging information; wherein, the non-contact infrared thermal imager is installed in the ceiling of the vehicle, facing the head and upper torso area of ​​each seat, and the target area includes the forehead and / or neck; the target parameters include target pixels and target frame rate; The passenger's chest and abdominal movements are detected by a millimeter-wave radar operating at the target frequency to calculate indirect indicators of heat stress; wherein, the indirect indicators of heat stress include at least respiratory rate; the millimeter-wave radar is installed on the seat back or the inside of the B-pillar. The system receives electrocardiogram (ECG) information from optional wearable devices via a preset communication method. The communication method includes at least Bluetooth and UWB, and the optional wearable devices include at least smartwatches and smart bracelets. The ECG information includes at least heart rate and skin conductance. The local airflow velocity of the passenger is collected by the miniature wind speed sensor, the local sound pressure level of the passenger is collected by the near-ear microphone, and the solar radiation intensity of the passenger is collected by the light sensor located on the roof.

[0015] In one possible implementation, the first acquisition module is specifically used for: Facial recognition is performed on the passenger to obtain facial recognition information, and the passenger's user identifier is determined based on the facial recognition information; The user identifier is used to match the passenger's preference profile stored in the cloud or locally; wherein each registered passenger has a unique user identifier; the preference profile includes at least temperature range, acoustic preferences, and physiological sensitivity tags.

[0016] In one possible implementation, the computing module is specifically used for: For the passenger, an environmental state vector is determined based on the skin temperature, respiratory rate, heart rate, skin conductance, local airflow velocity, local sound pressure level, and solar radiation intensity. The passenger's environmental state vector is input into a preset Gaussian process regression model, and the predicted comfort level of the passenger is output.

[0017] In one possible implementation, the second acquisition module is specifically used for: The comfort level selected by the passenger is determined, and the corresponding target comfort level is determined based on the comfort level; wherein, different comfort levels map to different values; The passenger's comfort weight is determined based on the passenger's type, the driver's driving status, or the seat priority mode selected by the passenger. The energy mode of the vehicle is determined, and the corresponding energy efficiency weight is determined based on the energy mode; wherein the energy mode includes at least a normal mode, an energy-saving mode, and a performance mode.

[0018] In one possible implementation, the device further includes: The fourth acquisition module is used to acquire the implicit abnormal data of the passenger in response to the passenger exhibiting implicit abnormal behavior. The update module is used to obtain the explicit feedback data of the passengers and to incrementally update the Gaussian process regression model online based on the explicit feedback data and implicit anomaly data of the passengers. The update module is specifically used for: Determine the comfort level selected by the passenger, and determine the corresponding explicit feedback data based on the comfort level.

[0019] In one possible implementation, the distributed actuator includes a distributed microclimate execution array characterizing the thermal management execution layer and a distributed acoustic execution array characterizing the acoustic management execution layer; The distributed microclimate execution array is used to provide each passenger with independently controllable local temperature, humidity and airflow; the distributed microclimate execution array corresponds to the first target installation location, the first target technical specifications and the first target control method, and the distributed microclimate execution array includes at least a directional adjustable air supply nozzle, a micro blower module and a central temperature and humidity control unit; The distributed acoustic execution array is used to simultaneously achieve local active noise cancellation, directional audio playback, and background sound masking within the same carriage; the distributed acoustic execution array corresponds to the installation location of the second target, the technical specifications of the second target, and the control method of the second target; the distributed microclimate execution array includes at least a directional speaker, an active noise cancellation microphone, and a multi-channel audio processing unit.

[0020] Thirdly, embodiments of this application provide an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the personalized optimization method for a multi-passenger vehicle environment as described in any of the first aspects.

[0021] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the personalized optimization method for a multi-passenger vehicle environment as described in any one of the first aspects.

[0022] This application provides a method and apparatus for personalized optimization of a passenger vehicle environment for multiple passengers. It acquires real-time perception data of each passenger through a multimodal perception device, identifies passengers, obtains their historical comfort preference models, calculates predicted comfort based on the passenger's perception data, determines the optimal set of execution instructions that satisfies the passenger's historical comfort preference model based on the predicted comfort, obtains the passenger's target comfort, comfort weight, and energy efficiency weight of the vehicle, and inputs the current number of passengers, target comfort, optimal execution instruction set, predicted comfort, comfort weight, energy efficiency weight, and total vehicle power consumption into a lightweight multi-objective genetic solver with target population size and target iteration number for multi-objective optimization. This yields Pareto control strategy candidate solutions, which are then weighted and scored based on the energy efficiency weight. A target solution for the control strategy is selected, and a corresponding control instruction package is generated based on the target solution and sent to a distributed actuator for execution. This application acquires thermal and acoustic perception data for each passenger and calculates their predicted comfort level. It then combines this with multi-dimensional information and multi-objective optimization to obtain candidate solutions for control strategies. Based on the objective solutions of these control strategies, corresponding control command packages are generated and sent to distributed actuators for execution. This achieves cross-domain collaboration between thermal and acoustic management. Zonal control is no longer based on individual areas; it can provide independent microenvironments for passengers of different body types or preferences within the same seat, improving spatial resolution. Furthermore, adaptive learning based on physiological signals or behavioral feedback enhances the feedback mechanism.

[0023] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a method for personalized optimization of a multi-passenger vehicle environment according to an embodiment of this application; Figure 2 This is a schematic diagram of the overall architecture of a personalized optimization method for passenger vehicle environments with multiple passengers. Figure 3 This is a schematic diagram of the structure of a personalized optimization device for a multi-passenger vehicle environment provided according to an embodiment of this application; Figure 4This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0027] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0028] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0029] Considering that passenger vehicle environmental control systems have evolved from basic temperature regulation to comprehensive intelligent cockpit management systems that integrate temperature and humidity management, air quality optimization, sound comfort, and personalized experiences, the core objective is to create a comfortable, healthy, and energy-efficient driving and riding space centered around the needs of "people."

[0030] Currently, the environmental control systems of mainstream passenger vehicles mainly include air conditioning systems and audio / noise control systems, which are functionally independent. The air conditioning system adopts a centralized air supply structure to control the average temperature and humidity in the vehicle cabin by adjusting parameters such as compressor power, damper opening, and blower speed. In the acoustic system, the in-vehicle audio system mainly plays throughout the cabin, and active noise cancellation (ANC) technology is mostly used to suppress low-frequency global noise such as engine or road noise.

[0031] However, current mainstream passenger vehicle environmental control systems lack cross-domain coordination between thermal and acoustic management, and zonal control is still based on individual zones, resulting in insufficient spatial resolution. Furthermore, most systems rely on manual user settings, leading to weak feedback mechanisms.

[0032] To address this issue, this application provides a method and apparatus for personalized optimization of passenger vehicle environments for multiple passengers. By acquiring thermal and acoustic perception data for each passenger and calculating their predicted comfort, multi-dimensional information is combined with multi-objective optimization to obtain candidate solutions for control strategies. Based on the objective solution of the control strategy, corresponding control command packages are generated and sent to distributed actuators for execution. This achieves cross-domain collaboration between thermal and acoustic management. Zonal control is no longer based on individual areas; it can provide independent microenvironments for passengers of different body types or preferences on the same seat, improving spatial resolution. Furthermore, adaptive learning based on physiological signals or behavioral feedback enhances the feedback mechanism.

[0033] Figure 1 This is a flowchart of a method for personalized optimization of a multi-passenger vehicle environment according to an embodiment of this application. Figure 1 As shown in the embodiments of this application, the personalized optimization method for multi-passenger passenger vehicle environments may specifically include: S101. Acquire the perception data of each passenger in real time through a multimodal perception device, identify the passengers, and obtain the passengers' historical comfort preference model.

[0034] S102. Calculate the predicted comfort level of passengers based on their perception data.

[0035] S103. Based on the predicted comfort level, determine the optimal set of execution instructions to meet the historical comfort preference model of passengers, and obtain the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located.

[0036] S104. Based on the current number of passengers in the vehicle, the target comfort level, the optimal execution instruction set, the predicted comfort level, the comfort level weight, the energy efficiency weight, and the total power consumption of the vehicle, a lightweight multi-objective genetic solver with target population size and target iteration number is used to perform multi-objective optimization and obtain candidate solutions for the Pareto control strategy.

[0037] S105. Based on the energy efficiency weight, the candidate solutions of the Pareto control strategy are weighted and scored, the target solution of the control strategy is selected, and the corresponding control instruction package is generated based on the target solution of the control strategy and sent to the distributed actuator for execution.

[0038] The aforementioned personalized optimization method for multi-passenger vehicle environments acquires thermal and acoustic perception data for each passenger and calculates their predicted comfort. It then combines multi-dimensional information with multi-objective optimization to obtain candidate solutions for control strategies. Based on the objective solution of the control strategy, corresponding control command packages are generated and sent to distributed actuators for execution. This achieves cross-domain collaboration between thermal and acoustic management. Zonal control is no longer based on individual areas; it can provide independent microenvironments for passengers of different body types or preferences on the same seat, improving spatial resolution. Furthermore, adaptive learning based on physiological signals or behavioral feedback enhances the feedback mechanism.

[0039] The exemplary steps described above in the embodiments of this application are illustrated below with specific examples: S101 acquires the perception data of each passenger in real time through a multimodal perception device, identifies the passengers, and obtains the passengers' historical comfort preference model.

[0040] In this embodiment, the multimodal sensing device includes at least a non-contact infrared thermal imager, millimeter-wave radar, optional wearable devices, a miniature wind speed sensor, a near-ear microphone, and a light sensor. The sensing data characterizes thermal and acoustic sensing data, and includes at least physiological state data. The multimodal sensing device acquires the sensing data of each passenger in real time, identifies the passenger, and obtains a historical comfort preference model for subsequent processing. For example, such as... Figure 2 As shown, passenger perception data is obtained through a multimodal perception layer.

[0041] It can be added here that TOF sensors can be used to replace millimeter-wave radar, and this application uses millimeter-wave radar as an example for description.

[0042] Optionally, while acquiring the perception data of each passenger in real time, the system measures the target area of ​​the passenger using a non-contact infrared thermal imager with target parameters to obtain the passenger's infrared thermal imaging information and outputs the passenger's body surface temperature distribution map based on the infrared thermal imaging information; it detects the passenger's chest and abdominal fluctuations using a millimeter-wave radar with the target operating frequency to calculate the passenger's indirect thermal stress index; it receives electrocardiogram information from optional wearable devices through a preset communication method; it collects the passenger's local airflow speed through a miniature wind speed sensor, collects the passenger's local sound pressure level through a near-ear microphone, and collects the passenger's solar radiation intensity through a light sensor located on the roof.

[0043] The non-contact infrared thermal imager is installed in the vehicle's ceiling, facing the head and upper torso area of ​​each seat, with the target area including the forehead and / or neck; target parameters include target pixels and target frame rate; the body surface temperature distribution map (i.e., skin temperature) characterizes the estimation of the passenger's local thermal sensation; indirect indicators of heat stress include at least respiratory rate; millimeter-wave radar is installed on the seat back or the inside of the B-pillar; communication methods include at least Bluetooth and UWB, and optional wearable devices include at least smartwatches and smart bracelets; electrocardiogram information includes at least heart rate (HR) and galvanic skin response (GSR).

[0044] For example, such as Figure 2 As shown, a non-contact infrared thermal imaging array is installed in the vehicle's ceiling, facing the head and upper torso areas of each seat. A 64×64 pixel, 10 Hz miniature thermal imager (such as the Heimann HTPA series) is used to output a Tskin map of body surface temperature distribution, used to estimate local thermal sensation. A millimeter-wave radar / TOF sensor, operating at 60 GHz, is installed on the seat back or inside the B-pillar, penetrating clothing to detect chest and abdominal movements and calculate respiratory rate (accuracy ±0.5 bpm), serving as an indirect indicator of heat stress. Optional wearable devices receive heart rate (HR), skin conductance (GSR), and other data from smartwatches / wristbands via Bluetooth 5.0 or UWB, which are then encrypted using AES-128 and transmitted to the central control unit. A miniature wind speed sensor collects local airflow velocity V. air (m / s), local sound pressure level L is collected through a near-ear microphone. noise (dB(A)), the solar radiation intensity t is collected by a light sensor located on the roof of the vehicle. sunlight (W / m²).

[0045] Optionally, when identifying passengers and obtaining their historical comfort preference models, facial recognition is performed on the passengers to obtain facial recognition information, and the passenger's user identifier is determined based on the facial recognition information; the user identifier is then used to match the passenger's preference profile stored in the cloud or locally. Each registered passenger has a unique user identifier; the preference profile includes at least temperature range, acoustic preferences, and physiological sensitivity labels; the preference profile represents the historical comfort preference model.

[0046] Specifically, identity matching is performed through facial recognition (e.g., front-facing camera). Each registered user has a unique ID, which is linked to a preference profile stored in the cloud or locally, including: temperature range (e.g., 22–24℃), acoustic preferences (e.g., "noise reduction priority", "background music type"), and physiological sensitivity tags (e.g., "elderly people are afraid of the cold", "children are easily disturbed by wind noise").

[0047] S102, calculates the passenger's predicted comfort level based on the passenger's perceived data.

[0048] In this embodiment of the application, the range of predicted comfort includes a first threshold and a second threshold, each threshold representing the passenger's predicted comfort level, for example, Xk∈[0,1], where 1 represents extreme comfort and 0 represents severe discomfort; the passenger's predicted comfort is calculated based on the passenger's perception data in step S101 for subsequent processing.

[0049] In some implementations, for each passenger, an environmental state vector is determined based on skin temperature, respiratory rate, heart rate, skin conductance, local airflow velocity, local sound pressure level, and solar radiation intensity. The passenger's environmental state vector is then input into a pre-defined Gaussian process regression (GPR) model to output the passenger's predicted comfort level.

[0050] Specifically, using a Gaussian process regression (GPR) model, for the k-th passenger, the input environment state vector of the GPR model is defined as:

[0051] The output is a comfort score Xk∈[0,1], where 1 represents extreme comfort and 0 represents severe discomfort. The meanings and acquisition methods of each parameter have been expanded, as shown in Table 1 below: Table 1

[0052] S103, based on the predicted comfort level, determine the optimal set of execution instructions to meet the passenger's historical comfort preference model, and obtain the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located.

[0053] In this embodiment of the application, the energy efficiency weight represents the penalty coefficient of total energy consumption. Based on the predicted comfort calculated in step S102, the optimal set of execution instructions that satisfies the historical comfort preference model of passengers, i.e., satisfies the comfort needs of all passengers, is determined. The target comfort of passengers, comfort weight, and energy efficiency weight of the vehicle in which the passengers are located are obtained for subsequent processing.

[0054] Optionally, when obtaining the passenger's target comfort level, comfort weight, and the energy efficiency weight of the vehicle in which the passenger is located, the passenger's selected comfort level is determined, and the corresponding target comfort level is determined based on the comfort level; the passenger's comfort weight is determined based on the passenger type, the driver's driving state, or the passenger's selected seat priority mode; the vehicle's energy mode is determined, and the corresponding energy efficiency weight is determined based on the energy mode. Different comfort levels map to different values; the passenger type includes elderly / children / pregnant women; the seat priority mode includes rear passenger care; and the energy mode includes at least normal mode, energy-saving mode, and performance mode.

[0055] Specifically, for example, the target comfort level (C) for the k-th passenger. ktarget The default value can be 0.9 (representing "high comfort"); it can be explicitly set by the user (e.g., selecting "Comfort Level: High / Medium / Low" via the slider → mapping to 0.95 / 0.85 / 0.75); or automatically inferred from historical data (e.g., if the historical average C for elderly users is 0.88, then set it as the target). The comfort weight (w) for the k-th passenger. k ), default to all w k =1.0 (Equal treatment); Dynamic adjustment: For the elderly / children / pregnant women: w k =1.3 (high priority); Driver: wdriver=1.2 (safety related) while driving; User manually increases the priority of a seat (e.g., "prioritize rear seats" mode). Energy efficiency weight (λ): How much comfort the control system is willing to sacrifice for energy saving. Normal mode: λ=0.05, Energy saving mode (EV low battery <20%): λ=0.2, Performance mode (charging / high-speed cruising): λ=0.01.

[0056] S104 uses a lightweight multi-objective genetic solver with the target population size and target iteration number as inputs to the current number of passengers in the vehicle, target comfort, optimal execution instruction set, predicted comfort, comfort weight, energy efficiency weight, and total power consumption of the vehicle to perform multi-objective optimization and obtain candidate solutions for the Pareto control strategy.

[0057] In this embodiment, each Pareto control strategy candidate solution is a complete control command vector (u This includes the wind speed / angle of all nozzles, the gain of all speakers, and ANC parameters. Based on the current number of passengers in the vehicle, target comfort, optimal execution instruction set, predicted comfort, comfort weight, energy efficiency weight, and the vehicle's total power consumption, the data is input into a lightweight multi-objective genetic solver to perform multi-objective optimization on the target population size and the number of iterations, resulting in a set (usually multiple) of Pareto-optimal control strategy candidate solutions. The lightweight multi-objective genetic solver can be the lightweight NSGA-II multi-objective genetic solver, with a population size of, for example, 50 and an iteration count of, for example, 20 generations.

[0058] It should be noted that the control command vector (u It can be in the following forms: u_candidate = { 'fan_speed': [2.1, 0.3, 1.8, 0.4], # Fan speed of the four nozzles (m / s) 'nozzle_angle_h': [12, -5, 8, -3], # Horizontal deflection angle (°) 'nozzle_angle_v': [-8, 0, -5, 2], 'audio_gain': [1.2, 0.9, 0.0, 0.7], # Four-seat audio gain 'anc_active': [False, False, True, True] }

[0059] S105: The candidate solutions of the Pareto control strategy are weighted and scored based on energy efficiency weights, the target solution of the control strategy is selected, and the corresponding control instruction package is generated based on the target solution of the control strategy and sent to the distributed actuator for execution.

[0060] In this embodiment, the control instruction package includes thermal instructions and acoustic instructions, which belong to different sub-objects. The candidate solutions of the Pareto control strategy are weighted and scored according to the energy efficiency weights, and the target solution of the control strategy that meets the current passenger's preference weights is selected. The corresponding control instruction package is generated based on the target solution of the control strategy and sent to the distributed actuator for execution.

[0061] It should be noted that the distributed actuators include a distributed microclimate execution array representing the thermal management execution layer and a distributed acoustic execution array representing the acoustic management execution layer. The distributed microclimate execution array is used to provide each passenger with independently controllable local temperature, humidity, and airflow, avoiding the overcooling / overheating problems caused by the "one-size-fits-all" approach of traditional air conditioning. The distributed microclimate execution array corresponds to the first target installation location, the first target technical specifications, and the first target control method, and includes at least a directional adjustable air nozzle, a micro blower module, and a central temperature and humidity control unit. The distributed acoustic execution array is used to simultaneously achieve local active noise cancellation (ANC), directional audio playback, and background sound masking within the same carriage, meeting the differentiated sound needs of different passengers. See Table 2 below for details. Table 2

[0062] Continuing, the distributed acoustic actuator array corresponds to the installation location, technical specifications, and control method of the second target. The distributed microclimate actuator array includes at least directional speakers, active noise cancellation (ANC) microphones, and a multi-channel audio processing unit. Details are shown in Table 3 below: Table 3

[0063] Optionally, a structured control command package is generated based on the target solution of the control strategy; the control command package is then sent to the corresponding distributed actuator via a preset communication method (e.g., CAN FD). For example, the control command package may include a command generation timestamp, the currently used Pareto solution identifier, wind speed, the seat's nozzle horizontal deflection angle (yaw angle), nozzle vertical deflection angle (pitch angle), light wind, audio channel gain, whether ANC is enabled, ANC operating mode, etc.

[0064] The personalized optimization method for multi-passenger passenger vehicle environments provided in this application embodiment acquires the perception data of each passenger in real time through a multimodal perception device, identifies the passengers, obtains the passengers' historical comfort preference model, calculates the passengers' predicted comfort based on the passengers' perception data, determines the optimal execution instruction set that satisfies the passengers' historical comfort preference model based on the predicted comfort, obtains the passengers' target comfort, comfort weight, and energy efficiency weight of the vehicle in which the passengers are located, and inputs the current number of passengers in the vehicle, target comfort, optimal execution instruction set, predicted comfort, comfort weight, energy efficiency weight, and total power consumption of the vehicle into a lightweight multi-objective genetic solver with target population size and target iteration number for multi-objective optimization, obtains Pareto control strategy candidate solutions, performs weighted scoring on the Pareto control strategy candidate solutions based on energy efficiency weight, selects the target solution of the control strategy, and generates corresponding control instruction packages based on the target solution of the control strategy and sends them to distributed actuators for execution. This application presents a personalized optimization method for multi-passenger passenger vehicle environments. It acquires thermal and acoustic perception data for each passenger and calculates their predicted comfort level. Combining multi-dimensional information with multi-objective optimization, it obtains candidate solutions for control strategies. Based on the objective solution of the control strategy, it generates corresponding control command packages and sends them to distributed actuators for execution. This achieves cross-domain collaboration between thermal and acoustic management. Zonal control is no longer based on individual areas; it can provide independent microenvironments for passengers of different body types or preferences within the same seat, improving spatial resolution. Furthermore, it can perform adaptive learning based on physiological signals or behavioral feedback, enhancing the feedback mechanism.

[0065] Furthermore, in response to passengers exhibiting implicit abnormal behavior, implicit abnormality data of passengers is acquired; explicit feedback data of passengers is acquired, and the Gaussian process regression model is updated online incrementally based on the explicit feedback data and implicit abnormality data of passengers.

[0066] Optionally, when obtaining the passenger's explicit feedback data, the passenger's selected comfort level is determined, and the corresponding explicit feedback data is determined based on the comfort level.

[0067] Specifically, each time explicit feedback (such as a sliding score (comfort level) or implicit abnormal signal (such as frequent temperature adjustment) is received, an online incremental learning Gaussian process regression model is triggered. For example, sparse variational inference SVGP is used to ensure real-time performance.

[0068] In summary, this application integrates non-contact infrared thermometry, millimeter-wave respiratory monitoring, and optional wearable physiological signals to construct a personalized comfort function that can be updated online. In the three-dimensional cockpit space, the air nozzles and speakers / ANC units are considered as coupled actuators, aiming to minimize comfort deviations for each passenger while constraining the impact of airflow noise on the auditory experience. An improved NSGA-II algorithm is employed to complete the Pareto optimal solution search and supports dynamic adjustment of user preference weights. Furthermore, implicit signals such as the frequency of user manual operations and changes in physiological stress indicators are combined with explicit ratings to trigger incremental model training, avoiding overfitting.

[0069] Therefore, this application has the following technical effects: 1. Achieve true personalized comfort for each individual. This application utilizes distributed micro-actuators and high-precision sensing to provide, for example, a 22°C quiet zone and a 26°C soft music zone for adjacent passengers (such as a mother and child sitting side by side) on the same seat, with a measured spatial resolution of 30cm×30cm.

[0070] 2. Deep coordination of thermal and acoustic fields to eliminate cross-interference. When this application activates strong cooling for the rear child seats (e.g., v=2.5m / s), airflow noise can reach, for example, 45dB(A), easily masking voice prompts. This application simultaneously enhances the ANC strength in this area and improves the background white noise masking effect in non-critical frequency bands (>2kHz), thereby improving the effective signal-to-noise ratio.

[0071] 3. Hardware reuse and energy efficiency optimization Directional airflow acts only on the human body area, which is more energy-efficient than whole-cabin cooling; the speakers also handle audio playback and ANC functions, eliminating the need for additional noise reduction hardware.

[0072] Figure 3 This is a schematic diagram of the structure of a personalized optimization device for a multi-passenger vehicle environment provided according to an embodiment of this application; as shown. Figure 3 As shown in the figure, the personalized optimization device 300 for multi-passenger vehicle environment according to an embodiment of this application may specifically include: The first acquisition module 301 is used to acquire the perception data of each passenger in real time through a multimodal perception device, identify the passenger, and acquire the passenger's historical comfort preference model; wherein the perception data represents thermal and acoustic perception data.

[0073] The calculation module 302 is used to calculate the passenger's predicted comfort level based on the passenger's perception data; wherein the range of the predicted comfort level includes a first threshold and a second threshold, and each threshold represents the passenger's predicted comfort level.

[0074] The second acquisition module 303 is used to determine the optimal set of execution instructions that satisfies the passenger's historical comfort preference model based on the predicted comfort level, and to acquire the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located.

[0075] The third acquisition module 304 is used to perform multi-objective optimization based on the current number of passengers in the vehicle, the target comfort level, the optimal execution instruction set, the predicted comfort level, the comfort level weight, the energy efficiency weight, and the total power consumption of the vehicle, inputting them into a lightweight multi-objective genetic solver with a target population size and a target iteration number, to obtain candidate solutions for the Pareto control strategy.

[0076] The execution module 305 is used to perform weighted scoring on the candidate solutions of the Pareto control strategy based on the energy efficiency weight, select the target solution of the control strategy, and generate a corresponding control instruction package based on the target solution of the control strategy and send it to the distributed actuator for execution.

[0077] In one possible implementation, the multimodal sensing device includes at least a non-contact infrared thermal imager, millimeter-wave radar, optional wearable device, miniature wind speed sensor, near-ear microphone, and light sensor; the first acquisition module is specifically used for: The target area of ​​the passenger is measured by a non-contact infrared thermal imager with target parameters to obtain the passenger's infrared thermal imaging information, and the passenger's body surface temperature distribution map is output based on the infrared thermal imaging information; wherein, the non-contact infrared thermal imager is installed in the ceiling of the vehicle, facing the head and upper torso area of ​​each seat, and the target area includes the forehead and / or neck; the target parameters include target pixels and target frame rate. The passenger's chest and abdominal movements are detected by millimeter-wave radar at the target operating frequency to calculate indirect indicators of heat stress; among which, indirect indicators of heat stress include at least respiratory rate; the millimeter-wave radar is installed on the seat back or the inside of the B-pillar. The system receives ECG information from selectable wearable devices via a preset communication method. The communication method includes at least Bluetooth and UWB, and the selectable wearable devices include at least smartwatches and smart bracelets. The ECG information includes at least heart rate and skin conductance. The system collects the local airflow speed of passengers using a miniature wind speed sensor, the local sound pressure level of passengers using a near-ear microphone, and the solar radiation intensity of passengers using a light sensor located on the roof.

[0078] In one possible implementation, the first acquisition module is specifically used for: Facial recognition is performed on passengers to obtain facial recognition information, and the passenger's user identifier is determined based on the facial recognition information; The system matches passengers' preference profiles stored in the cloud or locally based on their user identifiers. Each registered passenger has a unique user identifier. The preference profile includes at least temperature range, acoustic preferences, and physiological sensitivity tags.

[0079] In one possible implementation, the computing module is specifically used for: For passengers, their respective environmental state vectors are determined based on skin temperature, respiratory rate, heart rate, skin conductance, local airflow velocity, local sound pressure level, and solar radiation intensity. The passenger's environmental state vector is input into a pre-defined Gaussian process regression model, and the predicted comfort level of the passenger is output.

[0080] In one possible implementation, the second acquisition module is specifically used for: Determine the comfort level selected by the passenger, and determine the corresponding target comfort level based on the comfort level; different comfort levels map to different values; The passenger comfort weight is determined based on the passenger type, the driver's driving status, or the seat priority mode selected by the passenger. The energy mode of the vehicle is determined, and the corresponding energy efficiency weight is determined based on the energy mode; the energy mode includes at least normal mode, energy saving mode and performance mode.

[0081] In one possible implementation, the apparatus further includes: The fourth acquisition module is used to acquire the passenger's implicit abnormal data in response to the occurrence of implicit abnormal behavior. The update module is used to obtain explicit feedback data from passengers and to incrementally update the Gaussian process regression model online based on the explicit feedback data and implicit anomaly data from passengers. The update module is specifically used for: Determine the comfort level selected by the passenger, and then determine the corresponding explicit feedback data based on the comfort level.

[0082] In one possible implementation, the distributed actuator includes a distributed microclimate execution array characterizing the thermal management execution layer and a distributed acoustic execution array characterizing the acoustic management execution layer; The distributed microclimate execution array is used to provide each passenger with independently controllable local temperature, humidity and airflow; the distributed microclimate execution array corresponds to the first target installation location, the first target technical specifications and the first target control method, and the distributed microclimate execution array includes at least a directional adjustable air supply nozzle, a micro blower module and a central temperature and humidity control unit; The distributed acoustic actuator array is used to simultaneously achieve local active noise cancellation, directional audio playback, and background sound masking within the same carriage. The distributed acoustic actuator array corresponds to the installation location, technical specifications, and control method of the second target. The distributed microclimate actuator array includes at least a directional speaker, an active noise cancellation microphone, and a multi-channel audio processing unit.

[0083] The personalized optimization device for multi-passenger passenger vehicle environments provided in this application embodiment acquires the perception data of each passenger in real time through a multimodal perception device, identifies the passenger, obtains the passenger's historical comfort preference model, calculates the passenger's predicted comfort based on the passenger's perception data, determines the optimal execution instruction set that satisfies the passenger's historical comfort preference model based on the predicted comfort, obtains the passenger's target comfort, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located, and inputs the current number of passengers in the vehicle, target comfort, optimal execution instruction set, predicted comfort, comfort weight, energy efficiency weight, and total power consumption of the vehicle into a lightweight multi-objective genetic solver with target population size and target iteration number for multi-objective optimization, obtains Pareto control strategy candidate solutions, performs weighted scoring on the Pareto control strategy candidate solutions based on energy efficiency weight, selects the target solution of the control strategy, and generates corresponding control instruction packages based on the target solution of the control strategy and sends them to distributed actuators for execution. This application presents a personalized optimization device for multi-passenger passenger vehicle environments. It acquires thermal and acoustic perception data for each passenger and calculates their predicted comfort level. Combining multi-dimensional information with multi-objective optimization, it obtains candidate solutions for control strategies. Based on the objective solution of the control strategy, it generates corresponding control command packages and sends them to distributed actuators for execution. This achieves cross-domain collaboration between thermal and acoustic management. Zonal control is no longer based on individual areas; it can provide independent microenvironments for passengers of different body types or preferences on the same seat, improving spatial resolution. Furthermore, it can perform adaptive learning based on physiological signals or behavioral feedback, enhancing the feedback mechanism.

[0084] like Figure 4 As shown in the embodiment of this application, an electronic device 400 includes a processor 401, a memory 402, and a bus. The memory 402 stores machine-readable instructions executable by the processor 401. When the electronic device is running, the processor 401 communicates with the memory 402 via the bus. The processor 401 executes the machine-readable instructions to perform the steps of the above-described personalized optimization method for multi-passenger vehicle environments.

[0085] Specifically, the memory 402 and processor 401 mentioned above can be general-purpose memory and processor, without any specific limitations. When the processor 401 runs the computer program stored in the memory 402, it can execute the above-mentioned personalized optimization method for the multi-passenger vehicle environment.

[0086] Corresponding to the above-described method for personalized optimization of the passenger vehicle environment for multiple passengers, this application embodiment also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the above-described method for personalized optimization of the passenger vehicle environment for multiple passengers.

[0087] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.

[0088] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0089] In addition, 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.

[0090] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, 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 a portion 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 deployment methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0091] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for personalized optimization of a passenger vehicle environment for multiple passengers, characterized in that, The method includes: The sensory data of each passenger is acquired in real time through a multimodal sensing device, and the passengers are identified to obtain their historical comfort preference model; wherein the sensory data represents thermal and acoustic sensory data. The passenger's predicted comfort level is calculated based on the passenger's perceived data; wherein the range of the predicted comfort level includes a first threshold and a second threshold, each threshold representing the passenger's predicted comfort level; Based on the predicted comfort level, determine the optimal set of execution instructions that satisfies the passenger's historical comfort preference model, and obtain the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located; Based on the current number of passengers in the vehicle, the target comfort level, the optimal execution instruction set, the predicted comfort level, the comfort level weight, the energy efficiency weight, and the total power consumption of the vehicle, a lightweight multi-objective genetic solver with a target population size and a target iteration number is used for multi-objective optimization to obtain candidate solutions for the Pareto control strategy. The candidate solutions of the Pareto control strategy are weighted and scored based on the energy efficiency weights, the target solution of the control strategy is selected, and the corresponding control instruction package is generated based on the target solution of the control strategy and sent to the distributed actuator for execution.

2. The method according to claim 1, characterized in that, The multimodal sensing device includes at least a non-contact infrared thermal imager, millimeter-wave radar, optional wearable devices, a miniature wind speed sensor, a near-ear microphone, and a light sensor; the real-time acquisition of sensing data for each passenger includes: The target area of ​​the passenger is measured by a non-contact infrared thermal imager with target parameters to obtain the passenger's infrared thermal imaging information, and the passenger's body surface temperature distribution map is output based on the infrared thermal imaging information; wherein, the non-contact infrared thermal imager is installed in the ceiling of the vehicle, facing the head and upper torso area of ​​each seat, and the target area includes the forehead and / or neck; the target parameters include target pixels and target frame rate; The passenger's chest and abdominal movements are detected by a millimeter-wave radar operating at the target frequency to calculate indirect indicators of heat stress; wherein, the indirect indicators of heat stress include at least respiratory rate; the millimeter-wave radar is installed on the seat back or the inside of the B-pillar. The system receives electrocardiogram (ECG) information from optional wearable devices via a preset communication method. The communication method includes at least Bluetooth and UWB, and the optional wearable devices include at least smartwatches and smart bracelets. The ECG information includes at least heart rate and skin conductance. The local airflow velocity of the passenger is collected by the miniature wind speed sensor, the local sound pressure level of the passenger is collected by the near-ear microphone, and the solar radiation intensity of the passenger is collected by the light sensor located on the roof.

3. The method according to claim 2, characterized in that, The process of identifying the passenger and obtaining the passenger's historical comfort preference model includes: Facial recognition is performed on the passenger to obtain facial recognition information, and the passenger's user identifier is determined based on the facial recognition information; The user identifier is used to match the passenger's preference profile stored in the cloud or locally; wherein each registered passenger has a unique user identifier; the preference profile includes at least temperature range, acoustic preferences, and physiological sensitivity tags.

4. The method according to claim 3, characterized in that, The calculation of the passenger's predicted comfort level based on the passenger's perceived data includes: For the passenger, an environmental state vector is determined based on the skin temperature, respiratory rate, heart rate, skin conductance, local airflow velocity, local sound pressure level, and solar radiation intensity. The passenger's environmental state vector is input into a preset Gaussian process regression model, and the predicted comfort level of the passenger is output.

5. The method according to claim 1, characterized in that, The process of obtaining the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located includes: The comfort level selected by the passenger is determined, and the corresponding target comfort level is determined based on the comfort level; wherein, different comfort levels map to different values; The passenger's comfort weight is determined based on the passenger's type, the driver's driving status, or the seat priority mode selected by the passenger. The energy mode of the vehicle is determined, and the corresponding energy efficiency weight is determined based on the energy mode; wherein the energy mode includes at least a normal mode, an energy-saving mode, and a performance mode.

6. The method according to claim 4, characterized in that, The method further includes: In response to the passenger exhibiting implicit abnormal behavior, the implicit abnormal data of the passenger is obtained; Obtain the explicit feedback data of the passengers, and incrementally update the Gaussian process regression model online based on the explicit feedback data and implicit anomaly data of the passengers; The acquisition of explicit feedback data from the passenger includes: Determine the comfort level selected by the passenger, and determine the corresponding explicit feedback data based on the comfort level.

7. The method according to claim 1, characterized in that, The distributed actuator includes a distributed microclimate execution array characterizing the thermal management execution layer and a distributed acoustic execution array characterizing the acoustic management execution layer; The distributed microclimate execution array is used to provide each passenger with independently controllable local temperature, humidity and airflow; the distributed microclimate execution array corresponds to the first target installation location, the first target technical specifications and the first target control method, and the distributed microclimate execution array includes at least a directional adjustable air supply nozzle, a micro blower module and a central temperature and humidity control unit; The distributed acoustic execution array is used to simultaneously achieve local active noise cancellation, directional audio playback, and background sound masking within the same carriage; the distributed acoustic execution array corresponds to the installation location of the second target, the technical specifications of the second target, and the control method of the second target; the distributed microclimate execution array includes at least a directional speaker, an active noise cancellation microphone, and a multi-channel audio processing unit.

8. A personalized optimization device for the environment of a passenger vehicle with multiple passengers, characterized in that, The device includes: The first acquisition module is used to acquire the perception data of each passenger in real time through a multimodal perception device, identify the passenger, and obtain the passenger's historical comfort preference model; wherein, the perception data represents thermal and acoustic perception data; A calculation module is used to calculate the passenger's predicted comfort level based on the passenger's perception data; wherein the range of the predicted comfort level includes a first threshold and a second threshold, each threshold representing the passenger's predicted comfort level; The second acquisition module is used to determine the optimal set of execution instructions that satisfies the passenger's historical comfort preference model based on the predicted comfort level, and to acquire the passenger's target comfort level, comfort weight, and energy efficiency weight of the vehicle in which the passenger is located. The third acquisition module is used to perform multi-objective optimization by inputting the current number of passengers in the vehicle, the target comfort level, the optimal execution instruction set, the predicted comfort level, the comfort level weight, the energy efficiency weight, and the total power consumption of the vehicle into a lightweight multi-objective genetic solver with a target population size and a target iteration number, so as to obtain candidate solutions for the Pareto control strategy. The execution module is used to perform weighted scoring on the candidate solutions of the Pareto control strategy based on the energy efficiency weight, select the target solution of the control strategy, and generate a corresponding control instruction package based on the target solution of the control strategy and send it to the distributed actuator for execution.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of the personalized optimization method for a multi-passenger vehicle environment as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the method for personalized optimization of a multi-passenger vehicle environment as described in any one of claims 1 to 7.

Images