Intelligent environment adjusting method and device, storage medium and electronic device

Abstract

The application discloses an intelligent environment adjustment method and device, a storage medium and an electronic device. The intelligent environment adjustment method comprises the following steps: collecting an electroencephalogram signal through a brain-computer interface, analyzing the electroencephalogram signal, and generating a target environment adjustment instruction; collecting a current environment parameter through at least one environment sensor, and generating an environment state signal; mapping the target environment adjustment instruction into at least one comfort target, and determining a weight coefficient of each comfort target; selecting a target action from multiple candidate actions through a multi-target utility decision mechanism based on the comfort target, the weight coefficient and the environment state signal; and controlling a corresponding environment adjustment device according to the target action. The application can significantly improve the effectiveness, individualization degree and user satisfaction of environment adjustment in a complex environment.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) technology, specifically to an intelligent environmental control method, device, storage medium, and electronic device, and particularly to an environmental control method suitable for intelligent door and window systems. Background Technology

[0002] With the rapid development of IoT and AI technologies, intelligent environmental control has become a key means to improve the comfort, health, and energy efficiency of residential and office spaces. Especially in intelligent window and door systems, the accuracy and intelligence of environmental control directly affect indoor ventilation, lighting, insulation, and security. Traditional intelligent environmental control solutions mainly rely on preset programs or environmental sensor feedback, aiming to automatically regulate air conditioning, fresh air systems, lighting, windows, and other equipment based on objective physical parameters (such as temperature, humidity, light, and rain) to achieve basic goals such as constant temperature, humidity, oxygen levels, and security.

[0003] However, traditional intelligent environmental control solutions are essentially responses to environmental conditions. Their logic is usually unidirectional and isolated (e.g., controlling air conditioning by sensing the environment through temperature and humidity sensors, or automatically closing windows after detecting rain through rain sensors). They cannot perceive or integrate people's subjective feelings and real-time intentions in the environment, leading to a decoupling of decisions from users' actual needs and reducing the effectiveness, personalization, and user satisfaction of environmental control. Especially in door and window control scenarios, users' subjective intentions (such as wanting to introduce natural wind or avoid noise interference) are often difficult for traditional sensors to accurately capture, resulting in frequent contradictory situations such as "users feeling stuffy after automatically closing windows" or "rain entering due to windows not closing in time." Summary of the Invention

[0004] This application provides an intelligent environmental control method, apparatus, storage medium, and electronic device, which can improve the effectiveness, personalization, and user satisfaction of environmental control.

[0005] In a first aspect, embodiments of this application provide an intelligent environment adjustment method, including: Brain-computer interface is used to collect brain signals and analyze them to generate target environment regulation instructions; The current environmental parameters are collected by at least one environmental sensor to generate an environmental status signal; The target environment adjustment instructions are mapped to at least one comfort objective, and the weighting coefficients of each comfort objective are determined. Based on the comfort objective, the weighting coefficients, and the environmental state signal, a target action is selected from multiple candidate actions through a multi-objective utility decision-making mechanism. The corresponding environmental control equipment is controlled according to the target action.

[0006] In the intelligent environment adjustment method provided in this application embodiment, the step of selecting a target action from multiple candidate actions based on the comfort target, the weighting coefficient, and the environmental state signal through a multi-objective utility decision-making mechanism includes: Based on currently available environmental control equipment, generate multiple candidate actions; For each candidate action, based on a preset device performance model and environmental dynamics model, the change in the environmental state signal after executing the candidate action is predicted. Based on the change amount, the weighting coefficient, and the comfort objective, a multi-objective utility decision-making mechanism is used to select the target action from multiple candidate actions.

[0007] In the intelligent environment adjustment method provided in this application embodiment, the step of selecting a target action from multiple candidate actions based on the change amount, the weighting coefficient, and the comfort target using a multi-objective utility decision mechanism includes: Based on the amount of change, assess the degree to which the candidate action achieves each of the comfort goals; Based on the weight coefficient of each comfort goal and its degree of achievement, the comprehensive utility value of each candidate action is calculated through a utility function. Compare all the combined utility values ​​and determine the target action based on the comparison results.

[0008] In the intelligent environment adjustment method provided in this application embodiment, determining the weight coefficients of each comfort target includes: Based on the environmental state signal, calculate the degree of deviation between its value and the parameters related to each comfort target; The weighting coefficients for each comfort objective are determined based on the degree of deviation of the numerical values.

[0009] In the intelligent environment regulation method provided in this application embodiment, the step of collecting current environmental parameters and generating an environmental state signal through at least one environmental sensor includes: The current environmental parameters are preprocessed to generate a standardized environmental parameter vector; The standardized environmental parameter vector is fused with the generalized environmental data to obtain fused environmental data; Based on the fused environmental data, a structured environmental state signal is constructed.

[0010] In the intelligent environment regulation method provided in this application embodiment, the step of parsing the electroencephalogram (EEG) signal to generate a target environment regulation instruction includes: The EEG signals are preprocessed, and a multi-dimensional feature vector is constructed based on the preprocessed EEG signals. The multi-dimensional feature vector is input into the intent recognition model, and the intent recognition model outputs the corresponding candidate environment adjustment instructions and their confidence levels. The candidate environmental adjustment instructions are processed accordingly based on the confidence level to generate the target environmental adjustment instructions.

[0011] In the intelligent environment adjustment method provided in this application embodiment, the step of processing the candidate environment adjustment instructions based on the confidence level to generate the target environment adjustment instruction includes: If the confidence level is greater than the first threshold, then the candidate environmental adjustment instruction is taken as the target environmental adjustment instruction; If the confidence level is less than the first threshold and greater than the second threshold, the candidate environment adjustment instruction is modified in combination with historical instructions or the current environment context to generate the target environment adjustment instruction. If the confidence level is less than the second threshold, it is determined that the recognition has failed, and a process of requesting secondary confirmation or re-entry is triggered.

[0012] Secondly, embodiments of this application provide an intelligent environmental control device, comprising: The signal parsing unit is used to acquire brainwave signals through the brain-computer interface, parse the brainwave signals, and generate target environment regulation instructions; The parameter acquisition unit is used to acquire current environmental parameters through at least one environmental sensor and generate an environmental status signal. A weighting determination unit is used to map the target environment adjustment command to at least one comfort target and determine the weighting coefficient of each comfort target; An action selection unit is used to select a target action from multiple candidate actions based on the comfort objective, the weighting coefficient, and the environmental state signal through a multi-objective utility decision-making mechanism. An environmental control unit is used to control the corresponding environmental control equipment according to the target action.

[0013] Thirdly, this application provides a storage medium storing a plurality of instructions adapted for loading by a processor to execute the intelligent environment adjustment method described in any of the preceding claims.

[0014] Fourthly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the intelligent environment adjustment method described in any of the above claims.

[0015] In summary, the intelligent environment regulation method provided in this application collects brainwave signals through a brain-computer interface, analyzes the brainwave signals to generate a target environment regulation command; collects current environmental parameters through at least one environmental sensor to generate an environmental state signal; maps the target environment regulation command to at least one comfort goal and determines the weight coefficient of each comfort goal; selects a target action from multiple candidate actions based on the comfort goal, the weight coefficient, and the environmental state signal through a multi-objective utility decision-making mechanism; and controls the corresponding environment regulation device according to the target action. This application, by mapping the target environment regulation command analyzed from brainwave signals to comfort goals and introducing a multi-objective utility decision-making mechanism based on comfort goals, weight coefficients, and environmental state signals, enables the intelligent environment regulation system to comprehensively consider the user's subjective feelings and real-time intentions. Its decision-making is elevated from simply "executing instructions" to "meeting the user's fundamental needs," thereby significantly improving the effectiveness, personalization, and user satisfaction of environment regulation in complex real-world environments. Taking the intelligent door and window system as an example, this application can accurately identify the user's intention to introduce natural wind by opening the window. At the same time, combined with environmental parameters such as outdoor noise, air quality, and weather conditions, it can intelligently decide whether to perform the window opening action or start the fresh air system as an alternative, so as to achieve environmental regulation that truly meets the user's needs. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram illustrating an application scenario of the intelligent environmental control method provided in the embodiments of this application.

[0018] Figure 2 This is a flowchart illustrating the intelligent environmental regulation method provided in the embodiments of this application.

[0019] Figure 3 This is a schematic diagram of the structure of the intelligent environmental control device provided in the embodiments of this application.

[0020] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0022] 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 apparatus 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 apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0023] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0024] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0025] In the description of this application, it should be noted that the terms "upper," "lower," "left," "right," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. In addition, terms such as "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0026] Traditional intelligent environmental control solutions are essentially responses to environmental conditions. Their logic is usually unidirectional and isolated (e.g., controlling air conditioning through temperature and humidity sensors, or controlling window closing through rain sensors). They fail to perceive and integrate people's subjective feelings and real-time intentions within the environment, leading to a decoupling of decisions from users' actual needs and reducing the effectiveness, personalization, and user satisfaction of environmental control. Especially in door and window control scenarios, users' true intentions (such as wanting to introduce natural wind or avoid noise interference) are often difficult for traditional sensors to accurately capture, causing the system to frequently make decisions that contradict user expectations.

[0027] Based on this, embodiments of this application provide an intelligent environmental control method, apparatus, storage medium, and electronic device. Specifically, the intelligent environmental control apparatus can be integrated into an electronic device, which can be a server or a terminal, etc. The terminal can include mobile phones, wearable smart devices, tablets, laptops, and personal computers (PCs), etc., and other computers and auxiliary devices. The server can be a single server or a server cluster composed of multiple servers; it can be a physical server or a virtual server. This electronic device is particularly suitable for the control center of intelligent door and window systems and can work collaboratively with intelligent windows, doors, curtains, and other devices.

[0028] For example, such as Figure 1 As shown, the electronic device can collect brainwave signals via a brain-computer interface, analyze the brainwave signals, and generate target environmental adjustment instructions; it can also collect current environmental parameters through at least one environmental sensor to generate environmental state signals; map the target environmental adjustment instructions to at least one comfort objective and determine the weight coefficients of each comfort objective; based on the comfort objective, weight coefficients, and environmental state signals, it can select a target action from multiple candidate actions through a multi-objective utility decision-making mechanism; and control the corresponding environmental adjustment equipment according to the target action. In the door and window control scenario, the target action could be opening / closing a window, adjusting the window opening, starting a fresh air system, or opening / closing curtains, etc.

[0029] The technical solutions shown in this application will be described in detail below through specific embodiments. It should be noted that the order of description of the following embodiments is not intended to limit the priority of the embodiments.

[0030] Please see Figure 2 , Figure 2 This is a flowchart illustrating the intelligent environment control method provided in this application embodiment. The specific flow of the intelligent environment control method can be as follows: 101. Collect brain signals through brain-computer interface, analyze the brain signals, and generate target environment regulation instructions.

[0031] In this embodiment, the user's electroencephalogram (EEG) signals can be acquired in real time via a brain-computer interface (BCI). In some embodiments, a non-invasive EEG acquisition device worn by the user (such as a dry electrode-based EEG headband or headband) continuously acquires raw electrophysiological signals (i.e., EEG signals) from the cerebral cortex at a certain sampling rate (e.g., 250 Hz or higher). This EEG acquisition device typically includes multiple electrodes to acquire signals from different brain regions (such as the motor cortex, prefrontal cortex, etc.), providing rich spatial information for subsequent user intent recognition. In a door and window control scenario, the user's intent might include "I want to open the window for ventilation," "It's too noisy, close the window," or "The sunlight is too bright, draw the curtains," etc.

[0032] In practice, the step of "analyzing EEG signals and generating target environment regulation instructions" may include the following steps: 1011. Preprocess the EEG signals and construct a multi-dimensional feature vector based on the preprocessed EEG signals.

[0033] Understandably, the purpose of preprocessing EEG signals is to suppress noise, eliminate artifacts, and enhance signal components relevant to the user's intent. Preprocessing steps include, but are not limited to: using bandpass filters (e.g., 0.5-45 Hz) to filter out extremely low-frequency drift and high-frequency electromyographic interference; using Independent Component Analysis (ICA) or adaptive filtering algorithms to identify and remove artifacts caused by eye movements (EOG) and blinking; and applying power frequency notch filtering (e.g., 50Hz / 60Hz) to eliminate power supply interference. After preprocessing, a relatively "clean" EEG signal can be obtained.

[0034] In some embodiments, feature extraction can be performed first based on the preprocessed EEG signal, and then a multi-dimensional feature vector can be constructed based on the extracted features. In specific implementations, feature extraction of the preprocessed EEG signal can be performed from multiple domains: in the time domain, the amplitude, latency, or statistical characteristics (such as mean and variance) of the event-related potentials (ERPs) can be extracted; in the frequency domain, the power spectral density, energy, or power ratio of specific frequency bands (such as delta waves, theta waves, alpha waves, beta waves, and gamma waves) can be calculated using Fast Fourier Transform (FFT) or wavelet transform; in the time-frequency domain, the evolution characteristics of the signal in time and frequency can be analyzed simultaneously using methods such as wavelet packet transform. Finally, these features extracted from different domains are combined into a multi-dimensional feature vector, which serves as the input to the intent recognition model. In door and window control scenarios, the focus can be on brain region signal features related to spatial perception and motor intent.

[0035] 1012. Input the multi-dimensional feature vector into the intent recognition model, and the intent recognition model outputs the corresponding candidate environment adjustment instructions and their confidence scores.

[0036] The intent recognition model can be a deep learning model, such as a hybrid model combining convolutional neural networks (CNNs) and recurrent neural networks (RNNs, such as LSTMs). CNN layers excel at extracting local spatial-temporal patterns of signals, while RNN layers can capture the dynamic time-series dependencies generated by user intent. The intent recognition model is trained on a large amount of labeled data (EEG signals corresponding to specific commands), learning the mapping relationship from multi-dimensional feature vectors to different environmental adjustment commands (such as "open the window for ventilation," "adjust the temperature," "reduce the brightness," etc.). The output of the intent recognition model not only includes the most likely candidate environmental adjustment commands but also provides the confidence score of the prediction, reflecting the degree of certainty the intent recognition model has regarding the current recognition result. For door and window control scenarios, the intent recognition model can be specifically optimized and trained for commands such as "open the window," "close the window," "open the curtains," and "close the curtains."

[0037] 1013. Based on the confidence level, the candidate environmental adjustment instructions are processed accordingly to generate the target environmental adjustment instructions.

[0038] Specifically, if the confidence level is greater than the first threshold, it indicates that the recognition result of the intent recognition model is highly reliable, and the candidate environmental adjustment instruction can be used as the target environmental adjustment instruction.

[0039] For example, if a user strongly desires to "open the window" due to the stuffy heat, and the intent recognition model outputs the "open window for ventilation" instruction with a high confidence of 0.92, it can be adopted immediately.

[0040] If the confidence level is less than the first threshold but greater than the second threshold, it indicates that the recognition result has a certain degree of ambiguity or uncertainty. In this case, the electronic device will not directly adopt the result, but will instead modify the candidate environmental adjustment command by combining historical commands or the current environmental context, thereby generating the target environmental adjustment command.

[0041] The aforementioned current environmental context includes at least the environmental state context, user state context, and spatiotemporal and system context. The environmental state context is a quantified environmental state signal collected and generated in real time by environmental sensors, including physical parameters such as indoor and outdoor temperature and humidity, air quality, light intensity, and noise. The user state context includes the user's historical preference data, recent behavior sequences, and real-time available physiological or behavioral state indicators. The spatiotemporal and system context includes the current time, the spatial location of the user and device, and the real-time operating status of relevant environmental control equipment.

[0042] For example, the intent recognition model outputs the "open window for ventilation" command with a confidence level of 0.75. However, the electronic device, after checking the user's recent command history, finds that the user has attempted to "open the window" twice in the past half hour, even though the current environment indicates the window is already open. Based on this current environmental context, the electronic device can modify the candidate environmental adjustment command to an alternative command related to "ventilation," such as "increase the airflow of the fresh air system," or prompt the user to confirm their true intent. In another scenario, based on current environmental sensor data (e.g., outdoor noise reaching 80 decibels), the electronic device infers that the user's intent is more likely "ventilation" than simply "opening the window," thus modifying or mapping the candidate environmental adjustment command to a more fundamental comfort objective. For yet another example, in a door and window control scenario, if the user's intent is identified as "open the window," but it is raining outside, the system can modify the command to "start dehumidification mode" or "turn on the fresh air system" to replace the action of opening the window.

[0043] If the confidence level is less than the second threshold, it indicates that the intent recognition model has failed to provide a reliable recognition, and the recognition is judged as a failure, triggering a process of requesting secondary confirmation or re-entry.

[0044] In other words, no automatic commands are generated at this time; instead, an interactive process is triggered. For example, a voice prompt may be used to say, "Unable to recognize your intent, please reimagine the control command," or several of the most likely options (such as "Open the window?", "Cool down?", "Ventilate?") may be displayed on the user assistance interface for the user to manually select and confirm, thereby ensuring system security and user control.

[0045] In this embodiment, the first threshold can be set to a higher value, such as 0.85. The second threshold can be set to a lower value, such as 0.60.

[0046] 102. Collect current environmental parameters through at least one environmental sensor to generate an environmental status signal.

[0047] In this embodiment, a multi-layered, multi-dimensional environmental sensing network can be constructed by deploying at least one environmental sensor indoors and outdoors to comprehensively and accurately capture physical conditions affecting user comfort. Environmental parameters include, but are not limited to, temperature, humidity, light intensity, concentration of particulate matter (such as PM2.5 and PM10), concentration of volatile organic compounds, carbon dioxide concentration, and ambient noise level. In door and window control scenarios, special attention should also be paid to raindrop sensors, wind speed sensors, outdoor noise sensors, and door and window status sensors (such as open / closed status and opening degree).

[0048] The outdoor environment sensor can be deployed in a sheltered location on the outside of door and window openings, or integrated into the door and window structure. It is designed with high protection levels, weather-resistant materials, and a robust installation method to ensure long-term reliable data acquisition and extend the device's lifespan in complex outdoor environments. For example, a raindrop sensor can be installed on the outside of the window frame to detect in real time whether rainwater is contacting the window.

[0049] For example, in door and window control scenarios, key sensors may include: temperature sensors that monitor the temperature difference between indoors and outdoors; indoor and outdoor CO2 and PM2.5 sensors that determine air freshness; raindrop sensors on windows that detect rainwater; and outdoor noise sensors that assess external interference. Additionally, door and window magnetic sensors may be included to provide real-time feedback on the opening and closing status of doors and windows.

[0050] In practice, it can be implemented as follows: First, the current environmental parameters are preprocessed to generate standardized environmental parameter vectors.

[0051] It is understandable that current environmental parameters from different sensors and locations may be out of sync in time and differ in units and ranges. Therefore, it is necessary to perform data preprocessing such as timestamp alignment, anomaly detection, cleaning, and normalization on each current environmental parameter.

[0052] Specifically, timestamp alignment can be achieved by unifying the current environmental parameters of all sensors to the main clock reference of the electronic device through interpolation or resampling methods, ensuring that subsequent processing is based on a snapshot of the same moment. Anomaly detection and cleaning can be achieved by using statistical (e.g., the 3σ principle) or machine learning methods to identify anomalous data points (e.g., constant values ​​caused by sensor malfunctions or sudden physically unreliable values) from all current environmental parameters, and then removing them or repairing them through reasonable interpolation based on historical environmental data. Normalization can be achieved by mapping the actual measured values ​​of each current environmental parameter to a unified numerical range (e.g., [0,1]) to eliminate the influence of dimensions. The normalization benchmark can be a preset safe / comfort range (e.g., 0-1 corresponding to a temperature of 18-28℃). After the above data preprocessing, a standardized environmental parameter vector can be generated, where each dimension represents a normalized environmental factor state.

[0053] Next, the standardized environment parameter vector is fused with the generalized environment data to obtain fused environment data.

[0054] Understandably, environmental sensors alone cannot predict certain macroscopic or future environmental changes. Therefore, electronic devices can access or subscribe to generalized environmental data in real time via Internet APIs, for example: Weather forecasts and real-time data: Temperature, humidity, probability of precipitation, and wind speed and direction for the next few hours. This is crucial for deciding whether to open windows and when to close them to prevent rain. For example, even if it's not raining now, but the weather forecast indicates heavy rainfall in an hour, the system can decide to close the windows in advance.

[0055] Regional Air Quality Index (AQI) and Pollutant Details: City-level data on PM2.5, ozone, etc., can serve as a reliable reference for outdoor air quality, especially when local sensors are affected by transient factors.

[0056] Noise maps or traffic information: Understand the baseline noise levels in the surrounding area or the expected noise changes caused by special events (such as nearby construction or large events).

[0057] Pollen concentration forecast: This data can influence ventilation decisions for users with allergies.

[0058] The fusion process involves splicing or weighting standardized environmental parameter vectors with these generalized environmental data at the feature level to form a more comprehensive fused environmental data. For example, even if the instantaneous readings of local outdoor PM2.5 sensors are acceptable, integrating regional AQI early warning data on "pollution spread in the next two hours" will provide a more forward-looking assessment of the environmental situation.

[0059] Finally, based on the fused environmental data, a structured environmental state signal is constructed.

[0060] In some embodiments, the structured environmental state signal may include: Instantaneous state vector: directly reflects the environmental parameters at the current moment. For example: {Indoor temperature: 28℃, Indoor PM2.5: 25μg / m³, Outdoor PM2.5: 150μg / m³, Outdoor noise: 70dB}. In door and window control scenarios, it may also include: window status: 50% open, raindrop sensor status: dry, outdoor wind speed: 3m / s.

[0061] Trend and Forecast Information: Based on time series analysis or combined with generalized environmental data, describe the direction and rate of change of key parameters (such as "indoor temperature is rising rapidly" or "there is a probability of rain in one hour"). For example, it can be predicted that "indoor PM2.5 will continue to rise in the next 30 minutes, and it is recommended to turn on the air purifier."

[0062] Event and Warning Flags: Boolean flags or level information automatically generated according to preset rules. For example: RainAlert=True (rain or high probability of precipitation detected), Air Quality Alert=“Severe” (outdoor PM2.5 > 150µg / m³), Noise Alert=True (outdoor noise consistently above 70dB), Security Alert=True (door and window sensors detect abnormal prying; in this case, comfort adjustments may give way to security locking). For instance, when a rain sensor detects rain and the window is open, a “Rain Intrusion Risk” warning flag can be generated.

[0063] Structured environmental state signals can provide direct input for subsequent intention mapping and decision-making, enabling accurate assessment of specific issues such as whether "opening a window" will lead to rainwater intrusion, whether it can effectively improve air quality, and whether it will introduce intolerable noise, thereby making scientific environmental regulation decisions.

[0064] 103. Map the target environmental adjustment instructions to at least one comfort objective, and determine the weighting coefficient of each comfort objective.

[0065] In this embodiment of the application, an instruction-target mapping table can be preset to define the mapping relationship from target environment adjustment instructions to abstract comfort targets.

[0066] This mapping relationship is key to understanding users' deeper intentions and achieving the transition from "action execution" to "goal fulfillment." For example, a user's "open the window for ventilation" command expressed via a brain-computer interface is parsed into two comfort goals in the mapping table: [improving air quality] and [lowering room temperature]. Similarly, a user's "increase brightness" command when feeling dimly lit might be mapped to the goal of [improving visual comfort]. The mapping table is designed based on a general understanding of human comfort needs and can be fine-tuned using personalized user data. In window and door control scenarios, the "open the window for ventilation" command can be further refined into multiple sub-goals such as [reducing indoor CO2 concentration], [introducing natural wind], and [regulating indoor temperature].

[0067] In some embodiments, the determination of the weighting coefficients for each comfort objective needs to follow an adaptive principle. The weighting coefficients reflect the relative importance and urgency of each comfort objective to the user in the current specific context.

[0068] Specifically, firstly, the deviation of environmental state signals from the relevant parameters of each comfort target can be calculated. These relevant parameters refer to specific environmental physical quantities used to quantify the achievement of the corresponding comfort target. For example, for the target of **improving air quality**, the relevant parameters could be "indoor PM2.5 concentration" and / or "indoor CO2 concentration". For the target of **reducing room temperature**, the relevant parameter is "indoor temperature". For the target of **reducing noise**, the relevant parameter is "indoor ambient noise level in decibels". For the target of **improving visual comfort**, the relevant parameters could be "indoor light intensity" and "glare index". In the window and door control scenario, there may also be a target of **preventing rainwater intrusion**, with relevant parameters being "raindrop sensor status" and "outdoor precipitation probability"; and a target of **preventing strong wind interference**, with relevant parameters being "outdoor wind speed".

[0069] Each comfort target has a preset ideal value or comfort range. The degree of deviation can be obtained by calculating the difference between the actual value and the corresponding ideal value of the relevant parameter in the current environmental state signal. This difference can be an absolute difference, a relative difference, or a standardized score. For example, assuming the ideal temperature for "lowering room temperature" is 24°C, and the current indoor temperature is 30°C, the degree of deviation can be quantified as 6°C or a normalized score (such as 0.8, where a larger value indicates a more severe deviation and a more urgent need).

[0070] Next, the weighting coefficients for each comfort objective can be determined based on the degree of numerical deviation. Generally, a greater degree of numerical deviation indicates less than ideal environmental conditions and a stronger potential demand from users for that comfort objective; therefore, its weighting coefficient should be assigned a higher value. During implementation, linear or nonlinear functions can be used to map the degree of numerical deviation to initial weights.

[0071] For example, in a window and door control scenario, assuming a user issues a "open window for ventilation" command, two comfort goals are mapped: G1 (improving air quality) and G2 (reducing room temperature). The acquired current environmental status signals are: indoor PM2.5 = 20 μg / m³ (good, ideal value <35), indoor temperature = 30℃ (high, ideal value 24℃), outdoor PM2.5 = 180 μg / m³ (severely polluted), outdoor noise = 65 dB (high).

[0072] At this point, we can first calculate the degree of deviation of the values: For G1, the current value of the relevant parameter "indoor PM2.5" is 20, the ideal value is 35, and the degree of deviation is small (which can be quantified as 0.2); for G2, the current value of the relevant parameter "indoor temperature" is 30, the ideal value is 24, and the degree of deviation is large (which can be quantified as 0.9).

[0073] Next, the weighting coefficients for each comfort objective are determined: based on the degree of numerical deviation, a lower initial weight (e.g., 0.2) can be assigned to G1, and a higher initial weight (e.g., 0.8) to G2. This reflects that, under the current environment, cooling is more urgent than ventilation.

[0074] Finally, the weights are adjusted based on environmental constraints: when both severe outdoor pollution and high noise levels are detected simultaneously, opening the window, while potentially contributing to G2 (cooling), would severely impair G1 (air quality) and introduce new discomfort (noise). Therefore, in the final utility decision model, the prediction achievement of opening the window for G1 might be negative, effectively rejecting the simple window-opening option at the decision level. Instead, a high weight can be assigned to the implicit comfort objective of "preventing pollution / noise intrusion," guiding the decision to choose the alternative of "closing the window and starting the air conditioning and fresh air system," thus satisfying both the fundamental needs of cooling and air purification. In the window and door control scenario, the objective of "preventing rainwater intrusion" can also be introduced. If the rain sensor detects rainwater or a high probability of precipitation, the weight of this objective will be significantly increased, and the system will prioritize closing the window.

[0075] In this embodiment, through the above-mentioned mapping and weight allocation mechanism, not only can the surface meaning of user instructions be understood, but also the underlying purpose can be discerned. Combined with real-time environmental data, accurate input can be provided for subsequent decision-making, ensuring that the final target action is the optimal solution that truly meets the user's fundamental intentions and comfort needs.

[0076] In another embodiment, a fixed preset weight allocation can be made for each comfort target based on the type of environmental adjustment instruction.

[0077] In this embodiment, a basic weight table associated with the instruction-target mapping table can be preset to define an initial fixed weight allocation for the comfort target mapped by each type of environmental adjustment instruction, reflecting a general understanding of the primary intent usually implied by different environmental adjustment instructions.

[0078] For example, for commands like "open windows for ventilation," the preset weights might be: 0.7 for improving air quality and 0.3 for lowering room temperature. This reflects that, without additional context (environmental state context, user state context, spatiotemporal and system context), the primary purpose of "ventilation" is air exchange. For commands like "It's too sunny," the preset weights might be: 0.9 for reducing light intensity and 0.1 for preventing overheating. When users are using this command for the first time or lack sufficient personal data, these fixed weights can be used directly as the basis for decision-making. In door and window control scenarios, for the "open window" command, the preset weights for "introduce natural wind" and "improve air quality" could also be 0.5.

[0079] In another embodiment, the weight coefficients of each comfort objective can be dynamically adjusted based on the user's historical preference data.

[0080] In practice, by continuously learning from users' feedback on historical decisions and their subsequent behaviors, a personalized user model can be built and updated, thereby dynamically adjusting the weight coefficients to make decisions increasingly aligned with individual habits.

[0081] Specifically, after an electronic device makes a decision (e.g., activating fresh air ventilation instead of opening windows when there is outdoor pollution), if the user expresses "satisfaction" via voice, app, or EEG signals, or directly cancels the action (manifested as issuing the "open window" command again), the decision result and user feedback can be recorded. Then, the user's habitual choices under specific environmental conditions can be analyzed. For example, historical data shows that whenever the outdoor temperature is suitable in the evening (e.g., 26°C), regardless of air quality, users have an 80% probability of manually or via command opening the window. This indicates that during this time period and under these conditions, the user's preference for [natural ventilation] (which can be associated with the combined goals of [improving air quality] and [exiting the natural environment]) is significantly higher than their concern for [isolation from pollution]. Based on this historical user preference data, machine learning algorithms (such as reinforcement learning or Bayesian updates) can be used to iteratively optimize the weight adjustment strategy for that user. Ultimately, for the same "open window for ventilation" command, under similar conditions, the weight assigned to the user for "improving air quality" might be reduced from the default 0.7, while the weights for "lowering room temperature" or "introducing natural ventilation" would be increased, thus more accurately simulating their personal preferences. In window and door control scenarios, if historical data shows that users still tend to open windows during lightly polluted weather, the weight for "improving air quality" will be relatively reduced, while the weight for "introducing natural ventilation" will be relatively increased.

[0082] It should be noted that in the specific implementation process, a composite strategy can be adopted, combining at least one of the above three methods to generate the final weight. For example: Suppose user A has a long-standing strong preference for fresh air, often choosing to open windows even when there is slight outdoor pollution. One day, user A issues the command to "open windows for ventilation".

[0083] First, a fixed weight (air quality: 0.7, room temperature: 0.3) is used as the baseline weight. Then, based on user A's historical preference data, the weight for "improving air quality" is significantly increased (e.g., an additional personalized bias of 0.2), and the sensitivity weight for the risk factor "outdoor pollution" (i.e., the personalized bias) may be reduced. Simultaneously, the degree of deviation from the calculated values ​​based on the current environmental conditions (e.g., indoor 30°C, outdoor 28°C, and PM2.5 = 120) will assign a higher urgency weight to the goal of "reducing room temperature."

[0084] Ultimately, baseline weights, personalized biases, and urgency weights can be combined (e.g., by weighted averaging) to generate a set of final weighting coefficients tailored to the user at that moment. Decisions made based on this (e.g., still choosing to open the window for a short time because user preference is highly weighted and temperature differences help cool things down) will greatly improve user experience satisfaction.

[0085] 104. Based on comfort objectives, weighting coefficients, and environmental state signals, a target action is selected from multiple candidate actions through a multi-objective utility decision-making mechanism.

[0086] The purpose of this embodiment is to automatically select the optimal execution plan based on understanding the user's fundamental intent (reflected in comfort goals and their weighting coefficients) and grasping the current environmental conditions. This process simulates the human thought process of making trade-off judgments under complex constraints, thereby ensuring that the target action is the result of multi-objective optimization.

[0087] Specifically, step 104 may include the following steps: 1041. Based on the currently available environmental control equipment, generate multiple candidate actions.

[0088] First, all online and available environmental control devices can be inventoried, such as smart window motors, air conditioners, fresh air systems, air purifiers, humidifiers, dehumidifiers, smart doors and windows, and lighting systems. Based on the independent or collaborative capabilities of these devices, multiple candidate actions are generated. In a door and window control scenario, candidate actions may include: completely closing the window, opening the window to 30% / 50% / 100% opening, starting the fresh air system, opening / closing curtains, and starting the air conditioner.

[0089] The candidate actions can include simple operations of a single device (such as "opening the window to 50%" or "setting the air conditioner to 26°C cooling mode"), as well as complex coordinated sequences of multiple devices (such as "first completely closing the window, then simultaneously starting the air conditioner cooling and the fresh air system recirculation"). For example, when high outdoor temperatures and severe pollution are detected, the candidate actions "close the window + start the air conditioner + start the air purifier" can be generated.

[0090] During the generation of candidate actions, preliminary screening can be performed based on device status and basic physical logic to exclude actions that are obviously infeasible or contradictory (for example, if a window is detected to be locked or a security alarm is present, an "open window" action will not be generated). For instance, if a window sensor indicates that a window is damaged or stuck, no action involving that window will be generated.

[0091] 1042. For each candidate action, based on the preset equipment performance model and environmental dynamics model, predict the amount of change in the environmental state signal after executing the candidate action.

[0092] In some embodiments, a device performance model and an environmental dynamics model can be preset in advance, so that the effect after execution of each candidate action can be predicted in advance.

[0093] The equipment performance model describes the capabilities and characteristics of the equipment, such as: the cooling / heating power, energy consumption, and outlet air temperature of an air conditioner under different setting conditions; the air volume, purification efficiency, and heat exchange efficiency of a fresh air system; and the ventilation rate, sound insulation effect, and thermal conductivity of windows at different opening degrees. In door and window control scenarios, the equipment performance model may also include the response time, maximum opening degree, and operating noise of the window motor.

[0094] Environmental dynamics models describe the interactions and changing patterns of environmental parameters, such as: differential equations or data-driven models showing how indoor temperature is affected by outdoor temperature, sunlight, equipment heating / cooling, and human activity; models showing the attenuation of indoor pollutant concentrations as ventilation and purification equipment operates; and the propagation and attenuation characteristics of noise in space. In window and door control scenarios, environmental dynamics models can also describe the relationship between the rate of indoor and outdoor air exchange after windows are opened, the probability of rainwater entering, and wind speed / opening degree.

[0095] In practical implementation, the current environmental state signal (as initial conditions) and the candidate actions to be evaluated (as input control variables) can be input into the above model. Through simulation calculations, the changing trends and steady-state values ​​of key parameters of the environmental state within a future decision-making cycle (e.g., the next 5-10 minutes) can be predicted, thereby obtaining the predicted changes in the environmental state. For example, it can be predicted that after the "open windows to 50%" action is executed for 5 minutes, the indoor PM2.5 concentration will rise from the current value to near the outdoor level, the indoor temperature will approach the outdoor temperature by a certain margin, and the indoor noise level will increase. In the door and window control scenario, it can also be predicted that after the "open windows to 50%" action is executed, if the outdoor wind speed is 3 m / s, the probability of rainwater entering is 20%.

[0096] 1043. Based on the change quantity, weight coefficient and comfort objective, a multi-objective utility decision-making mechanism is adopted to select the target action from multiple candidate actions.

[0097] In practice, step 1043 may include the following steps: 1043a. Based on the amount of change, assess the degree to which the candidate actions achieve each comfort goal.

[0098] Specifically, the amount of change can be associated with each comfort goal. For each comfort goal, the improvement of the predicted value of its relevant parameters relative to the ideal value after performing the candidate action is calculated and quantified into an achievement score (usually between 0 and 1, where 1 indicates complete achievement, 0 indicates no improvement, and negative values ​​indicate deterioration).

[0099] For example, for the comfort objective of "reducing room temperature," the relevant parameter is indoor temperature. If the predicted action can reduce the room temperature from 30°C to 26°C (the ideal value is 24°C), then the achievement score might be (30-26) / (30-24) = 0.67. If the predicted action will raise the room temperature to 31°C, then the score is negative. In the door and window control scenario, for the objective of "preventing rainwater intrusion," if the predicted action of opening the window will cause rainwater to drift in, then the achievement score is negative (e.g., -0.5); if the predicted action of closing the window will completely prevent rainwater intrusion, then the achievement score is 1.0.

[0100] 1043b. Based on the weight coefficient of each comfort goal and its degree of achievement, calculate the comprehensive utility value of each candidate action through a utility function.

[0101] In this embodiment, a multi-objective utility function can be used to integrate the evaluation results from different dimensions. This multi-objective utility function can sum the achievement scores of each comfort goal according to their corresponding weight coefficients to obtain a scalarized comprehensive utility value. Specifically: Comprehensive utility value = Σ(weight coefficient_i × achievement level_i).

[0102] To be more comprehensive, this multi-objective utility function can also introduce a cost penalty term (such as the estimated energy consumption of candidate actions, equipment wear and tear costs, or potential new discomforts such as increased noise). This cost penalty term is subtracted from the overall utility value to form the net overall utility value. That is: Net Overall Utility Value = Σ(Weight Coefficient_i × Degree of Achievement_i) - Cost Penalty Term. In the door and window control scenario, the cost penalty term may include motor operating energy consumption, equipment wear and tear caused by frequent window opening and closing, and discomfort caused by introduced noise.

[0103] 1043c. Compare all the combined utility values ​​and determine the target action based on the comparison results.

[0104] After obtaining the (net) overall utility value of all candidate actions, they can be ranked. Typically, the candidate action with the highest overall utility value is selected as the target action. This indicates that, in the current environment, this target action best comprehensively and efficiently satisfies the user's weighted comfort needs.

[0105] For example, in a door and window control scenario, assuming the target environmental adjustment command is "ventilation," it is mapped to comfort objectives G1 [improve air quality] (weight 0.6) and G2 [reduce room temperature] (weight 0.4). The environmental state signals are: stuffy indoors (30°C), air quality fair (PM2.5=30), cool outdoors (26°C) but heavily polluted (PM2.5=200) with traffic noise (70dB). Additionally, the rain sensor status is dry, and the outdoor wind speed is 2m / s.

[0106] Candidate action A (opening the window): Predicted to significantly reduce temperature (achievement level of 0.9 for G2), but will severely deteriorate indoor air quality and introduce noise (achievement level of -1.0 for G1, high cost). Overall utility value: 0.6×(-1.0)+0.4×0.9-high cost≈very low negative value.

[0107] Candidate action B (starting the fresh air system and closing the windows): Predicted to effectively purify the air (achievement level of G1: 0.8), but with minimal contribution to cooling and possibly slight heat generation (achievement level of G2: 0.1). Overall utility value: 0.6 × 0.8 + 0.4 × 0.1 = 0.52.

[0108] Candidate action C (turn on the air conditioner and close the windows): Predicted to provide a strong cooling effect (achievement level of 0.95 for G2), but will not improve air quality (achievement level of 0.0 for G1). Overall utility value: 0.6 × 0.0 + 0.4 × 0.95 = 0.38.

[0109] Candidate action D (open windows 30% and activate the fresh air system's internal circulation): Predicted to slightly lower the temperature (achievement level of G2: 0.3), while partially improving air quality (achievement level of G1: 0.5), and introducing less noise (lower cost). Overall utility value: 0.6 × 0.5 + 0.4 × 0.3 - low cost = 0.42.

[0110] Comparing the three candidate actions, candidate action B has the highest overall utility value (0.52) and is therefore selected as the target action. Through the above multi-objective utility decision-making mechanism, this embodiment ensures that the selection of environmental adjustment actions is optimal and highly consistent with the user's true intentions and the reality of the environment.

[0111] 105. Control the corresponding environmental control equipment according to the target action.

[0112] This embodiment converts the target action into specific equipment control commands and drives the corresponding environmental control equipment to complete the corresponding actions, thereby achieving actual regulation of the physical environment.

[0113] In some embodiments, the target action can be converted into specific device control commands via home IoT protocols (such as Wi-Fi, Zigbee, Matter) and sent to the corresponding environmental control devices for execution. In door and window control scenarios, these commands can be sent to devices such as smart window motors, curtain motors, fresh air systems, and air conditioners.

[0114] For example, a command to "start to high power" will be sent to the "fresh air system," and a command to "keep closed" or "closed" will be sent to the "smart window motor." For example, if the option is to open the curtains, a command to "open to 100%" will be sent to the "smart curtain motor." Or, if the target action is "open the window to 50%," a command to "open to 50%" will be sent to the smart window motor.

[0115] After execution, the system can provide brief feedback to the user regarding the action and reason through a voice assistant, mobile app notification, or visual feedback interface via brain-computer interface. For example, "Severe outdoor air pollution has been detected, and the fresh air system has been activated to purify the air instead of opening the window," thereby enhancing system transparency and user trust. In door and window control scenarios, feedback can also be provided such as, "It is raining outside, and the windows have been automatically closed and dehumidification mode has been activated." In summary, the intelligent environment regulation method provided in this application collects brainwave signals through a brain-computer interface, analyzes the brainwave signals, and generates a target environment regulation command; collects current environmental parameters through at least one environmental sensor to generate an environmental state signal; maps the target environment regulation command to at least one comfort goal and determines the weight coefficient of each comfort goal; selects a target action from multiple candidate actions based on the comfort goals, weight coefficients, and environmental state signal through a multi-objective utility decision-making mechanism; and controls the corresponding environment regulation equipment according to the target action. This application first analyzes the user's EEG signals through a brain-computer interface to directly perceive their subjective feelings and real-time intentions, generating target environment adjustment instructions, thus overcoming the limitation of traditional systems that only respond to environmental states. Second, it maps these instructions to abstract comfort goals and determines their weighting coefficients, achieving an understanding and quantification of the user's deep needs and breaking the coupling between specific actions and fundamental objectives. Third, it combines the generated multi-dimensional environmental state signals with a multi-objective utility decision-making mechanism, simultaneously considering the user's intention weights and the real-time environmental context during decision-making. By calculating the degree to which candidate actions achieve various comfort goals and their comprehensive utility value, it proactively selects the target action that maximizes the satisfaction of the user's fundamental needs, thereby overcoming the shortcomings of traditional unidirectional, isolated control logic. Finally, it executes the target action, ensuring that the environmental adjustment results accurately match the user's personalized needs. This closed-loop process achieves a fundamental shift from "environment state-driven" to "user intention and environment fusion-driven," significantly improving the effectiveness, personalization, and user satisfaction of environmental adjustment. Taking intelligent door and window systems as an example, this application can intelligently weigh user intentions (such as "I want ventilation") against environmental constraints (such as rain, noise, and pollution) and select the best execution plan (such as slightly opening windows and starting fresh air, or completely closing windows and starting air conditioning), truly achieving "people-oriented" intelligent environmental regulation.

[0116] To facilitate better implementation of the intelligent environmental control method provided in this application, this application also provides an intelligent environmental control device. The meanings of the terms used are the same as in the intelligent environmental control method described above, and specific implementation details can be found in the descriptions within the method embodiments.

[0117] Please see Figure 3 , Figure 3 This is a schematic diagram of the intelligent environment control device provided in an embodiment of this application. The intelligent environment control device may include a signal analysis unit 201, a parameter acquisition unit 202, a weight determination unit 203, an action selection unit 204, and an environment control unit 205. The signal parsing unit 201 is used to collect brain signals through the brain-computer interface, parse the brain signals, and generate target environment regulation instructions; The parameter acquisition unit 202 is used to acquire current environmental parameters through at least one environmental sensor and generate an environmental status signal. The weight determination unit 203 is used to map the target environment adjustment command to at least one comfort target and determine the weight coefficient of each comfort target; Action selection unit 204 is used to select a target action from multiple candidate actions based on comfort goals, weight coefficients and environmental state signals through a multi-objective utility decision-making mechanism; The environmental control unit 205 is used to control the corresponding environmental control equipment according to the target action.

[0118] For specific implementation methods of each of the above units, please refer to the embodiments of the intelligent environment adjustment method described above, which will not be repeated here.

[0119] In summary, the intelligent environment adjustment device provided in this application embodiment can: a signal analysis unit 201 collects brainwave signals through a brain-computer interface and analyzes the brainwave signals to generate a target environment adjustment command; a parameter acquisition unit 202 collects current environmental parameters through at least one environmental sensor to generate an environmental state signal; a weight determination unit 203 maps the target environment adjustment command to at least one comfort goal and determines the weight coefficient of each comfort goal; an action selection unit 204 selects a target action from multiple candidate actions based on the comfort goal, weight coefficient, and environmental state signal through a multi-objective utility decision-making mechanism; and an environment adjustment unit 205 controls the corresponding environment adjustment equipment according to the target action. This application embodiment maps the target environment adjustment command obtained from brainwave signal analysis to comfort goals and introduces a multi-objective utility decision-making mechanism based on comfort goals, weight coefficients, and environmental state signals, enabling the intelligent environment adjustment system to comprehensively consider the user's subjective feelings and real-time intentions. Its decision-making process evolves from simply "executing commands" to "meeting the user's fundamental needs," thereby significantly improving the effectiveness, personalization, and user satisfaction of environment adjustment in complex real-world environments.

[0120] This application also provides an electronic device that may integrate the intelligent environmental control device of this application embodiment, such as... Figure 4 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 301 with one or more processing cores and a memory 302 with one or more computer-readable storage media. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs stored in the memory 302 and / or this application, and by calling data stored in the memory 302, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operation of the storage medium, user interface, and application programs, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 301.

[0121] The memory 302 can be used to store software programs and the intelligent environment adjustment method of this application. The processor 301 executes various functional applications and data processing by running the software programs stored in the memory 302 and the intelligent environment adjustment method of this application. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store applications required for operating the storage medium and at least one function; the data storage area may store data created according to the use of the electronic device. In addition, the memory 302 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0122] Although not shown, the electronic device may also include a display unit, an input unit, and a power supply, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 301 in the electronic device loads the executable files corresponding to the processes of one or more application programs into the memory 302 according to the following instructions, and the processor 301 runs the application programs stored in the memory 302 to realize various functions, as follows: Brain-computer interface is used to collect brain signals, analyze the brain signals, and generate target environment regulation instructions; The current environmental parameters are collected by at least one environmental sensor to generate an environmental status signal; The target environmental adjustment instructions are mapped to at least one comfort objective, and the weighting coefficients of each comfort objective are determined. Based on comfort objectives, weighting coefficients, and environmental state signals, a target action is selected from multiple candidate actions through a multi-objective utility decision-making mechanism. Control the corresponding environmental control equipment according to the target action.

[0123] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0124] Therefore, embodiments of this application provide a storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the methods provided in embodiments of this application. For example, the instructions can execute the following steps: Brain-computer interface is used to collect brain signals, analyze the brain signals, and generate target environment regulation instructions; The current environmental parameters are collected by at least one environmental sensor to generate an environmental status signal; The target environmental adjustment instructions are mapped to at least one comfort objective, and the weighting coefficients of each comfort objective are determined. Based on comfort objectives, weighting coefficients, and environmental state signals, a target action is selected from multiple candidate actions through a multi-objective utility decision-making mechanism. Control the corresponding environmental control equipment according to the target action.

[0125] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0126] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0127] Since the instructions stored in the storage medium can execute the steps of any method provided in the embodiments of this application, the beneficial effects that any method provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0128] The above provides a detailed description of the intelligent environmental adjustment method, device, storage medium, and electronic device provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of this application. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for intelligent environmental regulation, characterized in that, include: Brain-computer interface is used to collect brain signals and analyze them to generate target environment regulation instructions; The current environmental parameters are collected by at least one environmental sensor to generate an environmental status signal; The target environment adjustment instructions are mapped to at least one comfort objective, and the weighting coefficients of each comfort objective are determined. Based on the comfort objective, the weighting coefficients, and the environmental state signal, a target action is selected from multiple candidate actions through a multi-objective utility decision-making mechanism. The corresponding environmental control equipment is controlled according to the target action.

2. The intelligent environmental control method as described in claim 1, characterized in that, The step of selecting a target action from multiple candidate actions based on the comfort objective, the weighting coefficients, and the environmental state signal through a multi-objective utility decision-making mechanism includes: Based on currently available environmental control equipment, generate multiple candidate actions; For each candidate action, based on a preset device performance model and environmental dynamics model, the change in the environmental state signal after executing the candidate action is predicted. Based on the change amount, the weighting coefficient, and the comfort objective, a multi-objective utility decision-making mechanism is used to select the target action from multiple candidate actions.

3. The intelligent environmental control method as described in claim 2, characterized in that, The step of selecting a target action from multiple candidate actions based on the change amount, the weighting coefficient, and the comfort objective using a multi-objective utility decision-making mechanism includes: Based on the amount of change, assess the degree to which the candidate action achieves each of the comfort goals; Based on the weight coefficient of each comfort goal and its degree of achievement, the comprehensive utility value of each candidate action is calculated through a utility function. Compare all the combined utility values ​​and determine the target action based on the comparison results.

4. The intelligent environment control method as described in claim 1, characterized in that, The determination of the weighting coefficients for each of the comfort targets includes: Based on the environmental state signal, calculate the degree of deviation between its value and the parameters related to each comfort target; The weighting coefficients for each comfort objective are determined based on the degree of deviation of the numerical values.

5. The intelligent environmental control method as described in claim 1, characterized in that, The step of collecting current environmental parameters and generating an environmental state signal through at least one environmental sensor includes: The current environmental parameters are preprocessed to generate a standardized environmental parameter vector; The standardized environmental parameter vector is fused with the generalized environmental data to obtain fused environmental data; Based on the fused environmental data, a structured environmental state signal is constructed.

6. The intelligent environmental control method as described in claim 1, characterized in that, The step of parsing the EEG signals to generate target environment regulation instructions includes: The EEG signals are preprocessed, and a multi-dimensional feature vector is constructed based on the preprocessed EEG signals. The multi-dimensional feature vector is input into the intent recognition model, and the intent recognition model outputs the corresponding candidate environment adjustment instructions and their confidence levels. The candidate environmental adjustment instructions are processed accordingly based on the confidence level to generate the target environmental adjustment instructions.

7. The intelligent environmental control method as described in claim 6, characterized in that, The step of processing the candidate environmental adjustment instructions based on the confidence level to generate the target environmental adjustment instructions includes: If the confidence level is greater than the first threshold, then the candidate environmental adjustment instruction is taken as the target environmental adjustment instruction; If the confidence level is less than the first threshold and greater than the second threshold, the candidate environment adjustment instruction is modified in combination with historical instructions or the current environment context to generate the target environment adjustment instruction. If the confidence level is less than the second threshold, it is determined that the recognition has failed, and a process of requesting secondary confirmation or re-entry is triggered.

8. An intelligent environmental control device, characterized in that, include: The signal parsing unit is used to acquire brainwave signals through the brain-computer interface, parse the brainwave signals, and generate target environment regulation instructions; The parameter acquisition unit is used to acquire current environmental parameters through at least one environmental sensor and generate an environmental status signal. A weighting determination unit is used to map the target environment adjustment command to at least one comfort target and determine the weighting coefficient of each comfort target; An action selection unit is used to select a target action from multiple candidate actions based on the comfort objective, the weighting coefficient, and the environmental state signal through a multi-objective utility decision-making mechanism. An environmental control unit is used to control the corresponding environmental control equipment according to the target action.

9. A storage medium, characterized in that, The storage medium stores multiple instructions, which are adapted for loading by a processor to execute the intelligent environment adjustment method according to any one of claims 1-7.

10. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the intelligent environmental control method as described in any one of claims 1-7.

Images