Wearable emotion regulation system and method integrated with artificial intelligence algorithm

Abstract

The invention discloses a wearable emotion regulation system and method integrated with an artificial intelligence algorithm, and belongs to the technical field of intelligent wearing, system hardware comprises an Arduino microcontroller, a biosensor array, an audio output module, a man-machine interaction module and a communication module; software comprises a multi-mode emotion recognition model, a self-adaptive music recommendation engine and a self-healing logic controller, implementation steps include data acquisition and preprocessing, emotion state recognition, music interaction strategies and closed-loop feedback adjustment, the system is integrated to clothes, and the system has specific sensor layout and a safety mechanism. The method can be used in the fields of depression adjuvant therapy, motion monitoring and the like.

Description

Technical Field

[0001] The present invention relates to the field of intelligent wearable technology, and in particular to a wearable emotion regulation system and method integrating an artificial intelligence algorithm, and in particular to an intelligent interactive clothing system and method based on an Arduino microcontroller. Background Art

[0002] At a time when artificial intelligence and Internet of Things technologies are deeply integrated, emotional health management is shifting from passive intervention to active regulation. Traditional psychological intervention relies on subjective scales and clinical interviews, and there are problems with response delays and data fragmentation. Existing wearable devices mostly focus on monitoring physiological indicators and lack the ability to actively intervene and close-loop regulation. Existing music therapy equipment requires users to actively operate and lacks intelligent interactivity. To solve these problems, the present invention proposes an intelligent interactive clothing system based on an Arduino microcontroller, which collects data such as skin electrical response and heart rate, identifies the wearer's emotional state (such as depression, anxiety, happiness, etc.) through data analysis, and automatically plays music to relieve low mood or enhance pleasure. This technology can be used for auxiliary treatment clothing for depression, and can also be applied to the research and development of equipment, devices, and systems in the fields of sports monitoring, medical rehabilitation, etc. Summary of the Invention

[0003] In order to solve the above-mentioned problems in the prior art, the present invention provides a wearable emotion regulation system and method integrated with an artificial intelligence algorithm, which has technical characteristics such as being able to monitor the wearer's emotional state in real time, respond to emotional changes in a timely manner, alleviate emotional fluctuations or enhance pleasure.

[0004] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0005] The present invention provides a wearable emotion regulation system and method integrating an artificial intelligence algorithm, comprising a hardware system and a software system connectable to a power source; the hardware system comprises a core control module, a biosensor array, an audio output module, a human-computer interaction module, and a communication module, wherein the biosensor array, the audio output module, the human-computer interaction module, and the communication module are all connected to the core control module; the software system comprises a multimodal emotion recognition model, an adaptive music recommendation engine, and a self-healing logic controller.

[0006] Preferably, the core control module is an Arduino microcontroller, which serves as the core control unit of the system; the biosensor array includes a GSR skin galvanic response sensor, a flexible piezoelectric heart rate sensor, and an IMU motion sensor, which are used to collect physiological and motion data such as skin conductivity, heart rate variability, and three-dimensional acceleration; the audio output module includes a flexible piezoelectric speaker array, which is used to output audio for regulating emotions; the human-computer interaction module includes a touch-sensitive OLED display, which provides a user interaction interface; and the communication module includes low-power Bluetooth 5.0 for data communication.

[0007] Preferably, the multimodal emotion recognition model uses an LSTM neural network that integrates temporal physiological signals and motion features to identify the user's emotional state; the adaptive music recommendation engine uses a collaborative filtering algorithm based on emotional tags to recommend appropriate music based on the emotional state; and the self-healing logic controller uses a dynamic adjustment strategy with reinforcement learning capabilities to achieve self-regulation of the system.

[0008] Preferably, the implementation steps of the system include:

[0009] Data acquisition and preprocessing: Real-time collection of 0.5-20Hz skin conductivity, heart rate variability (HRV), and three-dimensional acceleration data, using moving average filtering and z-score normalization;

[0010] Emotional state recognition: Constructs a four-dimensional emotional space including pleasure, arousal, stress index, and exercise intensity, and applies a lightweight CNN-LSTM hybrid model for real-time classification with a response time of less than 200ms.

[0011] Music interaction strategy: Build a hierarchical music library with emotional attributes labeled by mode and spectral characteristics; implement dynamic playback strategies for different emotional states, prompting exercise for depression, playing meditation music or 40-60BPM low-frequency beat music for anxiety, and activating beat-synchronized LED lighting effects (optional) for joy.

[0012] Closed-loop feedback regulation: Real-time monitoring of the rate of change of physiological indicators ΔHRV / Δt, optimization of music recommendation strategies based on the Q-learning algorithm, and personalization of music preferences and sensitivity adjustment data through the personalized setting interface on the APP side.

[0013] Preferably, the wearable emotion regulation system can be integrated into clothing, with the following specific layout: a GSR sensor is placed on the left side of the garment's chest, a piezoelectric heart rate sensor is placed on the inside of the wrist, and flexible PCB traces are routed along the seams; an IMU sensor is placed in the center of the garment's back, and a Bluetooth module is placed on the inside of the garment's hem. Reinforced areas formed by multiple layers of fabric are provided at the shoulders and hem to enhance mechanical strength and protect the sensors and modules. The system can also incorporate safety mechanisms such as overheating protection and electromagnetic shielding.

[0014] Preferably, a GSR skin galvanic response sensor, a flexible piezoelectric heart rate sensor, and an IMU motion sensor form a sensing layer, the GSR skin galvanic response sensor obtains skin galvanic response, the flexible piezoelectric heart rate sensor monitors heart rate changes, and the IMU motion sensor captures motion information. The data collected by the sensor is transmitted to the CAN bus and then passed to the processing layer;

[0015] The Arduino microcontroller constitutes the processing layer, which processes the data sent by the perception layer. The processing layer is also responsible for feature extraction and generating data features. At the same time, it realizes Bluetooth communication through the Bluetooth protocol line and can exchange data with external devices.

[0016] The flexible piezoelectric speaker array, touch-sensitive OLED display, and cloud data interface constitute the interaction layer. The flexible piezoelectric speaker array is used to realize user interaction (such as playing audio to adjust emotions); the touch-sensitive OLED display provides a user interface to facilitate user operation and information viewing; the cloud data interface is used for data upload, and the data collected and processed by the system can be uploaded to the cloud for storage and further analysis.

[0017] The present invention provides a method for regulating a wearable emotion regulation system integrating an artificial intelligence algorithm, the method comprising the following steps:

[0018] Step 1) using a biosensor array to collect EEG, IMU, and GSR data;

[0019] Step 2) performing signal processing on the collected data, including wavelet transform noise reduction and normalization;

[0020] Step 3) feature extraction was performed to extract baseline EDA mean, acceleration standard deviation, motion duration ratio, SCR amplitude peak, SCR recovery time, and low-frequency to high-frequency power ratio;

[0021] Step 4) Classify the status by depression index, anxiety index, and happiness index;

[0022] Step 5) Based on the status classification result, audio adjustment is performed using a piezoelectric speaker and a visual alarm is performed using an OLED display through biofeedback control;

[0023] Step 6) Synchronize data storage with the cloud and repeat the above steps.

[0024] Preferably, the depression index includes skin conductivity changes and active movement; the anxiety index includes skin conductivity changes and HRV low-frequency components; and the pleasure index includes HRV high-frequency components and high-frequency small-amplitude movement.

[0025] Beneficial effects: This invention can achieve real-time monitoring and active adjustment of the wearer's emotional state, making up for the shortcomings of traditional psychological intervention and existing wearable devices; it has intelligent interactivity, and can alleviate emotional fluctuations, enhance pleasure, and enhance user experience through methods such as automatic music playback; it can be applied to the development of equipment in multiple fields such as auxiliary treatment of depression, exercise monitoring, and medical rehabilitation, and has broad application prospects BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a system framework diagram of the present invention.

[0027] Figure 2 It is a dynamic interaction diagram between modules of the present invention.

[0028] Figure 3 This is one of the integrated schematic diagrams of clothing of the present invention.

[0029] Figure 4 This is the second schematic diagram of clothing integration of the present invention.

[0030] Figure 5 It is a flowchart of the algorithm processing of the present invention.

[0031] Figure 1 The system framework diagram shows the system architecture with the Arduino microcontroller as the core, connecting the biosensor array, human-computer interaction module, communication module, and audio output module.

[0032] Figure 2 : A dynamic interaction diagram between modules, showing the data transmission and interaction relationship between the system perception layer (biosensor array), processing layer (Arduino microcontroller, etc.), and interaction layer (flexible piezoelectric speaker array, etc.).

[0033] Figure 3-4 The clothing integration diagram clearly shows the layout of sensors on the front view (GSR sensor, piezoelectric heart rate sensor) and the back view (IMU sensor, Bluetooth module), as well as the circuit routing, reinforcement areas, etc.

[0034] Figure 5 The algorithm processing flow chart describes in detail the complete process from data acquisition to signal processing, feature extraction, state classification, biofeedback control, and data storage and cloud synchronization. DETAILED DESCRIPTION

[0035] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0036] 1. Design principle of the present invention:

[0037] 1.1 Hardware installation and connection: Press Figure 3-4 As shown, place the GSR galvanic skin response sensor on the left side of the chest, the flexible piezoelectric heart rate sensor on the inside of the wrist, the IMU motion sensor in the center of the back, and the Bluetooth module on the inside of the hem. Connect the flexible PCB traces to ensure that each hardware module is firmly connected.

[0038] 1.2. Software Programming and Debugging: Develop a control program based on the Arduino platform to implement a multimodal emotion recognition model (LSTM neural network), an adaptive music recommendation engine (collaborative filtering algorithm), and a self-healing logic controller (reinforcement learning strategy). Debug the program to ensure that all modules function properly together.

[0039] 1.3. System Testing and Optimization: Test subjects were asked to wear clothing integrated with the system, simulating different emotional states. Data such as skin conductivity, heart rate variability, and three-dimensional acceleration were collected and analyzed. Based on the test results, algorithm parameters were adjusted to optimize the music recommendation strategy and system regulation performance, ensuring that the system can accurately identify emotional states and effectively regulate emotions.

[0040] 2. Design and technical solutions of the present invention (system architecture: hardware system, software system)

[0041] 2.1. Hardware system:

[0042] Core control module: Arduino microcontroller (such as Figure 1 、 Figure 2 As shown in Figure 2, it serves as the core of the system and coordinates the work of various modules.

[0043] Biosensor array: including GSR skin galvanic response sensor, flexible piezoelectric heart rate sensor, IMU motion sensor (see Figure 1 、 Figure 2 、 Figure 3 The GSR galvanic skin response sensor obtains skin galvanic response, the flexible piezoelectric heart rate sensor monitors heart rate changes, and the IMU motion sensor captures motion information, jointly collecting physiological and motion data.

[0044] Audio output module: flexible piezoelectric speaker array (see Figure 1 、 Figure 2 ), used to play mood-enhancing audio.

[0045] Human-computer interaction module: Touch OLED display (optional) (see Figure 1 、 Figure 2 ), providing a user interaction interface.

[0046] Communication module: Low power Bluetooth 5.0 (see Figure 1 、 Figure 2 、 Figure 3 ) to achieve data communication.

[0047] 2.2 Software system:

[0048] Multimodal emotion recognition model: An LSTM neural network that integrates temporal physiological signals and motion features to accurately identify the user's emotional state.

[0049] Adaptive music recommendation engine: A collaborative filtering algorithm based on emotional tags recommends appropriate music based on emotional state.

[0050] Self-healing logic controller: A dynamic adjustment strategy with reinforcement learning capabilities enables the system to self-adjust to different situations.

[0051] 3. Implementation steps of the present invention

[0052] Data acquisition and preprocessing: 0.5-20 Hz skin conductivity, heart rate variability (HRV), and three-dimensional acceleration data were collected in real time. Moving average filtering and z-score normalization were used to remove noise and unify the data scale.

[0053] Emotional state recognition: Constructs a four-dimensional emotional space (pleasure, arousal, stress index, and exercise intensity) and applies a lightweight CNN-LSTM hybrid model for real-time classification. With a response time of less than 200ms, it can quickly and accurately identify emotions.

[0054] Music interaction strategy: Build a hierarchical music library with emotional attributes labeled by mode and spectral features; prompt exercise (such as dancing, walking, jogging, etc.) for depressive states, play meditation music or 40-60BPM low-frequency beat music for anxious states, and activate beat-synchronized LED lighting effects (optional) for joyful states to achieve personalized music adjustment.

[0055] Closed-loop feedback regulation: Real-time monitoring of the rate of change of physiological indicators (ΔHRV / Δt), optimization of music recommendation strategies based on the Q-learning algorithm, and provision of personalized setting interfaces such as music preferences and sensitivity adjustment through the app to continuously optimize the regulation effect.

[0056] System integration and safety mechanism: The system is integrated into the clothing (see Figure 3-4 ). In the front view, the GSR sensor is located on the left side of the chest, the piezoelectric heart rate sensor is on the inside of the wrist, and the flexible PCB traces run along the seams. In the rear view, the IMU sensor is located in the center of the back, the Bluetooth module is on the inside of the hem, and reinforced areas are located on the shoulders and hem. It also features safety features such as overheat protection and electromagnetic shielding.

[0057] Algorithm processing flow: After data acquisition (EEG / IMU / GSR), signal processing (noise reduction, normalization), feature extraction (extraction of baseline EDA mean and other features), state classification (based on depression, anxiety, and pleasure index), biofeedback control (audio adjustment, visual alarm), and finally data storage and cloud synchronization and circulation (see Figure 5 ).

[0058] Example 1:

[0059] In a specific embodiment, the present invention introduces a wearable emotion regulation system and method integrating an artificial intelligence algorithm. The system combines hardware and software, and aims to monitor the user's physiological and emotional state in real time and provide personalized emotion regulation services accordingly.

[0060] 1. Hardware System Description

[0061] The hardware system includes a core control module, a biosensor array, an audio output module, a human-computer interaction module, and a communication module. These modules work together to achieve data collection, processing, and feedback.

[0062] Core control module: preferably an Arduino microcontroller, which serves as the "brain" of the system, responsible for processing data from the biosensor array and controlling the operation of other modules.

[0063] Biosensor array: This includes GSR galvanic skin response sensors, flexible piezoelectric heart rate sensors, and IMU motion sensors. These sensors can collect real-time physiological and motion data such as skin conductivity, heart rate variability, and three-dimensional acceleration, providing a foundation for emotion recognition.

[0064] Audio output module: composed of a flexible piezoelectric speaker array, which can output corresponding audio according to the emotion recognition results to adjust the user's emotions.

[0065] Human-computer interaction module: includes a touch-sensitive OLED display, providing an intuitive user interface for easy operation and information viewing.

[0066] Communication module: Uses low-power Bluetooth 5.0 to enable data communication between the system and other external devices, such as smartphones or cloud servers.

[0067] 2. Software System Description

[0068] The software system includes a multimodal emotion recognition model, an adaptive music recommendation engine, and a self-healing logic controller.

[0069] Multimodal emotion recognition model: Using an LSTM neural network that integrates temporal physiological signals and motion features, it can accurately identify the user's emotional state.

[0070] Adaptive music recommendation engine: A collaborative filtering algorithm based on emotion tags recommends appropriate music based on the user's emotional state. This engine can learn the user's music preferences and dynamically adjust its recommendation strategy based on changes in mood. For example, when a user is depressed, it can recommend energizing music; when a user is anxious, it can play meditation music or low-frequency beats to soothe the user. The adaptive music recommendation engine is a system component built based on a collaborative filtering algorithm based on emotion tags. Its main function is to recommend appropriate music based on the user's emotional state. It has the ability to learn the user's music preferences and dynamically adjust its recommendation strategy based on changes in the user's mood. An engine generally refers to a core module in a software architecture that processes data, executes algorithms, or provides specific functionality. For example, a search engine is used to search for information on the internet, while a game engine renders graphics, handles physics simulations, and provides game logic. In this application, an engine refers to an integrated collaborative filtering algorithm with emotion tags, specifically designed to recommend music based on the user's emotional state. This engine is a specific application of existing technical concepts, combining technologies from multiple fields such as machine learning, data analysis, and music recommendation.

[0071] Self-healing logic controller: A dynamic adjustment strategy with reinforcement learning capabilities that can continuously optimize emotion regulation strategies based on system performance and user feedback. This adaptive capability enables the system to become more intelligent and effective over time. (The self-healing logic controller is an advanced system component that integrates reinforcement learning capabilities. Its core is the ability to dynamically adjust and optimize emotion regulation strategies based on the system's real-time operating conditions and direct or indirect user feedback. This controller not only has the ability to self-learn and improve, but also automatically detects and fixes potential problems during system operation, thereby achieving system "self-healing." In specific implementations, the self-healing logic controller may include the following key elements. Reinforcement learning framework: This is the core algorithm of the self-healing logic controller. Reinforcement learning is a machine learning method. In this case, the self-healing logic controller takes actions in the environment and develops the best strategy based on these actions. In an emotion regulation system, the agent may try different music recommendations, lighting effects, or vibration patterns to observe the extent to which these interventions improve the user's emotional state and adjust its strategy accordingly. State representation and feature extraction: For effective reinforcement learning, the system needs to be able to accurately represent the current state and extract useful features from it. In an emotion regulation system, this may include the user's physiological data (such as Heart rate, skin conductivity, etc.), behavioral data (such as the frequency and duration of user-system interactions), and the emotional state output by the emotion recognition model. Strategy Optimization: The self-healing logic controller continuously optimizes its strategy based on the reinforcement learning algorithm. This involves adjusting the weights in the music recommendation algorithm, changing the parameters of the lighting effects, or adjusting the intensity of the vibration pattern. Through continuous trial and error and learning, the controller can gradually find the emotion regulation strategy that best suits the current user and environmental conditions. Adaptive and Self-Healing Mechanism: In addition to optimizing the strategy, the self-healing logic controller also needs to have adaptive and self-healing capabilities. When the system detects an abnormal situation, such as a continued deterioration of the user's emotional state or a sensor failure, it can automatically adjust the strategy or take remedial measures to restore the normal operation of the system. This mechanism can include steps such as fault detection, isolation, and recovery. User Feedback Integration: To improve the intelligence and effectiveness of the system, the self-healing logic controller also needs to be able to integrate direct user feedback, such as user satisfaction surveys and preference settings, and indirect feedback, such as user behavior data and changes in emotional state. This feedback will serve as an important source of information in the reinforcement learning process, helping the controller better understand user needs and optimize the strategy.

[0072] 3. Description of Security Protection Mechanism

[0073] This system can add safety protection mechanisms such as overheating protection and electromagnetic shielding to ensure the stable operation of the system and the safety of users.

[0074] Overheat protection: Through the built-in temperature sensor (connected to the Arduino microcontroller) and heat dissipation design, when the system temperature exceeds the safety threshold, the overheat protection mechanism is automatically triggered to reduce system power consumption or shut down some functions to prevent hardware damage.

[0075] Electromagnetic shielding: Use electromagnetic shielding materials and technologies to reduce the impact of external electromagnetic interference on the system, ensuring data accuracy and system stability.

[0076] 4. Description of system implementation steps

[0077] The implementation steps of the system include data collection and preprocessing, emotional state recognition, music interaction strategy and closed-loop feedback regulation.

[0078] Data collection and preprocessing: Real-time collection of users' physiological and motion data, as well as filtering and standardization processing to improve data accuracy and reliability.

[0079] Emotional state recognition: Construct a four-dimensional emotional space and apply a lightweight CNN-LSTM hybrid model for real-time classification to quickly identify the user's emotional state.

[0080] Music Interaction Strategy: Based on emotion recognition results, a dynamic playback strategy is implemented to provide users with personalized music recommendations. At the same time, visual elements such as LED lighting effects are combined to enhance the mood regulation effect.

[0081] Closed-loop feedback regulation: Real-time monitoring of the rate of change of physiological indicators, optimization of music recommendation strategies based on the Q-learning algorithm, and personalized settings interface on the APP side, allowing users to adjust according to their preferences and needs.

[0082] 5. System Integration and Layout Description

[0083] Wearable emotion regulation systems can be integrated into clothing, achieving a perfect combination of concealment and portability. The specific layout takes into account factors such as sensor location, wiring method, and mechanical strength.

[0084] Sensor Location: The GSR sensor is placed on the left side of the garment's chest, the piezoelectric heart rate sensor is placed on the inside of the wrist, and the IMU sensor is placed in the center of the garment's back. This layout ensures close contact between the sensors and the user's body, allowing for accurate data collection.

[0085] Routing method: Flexible PCB routing is arranged along the seams, which is both beautiful and practical.

[0086] Mechanical strength: The shoulders and hem are reinforced with multiple layers of fabric to improve mechanical strength and protect sensors and modules.

[0087] 6. Data Transmission and Processing Layer Description

[0088] The system adopts a layered design, including perception layer, processing layer and interaction layer.

[0089] Perception layer: Consists of a biosensor array, responsible for collecting the user's physiological and motion data.

[0090] The processing layer, composed of an Arduino microcontroller, performs signal processing, feature extraction, and data generation on the data transmitted from the perception layer. It also interacts with external devices via the Bluetooth protocol.

[0091] Interaction layer: It consists of a flexible piezoelectric speaker array, a touch-sensitive OLED display, and a cloud data interface, and is responsible for realizing functions such as user interaction, data visualization, and cloud synchronization.

[0092] 7. Description of adjustment method

[0093] The regulation method of the present invention includes the following steps: data acquisition, signal processing, feature extraction, state classification, biofeedback control and data storage and cloud synchronization.

[0094] Data acquisition: Use biosensor arrays to collect EEG (with extended functions), IMU, GSR and other data.

[0095] Signal processing: The collected data are subjected to wavelet transform noise reduction and normalization processing to improve the accuracy and comparability of the data.

[0096] Feature extraction: Extract emotion-related feature parameters, such as baseline EDA mean, acceleration standard deviation, etc.

[0097] State classification: State classification is carried out through depression index, anxiety index and pleasure index to provide a basis for subsequent biofeedback control.

[0098] Biofeedback control: Based on the status classification results, piezoelectric speakers are used for audio adjustment, and OLED displays are used for visual alarms and other feedback methods.

[0099] Data storage and cloud synchronization: The collected and processed data is stored locally or uploaded to the cloud for further analysis to achieve long-term data preservation and in-depth mining.

[0100] In summary, the present invention provides a highly integrated, intelligent wearable mood regulation system and method. By collecting data such as galvanic skin response and heart rate and performing a series of data analyses, the system identifies or determines the wearer's emotional state (such as depression, anxiety, or joy), automatically playing music to alleviate low moods or upbeat music to enhance joy. This technology can be used not only in depression-adjuvant clothing but also in sports monitoring, medical rehabilitation, and other applications, providing users with personalized mood regulation services, with broad application prospects and market value.

[0101] Finally, it should be noted that the present invention is not limited to the above embodiments and may be subject to many variations. All variations that can be directly derived or imagined by a person skilled in the art from the disclosure of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A wearable emotion regulation system integrating artificial intelligence algorithms, characterized in that: It includes a hardware system and a software system that can be connected to a power supply; the hardware system includes a core control module, a biosensor array, an audio output module, a human-computer interaction module, and a communication module, and the biosensor array, audio output module, human-computer interaction module, and communication module are all connected to the core control module; the software system includes a multimodal emotion recognition model, an adaptive music recommendation engine, and a self-healing logic controller.

2. A wearable emotion regulation system integrated with an artificial intelligence algorithm according to claim 1, characterized in that: Core control module: An Arduino microcontroller, serving as the system's core control unit; Biosensor array: Contains a GSR galvanic skin response sensor, a flexible piezoelectric heart rate sensor, and an IMU motion sensor, used to collect physiological and motion data such as skin conductivity, heart rate variability, and three-dimensional acceleration; Audio output module: Includes a flexible piezoelectric speaker array, used to output mood-modulating audio; Human-computer interaction module: Includes a touch-sensitive OLED display, providing a user interface; Communication module: includes Bluetooth 5.0 low energy for data communication.

3. A wearable emotion regulation system integrated with an artificial intelligence algorithm according to claim 1 or 2, characterized in that: Multimodal emotion recognition model: uses an LSTM neural network that integrates temporal physiological signals and motion features to identify the user's emotional state; adaptive music recommendation engine: uses a collaborative filtering algorithm based on emotional tags to recommend appropriate music based on emotional state; self-healing logic controller: uses a dynamic adjustment strategy with reinforcement learning capabilities to achieve system self-regulation.

4. A wearable emotion regulation system integrated with an artificial intelligence algorithm according to claim 3, characterized in that: The implementation steps of the system include: Data acquisition and preprocessing: Real-time collection of 0.5-20Hz skin conductivity, heart rate variability (HRV), and three-dimensional acceleration data, using moving average filtering and z-score normalization; Emotional state recognition: Constructs a four-dimensional emotional space including pleasure, arousal, stress index, and exercise intensity, and applies a lightweight CNN-LSTM hybrid model for real-time classification with a response time of less than 200ms. Music interaction strategy: Build a hierarchical music library with emotional attributes labeled by mode and spectral characteristics; implement dynamic playback strategies for different emotional states, prompting exercise for depression, playing meditation music or 40-60BPM low-frequency beat music for anxiety, and activating beat-synchronized LED lighting effects (optional) for joy. Closed-loop feedback regulation: Real-time monitoring of the rate of change of physiological indicators ΔHRV / Δt, optimization of music recommendation strategies based on the Q-learning algorithm, and personalization of music preferences and sensitivity adjustment data through the personalized setting interface on the APP side.

5. A wearable emotion regulation system integrated with an artificial intelligence algorithm according to claim 2, characterized in that: The wearable emotion regulation system can be integrated into clothing. The specific layout is as follows: the GSR sensor is placed on the left side of the chest of the clothing, the piezoelectric heart rate sensor is placed on the inside of the wrist, and the flexible PCB is routed along the seam; the IMU sensor is placed in the center of the back of the clothing, and the Bluetooth module is placed on the inside of the hem of the clothing. The shoulders and hem are provided with reinforcement areas formed by multiple layers of fabric to improve mechanical strength and protect each sensor and module.

6. A wearable emotion regulation system integrated with an artificial intelligence algorithm according to claim 2 or 5, characterized in that: The GSR galvanic skin response sensor, flexible piezoelectric heart rate sensor, and IMU motion sensor form the perception layer. The GSR galvanic skin response sensor obtains skin electrical response, the flexible piezoelectric heart rate sensor monitors heart rate changes, and the IMU motion sensor captures motion information. The data collected by the sensors is transmitted to the CAN bus and then passed to the processing layer. The Arduino microcontroller constitutes the processing layer, which processes the data sent by the perception layer. The processing layer is also responsible for feature extraction and generating data features. At the same time, it realizes Bluetooth communication through the Bluetooth protocol line and can exchange data with external devices. The flexible piezoelectric speaker array, touch-sensitive OLED display, and cloud data interface constitute the interaction layer. The flexible piezoelectric speaker array is used to realize user interaction; the touch-sensitive OLED display provides a user interface to facilitate user operation and information viewing; the cloud data interface is used for data upload, and the data collected and processed by the system can be uploaded to the cloud for storage and further analysis.

7. A method for regulating a wearable emotion regulation system integrated with an artificial intelligence algorithm according to any one of claims 1 to 6, characterized in that The method comprises the following steps: Step 1) using a biosensor array to collect EEG, IMU, and GSR data; Step 2) performing signal processing on the collected data, including wavelet transform noise reduction and normalization; Step 3) feature extraction was performed to extract baseline EDA mean, acceleration standard deviation, motion duration ratio, SCR amplitude peak, SCR recovery time, and low-frequency to high-frequency power ratio; Step 4) Classify the status by depression index, anxiety index, and happiness index; Step 5) Based on the status classification result, audio adjustment is performed using a piezoelectric speaker and a visual alarm is performed using an OLED display through biofeedback control; Step 6) Synchronize data storage with the cloud and repeat the above steps.

8. The method for regulating a wearable emotion regulation system integrated with an artificial intelligence algorithm according to claim 7, characterized in that: The depression index includes changes in skin conductivity and active exercise; the anxiety index includes changes in skin conductivity and low-frequency components of HRV; and the pleasure index includes high-frequency components of HRV and high-frequency small-amplitude exercise.

Images