Context awareness voice control lighting soothing method and system

Abstract

The invention discloses a context awareness voice control lighting relieving method and system, and relates to the technical field of intelligent lighting control, and the method comprises the steps: obtaining a user voice instruction and environment data; performing semantic analysis on the user voice instruction, and extracting a control intention and a parameter requirement from the user voice instruction; determining target lighting parameters according to the control intention, the parameter requirements and the environment data; obtaining a current lighting state, and calculating a transition path from the current lighting state to the target lighting parameters; the transition path comprises a gradual change sequence of illumination parameters; and controlling the lighting equipment to perform lighting adjustment according to the transition path. According to the method, through technologies of multi-dimensional environment data fusion, deep semantic analysis, smooth transition control and the like, precise matching of the illumination parameters and user requirements is realized, and the user comfort and the system intelligence level are greatly improved.

Description

Technical Field

[0001] This application relates to the field of intelligent lighting control technology, and particularly to a method and system for context-aware voice control of soothing lighting. Background Art

[0002] Intelligent lighting systems play an important role in modern home environments, and their development has undergone multiple stages of evolution from simple switch control and brightness adjustment to the integration of intelligent technologies. Traditional lighting systems were mainly controlled through physical switches or remote controls, and subsequently, automated control based on time or occupancy detection was introduced. In recent years, with the maturity of Internet of Things and artificial intelligence technologies, voice interaction has become the mainstream method for intelligent lighting control. Current voice-controlled lighting systems typically adopt a command-word recognition-based approach, mapping user voice commands to preset control actions and achieving control through simple operations such as switching, dimming, or scene switching. At the same time, some advanced systems have begun to integrate environmental perception functions, such as automatically adjusting the lighting intensity by detecting the ambient brightness through a light sensor, or adjusting lighting parameters based on external data such as time and weather. However, most of these systems adopt a fragmented technical architecture, with the voice control and environmental perception modules being independent of each other and lacking an effective information fusion mechanism.

[0003] Although existing intelligent lighting systems have made significant progress in terms of convenience, there are still many technical limitations. First, voice recognition systems are usually designed based on fixed command words and have limited natural language understanding capabilities, making it difficult to accurately extract the control intent in complex or ambiguous user instructions. Second, the environmental perception ability is one-sided, and most systems only focus on light intensity, ignoring multi-dimensional environmental information such as temperature, humidity, and user activity status, resulting in a deviation between lighting adjustment and actual needs. Third, there is a lack of a transition mechanism for lighting parameter adjustment, and direct switching causes visual stimulation and discomfort. In addition, existing systems are difficult to effectively fuse the voice control results with environmental perception data and cannot achieve intelligent decision-making based on comprehensive contexts. Existing context-aware and voice-controlled lighting systems usually suffer from problems such as insufficient semantic understanding, single environmental perception, and rigid lighting adjustment, making it difficult to provide a natural and comfortable user experience. Summary of the Invention

[0004] In view of the above problems, this application is proposed.

[0005] Therefore, this application provides a method and system for context-aware voice control of soothing lighting, which can solve the problems mentioned in the background art.

[0006] To solve the above technical problems, this application provides the following technical solutions:

[0007] In a first aspect, the present application provides a context-aware voice control soothing lighting method, including: obtaining a user voice command and environmental data, where the environmental data includes light intensity, temperature, humidity, and user activity status; performing semantic analysis on the user voice command to extract a control intention and parameter requirements from the user voice command; determining target lighting parameters according to the control intention, the parameter requirements, and the environmental data, where the target lighting parameters include brightness, color temperature, and light distribution; obtaining the current lighting state and calculating a transition path from the current lighting state to the target lighting parameters; the transition path includes a gradual change sequence of lighting parameters; and controlling a lighting device to perform lighting adjustment according to the transition path.

[0008] Preferably, the performing semantic analysis on the user voice command to extract a control intention and parameter requirements from the user voice command includes: performing speech recognition on the user voice command to convert a voice signal into text data; performing natural language processing on the text data to identify a command type; if the command type is a direct instruction, directly extracting control parameters in the text data as the parameter requirements; if the command type is a context instruction, converting the context instruction into corresponding control intention and default parameter requirements according to a preset context mapping relationship.

[0009] Preferably, the determining target lighting parameters according to the control intention, the parameter requirements, and the environmental data includes: determining an environmental lighting reference value and a current space activity type based on the environmental data; obtaining basic lighting parameters from a preset lighting scene library according to the current space activity type; adjusting the basic lighting parameters according to the control intention and the parameter requirements; and correcting the adjusted lighting parameters in combination with the environmental lighting reference value to obtain the target lighting parameters.

[0010] Preferably, the adjusting the basic lighting parameters according to the control intention and the parameter requirements includes: obtaining user historical lighting preference data, where the user historical lighting preference data includes records of lighting parameter selections by the user under different times and different environmental conditions; establishing a user preference model according to the user historical lighting preference data, and storing the user preference model in a user preference database; adjusting the basic lighting parameters in combination with the user preference model to generate preliminary adjusted parameters; and matching the preliminary adjusted parameters with the control intention and the parameter requirements to determine the adjusted lighting parameters.

[0011] Preferably, the obtaining of the current lighting state and calculating the transition path from the current lighting state to the target lighting parameters include: obtaining the working state of the current lighting device, including the current brightness, current color temperature, and current light distribution; calculating the difference between the working state of the current lighting device and the target lighting parameters; determining the transition time according to the magnitude of the difference and the user activity state; if the difference is greater than a first preset threshold, generating a segmented transition path, dividing the transition time into multiple time periods, and generating a gradient lighting parameter sequence within each time period; if the difference is less than or equal to the first preset threshold, generating a linear transition path and uniformly changing the lighting parameters within the transition time.

[0012] Preferably, the determining of the transition time according to the magnitude of the difference and the user activity state includes: during the execution of the lighting adjustment process, real-time monitoring of the user's feedback information on the lighting adjustment; the feedback information including voice evaluation or manual adjustment behavior; calculating the user comfort value according to the feedback information; when the user comfort value is lower than a second preset threshold, increasing the transition time in real-time and prolonging the gradient process of the remaining lighting parameters; when the user comfort value is not lower than the second preset threshold, maintaining or reducing the transition time; storing the feedback information and the corresponding environmental data, control intention, and lighting parameters in the user preference database for updating the user preference model during the next lighting adjustment.

[0013] Preferably, the obtaining of the user voice command and environmental data includes: collecting environmental sensor data and determining the number of users in the environment; if there is a single user in the environment, directly obtaining the voice command of the single user and the environmental data; if there are multiple users in the environment, obtaining the identity information and location information of each user, and determining the user priority according to the preset user priority rule and the location information of each user; in the order from high to low of the user priority, collecting the voice command of the user with the highest priority, and simultaneously obtaining the environmental data related to the user with the highest priority; when detecting a change in the user's location, recalculating the user priority and correspondingly adjusting the voice command collection target and the environmental data acquisition range.

[0014] In a second aspect, the present application also provides a context-aware voice control soothing lighting system, including: a data collection module for obtaining user voice commands and environmental data; a semantic analysis module for semantically analyzing the user voice commands and extracting control intentions and parameter requirements from the user voice commands; a parameter determination module for determining target lighting parameters according to the control intentions, the parameter requirements, and the environmental data; a path calculation module for obtaining the current lighting state and calculating the transition path from the current lighting state to the target lighting parameters; and a lighting control module for controlling the lighting device to perform lighting adjustment according to the transition path.

[0015] In a third aspect, the present application further provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: obtaining a user voice instruction and environmental data, where the environmental data includes light intensity, temperature, humidity, and user activity status; performing semantic analysis on the user voice instruction to extract a control intention and parameter requirements from the user voice instruction; determining target lighting parameters according to the control intention, the parameter requirements, and the environmental data, where the target lighting parameters include brightness, color temperature, and light distribution; obtaining the current lighting state and calculating a transition path from the current lighting state to the target lighting parameters; the transition path includes a gradual change sequence of lighting parameters; and controlling a lighting device to perform lighting adjustment according to the transition path.

[0016] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: obtaining a user voice instruction and environmental data, where the environmental data includes light intensity, temperature, humidity, and user activity status; performing semantic analysis on the user voice instruction to extract a control intention and parameter requirements from the user voice instruction; determining target lighting parameters according to the control intention, the parameter requirements, and the environmental data, where the target lighting parameters include brightness, color temperature, and light distribution; obtaining the current lighting state and calculating a transition path from the current lighting state to the target lighting parameters; the transition path includes a gradual change sequence of lighting parameters; and controlling a lighting device to perform lighting adjustment according to the transition path.

[0017] Implementing the present application has the following beneficial effects: The present application provides a context-aware voice control soothing lighting method and system. By obtaining multi-dimensional environmental data including light intensity, temperature, humidity, and user activity status and combining voice instruction analysis, it breaks through the limitation of a single data source in traditional lighting control systems and realizes a comprehensive perception of the user's situation. Semantic analysis can not only process direct and clear control instructions but also parse complex sentences containing implicit requirements, accurately converting the user's natural expression into control intentions and parameter requirements that can be understood by the system, improving the naturalness of human-computer interaction. The present application integrates control intentions, parameter requirements, and environmental data, and determines the target lighting parameters most suitable for the current environment and user needs through a context understanding algorithm, realizing the context adaptability of lighting control. In addition, the present application introduces a gradual transition mechanism to calculate a smooth transition path from the current lighting state to the target parameters, ensuring that lighting changes do not cause visual stimulation, thereby solving the problem of user discomfort caused by the abrupt parameter switching in traditional lighting systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0019] Figure 1 It is an application scenario diagram of the context-aware voice control soothing lighting method involved in the present application;

[0020] Figure 2 It is the overall flowchart of the context-aware voice control soothing lighting method involved in the present application;

[0021] Figure 3 It is the overall structural schematic diagram of the context-aware voice control soothing lighting system involved in the present application;

[0022] Figure 4 It is the computer device diagram of the context-aware voice control soothing lighting method involved in the present application. Specific Embodiments

[0023] In order to make the purpose, technical solutions and advantages of the present application more clear, the following further details the present application in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] The context-aware voice control soothing lighting system has broad application prospects in the field of smart home. For example, it can be applied to various scenarios such as smart home, hotel rooms, office spaces, and elderly care facilities. When the user expresses lighting needs through voice or the system senses environmental changes, the context-aware lighting system can combine the user's intention and environmental data to provide the most suitable soothing lighting effect for the current situation.

[0025] In related technologies, traditional lighting control systems usually can only recognize simple voice commands and execute preset lighting control actions. However, in actual use, users often have context-based lighting needs. How to make the lighting system understand the user's deep intentions and combine environmental data to provide a comfortable and natural lighting experience is a relatively complex problem.

[0026] Aiming at the problem that the related technologies cannot provide intelligent soothing lighting control according to user voice commands and environmental situations, the present application proposes a context-aware voice control soothing lighting method. By obtaining user voice commands and environmental data, semantic analysis is performed on the user voice commands to extract control intentions and parameter requirements, the target lighting parameters are determined in combination with environmental data, and a smooth transition path is calculated, and finally a soothing lighting effect adjustment is realized.

[0027] The context-aware voice control soothing lighting method provided by the embodiments of the present application can be applied to, for example, Figure 1 the application environment shown in the figure. Among them, the intelligent terminal 102 communicates with the control server 104 through the network. The environmental data acquisition system 108 and the lighting control system 106 are connected to the server 104. The environmental data acquisition system can be integrated in the intelligent terminal 102 or deployed as an independent sensor network. The intelligent terminal obtains the user voice command and environmental data, performs semantic analysis on the user voice command, extracts the control intention and parameter requirements from the user voice command; determines the target lighting parameters according to the control intention, parameter requirements and environmental data; obtains the current lighting state, calculates the transition path from the current lighting state to the target lighting parameters; and controls the lighting device to perform lighting adjustment according to the transition path.

[0028] Among them, the intelligent terminal 102 can be, but is not limited to, various intelligent speakers, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things device can be a smart home center, a smart switch panel, a smart TV, a smart lighting controller integrated with voice function, etc. The portable wearable device can be a smart watch, a smart bracelet, smart glasses, etc. The control server 104 can be an independent physical server, an edge computing server integrated in a local gateway device, or a cloud server providing cloud computing services. The environmental data acquisition system includes various sensing devices such as a light sensor, a temperature and humidity sensor, a motion sensor, and a human presence sensor. The lighting control system includes devices such as LED lamps, dimmers, and lighting control panels that support intelligent control.

[0029] In an exemplary embodiment, as Figure 2 shown, a context-aware voice control soothing lighting method is provided. Taking the terminal in Figure 1 as an example, the method includes the following steps 204 to 212. Among them:

[0030] Step 204, obtain the user voice command and environmental data.

[0031] Among them, the environmental data at least includes light intensity, temperature, humidity and user activity status. The light intensity can be obtained through a light sensor and is used to measure the current light intensity in the environment, and the unit is lux. The temperature and humidity can be obtained through a temperature and humidity sensor and are used to measure the real-time temperature (°C) and relative humidity (%) of the environment. The user activity status can be obtained through a human body sensor, a camera or other motion sensors and is used to judge different activity types of the user, such as stationary, walking, reading, watching movies, etc.

[0032] Exemplarily, the acquisition of environmental data can be achieved through a sensor network arranged everywhere in the space for real-time collection. These sensors can be independent devices or integrated into smart home hubs, smart speakers, or lighting controllers. The frequency of environmental data collection can be set according to the actual application scenario, for example, collecting once every 5 seconds, or triggering collection when a significant change in environmental conditions is detected.

[0033] User voice commands can be collected through a microphone array, and preprocessing technologies such as noise reduction and voice enhancement are used to improve the clarity and accuracy of voice commands.

[0034] In the embodiments of the present application, the steps of obtaining user voice commands and environmental data specifically include:

[0035] Collect environmental sensor data and determine the number of users in the environment. The determination of the number of users in the environment can be achieved in various ways, such as infrared human presence sensors, cameras combined with computer vision technology for people counting, user device identification based on Wi-Fi or Bluetooth signals, floor pressure sensors, etc. By fusing multiple sensor data, the accuracy of user number detection can be improved.

[0036] If there is a single user in the environment, directly obtain the voice commands of the single user and the environmental data. In a single-user scenario, it can be default to associate all sensor data in the environment with this user and directly process the user's voice commands without user identification and priority calculation.

[0037] If there are multiple users in the environment, obtain the identity information and location information of each user, and determine the user priority according to the preset user priority rules and the location information of each user.

[0038] Among them, the acquisition of user identity information can be achieved in various ways, including but not limited to: voice recognition technology for voiceprint identification; automatic pairing and recognition of user portable devices (such as smartphones, smartwatches) with lighting controllers; face recognition technology; users actively reporting their identities (such as through specific voice passwords).

[0039] User location information can be obtained through indoor positioning technology, including but not limited to: sound source localization based on multiple microphone arrays; position tracking of cameras combined with computer vision; infrared sensor arrays; triangulation of ultrasonic or radio frequency signals; position data provided by user devices (such as smartphones). The position information can be represented as three-dimensional coordinates in space or the distance and direction relative to lighting devices.

[0040] The preset user priority rules can include multiple dimensions, for example:

[0041] User role priority: Controller administrator > Family member > Visitor;

[0042] Location priority: The user closer to the controlled lighting device has a higher priority than the user farther away;

[0043] Historical priority: The user who issued the lighting control instruction most recently has the priority control right within a short period of time;

[0044] Activity status priority: The user who is performing a specific activity (such as reading, working) has a higher priority than the user in the idle state;

[0045] Custom priority: Assign priorities according to the rules set by the user.

[0046] In the embodiment of the present application, the implementation of determining the user priority according to the preset user priority rules and the location information of each user can adopt a context-based adaptive weight allocation method. Specifically, the total user priority score consists of five aspects: the basic score of the user role priority, the reciprocal value of the distance from the user to the lighting device, the historical operation priority based on time decay, the user activity status score, and the user-defined priority score. The scores of these five aspects will be multiplied by their respective corresponding weight coefficients and then added together to obtain the final priority score. The sum of all weight coefficients is equal to 1 to ensure the consistency of the scoring system. Among them, the user historical operation priority decays exponentially with time. The longer the time since the last operation, the smaller the weight of this item.

[0047] The weight coefficients of this implementation method are no longer fixed values, but are dynamically adjusted according to the current environmental context. For example, in the late-night reading scenario, the activity status weight will be automatically increased; in the multi-person meeting scenario, the role priority weight will be correspondingly increased. The weight adjustment is achieved by adding the basic weight of each factor to the weight adjustment amount based on the current context. After the weight adjustment, normalization processing is required to ensure that the sum of all weights is 1 and maintain the balance of the scoring system.

[0048] According to the order of the user priorities from high to low, collect the voice commands of the user with the highest priority, and at the same time obtain the environmental data related to the user with the highest priority.

[0049] It should be noted that the acquisition technology can enhance the voice signal from the direction of the user with the highest priority through spatial audio processing technologies such as beamforming, while suppressing the noise and interfering voices in other directions. For the environmental data, more consideration will be given to the sensor data in the area around the user with the highest priority. For example, the sensor data at different positions can be weighted by distance, so that the sensor data closer to the user with the highest priority has a higher weight.

[0050] The specific weight calculation adopts the inverse square relationship, that is, the weight value of the sensor data is equal to 1 divided by (1 plus the distance attenuation coefficient multiplied by the square of the distance from the sensor to the user). This design enables sensors closer to the user to obtain higher weights, and as the distance increases, the weights decrease rapidly, thus ensuring that the acquisition of environmental data is closer to the actual feelings of the user with the highest priority. The distance attenuation coefficient can be adjusted according to the size of the space, usually between 0.1 and 2. The larger the space, the smaller this coefficient should be to ensure an appropriate attenuation range.

[0051] When the user's location change is detected, recalculate the user priority and accordingly adjust the voice command acquisition target and the environmental data acquisition range.

[0052] Among them, a location change threshold can be set. For example, when the user's location changes by more than 0.5 meters or the relative angle changes by more than 30 degrees, trigger the recalculation of the user priority. The recalculation frequency can be set according to the scenario requirements. For example, in the normal usage scenario, it is calculated once every 10 seconds, and when rapid movement is detected, the calculation frequency is increased to once every 2 seconds.

[0053] When the user priority changes, smoothly switch the voice acquisition target to avoid sudden interruption of command recognition. Adopt a time-based smooth transition mechanism, that is, within a period of time after the priority switch, gradually increase the acquisition ratio of the voice signal of the user with the new priority. This ratio is obtained by calculating (the current time minus the priority switch time) divided by the preset transition time length. If the calculation result is greater than 1, the value is taken as 1. This can achieve a smooth switch from the original user to the new user and avoid sudden interruption of the voice signal when the user priority changes. The transition time length is usually set to 0.5 to 2 seconds and can be adjusted according to the actual application scenario. For the environmental data acquisition range, also perform a smooth transition to ensure data continuity and avoid unstable responses of the control method due to data mutations.

[0054] In a specific example, assume that there are two users A and B in the living room. User A is a family member and is reading in the sofa area, while user B is a visitor and is moving in the dining table area. It is determined that there are two users through the human body induction sensor and the camera, and their respective identities are determined through face recognition. Based on the user roles (family member > visitor), activity states (reading activity > general activity), and location factors (the distance of both from the lighting device), it is calculated that the priority of user A is higher than that of user B. Therefore, mainly collect the voice commands of user A and focus on considering the environmental data such as the light intensity and temperature in the sofa area. If user A moves from the sofa area to the kitchen area while user B remains in the dining table area unchanged, this location change will be detected, the priority will be recalculated, and it may be adjusted to make user B have a higher priority, thus starting to mainly respond to the voice commands of user B.

[0055] Step 206: Perform semantic analysis on the user voice command, and extract the control intention and parameter requirements from the user voice command.

[0056] Among them, the control intention refers to the basic lighting control purpose expressed in the user voice command, such as "turn on the light", "turn off the light", "brighten", "dim", "change the color temperature", etc. The parameter requirement refers to the user's setting requirement for specific lighting parameters, such as brightness value, color temperature value, light distribution, etc.

[0057] Exemplarily, the user voice command may be directly explicit, such as "set the brightness of the living room light to 50%", or it may be a descriptive situational requirement, such as "I'm going to start reading" or "I feel it's a bit dark". Different types of commands need to extract the control intention and parameter requirements through different processing methods.

[0058] In the embodiment of the present application, the operation steps of performing semantic analysis on the user voice command and extracting the control intention and parameter requirements from the user voice command specifically include:

[0059] Perform speech recognition on the user voice command to convert the voice signal into text data.

[0060] It should be noted that the speech recognition process is implemented by using a deep learning model, including three main steps: acoustic feature extraction, acoustic model processing, and language model decoding. First, preprocess the collected audio signal, including noise reduction, endpoint detection, and feature extraction. Feature extraction mainly uses Mel Frequency Cepstral Coefficients (MFCC), which can simulate the sensitivity difference of the human ear to sounds of different frequencies and extract the key features of the audio. Then, use models such as Deep Neural Network (DNN) or Long Short-Term Memory Network (LSTM) to convert the acoustic features into phoneme probability distributions. Finally, decode the probability distribution in combination with the language model to obtain the most likely text sequence.

[0061] In order to improve the accuracy of speech recognition in the lighting control scenario, domain adaptation technology can be adopted to fine-tune the general speech recognition model by collecting speech samples related to lighting control. In addition, the recognition of lighting device names can also be realized, and the user-defined device names are added to the vocabulary of speech recognition.

[0062] Perform natural language processing on the text data to identify the command type.

[0063] It should be noted that the natural language processing of text data includes three main links: syntactic analysis, semantic analysis, and intention recognition. Syntactic analysis performs part-of-speech tagging and dependency parsing on the text to identify key verbs, nouns, and modifiers in the text, such as "brighten" (verb), "living room light" (noun), "20%" (modifier). Semantic analysis further understands the specific meanings of these words in the field of lighting control. For example, it identifies that "living room light" is a lighting device and "20%" is a brightness parameter. Intention recognition is based on the above analysis results to judge the command type of the user's voice instruction.

[0064] In the embodiments of the present application, the command types are mainly divided into two categories: direct instructions and context instructions. A direct instruction refers to an instruction that clearly contains control parameters, such as "adjust the light brightness to 70%", "set the color temperature to 3000K", etc. A context instruction refers to an instruction that describes the user's activities, emotions, or environmental feelings, such as "I'm about to start working", "create a warm atmosphere", etc. The recognition of the command type can be achieved by combining a rule-based method with a machine learning model. The rule-based method is mainly based on keyword matching and template recognition. For example, it detects whether the text contains direct numerical parameters or percentage expressions; the machine learning method uses a trained classifier to judge the command type according to the semantic features of the text.

[0065] If the command type is a direct instruction, directly extract the control parameters in the text data as the parameter requirements.

[0066] It should be noted that for direct instructions, a rule-based parameter extraction algorithm can be used to identify and extract numerical parameters from the text. This algorithm first uses regular expressions to match the digital patterns in the text (such as 50%, 3000K, 70% brightness, etc.), and then performs unit conversion and numerical standardization to unify various expressions into standard parameter values. For example, convert "70% brightness" to a brightness value of 70%, and convert "warm color tone" to a color temperature value of about 2700K. For complex instructions with multiple parameters, such as "brighten the living room light by 20% and set it to warm color tone", the algorithm will identify two independent parameter requirements: increase the brightness by 20% and set the color temperature to warm color tone (about 2700K).

[0067] The extracted parameter requirements will include parameter types (such as brightness, color temperature, light distribution) and specific values, forming a structured parameter set. For parameters that are not clearly specified, the system will retain their current values without change. In addition, for instructions expressing relative changes, such as "a little brighter", the algorithm will calculate the corresponding target brightness value according to the current brightness value and the semantic meaning of "a little" (usually corresponding to a change range of 5% - 15%).

[0068] If the command type is a scenario instruction, according to the preset scenario mapping relationship, convert the scenario instruction into the corresponding control intention and default parameter requirements.

[0069] The processing of scenario instructions is based on a preset scenario mapping relationship library, which stores the corresponding relationships between various activity scenarios, emotional states, or environmental descriptions and lighting parameters. The scenario mapping relationship library contains mappings in multiple dimensions, such as activity scenario mapping, emotional state mapping, and environmental feeling mapping.

[0070] Activity scenario mapping: Associate activities such as "reading", "working", "dining", "resting", "watching movies" with corresponding lighting parameters. For example, the "reading" scenario is default mapped to a relatively high brightness (about 70 - 80%) and a neutral to cool color temperature (about 4000 - 5000K) to provide good visual clarity; the "dining" scenario is mapped to a medium brightness (about 50 - 60%) and a warm color temperature (about 2700 - 3000K) to create a warm atmosphere.

[0071] Emotional state mapping: Associate emotional states such as "relaxed", "energized", "focused", "romantic" with lighting parameters. For example, the "relaxed" state is mapped to a relatively low brightness (about 30 - 40%) and a warm color temperature (about 2500 - 2800K); the "energized" state is mapped to a relatively high brightness (about 70 - 90%) and a cool color temperature (about 5000 - 6000K).

[0072] Environmental feeling mapping: Associate environmental feeling descriptions such as "too bright", "too dark", "glare", "soft" with the direction of lighting parameter adjustment. For example, "too bright" is mapped to a 20 - 30% reduction in the current brightness; "glare" is mapped to a reduction in brightness and an adjustment to a warmer color temperature.

[0073] The scenario instruction conversion process uses a semantic similarity matching algorithm to match the user's scenario description with the preset scenarios in the scenario mapping relationship library, select the most similar scenario template, and extract its corresponding control intention and default parameter requirements. For fuzzy or polysemous scenario descriptions, combine environmental factors such as the current time, user location, and activity status for scenario disambiguation and precise matching.

[0074] In a specific example, when the user utters the context instruction "I'm about to start reading", the system first converts it into text data through speech recognition. Then, through natural language processing, it is recognized that this is a context instruction describing the activity that the user is about to start. Next, the system queries the context mapping relationship library and finds the lighting parameters corresponding to the "reading" activity: brightness 75%, color temperature 4500K, and the light distribution is concentrated at the location where the user is. These parameters are extracted as the control intention (optimize reading lighting) and the default parameter requirements. If the user is in the study, the parameters are further refined, perhaps increasing the light intensity in the desk area and reducing the brightness in the surrounding areas to create a focused reading environment.

[0075] For a composite context instruction, such as "I want a warm but bright enough environment to entertain guests", two key context descriptions, "warm" and "bright enough", and the activity scenario of "entertaining guests" are recognized. Through comprehensive analysis, it is decided that the control intention is to "create a social atmosphere", and the parameter requirements are medium-high brightness (about 65%) and warm color temperature (about 3000K), while ensuring that the light distribution evenly covers the entire social area.

[0076] Step 208, determine the target lighting parameters according to the control intention, parameter requirements, and environmental data.

[0077] Among them, the target lighting parameters refer to the specific parameter settings that are finally applied to the lighting device, mainly including three core dimensions: brightness, color temperature, and light distribution. Brightness refers to the luminous flux emitted by the light source, usually expressed as a percentage, with 0% being completely off and 100% being the maximum brightness. Color temperature refers to the warmth or coldness of the light source color, measured in Kelvin (K). The lower the value, the warmer the light (more yellowish), and the higher the value, the colder the light (more bluish). The color temperature range of general lighting systems is between 2000K and 6500K. Light distribution describes the way the light spreads in space, including characteristics such as irradiation angle, coverage range, and intensity distribution.

[0078] Exemplarily, the determination of the target lighting parameters needs to comprehensively consider the user's subjective needs and objective environmental factors. For example, for the same "reading" scenario, there should be differences in the parameter settings between daytime with sufficient light and night; for the same requirement of "70% brightness", the actual light output should also be adjusted under different environmental conditions to provide a consistent visual experience.

[0079] In the embodiments of the present application, determining the target lighting parameters according to the control intention, parameter requirements, and environmental data specifically includes:

[0080] Determine the environmental lighting reference value and the current spatial activity type based on the environmental data.

[0081] It should be noted that the ambient lighting reference value refers to the ambient light level of the current space, including the contributions of natural light and other light sources, with the unit of lux. In this embodiment, a distributed light sensor network is preferably used for ambient light collection. Compared with the traditional single-point light sensing method, the distributed sensing network can obtain the light intensity data at different positions in the space, and combine the space layout information to perform light distribution mapping, so as to obtain a more comprehensive and accurate overall ambient light distribution model. This model not only includes the light intensity value, but also the main source direction and change trend of the light, providing more accurate basic data for subsequent lighting parameter adjustment. In addition, the distributed light sensor network can also identify the contributions of different light sources, distinguish natural light and artificial light sources, which is of great significance for realizing the seamless integration of natural light and artificial lighting.

[0082] This distributed light sensing can also achieve light dynamic trend analysis. By detecting the time change rate of the light intensity, it can be judged whether there is a natural light fluctuation caused by cloud cover blocking the sun, and then predict the change of lighting conditions in the short term, providing a basis for forward-looking adjustment of the lighting system.

[0083] The current space activity type refers to the activity that the user may be currently engaged in inferred from the environmental data. The inference of the activity type adopts a multi-modal data fusion method, comprehensively analyzing the following data: user activity status (the user's movement pattern and posture obtained through a motion sensor), environmental spectral characteristics (the spectral distribution obtained through a spectral sensor, which can judge whether there are specific light sources such as a TV or a computer screen), environmental noise characteristics (the environmental sound characteristics obtained through a microphone array, such as whether there is music or conversation), and time information (the current time point is usually related to a specific activity).

[0084] The recognition of the space activity type preferably adopts a method combining rules and machine learning. Compared with the pure rule method or the pure machine learning method, this hybrid method can combine the interpretability of rules and the adaptability of machine learning to obtain more accurate recognition results in a complex environment. First, extract the feature vector according to the sensor data; then, use a trained classifier (such as a random forest or a support vector machine) to classify the feature vector to obtain a preliminary judgment of the activity type; finally, combine the context information such as time and location for rule-based verification and adjustment to obtain the final judgment result of the activity type. Common space activity types include "reading", "watching a movie", "dining", "resting", "socializing", "working", etc.

[0085] Obtain the basic lighting parameters from the preset lighting scene library according to the current space activity type.

[0086] It should be noted that the preset lighting scene library is a database containing various activity scenes and their corresponding lighting parameters, which is preset by lighting experts based on the research results of ergonomics and visual comfort. In the embodiments of the present application, a lighting scene library with a three-layer structure design is preferably adopted. Compared with the traditional flat scene library, the three-layer structure design can achieve more refined scene matching and parameter customization. The first layer is the classification of activity types, such as "reading", "watching movies", etc.; the second layer is the scene subdivision under each activity type. For example, "reading" can be subdivided into "paper reading", "electronic device reading", etc.; the third layer is the standard lighting parameter set for each specific scene, including the recommended brightness range, the recommended color temperature range, and the recommended light distribution mode.

[0087] The process of obtaining the basic lighting parameters first matches in the scene library according to the current spatial activity type to find the closest activity type; then, combined with context information such as time (day / night), spatial location (living room / bedroom / study, etc.), the scene matching is refined; finally, the standard lighting parameters of this scene are extracted as the basic lighting parameters. For example, if it is recognized that the user is performing the "paper reading" activity in the living room and it is night time, the standard parameters of the "night-time living room paper reading" scene will be extracted: an illuminance of approximately 500 - 700 lux, a color temperature of 4000 - 5000K, and a light distribution mode that focuses on illuminating the reading area.

[0088] Adjust the basic lighting parameters according to the control intention and the parameter requirements, which specifically includes the following steps:

[0089] Obtain the user's historical lighting preference data, where the user's historical lighting preference data includes the records of the user's lighting parameter selections under different times and different environmental conditions.

[0090] It should be noted that the collection of the user's historical lighting preference data is achieved by recording the user's lighting operation history. Each time the user adjusts the lighting by voice or other means, the following data is recorded: user identity, operation time, environmental conditions (including ambient light intensity, temperature and humidity, etc.), spatial location, activity type, lighting parameters before adjustment, lighting parameters after adjustment, and user feedback (such as whether to adjust again in a short time, whether there is a voice evaluation, etc.).

[0091] These data are organized according to the user ID and timestamp and stored in the user preference database. To ensure the timeliness of the data, higher weights are given to recent data, and outdated data is regularly cleared. In addition, the feedback data of the user on the automatic control results is also recorded, including whether the user accepts the recommended lighting parameters, whether manual adjustment has been made, etc., in order to continuously optimize the preference model.

[0092] Establish a user preference model according to the user's historical lighting preference data, and the user preference model is stored in the user preference database.

[0093] It should be noted that the embodiments of this application preferably use a method combining collaborative filtering and content filtering to establish a user preference model. Compared with a single model construction method, this hybrid method can utilize both user behavior patterns and environmental feature information to provide more accurate recommendations. The collaborative filtering part analyzes the lighting selection patterns of users under similar conditions and identifies the preference trends of users for parameters such as brightness and color temperature; the content filtering part focuses on the association rules between specific environmental conditions and lighting parameters, such as the consistent preferences of users during specific time periods or specific activities.

[0094] Specifically, first, perform feature engineering on historical data to extract key features such as time period classification (morning / afternoon / evening, etc.), activity type, ambient light intensity, etc.; then, use the multivariate regression analysis method to establish the mapping relationship between environmental features and lighting parameter preferences. For example, through analysis, it may be found that during the reading activity, as the ambient light intensity decreases, the preferred color temperature also shows a downward trend (that is, the darker the light, the more preferred the warm color tone). Finally, construct a parameterized user preference function that can predict the lighting parameter values most likely preferred by users according to the current environmental conditions.

[0095] In addition, the habitual patterns of users will also be identified, such as the lighting preference changes at fixed times every day, the lighting adjustment sequence before and after specific activities, etc. These patterns are stored as the rule part of the user preference model. The user preference model is updated regularly, and the model parameters are adjusted according to the newly added user operation data to ensure that the model can adapt to the changes in user preferences.

[0096] Adjust the basic lighting parameters in combination with the user preference model to generate preliminary adjustment parameters.

[0097] It should be noted that the adjustment process preferably uses a weighted fusion method to fuse the basic lighting parameters and the parameters predicted by the user preference model. Compared with simple replacement or average methods, weighted fusion can dynamically adjust the influence of each parameter according to data reliability, introducing elements while maintaining professional lighting standards. Specifically, for each lighting parameter (such as brightness and color temperature), calculate the adjustment value:

[0098] Adjusted parameter value = basic parameter value × (1 - weight) + user preference parameter value × weight;

[0099] Among them, the weight is dynamically calculated according to the reliability of the user preference model and the similarity between the current environment and historical records. When there is a large amount of historical data similar to the current situation, the weight will increase, making the adjustment result more inclined to the user's historical preferences; conversely, when the historical data is small or the similarity is low, the weight decreases, and the basic parameter value is retained more.

[0100] In addition, the consistency of user preferences is also considered. For preferences that show a high degree of consistency under specific conditions (such as always setting the color temperature to 2700K at night), a higher weight will be assigned; for parameters with large fluctuations, the impact of adjustment will be reduced.

[0101] Match the preliminary adjustment parameters with the control intention and the parameter requirements to determine the adjusted lighting parameters.

[0102] It should be noted that in the matching process, first check whether there are conflicts between the user's control intention and the clear parameter requirements and the preliminary adjustment parameters. For example, if the user clearly requests "adjust the brightness to 30%", while the brightness value in the preliminary adjustment parameters is 60%, there is a conflict.

[0103] For parameters with conflicts, give priority to the values clearly requested by the user, and record this deviation for future updates of the user preference model; for parameters not clearly specified by the user, retain the values of the preliminary adjustment parameters. In addition, the coordination between parameters will also be checked to avoid discordant combinations, such as combinations of extremely high brightness and extremely warm color temperature that may cause visual discomfort.

[0104] The matching of the control intention focuses on the high-level goals of the user's instructions. For example, if the control intention is "provide sufficient light for reading", ensure that the adjusted lighting parameters at least meet the minimum illuminance requirements for reading (usually above 300 lux); if the control intention is "create a relaxing atmosphere", then tend to choose warm tones and moderately low brightness.

[0105] Combine the adjusted lighting parameters with the ambient lighting reference value to correct the adjusted lighting parameters to obtain the target lighting parameters.

[0106] It should be noted that the correction of the ambient lighting reference value mainly considers the impact of ambient light on the final visual effect. The embodiments of the present application preferably use the principle of perceived light balance for correction. Compared with the simple ambient light compensation method, the perceived light balance can better simulate the non-linear perception characteristics of the human eye to light, ensuring that the total light level (ambient light + lighting light) perceived by the human eye reaches the target effect.

[0107] For the brightness parameter, the correction formula is: the output brightness of the lighting device = the target perceived brightness - the ambient light contribution brightness × the ambient light influence coefficient;

[0108] Among them, the ambient light influence coefficient is dynamically adjusted according to the direction, color and stability of the ambient light source, usually between 0.7 and 0.9. This correction ensures that when the ambient light is strong, the output of the lighting device will decrease accordingly, and when the ambient light is weak, the output will increase, thus maintaining a consistent visual experience.

[0109] For the color temperature parameter, considering the influence of the color temperature of the ambient light, the final output color temperature is calculated by means of weighted average, so that the comprehensive light color temperature perceived by the human eye is close to the target value. Especially in an environment where daylight and artificial light are mixed, this correction can reduce visual fatigue caused by inconsistent color temperatures.

[0110] For the light distribution parameter, the correction mainly focuses on the compensation of the shadow area of the ambient light. By analyzing the ambient light distribution map, the areas with insufficient light are identified, and the irradiation angle and intensity distribution of the lighting equipment are adjusted to provide a more balanced lighting effect.

[0111] Finally, considering all correction factors, the target lighting parameters are generated, including the exact brightness value, color temperature value, and light angle / distribution settings of each lighting device. These parameters are directly transmitted to the lighting control module for actual lighting adjustment operations.

[0112] In a specific example, assume that the user says the voice command "I'm going to start reading" in the living room at dusk. Through voice recognition and semantic analysis, it is identified that this is a context command, and the control intention is "provide lighting suitable for reading". Based on the ambient sensor data, it is determined that the current ambient light intensity is 150 lux (dusk natural light), the color temperature is about 4000K, and the user activity type is "preparing to read".

[0113] Query the preset lighting scene library to obtain the basic lighting parameters of the "reading" scene: the brightness is equivalent to 600 lux, the color temperature is 4500K, and the light distribution is concentrated in the reading area. Then, query the user's historical lighting preference data and find that the user tends to prefer warmer light (average 3800K) and slightly higher brightness (average 650 lux) when reading at dusk. Based on these data, the basic parameters are adjusted to obtain the preliminary adjusted parameters: the brightness is equivalent to 630 lux, and the color temperature is 4000K.

[0114] Considering the 150 lux contributed by the current ambient light (mainly the dusk light from the window), environmental light correction is performed, and it is calculated that the lighting equipment needs to provide about 500 lux of supplementary light (considering the ambient light influence coefficient of 0.8). At the same time, to balance the color temperature influence of the ambient light, the color temperature of the lighting equipment is set to about 3900K to ensure that the perceived comprehensive color temperature is close to the target value of 4000K.

[0115] Finally, the target lighting parameters are determined: the main lighting fixture brightness is set to 65% (equivalent to providing 500 lux), the color temperature is set to 3900K, and the light direction is appropriately adjusted to enhance the illuminance of the user's reading area and reduce the light output on the window side to balance the overall light distribution.

[0116] Step 210, obtain the current lighting state and calculate the transition path from the current lighting state to the target lighting parameters; the transition path includes a gradual change sequence of lighting parameters.

[0117] Among them, the gradual change sequence of lighting parameters refers to a series of intermediate state values between the current lighting state and the target lighting parameters. Through these intermediate states, a smooth transition of lighting parameters is achieved, avoiding visual discomfort caused by sudden changes. The transition path determines how the lighting parameters change over time and is a key technical link to achieve a soothing lighting experience.

[0118] Exemplarily, the change of lighting parameters may involve synchronous adjustment in multiple dimensions such as brightness, color temperature, and light distribution. Reasonable design of the transition path can reduce the stimulation of the user's visual system caused by lighting mutations, improve comfort, and can adjust the transition speed according to different scene requirements. For example, a slower transition can be adopted in the rest scene, while the transition speed can be appropriately increased in the work scene.

[0119] In the embodiments of the present application, the obtaining of the current lighting state and the calculation of the transition path from the current lighting state to the target lighting parameters include:

[0120] Obtain the working state of the current lighting device, including the current brightness, current color temperature, and current light distribution.

[0121] It should be noted that the working state of the lighting device can be obtained in various ways. In the embodiments of the present application, it is preferably to use the lighting device state feedback mechanism to obtain the real-time working state. Compared with the method of simply recording the previous control command, the device state feedback can provide more accurate actual working parameters, including parameter deviations caused by external factors (such as voltage fluctuations). The specific implementation methods include: directly reading the state data from intelligent lighting devices that support two-way communication; obtaining the parameter values set last time through the working records of the lighting controller; measuring the actual light output using a light sensor and inversely calculating the current working parameters in combination with the characteristic curve of the lighting device.

[0122] The current brightness is usually expressed as a percentage, ranging from 0% (fully off) to 100% (maximum brightness). The current color temperature is in units of Kelvin (K) and is usually in the range of 2000K to 6500K. The current light distribution includes parameters such as the irradiation angle, coverage range, and light intensity distribution of the lighting device, and can be a preset distribution mode number or a specific set of distribution parameters.

[0123] Calculate the difference between the working state of the current lighting device and the target lighting parameters.

[0124] It should be noted that the difference calculation adopts a multi-dimensional parameter distance evaluation method. Considering the differences in the dimensions and perceptual characteristics of different lighting parameters, each parameter is normalized and the comprehensive difference is calculated by combining the perceptual weights. In this embodiment, the weighted Euclidean distance is preferably used to calculate the comprehensive difference. Compared with the simple parameter difference calculation, this method can better reflect the perceptual differences of the human eye to the changes of different lighting parameters. The specific calculation method is as follows:

[0125] Calculation of brightness difference: Normalized brightness difference = |target brightness - current brightness| / 100%;

[0126] Calculation of color temperature difference: Normalized color temperature difference = |target color temperature - current color temperature| / 4500K;

[0127] Calculation of light distribution difference: Based on the similarity score of the distribution pattern or the weighted difference of specific distribution parameters;

[0128] Then, calculate the weighted comprehensive difference.

[0129] Determine the transition time according to the size of the difference and the user activity state.

[0130] It should be noted that the transition time refers to the time required to completely transition from the current lighting state to the target lighting parameters. In the embodiment of the present application, an adaptive transition time determination method is preferably used. Compared with the fixed transition time, the adaptive method can dynamically adjust the transition time according to the parameter change range and the user activity state, providing a more user-friendly lighting experience. The specific implementation includes two steps: basic transition time calculation and activity state adjustment.

[0131] The basic transition time is directly proportional to the comprehensive difference: Basic transition time = minimum transition time + (comprehensive difference / maximum possible difference) × (maximum transition time - minimum transition time);

[0132] Among them, the minimum transition time is usually set to 0.5 - 1 second, the maximum transition time is usually set to 10 - 30 seconds, and the maximum possible difference is 1 (the maximum difference after normalization).

[0133] Then, adjust according to the user activity state (such as static state, low activity state, high activity state):

[0134] Static state (such as reading, resting): Extend the transition time and use 1.2 - 1.5 times the basic transition time;

[0135] Low activity state (such as dining, socializing): Use the basic transition time;

[0136] High activity state (such as walking, exercising): Shorten the transition time and use 0.6 - 0.8 times the basic transition time.

[0137] In addition, time factors will also be considered, and the transition time will be automatically extended at night to adapt to the characteristic that the human eye is more sensitive to light changes in a dark environment.

[0138] If the difference is greater than the first preset threshold, a segmented transition path will be generated, dividing the transition time into multiple time periods, and generating a gradual change lighting parameter sequence within each time period.

[0139] It should be noted that the first preset threshold is the boundary value for judging whether the lighting parameter change is significant, usually set to a comprehensive difference of 0.3 - 0.4. When the difference is greater than this threshold, the lighting change is more obvious, and a more refined segmented transition strategy needs to be adopted.

[0140] The embodiment of the present application preferably adopts a three - segment transition path design. Compared with the traditional linear transition, the three - segment design can better simulate the adaptation process of the human eye to light changes and reduce visual discomfort. The three - segment transition path includes: a starting slow - change segment, a middle fast - change segment, and an ending slow - change segment.

[0141] Starting slow - change segment (about 20% of the total transition time): The parameter change rate gradually increases from zero to the rate of the middle fast - change segment, following an acceleration curve;

[0142] Middle fast - change segment (about 60% of the total transition time): The parameter changes at a relatively constant rate to complete most of the parameter adjustments;

[0143] Ending slow - change segment (about 20% of the total transition time): The parameter change rate gradually decreases from the rate of the middle fast - change segment to zero, following a deceleration curve.

[0144] Within each time period, according to the starting parameter, ending parameter, and time length of this segment, a sufficiently dense intermediate parameter sequence is generated to ensure the smoothness of the lighting change. Usually, the generation frequency of the intermediate parameter sequence is not less than 10 times per second to ensure a continuous visual transition effect.

[0145] If the difference is less than or equal to the first preset threshold, a linear transition path will be generated, and the lighting parameters will be uniformly changed within the transition time.

[0146] It should be noted that when the lighting parameter change is small, a simple linear transition path can already provide a good visual experience while reducing the consumption of computing resources. The linear transition path is based on the current lighting state and the target lighting parameters, and generates a sequence of intermediate parameter values at equal time intervals within the transition time.

[0147] The embodiment of the present application preferably adopts a parameter - independent linear interpolation method to independently perform linear interpolation calculations on the parameters of the three dimensions of brightness, color temperature, and light distribution.

[0148] Determining the transition time according to the magnitude of the difference and the user activity status includes:

[0149] During the execution of the lighting adjustment process, the feedback information of the user on the lighting adjustment is monitored in real time; the feedback information includes voice evaluation or manual adjustment behavior.

[0150] The acquisition of user feedback information adopts a multimodal monitoring method, including:

[0151] Voice evaluation monitoring: Through the microphone and speech recognition technology, capture the evaluative statements that the user may utter, such as "too bright", "changing too fast", etc.;

[0152] Manual adjustment monitoring: Detect whether the user makes manual intervention through physical switches, remote controls or applications during the automatic adjustment process, such as interrupting the current adjustment, manually adjusting the brightness, etc.;

[0153] Non-verbal feedback monitoring: Capture the user's facial expressions, blink frequency or body posture changes through a camera or other sensors to infer the discomfort reaction of the user to the lighting change.

[0154] The embodiment of the present application adopts a real-time feedback processing mechanism. Compared with the batch processing mechanism, real-time processing can respond to user feedback in a timely manner during the lighting adjustment process and provide a more personalized adaptive adjustment. The real-time monitoring system samples the feedback signal at a relatively high frequency (usually every 0.5 - 1 second) to ensure that the timely reaction of the user can be captured.

[0155] Calculate the user comfort value according to the feedback information.

[0156] It should be noted that the user comfort value is a quantitative index to measure the acceptance degree of the user to the current lighting change, and the range is usually from 0 (extremely uncomfortable) to 100 (completely comfortable). This embodiment adopts a multi-factor weighted scoring model to calculate the user comfort value. Compared with the single-index evaluation, the multi-factor model can comprehensively consider different types of feedback information and provide a more comprehensive comfort evaluation.

[0157] The comfort calculation considers the following factors:

[0158] Negative voice evaluation: If a clear negative evaluation (such as "too bright", "dazzling", etc.) is detected, the comfort value is greatly reduced (-30 to -50 points);

[0159] Degree of manual intervention: The greater the amplitude of the user's manual adjustment, the more the comfort value is reduced;

[0160] Intervention timing: Intervention occurring at the beginning of the transition indicates a higher degree of discomfort and a greater impact on the comfort value;

[0161] Non-verbal feedback: Detect discomfort manifestations such as facial frowning and frequent blinking, and moderately reduce the comfort value.

[0162] Lack of feedback: If there is no negative feedback within a certain period of time, gradually increase the comfort value until the default high comfort state (80 - 90 points) is reached.

[0163] When the user comfort value is lower than the second preset threshold, the transition time is increased in real time, and the gradual change process of the remaining lighting parameters is extended.

[0164] It should be noted that the second preset threshold is the boundary value for judging whether the user feels obvious discomfort with the lighting change, usually set at 60 - 70 points (out of 100). When the comfort value is lower than this threshold, it indicates that the user has discomfort with the current lighting change rate and the transition strategy needs to be adjusted.

[0165] This embodiment can adopt a dynamic transition time adjustment mechanism. Compared with a simple transition interruption or restart mechanism, dynamic adjustment can flexibly adjust the change rate while maintaining the continuous change of lighting parameters, providing a smoother adaptation experience. The specific adjustment method is as follows:

[0166] Calculate the extension ratio of the remaining transition time: Extension ratio = Basic extension coefficient × (Second preset threshold - User comfort value) / Second preset threshold;

[0167] Among them, the basic extension coefficient is usually set at 1.5 - 3, and the extension ratio is dynamically adjusted according to the gap between the comfort value and the threshold.

[0168] Then, update the remaining transition time: New remaining transition time = Original remaining transition time × (1 + Extension ratio);

[0169] At the same time, recalculate the remaining transition path to ensure the continuity and smoothness of the lighting parameter change. In extreme cases (very low comfort value), pause the transition process and wait for the user to adapt to the current state before slowly resuming the transition.

[0170] When the user comfort value is not lower than the second preset threshold, maintain or reduce the transition time.

[0171] When the user comfort value is within the acceptable range, this embodiment preferably adopts a "maintenance priority" strategy, that is, giving priority to maintaining the current transition time setting to avoid instability caused by frequent adjustments. Only when the comfort value continuously remains at a relatively high level (such as higher than 85 points in consecutive monitoring periods) will consideration be given to moderately reducing the transition time to accelerate the lighting adjustment process.

[0172] The amplitude of reducing the transition time is usually small (a 5%-15% reduction). Progressive adjustment is adopted to avoid new discomfort caused by sudden acceleration. In addition, record the user's acceptance of different change rates and gradually learn the range of transition rates most suitable for the user.

[0173] Store the feedback information, corresponding environmental data, control intention, and lighting parameters in the user preference database for updating the user preference model during the next lighting adjustment.

[0174] It should be noted that the storage of feedback information adopts a structured data recording method, including the following fields:

[0175] Timestamp: Record the exact time when the feedback occurred;

[0176] User ID: The user identity from which the feedback originated;

[0177] Environmental data summary: Record the key environmental parameters at the time of feedback, such as light intensity, time period, etc.;

[0178] Control intention: The purpose or scene type of the current lighting adjustment;

[0179] Lighting parameter trajectory: Record the sequence of lighting parameter changes from the starting state to the feedback point;

[0180] Feedback type: Distinguish different feedback forms such as voice evaluation and manual adjustment;

[0181] Feedback content: Record the specific feedback content, such as the text content of the voice evaluation;

[0182] Comfort score: The user comfort value calculated by the system;

[0183] System response: The adjustment measures taken by the system for this feedback.

[0184] This embodiment adopts an incremental learning model update mechanism. Compared with the batch retraining mechanism, incremental learning can gradually adjust the model parameters through new data while retaining the original model structure, realizing the continuous optimization of the user preference model. After storing new feedback data each time, evaluate the impact of the data on the current model and decide whether to trigger model update. For feedback data with significant impact (such as strong feedback that is significantly inconsistent with historical preferences), give priority to including it in the model update process.

[0185] In a specific example, assume that the user issues a voice command of "getting ready to go to bed" in the bedroom at night. The system recognizes it as the "before bedtime" scenario, with the target lighting parameters being a brightness of 20% and a color temperature of 2700K, while the current lighting state is a brightness of 70% and a color temperature of 4000K. The calculated normalized brightness difference is 0.5, the normalized color temperature difference is 0.29, and the comprehensive difference is approximately 0.42 (considering a brightness weight of 0.6 and a color temperature weight of 0.4).

[0186] Since the comprehensive difference of 0.42 is greater than the first preset threshold of 0.35 and the user is in a stationary state getting ready to fall asleep, the system decides to adopt a segmented transition path and extend the basic transition time, finally determining the total transition time to be 15 seconds. The transition path is divided into three segments: in the first 3 seconds, the brightness and color temperature are slowly decreased; in the middle 9 seconds, the main parameter adjustments are completed; and in the last 3 seconds, a smooth transition to the final state is achieved.

[0187] During the lighting adjustment process, when the transition reaches the 7th second (the brightness has dropped to approximately 40% and the color temperature is approximately 3200K), the system captures the voice evaluation of the user saying "it's changing too fast and a bit uncomfortable" through the microphone. The feedback is quickly analyzed, and the current user comfort value is calculated to be 55, which is lower than the second preset threshold of 65.

[0188] Based on this comfort evaluation, the system immediately adjusts the strategy, calculates the extension ratio to be 0.77 (basic extension coefficient 2×(65 - 55) / 65), and extends the remaining 8 - second transition time to approximately 14 seconds (8 seconds × 1.77). At the same time, the remaining transition path is re - planned to slow down the parameter change rate, especially reducing the brightness change rate, because brightness change is usually the main factor causing visual discomfort.

[0189] Finally, the entire lighting adjustment process is extended from the original planned 15 seconds to approximately 21 seconds, and the user no longer expresses discomfort. This feedback information and related data are stored in the user preference database as a reference for future lighting adjustments in the "before bedtime" scenario. In several subsequent similar scenarios, a more gentle transition strategy is automatically adopted, and the transition time from the initial state to the "before bedtime" lighting is directly set to 20 - 25 seconds, avoiding similar discomfort reactions.

[0190] Step 212, control the lighting device to perform lighting adjustment according to the transition path.

[0191] Among them, lighting adjustment means sending a series of control commands to the lighting device according to the transition path calculated in the foregoing steps, so that it gradually changes parameters such as brightness, color temperature, and light distribution according to a predetermined gradual change sequence, and finally reaches the target lighting state.

[0192] Exemplarily, the control of lighting devices can be achieved through various communication protocols, such as Zigbee, Wi-Fi, Bluetooth, or dedicated lighting control protocols (such as DALI, DMX), etc. The control methods of devices under different protocols will vary, but the core principle is the same, which is to convert the parameter sequence in the transition path into control instructions that the device can understand and send them to the corresponding device according to the time sequence.

[0193] In one embodiment, controlling a lighting device to perform lighting adjustment according to a transition path specifically includes:

[0194] Converting the lighting parameter sequence in the transition path into device-specific control instructions. Different lighting devices may use different parameter formats and control methods, and protocol adaptation and parameter mapping are required. For example, some devices use integers from 0 to 255 to represent brightness, while others use percentages from 0% to 100%; some devices support directly setting the color temperature value, while other devices may need to adjust the RGB or warm-cold light ratio to achieve color temperature adjustment.

[0195] Establishing the sending time sequence of control instructions. According to the timestamps of each parameter point in the transition path, determine the sending timing of control instructions. In this embodiment, a predictive instruction scheduling mechanism is preferably adopted. Compared with simple timed sending, predictive scheduling takes into account network latency and device response time, and sends instructions at an appropriate time in advance to ensure that the actual time sequence of parameter changes is consistent with the designed transition path.

[0196] Executing the gradual change control process. Sending control instructions to the lighting device through the control center according to the established time sequence. During the instruction sending process, continuously monitor the status feedback of the device to confirm whether the parameter changes are executed as expected. If abnormal device responses are detected, such as instruction loss or excessive delay, the sending strategy of subsequent instructions will be dynamically adjusted to ensure that the overall lighting effect is not affected.

[0197] Coordinating multi-device linkage control. When the lighting environment involves multiple lighting devices, it is necessary to ensure that the parameter changes between devices are coordinated. This embodiment implements a device group synchronization control mechanism, which groups devices according to their physical distribution and functional attributes, sends synchronous control instructions to devices in the same group to maintain consistency in changes; for devices in different groups, according to the design in the transition path, an expected control sequence is achieved, such as a progressive lighting change effect from one side of the room to the other side.

[0198] Since the design and adjustment mechanism of the transition path have been described in detail in step 210, this step mainly focuses on how to implement these designs, converting the theoretically parameter change sequence into an actual device control effect. By precisely controlling the parameter changes of each lighting device in the time dimension, a smooth and natural transition from the current lighting state to the target lighting parameters is ultimately achieved, providing users with a visually comfortable lighting experience.

[0199] After the entire lighting adjustment process is completed, the lighting system will continue to monitor environmental changes and the user's activity status, and be ready to respond to the next lighting adjustment requirement, forming a complete context-aware voice-controlled soothing lighting closed-loop system.

[0200] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0201] Based on the same inventive concept, the embodiments of the present application also provide a context-aware voice-controlled soothing lighting system. The implementation solutions for solving problems provided by this system are similar to those described in the above method. Therefore, the specific limitations in one or more embodiments of the context-aware voice-controlled soothing lighting system provided below can refer to the limitations on the context-aware voice-controlled soothing lighting method in the above text, and will not be repeated here.

[0202] In an exemplary embodiment, as Figure 3 shown, a context-aware voice-controlled soothing lighting system is provided, including:

[0203] A data acquisition module for obtaining user voice commands and environmental data;

[0204] A semantic analysis module for performing semantic analysis on user voice commands and extracting control intentions and parameter requirements from the user voice commands;

[0205] A parameter determination module for determining target lighting parameters according to the control intention, parameter requirements, and environmental data;

[0206] A path calculation module for obtaining the current lighting state and calculating the transition path from the current lighting state to the target lighting parameters;

[0207] A lighting control module for controlling lighting devices to perform lighting adjustments according to the transition path.

[0208] Each module in the above-mentioned context-aware voice control soothing lighting system can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0209] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 4 shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used for the processor to exchange information with external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be implemented through Wi-Fi, a mobile cellular network, near field communication (NFC), or other technologies. The computer program, when executed by the processor, implements the context-aware voice control soothing lighting method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0210] Those skilled in the art can understand that Figure 4 the structure shown in

[0211] is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0212] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the foregoing method embodiments are implemented.

[0213] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the foregoing method embodiments are implemented.

[0214] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0215] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0216] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in the present application.

[0217] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A situation-aware voice-controlled soothing lighting method, characterized in that: include: Acquire user voice commands and environmental data, wherein the environmental data includes light intensity, temperature, humidity, and user activity status; Performing semantic analysis on the user voice command to extract control intent and parameter requirements from the user voice command; Determine target lighting parameters according to the control intention, the parameter requirements and the environmental data; the target lighting parameters include brightness, color temperature and light distribution; Acquire a current lighting state, and calculate a transition path from the current lighting state to the target lighting parameter; the transition path includes a gradual change sequence of the lighting parameter; The lighting device is controlled to adjust the lighting according to the transition path.

2. The context-aware voice-controlled soothing lighting method according to claim 1, characterized in that: The performing semantic analysis on the user voice command to extract control intent and parameter requirements from the user voice command includes: Performing voice recognition on the user's voice command and converting the voice signal into text data; Performing natural language processing on the text data to identify the command type; If the command type is a direct instruction, directly extracting the control parameters in the text data as the parameter requirements; If the command type is a situational instruction, the situational instruction is converted into corresponding control intentions and default parameter requirements according to a preset situational mapping relationship.

3. The context-aware voice-controlled soothing lighting method according to claim 2, characterized in that: The determining the target lighting parameter according to the control intention, the parameter requirement and the environmental data comprises: Determine an ambient lighting baseline value and a current spatial activity type based on the ambient data; Acquire basic lighting parameters from a preset lighting scene library according to the current space activity type; Adjusting the basic lighting parameters according to the control intention and the parameter requirements; The adjusted lighting parameters are corrected in combination with the ambient lighting reference value to obtain the target lighting parameters.

4. The context-aware voice-controlled soothing lighting method according to claim 3, characterized in that: The adjusting the basic lighting parameters according to the control intention and the parameter requirement includes: Acquire historical lighting preference data of the user, wherein the historical lighting preference data of the user includes lighting parameter selection records of the user at different times and under different environmental conditions; Establishing a user preference model according to the user historical lighting preference data, wherein the user preference model is stored in a user preference database; adjusting the basic lighting parameters in combination with the user preference model to generate preliminary adjustment parameters; The preliminary adjustment parameters are matched with the control intention and the parameter requirements to determine the adjusted lighting parameters.

5. The context-aware voice-controlled soothing lighting method according to claim 4, characterized in that: The acquiring the current lighting state and calculating the transition path from the current lighting state to the target lighting parameter comprises: Get the current working status of the lighting equipment, including the current brightness, current color temperature and current light distribution; Calculating the difference between the current working state of the lighting device and the target lighting parameter; Determine the transition time according to the difference value and the user activity state; If the difference is greater than a first preset threshold, a segmented transition path is generated, the transition time is divided into a plurality of time periods, and a gradual lighting parameter sequence is generated in each time period; If the difference is less than or equal to the first preset threshold, a linear transition path is generated to uniformly change the lighting parameter within the transition time.

6. The context-aware voice-controlled soothing lighting method according to claim 5, characterized in that: The determining the transition time according to the difference value and the user activity state includes: During the lighting adjustment process, real-time monitoring of user feedback on the lighting adjustment; the feedback information includes voice evaluation or manual adjustment behavior; Calculating a user comfort value according to the feedback information; When the user comfort value is lower than a second preset threshold, increasing the transition time in real time to extend the gradual change process of the remaining lighting parameters; When the user comfort value is not lower than the second preset threshold, maintaining or reducing the transition time; The feedback information and the corresponding environmental data, control intention and lighting parameters are stored in a user preference database for updating the user preference model during the next lighting adjustment.

7. The context-aware voice-controlled soothing lighting method according to claim 1, characterized in that: The obtaining of user voice commands and environmental data includes: Collect environmental sensor data to determine the number of users in the environment; If there is a single user in the environment, directly obtaining the voice command of the single user and the environment data; If there are multiple users in the environment, the identity information and location information of each user are obtained, and the user priority is determined according to the preset user priority rules and the location information of each user; Collecting the voice commands of the user with the highest priority in descending order of priority of the users, and acquiring the environmental data related to the user with the highest priority; When a change in the user's position is detected, the user priority is recalculated, and the voice command collection target and environmental data acquisition range are adjusted accordingly.

8. A situation-aware voice-controlled soothing lighting system, using the situation-aware voice-controlled soothing lighting method according to any one of claims 1 to 7, characterized in that: Data collection module, used to obtain user voice commands and environmental data; A semantic analysis module, used to perform semantic analysis on the user voice command and extract control intent and parameter requirements from the user voice command; A parameter determination module, used to determine target lighting parameters according to the control intention, the parameter requirements and the environmental data; A path calculation module, used to obtain a current lighting state and calculate a transition path from the current lighting state to the target lighting parameter; The lighting control module is used to control the lighting equipment to adjust the lighting according to the transition path.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the context-aware voice-controlled soothing lighting method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the context-aware voice-controlled soothing lighting method described in any one of claims 1 to 7 are implemented.

Images