Intelligent illumination projection system and control method thereof

Abstract

The invention provides an intelligent illumination projection system and a control method thereof, and relates to the technical field of intelligent illumination and projection, and the system comprises a main control module which comprises a main controller as a system center and an auxiliary interface chip connected with the main controller, and is used for processing a sensor signal, executing module control logic and operating a data processing task; the power supply management module comprises a charging controller used for receiving the energy input of the solar charging panel, an energy storage battery / UPS connected with the charging controller, and a multi-stage voltage stabilizing circuit used for converting the output voltage of the energy storage battery / UPS into a plurality of voltage grades; and the function module driving circuit comprises an LED illumination module, a projection module, an intelligent voice module and a monitoring module which are connected with the power supply management module and the main control module, and is used for realizing illumination, projection, voice playing and monitoring functions under the control of the main control module. According to the invention, full-automatic projection management can be realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent lighting and projection technology, and in particular to an intelligent lighting projection system and its control method. Background Technology

[0002] Projection and lighting systems, as important tools for information display and environmental creation, have been widely used in many fields such as commercial displays, education and training, and home entertainment. However, current systems have significant limitations in terms of integration, energy efficiency, and intelligence, making it difficult to meet the growing demand for multi-scenario and adaptive applications.

[0003] Traditional projection systems typically rely entirely on mains power. This not only leads to complex wiring issues during deployment, increasing installation costs and engineering difficulty, but also severely restricts their application in special scenarios such as outdoor plazas, temporary exhibition halls, and remote areas without a stable power grid. These systems are often single-function, centered on projection display, and lack effective integration with other environmental control equipment. Their energy consumption is relatively concentrated, and energy efficiency management is inefficient, which is inconsistent with the trend of green and low-carbon development.

[0004] On the other hand, while independent power supply systems based on solar energy (such as some LED streetlights or courtyard lights) solve the power supply problem in specific scenarios, their functions are usually limited to lighting and lack extended capabilities such as information projection, thus limiting their application value. Furthermore, most intelligent sensing systems on the market are limited to simple "lights on when people are present, lights off when people leave" or illuminance threshold switching control. Their sensing dimensions are singular, their control logic is rigid, and they cannot perform complex environmental feature analysis (such as distinguishing activity patterns at different times and identifying specific scenario needs) and execute adaptive, multi-dimensional collaborative control strategies.

[0005] A more prominent problem is that existing technological solutions generally suffer from scenario-specific limitations. The demands of commercial venues—such as long-term stable operation, programmed content playback, and high-brightness ambient lighting—are difficult to flexibly adapt to the comfort, personalized content interaction, and convenient voice control experiences emphasized in home environments through a single system. Devices often lack a unified intelligent hub for cross-module (lighting, projection, voice, monitoring) data aggregation and decision-making, resulting in poor system scalability and an inability to achieve truly integrated intelligent management of lighting, information display, human-computer interaction, and environmental perception. Summary of the Invention

[0006] To address the aforementioned technical problems, the present invention provides a smart lighting projection system and its control method, comprising: The main control module includes a main controller that serves as the central hub of the system and an auxiliary interface chip connected to the main controller. It is used to process sensor signals, execute module control logic, and run data processing tasks. The power management module includes a charging controller for receiving energy input from a solar charging panel, an energy storage battery / UPS connected to the charging controller, and a multi-stage voltage regulator circuit that converts the output voltage of the energy storage battery / UPS into multiple voltage levels. The functional module driving circuit includes an LED lighting module, a projection module, an intelligent voice module, and a monitoring module connected to the power management module and the main control module, which are used to realize lighting, projection, voice playback and monitoring functions under the control of the main control module; The data processing module, executed by the main control module, is used to collect raw data from the functional module drive circuit and environmental sensors, and perform feature extraction, pattern recognition, and adaptive control decision-making, wherein: The main control module automatically controls the start and stop of the LED lighting module and the projection module based on the output of the data processing module, and supports switching between commercial and home working modes through software logic. The sensor signals are acquired through an ambient light sensor and a human infrared sensor.

[0007] Preferably, the data processing module includes a sensor data fusion unit, which is used to fuse at least one of the following data: ambient light intensity, human infrared signal, audio input, and camera image, to generate multi-dimensional environmental state features for system control.

[0008] Preferably, the data processing module further includes an anomaly detection and diagnosis unit, which is used to monitor the current, voltage, temperature and signal continuity of each module of the system in real time, and to establish a normal operating state model through unsupervised learning methods. Furthermore, when data is detected to deviate from the normal operating state model by more than a set threshold, an abnormal alarm is generated and a corresponding protection or correction mechanism is triggered. Then, after detecting the anomaly, a fault tracing analysis is performed. By comparing the abnormal data pattern with the preset fault feature library, the potential fault module is located, and the diagnostic results are broadcast through the intelligent voice module or a fault report is sent to the remote monitoring terminal.

[0009] Preferably, the data processing module further includes a behavior learning unit, which is used to train a behavior prediction model based on historical operation data and environmental state characteristics using supervised or semi-supervised machine learning methods. The behavior prediction model is used to predict users' device usage preferences in specific time periods and environments, and accordingly adaptively adjust lighting brightness, projection duration, or voice response strategies.

[0010] Preferably, the behavior learning unit also dynamically optimizes the startup power of the projection module, the operating level of the LED lighting module, and the broadcast frequency of the intelligent voice module based on the predicted device usage preferences and in conjunction with the real-time energy storage battery / UPS power information provided by the power supply system circuit.

[0011] Preferably, the data processing module includes a voice and image processing unit, which is used to perform voice recognition and semantic analysis on the audio data input by the intelligent voice module, and the recognized control commands or interactive content is used to update the system control strategy; and to perform image content recognition on the video data input by the monitoring module, so as to realize at least one of the functions of dynamic target detection, human tracking or scene understanding. Furthermore, when a preset specific gesture or action is detected in the video frame, it is converted into a corresponding device control command for contactless control of switching projected content, adjusting lighting modes, or retrieving system settings.

[0012] Preferably, the data processing module includes a scene adaptive decision unit, which is used to automatically select and execute a matching lighting scheme, projection triggering logic and voice interaction process by calling a preset scene rule base or decision model based on the fused environmental state characteristics, real-time time information and the currently set working mode.

[0013] Preferably, the data processing module is also used to upload system operation data, receive updated control models or algorithms, retrain the model in the cloud based on the uploaded operation data, and send the optimized model parameters or algorithm strategies to the local system to achieve continuous optimization of the control strategy. A control method for intelligent lighting projection, applied to the intelligent lighting projection system described above, includes the following steps: The main control module continuously collects environmental signals through the environmental sensors, including ambient light illuminance data collected by the ambient light sensor and human presence status data within the area collected by the human infrared sensor. The main control module processes the collected environmental signals to extract environmental features and generate control decisions accordingly. Based on the control decision, the main control module adaptively controls the start / stop and operation parameters of the connected LED lighting module and projection module through the functional module drive circuit; The main control module supports switching between commercial and home working modes based on preset conditions or external commands.

[0014] Preferably, the main control module processes the collected environmental signals to extract environmental features and generate control decisions accordingly, specifically including: The ambient light illuminance data is analyzed to determine whether the current ambient light is below a first preset threshold. The data on the presence of human beings are analyzed to determine whether human activity exists within the target area. The system combines at least the ambient light determination result and the human presence determination result to generate instructions for controlling the LED lighting module and / or the projection module. Specifically, the control decision is based on the adaptive control of the start / stop and operation parameters of the connected LED lighting module and projection module by the main control module through the functional module drive circuit, including: If it is determined that the current ambient light is below the first preset threshold and there is human activity in the target area, a first control command is generated to control the LED lighting module to turn on and adjust its brightness according to the ambient light illuminance data. When the LED lighting module is turned on or the ambient light meets the preset conditions, if a projection start signal is received, a second control command is generated to control the projection module to start and project the predetermined content. The present invention has at least the following beneficial effects: 1. By integrating and processing information from lighting, human body, audio, and video through the sensor data fusion unit, the system can construct multi-dimensional environmental state features in real time. Based on the prediction model trained by the behavior learning unit, it can proactively predict user intentions and scene requirements. This allows modules such as lighting, projection, and voice to move beyond simple threshold triggering modes and instead make collaborative decisions based on fused features and prediction results, significantly improving the system's human-computer interaction intelligence and scene adaptability.

[0015] 2. The anomaly detection and diagnosis unit establishes a normal system operation model through unsupervised learning. It can monitor electrical parameters and signal continuity in real time, promptly detect hardware faults or performance degradation, and achieve rapid localization through fault mode matching, effectively improving system maintenance efficiency and operational reliability. Simultaneously, leveraging cloud collaboration and incremental learning mechanisms, the system can continuously collect operational data to optimize local behavior models and control strategies. This allows projection start-stop logic and energy consumption scheduling schemes to continuously evolve with changes in usage habits and the environment, achieving self-improvement of system performance and long-term stability.

[0016] 3. The voice and image processing unit not only supports semantic understanding and video analysis, but also converts specific gestures or actions into control commands, providing users with an intuitive and hygienic contactless operation method. Meanwhile, the behavior learning unit can dynamically adjust the operating parameters of each module based on real-time power information, maximizing system battery life while ensuring core functionality. This achieves an effective balance between feature richness and energy sustainability under limited solar power supply conditions. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart provided for Embodiment 1 of the present invention; Figure 2 This is a unit diagram of the data processing module provided in Embodiment 1 of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "including," "having," and any variations are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. Example

[0021] This embodiment provides a smart lighting projection system and its control method, such as Figures 1-2 As shown, it includes: The main control module includes a main controller, which serves as the central hub of the system, and an auxiliary interface chip connected to the main controller. It is used to process sensor signals (the sensor signals are acquired through an ambient light sensor and a human infrared sensor), execute module control logic, and run data processing tasks. The power management module includes a charging controller for receiving energy input from the solar charging panel, an energy storage battery / UPS connected to the charging controller, and a multi-stage voltage regulator circuit that converts the output voltage of the energy storage battery / UPS into multiple voltage levels. The functional module driving circuit includes an LED lighting module, a projection module, an intelligent voice module, and a monitoring module connected to the power management module and the main control module, which are used to realize lighting, projection, voice playback and monitoring functions under the control of the main control module; The data processing module, executed by the main control module, is used to collect raw data from the functional module drive circuits and environmental sensors, and to perform feature extraction, pattern recognition, and adaptive control decisions. The main control module automatically controls the start and stop of the LED lighting module and projection module based on the output of the data processing module, and supports switching between commercial and home working modes through software logic.

[0022] Specifically, the system uses the ESP32-S3 main controller, which integrates PSRAM and FLASH, as its core to form the main control module, responsible for overall logic scheduling and computation tasks. It connects to various sensors and drive circuits via I2C and GPIO interfaces, and expands USB connectivity through auxiliary interface chips such as CH340K and CH334F, providing access points for cameras, debugging equipment, etc. The power management module begins with an 18V / 30W monocrystalline silicon solar panel. Its output efficiently charges a 12.6V energy storage lithium battery via a CN3791 MPPT charging controller, and features automatic shutdown protection logic when fully charged (SOC≥95%). The output of the energy storage battery / UPS is then precisely converted to the required 12V (for projection lamps and LED lighting), 5V (for cameras and USB devices), and 3.3V (for MCU and sensors) voltages through a multi-stage voltage regulation circuit built around chips such as EG1192L (step-down) and SY8088AAC, providing stable and adaptable energy for the entire system.

[0023] The functional module drive circuits are implemented as follows: the LED lighting module uses a PT4115 driver chip to control a 12V LED group and connects to a BH1750FVI light sensor; the projection module's 12V LED projection lamp is controlled by a switching circuit consisting of a transistor (such as Q4) and a relay (RLY1) and connects to a BISS0001 human infrared sensor; the intelligent voice module integrates an ES8311 audio codec and an NS4150B power amplifier to drive the speaker; the monitoring module connects to a camera via a USB TYPE-C interface and can be supported by a multi-channel USB hub expanded by the CH334F. All these modules are powered by the power management module, and their control signals are controlled by the corresponding GPIO of the main control module (such as IO10_RELAY controlling the relay).

[0024] The data processing module, running as a software layer on the main controller, is primarily responsible for continuously collecting raw data from sources such as the BH1750FVI, BISS0001, camera, and audio input, as well as the system's own voltage and current parameters. By performing real-time feature extraction (e.g., calculating average illumination, determining human presence, and recognizing simple voice commands) and pattern recognition (e.g., determining whether it's day or night, and recognizing movement in the monitored area), this module can make adaptive control decisions. For example, when the data processing module determines that the ambient light is below 10 Lux and the human infrared sensor is triggered, it outputs a decision to start projection to the main control module, which then controls the corresponding relay to activate the projection. The entire system, through software configuration, can switch between two preset operating modes: commercial and home. This allows for changes to advanced functions such as the voice announcements after human detection, lighting timing strategies (7:00-6:00 for commercial use, 18:00-22:00 for home use), and whether Wi-Fi projection is enabled, achieving flexible adaptation of a single hardware platform to different application scenarios.

[0025] As a further embodiment of the present invention, the data processing module includes: The sensor data fusion unit is used to fuse at least one type of data, including ambient light intensity, human infrared signal, audio input, and camera image, to generate multi-dimensional environmental state features for system control.

[0026] The anomaly detection and diagnosis unit is used to monitor the current, voltage, temperature and signal continuity of each module of the system in real time, and to establish a normal operating state model through unsupervised learning methods. Furthermore, when data is detected to deviate from the normal operating state model by more than a set threshold, an abnormal alarm is generated and corresponding protection or correction mechanisms are triggered. Then, after detecting the anomaly, a fault tracing analysis is performed. By comparing the abnormal data pattern with the preset fault feature library, the potential fault module is located, and the diagnostic results are broadcast through the intelligent voice module or a fault report is sent to the remote monitoring terminal.

[0027] Specifically, the software algorithm of the main control module (ESP32-S3) synchronously acquires, preprocesses, and correlates raw data from each sensor. Specifically, the system periodically reads the Lux value output by the BH1750FVI light sensor via the I2C bus, receives the digital trigger signal from the BISS0001 human infrared sensor via GPIO interrupt, acquires audio waveform data from the microphone input via the ES8311 codec, and obtains compressed video streams (such as MJPEG format) from the camera via the USB interface. The data fusion unit first performs timestamp alignment and normalization on these heterogeneous data, then extracts key features such as light intensity values, Boolean states indicating the presence of a human body, sound pressure level of the audio and confidence levels for specific keyword recognition, and the area and brightness distribution of moving areas in the video frame. These features are combined into a structured environmental state vector. For example, at dusk (when the light value drops to 15 Lux), the system detects a human infrared trigger, the audio module recognizes the voice command to turn on the projector, and the camera image analysis shows the presence of multiple people's silhouettes within the area. The fusion unit correlates this information to generate a multi-dimensional feature state that includes low light conditions, human activity, valid voice commands, and multi-person scenarios. This fused state serves as the core input for subsequent decisions, enabling the system to make more precise and robust control judgments than those triggered by a single sensor. In this case, the system can prioritize responding to voice commands to immediately start the projection and simultaneously adjust the LED lighting module to a moderate brightness suitable for multiple viewers, rather than relying solely on light thresholds or human body sensors.

[0028] During initial system installation or periodic maintenance, the unit will collect typical operating data from each module as a training set over a normal working period (e.g., one week). This includes: the charging and discharging current and voltage ripple of the energy storage battery / UPS obtained through ADC pin sampling; the voltage at the output terminals of each voltage regulator circuit; temperature sensor readings of key chips (e.g., ESP32-S3, PT4115); and the pulse frequency of the output signal from the human infrared sensor, as well as the frame rate and packet loss rate of the camera video stream. Unsupervised learning algorithms (e.g., based on Gaussian mixture models or a type of support vector machine) are used to model these multi-dimensional data sequences, learning their normal fluctuation range and correlation patterns to form a normal operating state model.

[0029] During online monitoring, the unit calculates in real time the deviation (or reconstruction error) of the currently collected data relative to the model. For example, the system may detect that the power supply circuit current of the projection module should be 1.2A ± 0.1A in steady state (according to the model), but the current continuous reading is 0.8A. At the same time, the temperature sensor reading of the EG1192L chip responsible for the 12V output in the multi-stage voltage regulator circuit suddenly rises from the normal 45℃ to 70℃, and the human infrared sensor signal remains high (without normal high-low level alternation). When the combined deviation of these parameters exceeds a preset threshold, the unit determines it as abnormal and triggers an alarm.

[0030] Subsequently, fault tracing and analysis are initiated. The unit compares the current abnormal patterns (low projection current, overheating of the 12V voltage regulator chip, and abnormal human body sensor signal) with a preset fault feature database. The fault feature database stores known fault patterns; for example, persistently low current + chip overheating may correspond to LED aging or a driver circuit failure in the projection lamp; persistently high human body sensor signal may correspond to the sensor being obstructed or damaged by foreign objects. Through pattern matching and probability calculation, the unit may pinpoint two potential faults: an abnormal projection module driver circuit and an abnormal human body infrared sensor. Finally, the system can announce via the intelligent voice module: "A possible decrease in projection brightness has been detected, and the human body sensor is abnormal. Please check." Simultaneously, a report containing detailed diagnostic data and possible fault points is sent to the administrator's remote terminal via Wi-Fi, thereby automating the process from anomaly detection to preliminary diagnosis and greatly improving system maintainability.

[0031] As a further embodiment of the present invention, the data processing module also includes a behavior learning unit, which is used to train a behavior prediction model based on historical operation data and environmental state characteristics through supervised or semi-supervised machine learning methods. Behavioral prediction models are used to predict users' device usage preferences in specific time periods and environments, and to adaptively adjust lighting brightness, projection duration, or voice response strategies accordingly.

[0032] Secondly, the behavior learning unit also dynamically optimizes the startup power of the projection module, the operating level of the LED lighting module, and the broadcast frequency of the intelligent voice module based on the predicted device usage preferences and the real-time energy storage battery / UPS power information provided by the power supply system circuit.

[0033] Specifically, by deploying a machine learning framework (such as TensorFlowLite for Microcontrollers) in the main control module (such as ESP32-S3), historical operation data and environmental status features are continuously collected and structured for storage. This data includes user records of turning lights on and off during specific time periods, projection usage duration, voice interaction frequency, and corresponding ambient light intensity, human body detection signals, and time period information. For example, after a week of learning, the system may discover that users have an 80% probability of turning on the projector and dimming the lighting when the ambient light is below 30 Lux and human activity is detected between 7:00 PM and 10:00 PM on weekday evenings. Based on this data, the system uses lightweight supervised learning algorithms (such as decision trees or shallow neural networks) to train a behavior prediction model. During runtime, this model predicts user intentions based on real-time collected environmental features (such as the current time being 8:00 PM, light intensity being 20 Lux, and human movement), and automatically executes control strategies accordingly: such as pre-setting the projection module to standby mode, adjusting the lighting brightness to a preset 50%, and preparing to respond to voice commands.

[0034] Furthermore, real-time battery / UPS power information provided by the power management module is used as a decision variable. The system internally sets up an energy efficiency strategy library for different power levels. For example, when the behavior prediction model indicates that projection is about to start, but the real-time battery power is below 40%, the behavior learning unit dynamically optimizes control parameters: limiting the projection module's startup power to 70% of its rated value, adjusting the LED lighting module's operating level from medium brightness to low brightness, and changing the intelligent voice module's prompt tone broadcast frequency from every trigger to once every three triggers. This optimization process is implemented through a strategy arbitrator in the embedded software, which queries a pre-set optimization mapping table based on a combination of power thresholds and behavior prediction results. This significantly extends the system's sustainable operating time during periods of intermittent or insufficient solar power supply, while closely aligning with user habits.

[0035] As a further embodiment of the present invention, the data processing module includes a voice and image processing unit, which is used to perform voice recognition and semantic analysis on the audio data input by the intelligent voice module, and the recognized control commands or interactive content is used to update the system control strategy; and to perform image content recognition on the video data input by the monitoring module, so as to realize at least one of the functions of dynamic target detection, human tracking or scene understanding. Furthermore, when a preset specific gesture or action is detected in the video frame, it is converted into a corresponding device control command for contactless control of switching projected content, adjusting lighting modes, or retrieving system settings.

[0036] Specifically, in terms of speech processing, the system uses an ES8311 audio codec to capture analog audio signals from the microphone, converts them into digital signals, and then feeds them into an integrated open-source speech recognition framework (such as Vosk or a custom model based on TensorFlow Lite Micro) for real-time recognition. This model converts the audio stream into text, and then uses a built-in semantic parser to extract key commands (such as turning on the projector or turning on the lights) or interactive intents. In terms of image processing, a camera connected to a USB hub continuously captures video streams. The system uses a cropped and quantized visual model adapted to the embedded device (such as an SSD object detection model based on MobileNetV2) to analyze video frames in real time, enabling basic functions such as dynamic object detection and human tracking.

[0037] Furthermore, this system predefines a series of control gestures (such as spreading fingers, clenching a fist, and waving to the left) and trains a specialized keypoint detection model (such as a simplified version of MediaPipe Hands) to identify the position and movement trajectory of hand skeletal points. For example, when a user makes a leftward waving gesture in front of the projection area, the camera captures the horizontal leftward displacement of the hand's keypoints in consecutive frames. The visual model recognizes this pattern and successfully matches it with the preset gesture library. Subsequently, this gesture is mapped to a device control command to switch to the previous page. This command is sent to the main control module, and the central module then performs a rollback operation of the projected content through the corresponding interface (such as simulating infrared signals or directly controlling the projector drive circuit). Similarly, the spread-finger gesture can be mapped to turn on ambient lighting, and the clenched fist gesture can be mapped to mute. This process achieves completely contactless and intuitive device control, greatly enriching the interaction dimensions, and is particularly suitable for educational demonstrations, smart home control, and other scenarios that prioritize convenience and hygiene.

[0038] As a further embodiment of the present invention, the data processing module includes a scene adaptive decision unit, which is used to automatically select and execute a matching lighting scheme, projection triggering logic and voice interaction process by calling a preset scene rule base or decision model based on the fused environmental state characteristics, real-time time information and the currently set working mode.

[0039] Specifically, the sensor data fusion unit synchronously collects heterogeneous data, including ambient light intensity (acquired by BH1750FVI), human infrared signals (detected by BISS0001), audio input (acquired by ES8311 codec), and camera footage (transmitted via USB interface). After timestamping and normalizing these data, the data fusion unit extracts key features, such as light intensity values, Boolean states indicating human presence, sound pressure levels and confidence levels for specific keyword recognition, and the area and brightness distribution of moving regions in the video footage, combining them into a structured environmental state vector. For example, in a specific scenario, if the system detects low light (15 Lux), human activity (triggered by human infrared), recognizes a voice command to turn on the projector, and the camera displays multiple silhouettes, the fusion unit correlates this information to generate a multi-dimensional feature state that includes low light, human activity, valid voice command, and multiple-person scene conditions, providing a comprehensive basis for scene adaptive decision-making.

[0040] Secondly, the scenario rule base is developed based on extensive experience from real-world application scenarios, encompassing standard response strategies for different combinations of environmental states, times, and work modes. For example, the rule base might stipulate that during business hours, when ambient light is below 20 Lux and there is human activity, the projector should be activated and the lighting brightness set to medium; during evening hours in home mode, under the same lighting and human activity conditions, in addition to activating the projector, the lighting brightness should be adjusted to a lower level to create a comfortable atmosphere. The decision model is trained using machine learning algorithms (such as decision trees and shallow neural networks), enabling it to automatically learn and predict the optimal lighting scheme, projection trigger logic, and voice interaction process based on input environmental state characteristics, time information, and work mode. For instance, by analyzing historical data, the model might discover that under specific time periods and ambient light conditions, users tend to adjust the lighting brightness to a specific range after activating the projector, and the model will provide corresponding suggestions during decision-making.

[0041] This allows the system to automatically select and execute matching lighting schemes, projection triggering logic, and voice interaction processes based on rule base matching results or decision model predictions. For example, in the scenario described above, where the fusion state is low light, people are active, voice commands are valid, and multiple people are present, if it is currently in commercial mode and during working hours, the decision unit will activate the projection based on the rule base, adjust the LED lighting module to medium brightness, and play a preset commercial scenario welcome message through the intelligent voice module. If it is in home mode and during evening hours, the decision unit will activate the projection, adjust the lighting brightness to a lower level, and play a more welcoming home scenario voice prompt. Furthermore, the decision unit has dynamic adjustment capabilities, continuously monitoring changes in environmental conditions during execution, such as further reductions in light intensity or changes in the range of human activity. It will adjust decisions in real time based on these new circumstances to ensure the system always operates in optimal condition, achieving flexible adaptation of a single hardware platform to different application scenarios.

[0042] As another embodiment of the present invention, the data processing module is also used to upload system operation data, receive updated control models or algorithms, retrain the model in the cloud based on the uploaded operation data, and send the optimized model parameters or algorithm strategies to the local system to realize continuous optimization of the control strategy.

[0043] Specifically, firstly, during local system operation, the data processing module continuously collects various types of data. This includes not only raw data from environmental sensors (such as light sensors and human infrared sensors) and functional modules (such as cameras and audio input devices), but also the system's own operating parameters, such as the charging and discharging current and voltage of the energy storage battery / UPS, the output voltage of each voltage regulator circuit, the temperature of key chips, and the operating status data of each module. After initial processing, this data is uploaded to the cloud server via the system's built-in communication module (such as a chip with Wi-Fi functionality). The upload process can be performed periodically according to preset time intervals (such as hourly or daily), or triggered when a specific event is detected (such as a system anomaly or the completion of an important task).

[0044] After receiving operational data uploaded from the local system, the cloud server stores it in a dedicated data storage system, forming a large-scale training dataset. Subsequently, leveraging the powerful computing resources of the cloud, advanced machine learning algorithms and deep learning frameworks are used to perform in-depth analysis and mining of this data. For example, deep neural network algorithms are used to retrain historical operational data and environmental state features to build a more accurate and comprehensive behavior prediction model; or the normal operation status model in the anomaly detection and diagnosis unit is optimized to improve its accuracy and sensitivity in identifying abnormal situations. During model retraining, the cloud server can continuously adjust the model's parameters and structure, enabling the model to better adapt to various complex application scenarios and user needs through extensive data iteration and optimization.

[0045] After the optimized model parameters or algorithm strategies are obtained through retraining in the cloud, the cloud server sends these optimization results to the local system via the communication network. The local system's data processing module receives these updated contents, parses and verifies them to ensure data integrity and accuracy. Then, it updates the corresponding local modules with these optimized model parameters or algorithm strategies. For example, the optimized behavior prediction model parameters are loaded into the behavior learning unit, enabling it to more accurately predict user device usage preferences during specific time periods and environments; the updated anomaly detection algorithm is applied to the anomaly detection and diagnosis unit, improving the system's ability to monitor and handle abnormal situations. In this way, the local system can continuously absorb the optimization results from cloud training, achieving continuous optimization of the control strategy, thereby improving the performance and stability of the entire smart lighting projection system and its control methods, and better meeting user needs in different scenarios. Example

[0046] This embodiment aims to provide a control method for the system of Embodiment 1, the method including the following steps: The main control module continuously collects environmental signals through environmental sensors, including ambient light illuminance data collected by an ambient light sensor and human presence status data collected by a human infrared sensor within the area. The main control module processes the collected environmental signals to extract environmental features and generate control decisions accordingly. Based on control decisions, the main control module adaptively controls the start-stop and operation parameters of the connected LED lighting module and projection module through the functional module drive circuit. The main control module supports both commercial and home working modes based on preset conditions or external commands.

[0047] Preferably, the main control module processes the collected environmental signals to extract environmental features and generate control decisions accordingly, specifically including: Analyze ambient light illuminance data to determine whether the current ambient light is below a first preset threshold; Analyze human presence data to determine whether human activity exists within the target area; At least by combining the results of ambient light assessment and human presence assessment, instructions are generated to control the LED lighting module and / or projection module; Specifically, based on control decisions, the main control module adaptively controls the start / stop and operating parameters of the connected LED lighting module and projection module through the functional module drive circuit, including: If it is determined that the current ambient light is below the first preset threshold and there is human activity in the target area, a first control command is generated to control the LED lighting module to turn on and adjust its brightness according to the ambient light illuminance data. When the LED lighting module is turned on or the ambient light meets the preset conditions, if a projection start signal is received, a second control command is generated to control the projection module to start and project the predetermined content. Example

[0048] This invention provides a non-transitory computer-readable storage medium storing at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the following steps: It processes signals from environmental sensors and generates control commands based on preset module control logic; It receives energy input from solar charging panels and uses a charging controller to charge the energy storage battery / UPS. At the same time, it uses a multi-stage voltage regulation circuit to convert the output voltage of the energy storage battery / UPS into multiple voltage levels to power various modules of the system. The driving function module driving circuit includes an LED lighting module, a projection module, an intelligent voice module, and a monitoring module, to respectively realize the functions of lighting, projection, voice playback, and monitoring; Raw data is collected from the functional module drive circuit and environmental sensors, and feature extraction, pattern recognition and adaptive control decision-making are performed on the raw data to generate data processing results; Based on the data processing results, the system automatically controls the start and stop of the LED lighting module and projection module, and switches between commercial and home operating modes according to software logic.

[0049] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0050] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. Example

[0051] This invention provides an electronic device, including a processor and a memory, wherein the memory stores at least one instruction or at least one program segment, and the at least one instruction or the at least one program segment is loaded and executed by the processor to implement the following steps: It processes signals from environmental sensors and generates control commands based on preset module control logic; It receives energy input from solar charging panels and uses a charging controller to charge the energy storage battery / UPS. At the same time, it uses a multi-stage voltage regulation circuit to convert the output voltage of the energy storage battery / UPS into multiple voltage levels to power various modules of the system. The driving function module driving circuit includes an LED lighting module, a projection module, an intelligent voice module, and a monitoring module, to respectively realize the functions of lighting, projection, voice playback, and monitoring; Raw data is collected from the functional module drive circuit and environmental sensors, and feature extraction, pattern recognition and adaptive control decision-making are performed on the raw data to generate data processing results; Based on the data processing results, the system automatically controls the start and stop of the LED lighting module and projection module, and switches between commercial and home operating modes according to software logic.

[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A smart lighting projection system, characterized in that, include: The main control module includes a main controller that serves as the central hub of the system and an auxiliary interface chip connected to the main controller. It is used to process sensor signals, execute module control logic, and run data processing tasks. The power management module includes a charging controller for receiving energy input from a solar charging panel, an energy storage battery / UPS connected to the charging controller, and a multi-stage voltage regulator circuit that converts the output voltage of the energy storage battery / UPS into multiple voltage levels. The functional module driving circuit includes an LED lighting module, a projection module, an intelligent voice module, and a monitoring module connected to the power management module and the main control module, which are used to realize lighting, projection, voice playback and monitoring functions under the control of the main control module; The data processing module, executed by the main control module, is used to collect raw data from the functional module drive circuit and environmental sensors, and perform feature extraction, pattern recognition, and adaptive control decision-making, wherein: The main control module automatically controls the start and stop of the LED lighting module and the projection module based on the output of the data processing module, and supports switching between commercial and home working modes through software logic. The sensor signals are acquired through an ambient light sensor and a human infrared sensor.

2. The intelligent lighting projection system according to claim 1, characterized in that, The data processing module includes a sensor data fusion unit, which is used to fuse at least one of the following data: ambient light intensity, human infrared signal, audio input, and camera image, to generate multi-dimensional environmental state features for system control.

3. The intelligent lighting projection system and its control method according to claim 2, characterized in that, The data processing module also includes an anomaly detection and diagnosis unit, which is used to monitor the current, voltage, temperature and signal continuity of each module of the system in real time, and to establish a normal operating state model through unsupervised learning methods. Furthermore, when data is detected to deviate from the normal operating state model by more than a set threshold, an abnormal alarm is generated and a corresponding protection or correction mechanism is triggered. Then, after detecting the anomaly, a fault tracing analysis is performed. By comparing the abnormal data pattern with the preset fault feature library, the potential fault module is located, and the diagnostic results are broadcast through the intelligent voice module or a fault report is sent to the remote monitoring terminal.

4. The intelligent lighting projection system according to claim 1, characterized in that, The data processing module also includes a behavior learning unit, which is used to train a behavior prediction model based on historical operation data and environmental state characteristics using supervised or semi-supervised machine learning methods. The behavior prediction model is used to predict users' device usage preferences in specific time periods and environments, and accordingly adaptively adjust lighting brightness, projection duration, or voice response strategies.

5. The intelligent lighting projection system according to claim 4, characterized in that, The behavior learning unit also dynamically optimizes the startup power of the projection module, the operating level of the LED lighting module, and the broadcast frequency of the intelligent voice module based on the predicted device usage preferences and the real-time energy storage battery / UPS power information provided by the power supply system circuit.

6. The intelligent lighting projection system according to claim 1, characterized in that, The data processing module includes a voice and image processing unit, which is used to perform voice recognition and semantic analysis on the audio data input by the intelligent voice module, and the recognized control commands or interactive content is used to update the system control strategy; and to perform image content recognition on the video data input by the monitoring module, so as to realize at least one of the functions of dynamic target detection, human tracking or scene understanding. Furthermore, when a preset specific gesture or action is detected in the video frame, it is converted into a corresponding device control command for contactless control of switching projected content, adjusting lighting modes, or retrieving system settings.

7. The intelligent lighting projection system according to claim 1, characterized in that, The data processing module includes a scene adaptive decision unit, which is used to automatically select and execute matching lighting schemes, projection triggering logic and voice interaction processes by calling a preset scene rule base or decision model based on the fused environmental state characteristics, real-time time information and the currently set working mode.

8. The intelligent lighting projection system according to claim 1, characterized in that, The data processing module is also used to upload system operation data, receive updated control models or algorithms, retrain the model in the cloud based on the uploaded operation data, and send the optimized model parameters or algorithm strategies to the local system to achieve continuous optimization of the control strategy.

9. A control method for intelligent lighting projection, characterized in that, The application of the intelligent lighting projection system according to any one of claims 1-8 includes the following steps: The main control module continuously collects environmental signals through the environmental sensors, including ambient light illuminance data collected by the ambient light sensor and human presence status data within the area collected by the human infrared sensor. The main control module processes the collected environmental signals to extract environmental features and generate control decisions accordingly. Based on the control decision, the main control module adaptively controls the start / stop and operation parameters of the connected LED lighting module and projection module through the functional module drive circuit; The main control module supports software logic switching between commercial and home working modes based on preset conditions or external instructions, thereby changing the adaptive control strategy.

10. The control method for intelligent lighting projection according to claim 9, characterized in that, The main control module processes the collected environmental signals to extract environmental features and generate control decisions accordingly, specifically including: The ambient light illuminance data is analyzed to determine whether the current ambient light is below a first preset threshold. The data on the presence of human beings are analyzed to determine whether human activity exists within the target area. The system combines at least the ambient light determination result and the human presence determination result to generate instructions for controlling the LED lighting module and / or the projection module. Specifically, the control decision is based on the adaptive control of the start / stop and operation parameters of the connected LED lighting module and projection module by the main control module through the functional module drive circuit, including: If it is determined that the current ambient light is below the first preset threshold and there is human activity in the target area, a first control command is generated to control the LED lighting module to turn on and adjust its brightness according to the ambient light illuminance data. When the LED lighting module is turned on or the ambient light meets the preset conditions, if a projection start signal is received, a second control command is generated to control the projection module to start and project the predetermined content.

Images