Intelligent cabin air conditioning method, device, equipment and storage medium

CN122976270APending Publication Date: 2026-10-27SAIC GM WULING AUTOMOBILE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611397560.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-09
Publication Date
2026-10-27

AI Technical Summary

Technical Problem

[0005]本发明的主要目的在于提供了一种智能座舱空气调节方法、装置、设备及存储介质,旨在解决现有技术的座舱空气调节的精准性和适应性不足的技术问题

Benefits of technology

[0016]本发明公开了获取多模态感知数据,并对所述多模态感知数据进行融合处理,生成融合特征序列,所述多模态感知数据至少包括座舱内外空气质量数据、乘员健康特征数据和驾驶场景数据;基于所述融合特征序列,生成空气净化策略和风场控制策略;基于所述空气净化策略和所述风场控制策略,按照预设多级优先级规则生成座舱空气调节策略;根据所述座舱空气调节策略驱动分级净化组件和场景化风场控制组件在座舱区域内进行空气调节。由于本发明通过多模态感知数据融合,基于融合特征序列生成空气净化策略和风场控制策略,并依据预设多级优先级规则生成座舱空气调节策略,驱动分级净化与场景化风场组件执行调节,相比于现有技术,本发明提升了座舱空气调节的精准性和适应性。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122976270A_ABST
    Figure CN122976270A_ABST
Patent Text Reader

Abstract

The application discloses an intelligent cabin air conditioning method, device, equipment and storage medium, the method comprises: acquiring multi-modal sensing data, and carrying out fusion processing on the multi-modal sensing data to generate a fusion feature sequence, the multi-modal sensing data at least includes cabin internal and external air quality data, occupant health feature data and driving scene data; based on the fusion feature sequence, generate air purification strategy and wind field control strategy; based on the air purification strategy and the wind field control strategy, generate the cabin air conditioning strategy according to the preset multi-level priority rule; according to the cabin air conditioning strategy, drive the hierarchical purification component and the scene wind field control component to carry out air conditioning in the cabin area. Since the application generates air purification strategy and wind field control strategy by fusing multi-modal sensing data, and generates cabin air conditioning strategy according to preset multi-level priority rule, compared with the prior art, the application improves the accuracy and adaptability of cabin air conditioning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to an intelligent cockpit air conditioning method, apparatus, device, and storage medium. Background Technology

[0002] With the continuous improvement of automotive intelligence, intelligent cabin air conditioning systems have become one of the core configurations for enhancing driving comfort and safety. Existing cabin air conditioning solutions typically rely on single or limited environmental monitoring devices such as temperature sensors and PM2.5 sensors to achieve basic temperature regulation, particulate matter filtration, and airflow control. Some improved solutions further incorporate an external environment sensing module, which can automatically switch between internal and external air circulation modes based on outside air quality. In addition, a few existing technologies attempt to integrate visual perception with environmental data to recommend air purification or airflow adjustment suggestions to users, aiming to improve the intelligence level of the cabin environment.

[0003] However, the existing solutions still have significant limitations. First, existing solutions generally lack the perception and utilization of the personalized health characteristics of cabin occupants. Air purification strategies typically employ a uniform filtration mode, failing to provide differentiated targeted adjustments based on different occupant allergen types, special physical conditions (such as children and the elderly), or real-time health status, resulting in severely insufficient precision in air conditioning. Second, the wind field control logic of existing solutions is relatively rigid, only supporting fixed wind direction and airflow levels, failing to dynamically optimize wind field parameters based on the occupant's current state (such as fatigue level and seating position) and specific driving scenarios, easily causing occupant discomfort. Third, most existing systems control purification and wind field components independently, lacking a unified decision-making mechanism based on multi-dimensional information fusion, making it difficult to coordinate the comprehensive output of purification and wind field in complex scenarios, resulting in poor adaptability of the overall adjustment strategy. Furthermore, although a few solutions have introduced user interaction recommendation functions, their decision-making process still requires user confirmation, resulting in delayed response and the inability to achieve automatic iterative optimization of strategies, further limiting the level of intelligence in cabin air conditioning.

[0004] Therefore, there is an urgent need for an intelligent cockpit air conditioning method that can improve the accuracy and adaptability of cockpit air conditioning. Summary of the Invention

[0005] The main objective of this invention is to provide an intelligent cockpit air conditioning method, apparatus, device, and storage medium, aiming to solve the technical problems of insufficient accuracy and adaptability of existing cockpit air conditioning technologies.

[0006] To achieve the above objectives, the present invention provides an intelligent cockpit air conditioning method, the method comprising the following steps: Acquire multimodal perception data and perform fusion processing on the multimodal perception data to generate a fusion feature sequence. The multimodal perception data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. Based on the fused feature sequence, an air purification strategy and a wind field control strategy are generated; Based on the air purification strategy and the wind field control strategy, a cabin air conditioning strategy is generated according to a preset multi-level priority rule. The cabin air conditioning strategy drives the graded purification components and the scenario-based airflow control components to perform air conditioning within the cabin area.

[0007] Optionally, the step of acquiring multimodal sensing data and fusing the multimodal sensing data to generate a fused feature sequence includes: Collect data on air quality inside and outside the cabin, occupant health characteristics, and driving scenario data; The cabin air quality data, occupant health characteristic data, and driving scenario data are time-stamped and synchronized to obtain synchronized multimodal perception data; The synchronized multimodal sensing data is subjected to feature weighted fusion according to preset weights to generate a fused feature sequence.

[0008] Optionally, the step of generating an air purification strategy and a wind field control strategy based on the fused feature sequence includes: Based on the fused feature sequence, the occupant's health status, current pollution type, and driving scenario information are determined; Based on a preset health-pollution matching model, a purification label is determined according to the occupant's health status and the current pollution type, and an air purification strategy is generated based on the purification label. Based on a preset wind field-user status matching model, wind field labels are determined according to the occupant's health status and the driving scenario information, and a wind field control strategy is generated based on the wind field labels.

[0009] Optionally, the step of generating a cabin air conditioning strategy based on the air purification strategy and the wind field control strategy according to a preset multi-level priority rule includes: Based on the fused feature sequence, determine the health protection needs, scenario response needs, and comfort optimization needs in the current scenario; Based on a preset multi-level priority rule, the health protection requirements, the scenario response requirements, and the comfort optimization requirements are prioritized to obtain a requirement priority sequence. Based on the required priority sequence, the air purification strategy and the wind field control strategy are combined to generate a cabin air conditioning strategy.

[0010] Optionally, the step of driving the graded purification components and the scenario-based airflow control components to perform air conditioning in the cabin area according to the cabin air conditioning strategy includes: Based on the purification control parameters in the cabin air conditioning strategy, the collaborative working mode of the multi-stage filter element in the graded purification component is determined. The multi-stage filter element includes a first filtration layer, a second filtration layer, and a catalytic adsorption layer. Based on the aforementioned collaborative working mode, the operating parameters of each filter element are adjusted to purify the air within the cabin area. Based on the airflow control parameters in the cabin air conditioning strategy, the guide angle of the electric air guide vane and the opening / closing position of the airflow valve are adjusted to control the airflow within the cabin area.

[0011] Optionally, after the step of driving the graded purification component and the scenario-based airflow control component to perform air conditioning in the cabin area according to the cabin air conditioning strategy, the method further includes: Collect user feedback data, which includes at least one of touch feedback data, voice feedback data, and behavioral feedback data; Based on the user feedback data, determine whether there are any parameter optimization trigger conditions; When the parameter optimization triggering condition exists, the cabin air conditioning strategy is adjusted based on the user feedback data to obtain the adjusted cabin air conditioning strategy. Extract the scene tag of the current driving scenario, and associate and store the scene tag, the control parameters in the adjusted cabin air conditioning strategy, and the feedback score corresponding to the user feedback data as a scene record.

[0012] Optionally, the method further includes: Obtain the scene features of the driving scenario to be adjusted, and determine the similarity between the scene features and the scene tags in each scene record; The scene record with the highest similarity and whose similarity exceeds a preset similarity threshold is selected as the target scene record; Based on the control parameters recorded in the target scenario, a cabin air conditioning strategy to be executed corresponding to the driving scenario to be adjusted is generated.

[0013] Furthermore, to achieve the above objectives, the present invention also proposes an intelligent cockpit air conditioning device, the device comprising: The data processing module is used to acquire multimodal perception data and perform fusion processing on the multimodal perception data to generate a fusion feature sequence. The multimodal perception data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. The strategy generation module is used to generate air purification strategies and wind field control strategies based on the fused feature sequence; The strategy adjustment module is used to generate a cabin air conditioning strategy based on the air purification strategy and the wind field control strategy, according to a preset multi-level priority rule. An air conditioning module is used to drive the graded purification components and the scenario-based airflow control components to perform air conditioning in the cabin area according to the cabin air conditioning strategy.

[0014] Furthermore, to achieve the above objectives, the present invention also proposes an intelligent cockpit air conditioning device, the device comprising: a memory, a processor, and an intelligent cockpit air conditioning program stored in the memory and executable on the processor, the intelligent cockpit air conditioning program being configured to implement the steps of the intelligent cockpit air conditioning method described above.

[0015] Furthermore, to achieve the above objectives, the present invention also proposes a storage medium storing an intelligent cockpit air conditioning program, which, when executed by a processor, implements the steps of the intelligent cockpit air conditioning method described above.

[0016] This invention discloses a method for acquiring multimodal sensing data and fusing the data to generate a fused feature sequence. The multimodal sensing data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. Based on the fused feature sequence, an air purification strategy and a wind field control strategy are generated. Based on the air purification strategy and the wind field control strategy, a cabin air conditioning strategy is generated according to a preset multi-level priority rule. The cabin air conditioning strategy drives the graded purification components and the scenario-based wind field control components to perform air conditioning within the cabin area. Because this invention uses multimodal sensing data fusion to generate air purification and wind field control strategies based on the fused feature sequence, and generates a cabin air conditioning strategy according to preset multi-level priority rules, driving the graded purification and scenario-based wind field components to perform adjustments, compared to existing technologies, this invention improves the accuracy and adaptability of cabin air conditioning. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the first embodiment of the intelligent cockpit air conditioning method of the present invention; Figure 2 This is a diagram illustrating the structure and collaborative workflow of the graded purification components in the intelligent cabin air conditioning method of the present invention. Figure 3 This is a flowchart illustrating the second embodiment of the intelligent cockpit air conditioning method of the present invention; Figure 4 This is a flowchart illustrating the third embodiment of the intelligent cockpit air conditioning method of the present invention; Figure 5 This is a schematic diagram of the system framework in the intelligent cockpit air conditioning method of the present invention; Figure 6 This is a structural block diagram of the first embodiment of the intelligent cockpit air conditioning device of the present invention; Figure 7 This is a schematic diagram of the structure of the intelligent cockpit air conditioning device in the hardware operating environment involved in the embodiments of the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0020] This invention provides an intelligent cockpit air conditioning method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the intelligent cockpit air conditioning method of the present invention.

[0021] In this embodiment, the intelligent cockpit air conditioning method includes steps S10 to S40: Step S10: Acquire multimodal perception data and perform fusion processing on the multimodal perception data to generate a fusion feature sequence. The multimodal perception data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data.

[0022] It should be noted that the executing entity in this embodiment can be a computer server device with data processing, network communication, and program execution functions applied in a cabin air conditioning scenario, such as a server, tablet computer, or personal computer, or an electronic device capable of performing the above functions (such as an intelligent cabin air conditioning device). The following uses a system including an intelligent cabin air conditioning device (hereinafter referred to as the system) as an example to illustrate this embodiment and the following embodiments.

[0023] It should be understood that the aforementioned multimodal sensing data can refer to environmental and state information from various types of sensors or data sources, with different data formats and physical meanings.

[0024] For example, cabin air quality data may include, but is not limited to, air parameters such as PM2.5 concentration, volatile organic compound concentration, carbon dioxide concentration, and pollen concentration. Occupant health characteristic data may refer to information reflecting the physiological state or health sensitivities of occupants within the cabin, such as whether an occupant has pollen allergies, is susceptible to asthma, or has a special physical condition such as being a child or elderly person. Driving scenario data may refer to information reflecting the vehicle's current external environment and driving conditions, such as whether the vehicle is currently traveling in different areas or road conditions, such as industrial areas, botanical gardens, highways, or urban roads.

[0025] It should be explained that the aforementioned fusion feature sequence can refer to a set of feature vectors with a unified time reference and standardized format generated after preprocessing, time alignment and weighted merging of the above-mentioned multiple sensing data. This fusion feature sequence is used as a unified data input for subsequent strategy generation.

[0026] It should be noted that the multimodal sensing data in this embodiment includes at least the above-mentioned cabin air quality data, occupant health characteristic data, and driving scenario data. However, in specific implementations, other types of sensing data, such as in-vehicle temperature and in-vehicle humidity, can also be added according to the actual sensor configuration.

[0027] To ensure the accuracy and reliability of the fused feature sequence and thus improve the precision of subsequent strategy generation, the steps of acquiring multimodal perception data and fusing the multimodal perception data to generate a fused feature sequence may include: collecting cabin air quality data, occupant health characteristic data, and driving scenario data; synchronizing the cabin air quality data, occupant health characteristic data, and driving scenario data with timestamps to obtain synchronized multimodal perception data; and performing feature weighted fusion on the synchronized multimodal perception data according to preset weights to generate a fused feature sequence.

[0028] It should be explained that the aforementioned timestamp synchronization refers to a processing method that aligns multimodal sensing data from different data sources to a unified time reference. Since cabin air quality data, occupant health characteristic data, and driving scenario data come from different sensors or data interfaces, the data acquisition time and sampling frequency of each data source may be inconsistent. Therefore, it is necessary to align the data from different sources to the same reference time point in the time dimension so that subsequent joint processing of multi-dimensional features can be performed.

[0029] Understandably, the aforementioned preset weights can refer to fusion coefficients pre-set for different categories of multimodal sensing data. These fusion coefficients reflect the importance of each type of data in generating the fused feature sequence. For example, in a specific implementation, the weight of occupant health feature data can be set to 0.4, the weight of cabin air quality data can be set to 0.35, and the weight of driving scenario data can be set to 0.25. The specific values ​​of these weights can be calibrated and adjusted according to the actual application scenario. The aforementioned feature weighted fusion can refer to a data processing method that combines various types of sensing data after time-stamp synchronization, each multiplied by its corresponding preset weight, to generate a comprehensive fused feature sequence.

[0030] In its implementation, the system acquires multimodal sensing data and fuses it to generate a fused feature sequence. Specifically, the system first acquires cabin air quality data, occupant health data, and driving scenario data through various vehicle sensors and data interfaces. After acquiring these three types of data, the system performs necessary data preprocessing, such as removing invalid data, filling missing values, or normalizing the data format. Next, the system fuses the preprocessed multimodal sensing data. Using a unified time reference, the system aligns the multimodal sensing data from different data sources with different timestamps, and then merges the time-aligned sensing data into a fused feature sequence according to preset fusion weights. This fused feature sequence integrates information from three dimensions: air quality, occupant health, and driving scenario, serving as input data for subsequently generating air purification and wind farm control strategies.

[0031] To facilitate understanding, the following examples are provided for illustration, but they do not constitute a specific limitation on this embodiment. For instance, in a real-world driving scenario, the vehicle is passing through a botanical garden area. The system described above uses an onboard PM2.5 sensor to obtain the current PM2.5 concentration inside the vehicle as 35 μg / m³ and the outside PM2.5 concentration as 42 μg / m³. It also uses a VOCs sensor to obtain the outside VOCs concentration as 0.1 mg / m³ and an onboard pollen sensor to obtain the outside pollen concentration as 150 grains / m³. Simultaneously, the system obtains occupant health characteristic data from a user-pre-entered health record, indicating that the current driver and passenger have pollen allergies. The system also obtains driving scenario data from the onboard navigation system, indicating that the vehicle is currently located within the botanical garden area. After normalizing the three types of data, the system aligns them according to a unified timestamp (e.g., 14:30:25 on August 28, 2026), and then weights and merges them according to preset weights of 0.4 for health data, 0.35 for air quality data, and 0.25 for scene data to generate a set of fused feature sequences. The fused feature sequences include timestamps, normalized PM2.5 concentration values, VOCs concentration values, pollen concentration values, occupant allergy status indicators, and botanical garden scene indicators.

[0032] Step S20: Based on the fused feature sequence, generate an air purification strategy and a wind field control strategy.

[0033] It should be explained that the aforementioned air purification strategy may refer to a set of control parameters used to control the air purification device in the cabin to perform purification actions. The set of control parameters describes the purification mode, purification intensity, or purification target to be executed, and may include, but is not limited to, one or more of the following: purification mode identifier, filtration level, filter cartridge operating frequency, and target pollutant concentration threshold.

[0034] The aforementioned wind field control strategy can refer to a set of control parameters used to control the wind field adjustment device in the cabin to perform wind field adjustment actions. The set of control parameters describes the wind field mode, wind speed, wind direction angle or noise control target to be executed. For example, it may include, but is not limited to, one or more of the following: wind field mode identifier, target air volume level, target angle of air guide plate, and noise upper limit value.

[0035] In its implementation, after generating the fused feature sequence, the system uses this sequence as input data to generate air purification and wind farm control strategies. The system parses the feature vectors in the fused feature sequence to identify various environmental and occupant status information. Subsequently, based on this parsed information, the system generates air purification and wind farm control strategies that match the current status.

[0036] To facilitate understanding, the following examples are provided for illustration, but do not impose specific limitations on this embodiment. For instance, suppose the current feature vector in the fused feature sequence obtained by the above system contains the following information: the PM2.5 concentration inside the cabin is 35 μg / m³, the pollen concentration outside the cabin is 500 grains / m³, and the occupant's pollen allergy level is moderate. After parsing the above feature vector, the above system generates an air purification strategy of "anti-allergy purification mode, target pollen as pollen, and purification intensity level as high". At the same time, based on the vehicle speed information of 60 km / h and the number of occupants of 2 from the same feature vector, the above system generates a wind field control strategy of "low-noise air supply mode, target air volume level of 5, and air guide angle in the direction of face air supply".

[0037] Step S30: Based on the air purification strategy and the wind field control strategy, generate a cabin air conditioning strategy according to a preset multi-level priority rule.

[0038] It should be explained that the aforementioned preset multi-level priority rules refer to a set of pre-defined rules used to determine the execution priority among multiple control requirements. These preset multi-level priority rules divide control requirements into multiple priority levels, with higher-priority control requirements having priority in strategy generation. For example, in the aforementioned preset multi-level priority rules, health protection requirements can be set as the first priority, scenario response requirements as the second priority, and comfort optimization requirements as the third priority. When control requirements corresponding to different needs conflict, the final control direction is determined according to the aforementioned priority order.

[0039] It should be noted that the above-mentioned cabin air conditioning strategy can refer to a set of comprehensive control parameters used to simultaneously control the air purification device and the air field device in the cabin to perform joint adjustment actions. The above-mentioned cabin air conditioning strategy is obtained by combining and adjusting the air purification strategy and the air field control strategy according to the above-mentioned preset multi-level priority rules, and includes two parts: purification control parameters and air field control parameters.

[0040] To avoid execution confusion caused by strategy conflicts and improve the rationality and adaptability of cabin air conditioning strategies, the step of generating a cabin air conditioning strategy based on the air purification strategy and the wind field control strategy according to a preset multi-level priority rule may include: determining health protection needs, scenario response needs, and comfort optimization needs under the current scenario based on the fused feature sequence; prioritizing the health protection needs, scenario response needs, and comfort optimization needs based on the preset multi-level priority rule to obtain a demand priority sequence; and combining the air purification strategy and the wind field control strategy according to the demand priority sequence to generate a cabin air conditioning strategy.

[0041] It should be noted that the aforementioned health protection requirements may refer to the control objectives that need to be implemented during cabin air conditioning to meet the health protection requirements of occupants. The aforementioned health protection requirements include at least one or more of the following: anti-allergy purification requirements, respiratory protection requirements, and irritant gas removal requirements.

[0042] The aforementioned scenario response requirements can refer to the control objectives that need to be executed during the cabin air conditioning process in response to changes in the external driving scenario. The aforementioned scenario response requirements include at least one or more of the following: rapid response to sudden pollution, wind noise suppression at high speed, and internal circulation switching.

[0043] The aforementioned comfort optimization requirements can refer to the control objectives that need to be implemented during cabin air conditioning to improve the subjective comfort of passengers. These comfort optimization requirements include at least one or more of the following: temperature comfort requirements, wind speed comfort requirements, wind direction comfort requirements, and noise comfort requirements.

[0044] It should be explained that the above-mentioned priority sequence of requirements can refer to an ordered list of requirements obtained by sorting the above-mentioned health protection requirements, the above-mentioned scenario response requirements, and the above-mentioned comfort optimization requirements according to a preset multi-level priority rule. The requirements in the above-mentioned priority sequence are arranged in order from high priority to low priority.

[0045] To facilitate understanding, the following explanation uses examples, but does not impose specific limitations on this embodiment. For example, suppose that after parsing and fusing the feature sequence, the system determines that the health protection requirement in the current scenario is "occupants are allergic to pollen and need anti-allergy purification," the scenario response requirement is "the vehicle is driving in a botanical garden area with high pollen concentration and needs to quickly switch to internal circulation," and the comfort optimization requirement is "occupants want lower wind noise." The system sorts these requirements based on a preset multi-level priority rule. Since the health protection requirement is the first priority, the scenario response requirement is the second priority, and the comfort optimization requirement is the third priority, the system obtains the following priority sequence: "anti-allergy purification requirement (first priority) → internal circulation switching requirement (second priority) → low wind noise requirement (third priority)." Subsequently, the system combines the air purification strategy "anti-allergy purification mode, HEPA ventilation frequency 2.0Hz" and the wind field control strategy "low noise air supply mode, target air volume level 4" according to the above priority sequence. The system determines that the HEPA ventilation frequency parameter in the air purification strategy and the target airflow parameter in the wind field control strategy do not conflict. It then merges these two strategies into a single cabin air conditioning strategy, which includes the purification control parameter "anti-allergy purification mode, HEPA ventilation frequency 2.0Hz" and the wind field control parameter "low-noise air supply mode, target airflow level 4". For example, in the scenario described above, if the system-analyzed air purification strategy requires internal circulation to prevent external pollen from entering, while the wind field control strategy requires external circulation to reduce carbon dioxide concentration, the system prioritizes anti-allergy purification over low-noise requirements. It then adjusts the external circulation activation parameter in the wind field control strategy to an internal circulation activation parameter, generating a cabin air conditioning strategy that includes both the purification control parameters and the adjusted wind field control parameters.

[0046] It should be noted that by introducing a three-tiered priority ranking mechanism of health protection, scenario response, and comfort optimization, the purification strategy and the air field control strategy can be rationally combined and coordinated according to the importance of actual needs. This can prioritize the health and safety of passengers when resources are limited or strategies conflict, while also taking into account rapid response to emergencies. This effectively avoids execution chaos caused by strategy conflicts and improves the decision-making rationality and overall adaptability of the cabin air conditioning strategy.

[0047] Step S40: Drive the graded purification component and the scenario-based air field control component to perform air conditioning in the cabin area according to the cabin air conditioning strategy.

[0048] It should be explained that the aforementioned graded purification components can refer to a collection of hardware devices installed in the air circulation path of a vehicle cabin for multi-stage filtration and purification of the air in the cabin. The aforementioned graded purification components typically include multiple filter layers with different filtration functions and filtration precisions. Each filter layer is arranged in sequence according to the airflow direction, and may include, but is not limited to, a pre-filter layer, a high-efficiency filter layer, and a catalytic adsorption layer.

[0049] Furthermore, the aforementioned primary filter layer is used to intercept large particulate matter, the aforementioned high-efficiency filter layer is used to filter PM2.5 and fine particulate matter such as pollen, and the aforementioned catalytic adsorption layer is used to adsorb and decompose volatile organic compounds and odorous gases.

[0050] It should be explained that the aforementioned scenario-based airflow control components can refer to a collection of hardware devices installed in the vehicle cabin for adjusting the airflow direction and volume. These scenario-based airflow control components can typically include air guides for changing the airflow direction and airflow valves for adjusting the ventilation volume, such as electric air guides and stepper motor airflow valves.

[0051] Understandably, the aforementioned cabin area can refer to the space inside the vehicle cabin. The aforementioned cabin area can be divided into multiple sub-areas according to the position of the occupants' seats, such as the front left area, the front right area, the rear left area, the rear right area, and the rear middle area. Each sub-area is independent of the others or partially related to them.

[0052] It should be noted that the step of driving the graded purification component and the scenario-based airflow control component to perform air conditioning in the cabin area according to the cabin air conditioning strategy may include: determining the collaborative working mode of the multi-stage filter elements in the graded purification component according to the purification control parameters in the cabin air conditioning strategy, wherein the multi-stage filter elements include a first filter layer, a second filter layer, and a catalytic adsorption layer; adjusting the working parameters of each stage of the filter elements based on the collaborative working mode to purify the air in the cabin area; and adjusting the guiding angle of the electric air guide plate and the opening and closing position of the airflow valve according to the airflow control parameters in the cabin air conditioning strategy to control the airflow in the cabin area.

[0053] It should be noted that the aforementioned multi-stage filter element can refer to multiple filter layers arranged sequentially along the airflow direction. Each filter layer has different filtration precision and filtration function, and through multi-layer synergy, it achieves graded filtration of pollutants of different particle sizes and types.

[0054] Furthermore, the aforementioned first filter layer can refer to the filter layer located at the forefront of the airflow direction, used to intercept larger particulate matter in the air, such as a pre-filter. The aforementioned second filter layer can refer to the filter layer located after the first filter layer, used to filter smaller fine particulate matter in the air, such as a HEPA H13 high-efficiency filter layer. The aforementioned catalytic adsorption layer can refer to the filter layer located after the second filter layer, used to adsorb and decompose volatile organic compounds and odor gases in the air, such as an activated carbon photocatalytic composite filter layer.

[0055] It should be noted that the above-mentioned collaborative working mode refers to the combination of cooperation and operating states of each level of filter element in the above-mentioned multi-stage filter element during the purification process. The above-mentioned collaborative working mode describes the working intensity of each level of filter element and the combination relationship between each level of filter element.

[0056] Furthermore, the aforementioned operating parameters may refer to the specific operating state values ​​adopted by each level of filter element when performing air purification, such as including but not limited to ventilation frequency value, air volume distribution ratio, and one or more of the on or off operating states.

[0057] It should be explained that the guiding angle of the aforementioned electric air guide plate refers to the rotation angle of the electric air guide plate relative to the reference direction, used to control the output direction of the airflow. The opening / closing position of the aforementioned air volume valve refers to the opening degree level of the aforementioned air volume valve, used to control the output volume of the airflow.

[0058] In its implementation, after obtaining the cabin air conditioning strategy, the system extracts purification control parameters from the strategy. Based on these parameters, the system determines the collaborative operating mode of the multi-stage filters in the graded purification assembly. The multi-stage filters include a first filter layer, a second filter layer, and a catalytic adsorption layer. The collaborative operating mode specifies the individual operating states of the first filter layer, the second filter layer, and the catalytic adsorption layer, as well as the synergistic relationship between them.

[0059] Next, based on the aforementioned collaborative working mode, the system adjusts the operating parameters of each stage of the multi-stage filter element. These operating parameters include the ventilation frequency or on / off status of each filter element. By adjusting these operating parameters, the system controls the tiered purification components to perform air purification actions within the cabin area. Simultaneously, the system extracts airflow control parameters from the cabin air conditioning strategy. Based on these airflow control parameters, the system adjusts the guide angle of the electric air guide vanes and the opening / closing position of the airflow valves. Specifically, the system drives each electric air guide vane to rotate to the corresponding guide angle according to the target angle specified in the airflow control parameters; and drives the airflow valves to open to the corresponding opening / closing position according to the target position specified in the airflow control parameters. The system controls the airflow direction by adjusting the guide angle of the electric air guide vanes and controls the airflow volume by adjusting the opening / closing position of the airflow valves, thereby performing airflow control actions within the cabin area. The air purification and airflow control actions are performed collaboratively under the drive of the system to jointly complete the air conditioning within the cabin area.

[0060] To facilitate understanding, the following explanation uses examples, but does not limit the specific implementation of this embodiment. For example, suppose that the cabin air conditioning strategy obtained by the above system includes purification control parameters such as "anti-allergy purification mode, HEPA ventilation frequency 2.0Hz, internal circulation on", and airflow control parameters such as "low-noise air supply mode, airflow level 4, and air guide plate angle in each area facing the face". Based on the "anti-allergy purification mode" in the purification control parameters, the above system determines the collaborative working mode of the multi-stage filters as "the first filter layer is normally on, the second filter layer is operating at enhanced speed, and the catalytic adsorption layer is normally on", and sets the HEPA ventilation frequency of the second filter layer to 2.0Hz. According to the above collaborative working mode, the above system keeps the first filter layer in a normally open state, adjusts the second filter layer to a ventilation frequency of 2.0Hz, keeps the catalytic adsorption layer in a normally open state, and switches to internal circulation mode, thereby performing air purification. Simultaneously, based on the airflow control parameters that specify "the angle of the air guide vanes in each area is the direction of facial airflow," the system rotates five sets of electrically operated air guide vanes—one for the front left and right, one for the rear left and right, and one for the rear center—to their corresponding facial airflow angles. Based on the "airflow level 4," the system opens the stepper motor airflow valve to level 4, thus executing the airflow control action. Driven in parallel by the system, these purification and airflow control actions achieve air conditioning within the cabin area that combines anti-allergy purification with facial airflow.

[0061] It should be understood that by determining the collaborative working mode of multi-stage filters based on purification control parameters and adjusting the working parameters of each stage of filters, graded and targeted treatment of different types of pollutants is achieved, avoiding the problem of low purification efficiency when a single filter works uniformly. At the same time, by adjusting the angle of the air guide plate and the air volume valve setting in conjunction with the air field control parameters, purification and air field are executed collaboratively under the same strategy framework, improving the coordination and accuracy of air conditioning actions.

[0062] To avoid premature filter replacement leading to waste or delayed replacement causing purification failure, and to improve purification reliability and economy, after the step of driving the graded purification components and the scenario-based airflow control components to perform air conditioning in the cabin area according to the cabin air conditioning strategy, the method may further include: obtaining the cumulative purification amount of each filter in the graded purification components and the current pollutant concentration in the cabin; when the cumulative purification amount of any filter reaches a preset cumulative threshold, or the current pollutant concentration in the cabin exceeds a preset concentration threshold, generating a replacement reminder signal for the corresponding filter.

[0063] For example, refer to Figure 2 , Figure 2 This diagram illustrates the structure and collaborative workflow of the graded purification component in the intelligent cabin air conditioning method of this invention. The diagram shows the structure and control process of the graded purification component, corresponding to the steps that drive it to purify the air. The component has an air inlet and an air outlet, and internally includes a pre-filter layer, a core adjustable HEPA H13 filter layer, and an activated carbon and photocatalyst layer. During control, inputting purification requirement tags such as anti-allergy, powerful deodorization, or CO2 reduction triggers the corresponding strategy: for anti-allergy, the HEPA high-frequency filter is controlled and internal circulation is enabled while the photocatalyst is disabled; for powerful deodorization, the HEPA medium-frequency filter is controlled and photocatalyst and internal circulation are enabled; for CO2 reduction, the HEPA low-frequency filter is controlled and external circulation is enabled while the photocatalyst is disabled, thus driving the three filter layers to purify collaboratively. After purification, an effect test is performed. The filter life monitoring module determines the lifespan based on the cumulative purification volume and concentration threshold. If the target is met, operation continues; otherwise, a replacement reminder is output and the module resets after replacement, thereby realizing parameter adjustment and lifespan management for the multi-stage filter collaborative working mode.

[0064] This embodiment discloses the acquisition of multimodal sensing data and the fusion processing of the multimodal sensing data to generate a fused feature sequence. The multimodal sensing data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. Based on the fused feature sequence, an air purification strategy and a wind field control strategy are generated. Based on the air purification strategy and the wind field control strategy, a cabin air conditioning strategy is generated according to a preset multi-level priority rule. The cabin air conditioning strategy drives the graded purification components and the scenario-based wind field control components to perform air conditioning within the cabin area. Because this embodiment uses multimodal sensing data fusion to generate air purification and wind field control strategies based on the fused feature sequence, and generates a cabin air conditioning strategy according to a preset multi-level priority rule, driving the graded purification and scenario-based wind field components to perform adjustments, compared with the prior art, this embodiment improves the accuracy and adaptability of cabin air conditioning.

[0065] refer to Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the intelligent cockpit air conditioning method of the present invention.

[0066] Based on the first embodiment described above, in this embodiment, step S20 includes steps S201 to S203: Step S201: Determine the occupant's health status, current pollution type, and driving scenario information based on the fused feature sequence.

[0067] It should be noted that the aforementioned occupant health status can refer to the physiological health conditions of the occupants in the cabin at the current moment and their special needs for the air environment. The aforementioned occupant health status includes at least one or more of the following: allergy type information, age group information, and respiratory system sensitivity information. For example, it can include specific conditions such as pollen allergy, asthma, child constitution, and elderly constitution.

[0068] The aforementioned current pollution type can refer to the main types of pollutants present in the air inside or outside the cabin. The aforementioned current pollution type includes at least one or more of the following: particulate matter pollution type, gaseous pollutant type, and biological allergen type. For example, it can include specific types such as PM2.5 pollution, volatile organic compound pollution, pollen pollution, and nitrogen dioxide pollution.

[0069] The aforementioned driving scenario information may refer to the current driving environment and driving conditions of the vehicle. The aforementioned driving scenario information includes at least one or more of the following: geographical location environment information, vehicle speed information, and road type information. For example, it may include specific scenarios such as botanical garden areas, industrial areas, highways, and urban roads.

[0070] Step S202: Based on the preset health-pollution matching model, determine the purification label according to the occupant's health status and the current pollution type, and generate an air purification strategy based on the purification label.

[0071] It should be explained that the aforementioned preset health-pollution matching model can refer to a set of pre-established mapping rules used to describe the correspondence between the health status of occupants and the types of pollutants. The preset health-pollution matching model defines the purification response level and purification method triggered by various pollutants under different health statuses of occupants.

[0072] It should be noted that the aforementioned purification label may refer to the identification information output by the aforementioned preset health-pollution matching model based on the input occupant health status and the current pollution type. The aforementioned purification label is used to characterize the type of purification target and the purification intensity level to be performed. For example, it may include anti-allergy purification label, strong deodorization label, CO2 emission reduction label, and regular filtration label, etc.

[0073] Step S203: Based on the preset wind field-user status matching model, determine the wind field label according to the occupant's health status and the driving scenario information, and generate a wind field control strategy based on the wind field label.

[0074] It should be explained that the aforementioned preset wind field-user state matching model can refer to a set of pre-established mapping rules used to describe the correspondence between occupant health status and driving scenario information and wind field adjustment methods. The aforementioned preset wind field-user state matching model defines the wind field modes adapted to different occupant states and different driving scenarios.

[0075] It should be noted that the aforementioned wind field label can refer to the identification information output by the aforementioned preset wind field-user status matching model based on the input occupant health status and driving scenario information. The aforementioned wind field label is used to characterize the type of wind field mode to be executed and the direction of wind field adjustment. For example, it may include anti-direct blowing wind field label, low noise wind field label, rapid cooling wind field label, and uniform air supply label, etc.

[0076] In its specific implementation, when generating air purification strategies and wind field control strategies, the system first takes the aforementioned fused feature sequence as input, analyzes the feature vector of the current moment in the aforementioned fused feature sequence, and extracts feature components related to occupant health, feature components related to pollutant type, and feature components related to driving scenario from the aforementioned feature vector, thereby determining the occupant health status, current pollution type, and driving scenario information.

[0077] Next, the system inputs the occupant's health status and the current pollution type into a preset health-pollution matching model. The preset health-pollution matching model performs a matching determination based on the occupant's health status and the current pollution type, and outputs a purification label. This purification label indicates the type of purification target and the purification intensity level that should be prioritized in the current state. After obtaining the purification label, the system generates an air purification strategy based on it. This air purification strategy includes a purification mode identifier and purification intensity parameters corresponding to the purification label.

[0078] Simultaneously, the system inputs the occupant health status and driving scenario information into a preset wind field-user status matching model. This model performs a matching determination based on the occupant health status and driving scenario information, outputting a wind field label. This wind field label indicates the wind field mode type and wind field adjustment direction to be executed in the current state. After obtaining the wind field label, the system generates a wind field control strategy based on it. This strategy includes a wind field mode identifier and wind field adjustment parameters corresponding to the wind field label.

[0079] To facilitate understanding, the following explanation uses examples, but does not impose specific limitations on this embodiment. For example, suppose that after parsing and fusing the feature sequence, the system determines that the occupant's health status is "pollen allergy," the current pollution type is "outdoor air pollen concentration 500 grains / m³," and the driving scenario information is "botanical garden area, vehicle speed 30km / h." The system inputs "pollen allergy" and "outdoor air pollen concentration 500 grains / m³" into a preset health-pollution matching model. After matching, the preset health-pollution matching model outputs a purification label of "anti-allergy purification (high-efficiency level)." The system generates an air purification strategy based on the purification label, which includes control parameters such as "anti-allergy purification mode, HEPA ventilation frequency 2.0Hz, internal circulation on." Simultaneously, the system inputs "pollen allergy" and "botanical garden area, vehicle speed 30km / h" into a preset wind field-user status matching model. After matching, the preset wind field-user status matching model outputs a wind field label of "anti-direct-blow wind field." The system generates an airflow control strategy based on the aforementioned airflow labels. This strategy includes control parameters such as "anti-direct-blow airflow mode, target airflow level 3, and air deflector angle to avoid the direction of the occupant's face." The system then outputs both the air purification strategy and the airflow control strategy separately.

[0080] This embodiment discloses a method for determining occupant health status, current pollution type, and driving scenario information based on the fused feature sequence; determining a purification label based on the occupant health status and current pollution type using a preset health-pollution matching model, and generating an air purification strategy based on the purification label; and determining a wind field label based on the occupant health status and driving scenario information using a preset wind field-user status matching model, and generating a wind field control strategy based on the wind field label. Because this embodiment determines occupant health status, pollution type, and driving scenario information separately based on the fused feature sequence, and generates purification labels based on the health-pollution matching model and wind field labels based on the wind field-user status matching model, it achieves precise matching of the purification strategy with occupant health characteristics and pollution type, as well as dynamic matching of the wind field strategy with occupant status and scenario information. Compared to existing technologies, this embodiment improves the targeting and accuracy of strategy generation.

[0081] refer to Figure 4 , Figure 4 This is a flowchart illustrating the third embodiment of the intelligent cockpit air conditioning method of the present invention.

[0082] Based on the above embodiments, in this embodiment, after step S40, steps S50 to S80 are further included: Step S50: Collect user feedback data, wherein the user feedback data includes at least one of touch feedback data, voice feedback data, and behavioral feedback data.

[0083] Understandably, the aforementioned user feedback data can refer to a collection of information provided by cabin occupants, either actively or passively, to express their subjective feelings or operational intentions regarding the current air conditioning effect.

[0084] Furthermore, the aforementioned touch feedback data can refer to the feedback information input by the passenger through the preset interactive controls on the touch screen. For example, it can include the rating data generated by the passenger clicking the "Comfortable", "Neutral", and "Uncomfortable" buttons on the central control screen or the rear armrest touch screen. The touch feedback data usually includes the feedback rating value and the seat area identifier that generated the feedback.

[0085] The aforementioned voice feedback data can refer to the feedback information expressed by the occupant through voice. The voice feedback data is generated by the vehicle microphone array picking up the natural language statements spoken by the occupant and then processing them through speech recognition and semantic analysis. For example, it may include the analysis results such as "pollen or allergy", "the wind is too noisy", "the wind noise in the left rear area is too loud". The voice feedback data usually includes the semantic content of the feedback and the location information of the seat area that generated the feedback.

[0086] The aforementioned behavioral feedback data can refer to implicit feedback information expressed by occupants through manual operation of cabin air conditioning-related controls. For example, it may include operation records such as occupants manually increasing the fan speed, manually adjusting the airflow angle, and manually switching between internal and external circulation modes. Behavioral feedback data is used to characterize occupants' dissatisfaction with the current automatic adjustment strategy or their intention to adjust it.

[0087] Step S60: Based on the user feedback data, determine whether there are parameter optimization trigger conditions.

[0088] It should be explained that the above-mentioned parameter optimization triggering conditions can refer to preset judgment conditions used to determine whether the control parameters in the current cabin air conditioning strategy need to be corrected. The above-mentioned parameter optimization triggering conditions use the feedback distribution characteristics of a specific dimension in user feedback data as the judgment basis. For example, it can be set to trigger parameter optimization when the proportion of "discomfort" feedback in a certain seat area or a certain dimension exceeds a preset percentage threshold.

[0089] Step S70: When the parameter optimization triggering condition exists, the cabin air conditioning strategy is adjusted based on the user feedback data to obtain the adjusted cabin air conditioning strategy.

[0090] Step S80: Extract the scene label of the current driving scenario, and associate and store the scene label, the control parameters in the adjusted cabin air conditioning strategy, and the feedback score corresponding to the user feedback data as a scene record.

[0091] It should be noted that the above-mentioned scene labels can refer to a set of feature identifiers used to uniquely identify the current driving scene type. The above-mentioned scene labels include at least one or more of the following: current geographical location environment type, current pollutant type, and current driving condition type. For example, "Botanical Garden - High Pollen - Low Speed ​​Driving" can be used as a set of scene labels.

[0092] Understandably, the aforementioned feedback score can refer to the result of converting the aforementioned user feedback data into numerical evaluation values ​​according to preset scoring rules. The aforementioned feedback score is used to quantitatively represent the passenger's satisfaction with the current air conditioning effect. For example, it can be converting "comfortable" in the three-level touch feedback into 3 points, "average" into 2 points, and "uncomfortable" into 1 point. Alternatively, it can be mapping the positive or negative semantics in the voice feedback into corresponding scores after performing sentiment analysis.

[0093] It should be noted that the above-mentioned scenario record can refer to the data entry formed by associating and storing the above-mentioned scenario label, the control parameters in the above-mentioned adjusted cabin air conditioning strategy, and the above-mentioned feedback score. The above-mentioned scenario record can be stored in the system storage unit in the form of structured data entries. Each scenario record contains a complete set of scenario label-strategy parameter-feedback score triplet information.

[0094] To facilitate understanding, the following explanation uses examples, but does not impose specific limitations on this embodiment. For example, suppose that during an air conditioning process, the above system has driven the graded purification component and the scenario-based airflow control component to execute the cabin air conditioning strategy (which includes "anti-allergy purification mode, HEPA ventilation frequency 2.0Hz" and "low-noise air supply mode, airflow level 4"). After the adjustment is completed, the above system collects the touch feedback data of the rear left occupant through the central control screen, which is "discomfort" (corresponding to a score of 1 point). At the same time, it collects the occupant's voice feedback data through the vehicle microphone array, which is "excessive wind noise in the rear left area". The above system statistics show that the "discomfort" feedback in the rear area accounts for 20%, exceeding the preset 15% trigger threshold, thus determining that there is a parameter optimization trigger condition. Based on the user feedback data regarding "high wind noise" and the location information of the "rear left area," the system determined that the airflow parameters for the rear left area needed adjustment. The airflow valve's opening / closing level in the airflow control parameters was adjusted from level 4 to level 3, and the noise-wind speed linkage parameter was lowered by 2dB, resulting in the adjusted cabin air conditioning strategy. Subsequently, the system extracted the current scene label as "Botanical Garden Area - Pollen Pollution - Low-Speed ​​Driving," and associated and stored the scene label, the control parameters in the adjusted cabin air conditioning strategy, and the feedback score (1 point) as a scene record. This scene record includes the scene identifier, the adjusted HEPA ventilation frequency value, the adjusted airflow level value, and the corresponding feedback score.

[0095] To significantly shorten scenario response latency and improve system control efficiency in repetitive or similar scenarios while ensuring strategy adaptability, the method may further include: acquiring scenario features of the driving scenario to be adjusted; determining the similarity between the scenario features and scenario tags in each scenario record; selecting the scenario record with the highest similarity and the similarity exceeding a preset similarity threshold as the target scenario record; and generating a cabin air conditioning strategy to be executed corresponding to the driving scenario to be adjusted based on the control parameters in the target scenario record.

[0096] It is understandable that the aforementioned driving scenario to be adjusted may refer to the driving scenario in which the system is currently preparing to match the air conditioning strategy, and the aforementioned driving scenario to be adjusted is the current driving scenario in which the corresponding cabin air conditioning strategy has not yet been generated.

[0097] It should be noted that the above-mentioned scene features can refer to a set of feature information used to describe the above-mentioned driving scenario to be adjusted. The above-mentioned scene features include at least one or more of the following: current pollutant type information, current geographical location environment type information, and current driving condition type information. The above-mentioned scene features are used to compare the similarity with the scene tags in the stored scene records.

[0098] Furthermore, the aforementioned similarity can refer to the quantitative value of the matching degree between the aforementioned scene features and the scene labels in each scene record. The aforementioned similarity is used to measure the closeness of two driving scenes in key dimensions. The higher the similarity value, the closer the two driving scenes are.

[0099] It should be understood that the aforementioned preset similarity threshold may refer to a pre-set minimum similarity value used to determine whether two driving scenarios belong to similar scenarios. When the aforementioned similarity exceeds the aforementioned preset similarity threshold, the aforementioned driving scenario to be adjusted is determined to be a similar scenario to the historical scenario corresponding to the scenario record.

[0100] Understandably, the aforementioned target scenario record can refer to the scenario record selected from all scenario records that has the highest similarity to the aforementioned driving scenario to be adjusted and whose similarity exceeds the aforementioned preset similarity threshold. The aforementioned target scenario record serves as the source of strategy invocation for the aforementioned driving scenario to be adjusted. The aforementioned control parameters can refer to the specific parameter values ​​stored in the aforementioned scenario record that correspond to the cabin air conditioning strategy associated with the scenario tag. The aforementioned control parameters include purification control parameters and airflow control parameters.

[0101] Furthermore, the aforementioned cabin air conditioning strategy to be executed can refer to a comprehensive control strategy generated based on the control parameters recorded in the aforementioned target scenario, used to execute cabin air conditioning in the aforementioned driving scenario to be regulated. The aforementioned cabin air conditioning strategy to be executed includes two parts: purification control parameters and airflow control parameters.

[0102] In its implementation, after storing the scene records, the system acquires the scene features of a new driving scenario to be adjusted when such a scenario exists. These scene features include current pollutant type information, current geographical location environment type information, and current driving condition type information. After acquiring these scene features, the system calculates the similarity between each scene feature and the scene tag in each stored scene record to determine the similarity between the scene feature and the scene tag in each scene record. Specifically, the system calculates the matching degree for each of the three dimensions: pollutant type, pollutant concentration, and geographical location type. The matching degrees of the three dimensions are then weighted and merged according to preset dimension weights to obtain the similarity between the scene feature and each scene tag. After calculating the similarity for all scene records, the system selects the scene record with the highest similarity exceeding a preset similarity threshold as the target scene record. If no scene record with a similarity exceeding the preset similarity threshold exists, the system determines that there is no matching historical scene record for the current driving scenario to be adjusted and does not execute the strategy invocation operation. After successfully selecting a target scenario record, the system extracts control parameters from the record, including purification control parameters and airflow control parameters. Based on these parameters, the system generates a cabin air conditioning strategy to be executed for the specific driving scenario. This strategy includes purification and airflow control parameters identical to those in the target scenario record and serves as the air conditioning control command for the driving scenario.

[0103] For example, refer to Figure 5 , Figure 5This is a schematic diagram of the system framework of the intelligent cockpit air conditioning method of the present invention. The system includes a multimodal perception unit, a health-scenario decision-making module, an adaptive execution module, and a self-learning optimization unit. The multimodal perception unit collects air quality data inside and outside the cockpit and occupant health characteristic data through an air quality sensor group and a user health perception component, and combines it with driving scenario data obtained by the scenario perception component. The data preprocessing module then fuses these data to generate a fused feature sequence. The health-scenario decision-making module uses a health-pollution matching model and a wind field-user state matching model to generate an air purification strategy and a wind field control strategy, and generates a cockpit air conditioning strategy based on priority decision rules. The adaptive execution module drives the graded purification component (including multi-stage filters) and the scenario-based wind field control component (including electric air guides and air volume valves) to perform air conditioning within the cockpit area according to the strategy. The self-learning optimization unit collects touch, voice, and behavioral feedback data through a user feedback interface, adjusts the strategy through a parameter self-optimization unit, and stores the scenario label, control parameters, and feedback score in a scenario memory unit for recall in similar scenarios to achieve personalized air conditioning.

[0104] This embodiment discloses the collection of user feedback data, which includes at least one of touch feedback data, voice feedback data, and behavioral feedback data. Based on the user feedback data, it determines whether parameter optimization trigger conditions exist. When such conditions exist, the cabin air conditioning strategy is adjusted based on the user feedback data to obtain an adjusted cabin air conditioning strategy. Scene tags for the current driving scenario are extracted, and the scene tags, control parameters in the adjusted cabin air conditioning strategy, and feedback scores corresponding to the user feedback data are associated and stored as scene records. Because this embodiment establishes a closed-loop feedback mechanism by collecting multi-dimensional user feedback data such as touch, voice, and behavioral feedback, the system can automatically trigger parameter adjustments based on the user's actual experience, achieving continuous optimization of the air conditioning strategy. Furthermore, by associating scene tags, control parameters, and feedback scores as scene records, subsequent similar scenarios can directly call upon historical optimization strategies. Compared to existing technologies, this embodiment improves the system's adaptability to different user preferences and usage scenarios.

[0105] Furthermore, this embodiment of the invention also proposes a storage medium storing an intelligent cockpit air conditioning program, which, when executed by a processor, implements the steps of the intelligent cockpit air conditioning method described above.

[0106] Reference Figure 5 , Figure 5 This is a structural block diagram of the first embodiment of the intelligent cockpit air conditioning device of the present invention.

[0107] like Figure 5As shown, the intelligent cockpit air conditioning device proposed in this embodiment of the invention includes: a data processing module 501, a strategy generation module 502, a strategy adjustment module 503, and an air conditioning module 504.

[0108] The data processing module 501 is used to acquire multimodal perception data and perform fusion processing on the multimodal perception data to generate a fusion feature sequence. The multimodal perception data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data.

[0109] The strategy generation module 502 is used to generate an air purification strategy and a wind field control strategy based on the fused feature sequence.

[0110] The strategy adjustment module 503 is used to generate a cabin air conditioning strategy based on the air purification strategy and the wind field control strategy, according to a preset multi-level priority rule.

[0111] The air conditioning module 504 is used to drive the graded purification component and the scenario-based air field control component to perform air conditioning in the cabin area according to the cabin air conditioning strategy.

[0112] The data processing module 501 is also used to collect cabin air quality data, occupant health characteristic data, and driving scenario data; to synchronize the cabin air quality data, occupant health characteristic data, and driving scenario data with timestamps to obtain synchronized multimodal perception data; and to perform feature weighted fusion on the synchronized multimodal perception data according to preset weights to generate a fused feature sequence.

[0113] The strategy adjustment module 503 is further configured to determine the health protection requirements, scenario response requirements, and comfort optimization requirements under the current scenario based on the fusion feature sequence; prioritize the health protection requirements, scenario response requirements, and comfort optimization requirements based on preset multi-level priority rules to obtain a requirement priority sequence; and combine the air purification strategy and the wind field control strategy according to the requirement priority sequence to generate a cabin air conditioning strategy.

[0114] The air conditioning module 504 is further configured to determine the collaborative working mode of the multi-stage filter elements in the graded purification component according to the purification control parameters in the cabin air conditioning strategy, wherein the multi-stage filter elements include a first filtration layer, a second filtration layer, and a catalytic adsorption layer; adjust the working parameters of each stage of the filter elements based on the collaborative working mode to purify the air in the cabin area; and adjust the guiding angle of the electric air guide vane and the opening and closing position of the air volume valve according to the air field control parameters in the cabin air conditioning strategy to control the air field in the cabin area.

[0115] This device embodiment discloses acquiring multimodal sensing data and fusing the multimodal sensing data to generate a fused feature sequence. The multimodal sensing data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. Based on the fused feature sequence, an air purification strategy and a wind field control strategy are generated. Based on the air purification strategy and the wind field control strategy, a cabin air conditioning strategy is generated according to a preset multi-level priority rule. The cabin air conditioning strategy drives the graded purification components and the scenario-based wind field control components to perform air conditioning within the cabin area. Because this device embodiment uses multimodal sensing data fusion to generate air purification and wind field control strategies based on the fused feature sequence, and generates a cabin air conditioning strategy according to a preset multi-level priority rule, driving the graded purification and scenario-based wind field components to perform adjustments, compared to existing technologies, this device embodiment improves the accuracy and adaptability of cabin air conditioning.

[0116] Based on the first embodiment of the intelligent cockpit air conditioning device of the present invention described above, a second embodiment of the intelligent cockpit air conditioning device of the present invention is proposed.

[0117] In this embodiment, the data processing module 501 is further configured to collect cabin air quality data, occupant health characteristic data, and driving scenario data; to timestamp and synchronize the cabin air quality data, occupant health characteristic data, and driving scenario data to obtain synchronized multimodal perception data; and to perform feature weighted fusion on the synchronized multimodal perception data according to preset weights to generate a fused feature sequence.

[0118] Other embodiments or specific implementations of the intelligent cockpit air conditioning device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0119] This application provides an intelligent cockpit air conditioning device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the intelligent cockpit air conditioning method in the first embodiment described above.

[0120] The following is for reference. Figure 7The diagram illustrates a structural schematic suitable for implementing the intelligent cockpit air conditioning device in the embodiments of this application. The intelligent cockpit air conditioning device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 7 The intelligent cockpit air conditioning device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of this application.

[0121] like Figure 7 As shown, the intelligent cockpit air conditioning device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the intelligent cockpit air conditioning device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the intelligent cockpit air conditioning system to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows an intelligent cockpit air conditioning system with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0122] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0123] The intelligent cockpit air conditioning device provided in this application, employing the intelligent cockpit air conditioning method described in the above embodiments, can solve the technical problems of insufficient accuracy and adaptability of existing cockpit air conditioning systems. Compared with the prior art, the beneficial effects of the intelligent cockpit air conditioning device provided in this application are the same as those of the intelligent cockpit air conditioning method provided in the above embodiments, and other technical features of this intelligent cockpit air conditioning device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0124] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0125] The above description is merely a specific embodiment 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.

[0126] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0127] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0128] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0129] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for intelligent cockpit air conditioning, characterized in that, The method includes: Acquire multimodal perception data and perform fusion processing on the multimodal perception data to generate a fusion feature sequence. The multimodal perception data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. Based on the fused feature sequence, an air purification strategy and a wind field control strategy are generated; Based on the air purification strategy and the wind field control strategy, a cabin air conditioning strategy is generated according to a preset multi-level priority rule. The cabin air conditioning strategy drives the graded purification components and the scenario-based airflow control components to perform air conditioning within the cabin area.

2. The intelligent cockpit air conditioning method as described in claim 1, characterized in that, The step of acquiring multimodal sensing data and fusing the multimodal sensing data to generate a fused feature sequence includes: Collect data on air quality inside and outside the cabin, occupant health characteristics, and driving scenario data; The cabin air quality data, occupant health characteristic data, and driving scenario data are time-stamped and synchronized to obtain synchronized multimodal perception data; The synchronized multimodal sensing data is subjected to feature weighted fusion according to preset weights to generate a fused feature sequence.

3. The intelligent cockpit air conditioning method as described in claim 1, characterized in that, The step of generating air purification strategy and wind field control strategy based on the fused feature sequence includes: Based on the fused feature sequence, the occupant's health status, current pollution type, and driving scenario information are determined; Based on a preset health-pollution matching model, a purification label is determined according to the occupant's health status and the current pollution type, and an air purification strategy is generated based on the purification label. Based on a preset wind field-user status matching model, wind field labels are determined according to the occupant's health status and the driving scenario information, and a wind field control strategy is generated based on the wind field labels.

4. The intelligent cockpit air conditioning method as described in claim 1, characterized in that, The step of generating a cabin air conditioning strategy based on the air purification strategy and the wind field control strategy according to a preset multi-level priority rule includes: Based on the fused feature sequence, determine the health protection needs, scenario response needs, and comfort optimization needs in the current scenario; Based on a preset multi-level priority rule, the health protection requirements, the scenario response requirements, and the comfort optimization requirements are prioritized to obtain a requirement priority sequence. Based on the required priority sequence, the air purification strategy and the wind field control strategy are combined to generate a cabin air conditioning strategy.

5. The intelligent cockpit air conditioning method as described in claim 1, characterized in that, The step of driving the graded purification components and the scenario-based airflow control components to perform air conditioning in the cabin area according to the cabin air conditioning strategy includes: Based on the purification control parameters in the cabin air conditioning strategy, the collaborative working mode of the multi-stage filter element in the graded purification component is determined. The multi-stage filter element includes a first filtration layer, a second filtration layer, and a catalytic adsorption layer. Based on the aforementioned collaborative working mode, the operating parameters of each filter element are adjusted to purify the air within the cabin area. Based on the airflow control parameters in the cabin air conditioning strategy, the guide angle of the electric air guide vane and the opening / closing position of the airflow valve are adjusted to control the airflow within the cabin area.

6. The intelligent cockpit air conditioning method as described in claim 1, characterized in that, After the step of driving the graded purification component and the scenario-based airflow control component to perform air conditioning in the cabin area according to the cabin air conditioning strategy, the method further includes: Collect user feedback data, which includes at least one of touch feedback data, voice feedback data, and behavioral feedback data; Based on the user feedback data, determine whether there are any parameter optimization trigger conditions; When the parameter optimization triggering condition exists, the cabin air conditioning strategy is adjusted based on the user feedback data to obtain the adjusted cabin air conditioning strategy. Extract the scene tag of the current driving scenario, and associate and store the scene tag, the control parameters in the adjusted cabin air conditioning strategy, and the feedback score corresponding to the user feedback data as a scene record.

7. The intelligent cockpit air conditioning method as described in claim 6, characterized in that, The method further includes: Obtain the scene features of the driving scenario to be adjusted, and determine the similarity between the scene features and the scene tags in each scene record; The scene record with the highest similarity and whose similarity exceeds a preset similarity threshold is selected as the target scene record; Based on the control parameters recorded in the target scenario, a cabin air conditioning strategy to be executed corresponding to the driving scenario to be adjusted is generated.

8. An intelligent cockpit air conditioning device, characterized in that, The device includes: The data processing module is used to acquire multimodal perception data and perform fusion processing on the multimodal perception data to generate a fusion feature sequence. The multimodal perception data includes at least cabin air quality data, occupant health characteristic data, and driving scenario data. The strategy generation module is used to generate air purification strategies and wind field control strategies based on the fused feature sequence; The strategy adjustment module is used to generate a cabin air conditioning strategy based on the air purification strategy and the wind field control strategy, according to a preset multi-level priority rule. An air conditioning module is used to drive the graded purification components and the scenario-based airflow control components to perform air conditioning in the cabin area according to the cabin air conditioning strategy.

9. An intelligent cockpit air conditioning device, characterized in that, The device includes: a memory, a processor, and an intelligent cockpit air conditioning program stored in the memory and executable on the processor, the intelligent cockpit air conditioning program being configured to implement the steps of the intelligent cockpit air conditioning method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores an intelligent cockpit air conditioning program, which, when executed by a processor, implements the steps of the intelligent cockpit air conditioning method as described in any one of claims 1 to 7.