AI-based collaborative innovation system and intelligent wearable terminal

Abstract

The invention relates to the technical field of AI collaborative innovation, in particular to an AI-based collaborative innovation system and intelligent wearable terminals.The AI-based collaborative innovation system comprises a plurality of intelligent wearable terminals and a collaborative innovation platform connected with the terminals, and the intelligent wearable terminals are used for collecting physiological data, action data and voice data of employees; the information is sent to the collaborative innovation platform; the collaborative innovation platform comprises a data processing module, an AI analysis module and an innovation auxiliary module, and the data processing module is used for preprocessing data sent by the intelligent wearable terminal; the AI-based collaborative innovation system is combined with the intelligent wearable terminal, so that the physiological data, the action data and the voice data of the employees can be acquired in real time, the physical and mental states of the employees in the innovation process can be comprehensively known by the collaborative innovation platform, and a data basis is provided for personalized innovation support and service.

Description

Technical Field

[0001] This invention relates to the field of collaborative innovation technology in AI, and more particularly to an AI-based collaborative innovation system and a smart wearable terminal. Background Technology

[0002] In today's rapidly evolving work environment, team collaboration and innovation play a crucial role in enhancing a company's competitiveness. Existing collaborative innovation tools, primarily traditional office software, online collaboration platforms, and some simple wearable devices, do not fully consider the physiological and psychological states of employees during the innovation process, failing to monitor their status in real time and provide personalized innovation support. Furthermore, while existing wearable devices such as smart bracelets and smartwatches can collect some physiological data, such as heart rate and steps, these devices typically have limited functionality, weak correlation with the collaborative innovation process, and lack deep integration with team collaboration and innovation platforms, failing to provide effective intelligent assistance and guidance for employees in team collaboration and innovation. Therefore, existing collaborative innovation tools and technologies are insufficient in fully tapping into employee innovative potential and providing intelligent collaborative innovation support, making it difficult to meet the complex needs of modern enterprises in team collaboration and innovation. To address these issues, this invention proposes an AI-based collaborative innovation system and an intelligent wearable terminal. Summary of the Invention

[0003] To achieve the above objectives, this invention provides an AI-based collaborative innovation system, comprising multiple smart wearable terminals and a collaborative innovation platform connected to the terminals. The smart wearable terminals are used to collect employees' physiological data, motion data, and voice data, and send them to the collaborative innovation platform. The collaborative innovation platform includes a data processing module, an AI analysis module, and an innovation assistance module. The data processing module is used to preprocess the data sent by the smart wearable terminals. The AI ​​analysis module is used to evaluate the employees' innovation status based on the preprocessed data and generate innovation suggestions. The innovation assistance module is used to provide employees with corresponding innovation resources and collaborative tools based on the innovation suggestions.

[0004] Optionally, the smart wearable terminal includes a data acquisition unit, a data transmission unit, and a human-computer interaction unit. The data acquisition unit is used to collect employees' physiological data, motion data, and voice data. The data transmission unit is used to send the collected data to the collaborative innovation platform. The human-computer interaction unit is used to receive innovation suggestions and collaborative information sent by the collaborative innovation platform and display them to employees.

[0005] Optionally, the collaborative innovation platform also includes a user management module for managing employee user accounts, including account creation, permission allocation, and information maintenance functions, to ensure data security and the orderly operation of the system.

[0006] Optionally, the AI ​​analysis module uses deep learning algorithms to evaluate the innovation status of employees. By constructing a neural network model and training it on a large amount of historical innovation data of employees, it can accurately judge the current innovation status of employees and generate personalized innovation suggestions.

[0007] The AI ​​analysis module uses deep learning algorithms to evaluate employees' innovation status. Its training and application process is as follows:

[0008] Data preprocessing:

[0009] Normalization method: The collected physiological data (such as heart rate, blood pressure, body temperature, etc.), motion data (such as acceleration, angular velocity, etc.), and speech data (such as speech rate, pitch, etc.) are normalized to scale each feature value to the range [0,1]. The Min-Max normalization method is used, and the formula is:

[0010]

[0011] Where x is the original data, x min and x max ...

[0012] Feature extraction: Features such as heart rate variability (HRV), blood pressure stability (measured by the standard deviation of blood pressure fluctuations), and body temperature change rate are extracted from physiological data; features such as activity intensity (calculated using the magnitude of acceleration), frequency of movement pattern changes (based on the statistical analysis of the number of posture transitions), and limb coordination (derived through the analysis of the synchronicity of movements in different body parts) are extracted from motion data; features such as speech rate (number of words spoken per minute), pitch variation range (derived through the analysis of frequency changes in the speech signal), and keyword frequency (based on word frequency statistics of innovation-related words) are extracted from speech data. These extracted features are combined into a comprehensive feature vector with [specific numerical] dimensions (e.g., 30 dimensions), which serves as the input to the model.

[0013] Deep Learning Model Architecture and Parameter Settings: A Long Short-Term Memory (LSTM) network model is used to model and analyze the innovation status of employees. The LSTM model contains a [specific number] (e.g., 3) hidden layers, and each hidden layer has a [specific number] (e.g., 128) neurons. The input gate, forget gate, and output gate of the LSTM all use the sigmoid activation function, and the candidate memory units use the tanh activation function. The output layer of the model uses the softmax activation function to map the output to different innovation status categories (such as high innovation enthusiasm, innovation bottleneck period, innovation fatigue state, etc.).

[0014] During model training, the Adam optimization algorithm was used to train the LSTM model, with an initial learning rate of 0.001, beta1 (exponential decay rate for first-order moment estimation) set to 0.9, and beta2 (exponential decay rate for second-order moment estimation) set to 0.999. The cross-entropy loss function was used, with the following formula:

[0015]

[0016] Where yi represents the real label. Here, n represents the probability value predicted by the model, and n is the number of categories.

[0017] Furthermore, the model optimization strategy includes: Hyperparameter tuning: By performing multiple iterations of training on the training set and validation on the validation set, the model's hyperparameters are adjusted based on the loss function value and accuracy on the validation set. For example, a grid search method is used to search for hidden layer neurons with 64, 128, and 256 neurons and dropout rates of 0.2, 0.3, 0.4, and 0.5. Ultimately, a model with 128 hidden layer neurons and a dropout rate of 0.3 achieves an accuracy of 92.5% on the validation set. Regularization and overfitting prevention: An L2 regularization term is added to the model with a regularization coefficient of 0.001 to prevent overfitting. Simultaneously, a dropout layer is added after each hidden layer to randomly discard the output of some neurons, further improving the model's generalization ability.

[0018] Optionally, the innovation support module provides innovation resources including industry reports, technical literature, and case studies, and collaboration tools including online document editing, real-time communication, task assignment, and tracking functions to meet the various needs of employees during the innovation process. The innovation support module provides employees with abundant innovation resources and convenient collaboration tools, enabling them to more efficiently obtain the necessary information and collaborate during the innovation process.

[0019] The innovation support module provides employees with abundant innovation resources and convenient collaboration tools, and its functions are as follows:

[0020] Innovation Resource Management and Recommendation: Resource Repository Construction: Establish an innovation resource repository on the collaborative innovation platform, encompassing various types of resources, including industry reports, technical literature, case studies, and creative materials. Resources are meticulously categorized and tagged, for example, by industry sector (e.g., artificial intelligence, biomedicine, intelligent manufacturing), technology category (e.g., algorithms, materials, machinery), and innovation type (e.g., product innovation, process innovation, business model innovation). Collaborative Filtering-Based Approach: Data Collection and Analysis: Collect historical behavioral data of employees on the collaborative innovation platform, including browsing, downloading, saving, and liking records, as well as interest tags actively marked by employees. Analyze this data to construct employee interest profiles. For example, determine employees' focus on specific technology areas based on the frequency and category of their browsing of technical literature; extract their industry interests based on the topics of industry reports they save. Collaborative Filtering Model Construction: Calculate the interest similarity among employees using a cosine similarity algorithm. Resource Matching and Recommendation: When recommending resources, first determine the scope of recommended topics based on the employee's innovation status assessment results. For example, if an employee is currently experiencing an innovation bottleneck and the innovation task involves the field of artificial intelligence, the recommended topics would focus on expanding innovative ideas and case studies related to artificial intelligence. Then, resources used by other employees with similar interests would be selected, and a weighted average algorithm would be used to calculate the matching degree between resources and employees, combining resource tagging information and the employee's interest profile.

[0021] To achieve the above objectives, the present invention provides an intelligent wearable terminal, including the aforementioned AI-based collaborative innovation system, and further including a data acquisition unit, a data processing unit, a data transmission unit, and a human-computer interaction unit. The data acquisition unit is used to collect employees' physiological data, motion data, and voice data. The data processing unit is used to perform preliminary processing on the collected data. The data transmission unit is used to send the processed data to the collaborative innovation platform. The human-computer interaction unit is used to receive innovation suggestions and collaborative information sent by the collaborative innovation platform and display them to employees.

[0022] Optionally, the data acquisition unit includes a physiological sensor, a motion sensor, and a voice sensor. The physiological sensor is used to collect employees' heart rate, blood pressure, and body temperature physiological data. The motion sensor is used to collect employees' limb movements and posture changes. The voice sensor is used to collect employees' voice information.

[0023] The system employs a distributed motion sensor layout, placing multiple small motion sensors on different parts of the smart wearable device (such as the arms, legs, and torso) to accurately capture the employee's full-body movements. This helps to more accurately analyze employees' posture, movement fluency, and behavioral habits during work. For example, in team-based creative activities, full-body motion data can be used to determine the interaction patterns and level of teamwork among employees, providing AI with more multi-dimensional behavioral characteristic information.

[0024] Improving the sampling frequency and accuracy of motion sensors enables them to capture more subtle changes in movement, such as minute finger movements and eye movements. This detailed motion data is particularly important in work scenarios that require precise operation and creative thinking, such as when designers are drawing or engineers are operating complex equipment; their finger movements and eye focus can reflect their creative process and concentration on their work.

[0025] Furthermore, in addition to existing heart rate, blood pressure, and body temperature sensors, the data acquisition unit can also incorporate an electroencephalogram (EEG) sensor to monitor employees' brain activity levels, concentration, and cognitive activity, providing a more accurate understanding of their cognitive state and level of creativity. For example, the EEG sensor can capture neural activity patterns during the creative process, providing deeper physiological evidence for AI analysis of employees' innovative states.

[0026] Furthermore, a muscle electrophysiological activity sensor is added to the data acquisition unit to detect minute electrical changes in the employee's limb muscles, thereby inferring the employee's level of physical fatigue, stress level, etc., since physical condition also affects creativity. For example, when an employee's muscles are in a state of tension for a long time, it may affect their mental flexibility and creativity. This data can provide a more comprehensive reference for the AI ​​system to comprehensively evaluate the employee's innovation status.

[0027] Preferably, the data acquisition unit integrates environmental sensing components such as ambient light sensors and sound sensors to monitor environmental factors such as light intensity and noise levels in the employee's work environment in real time. This is because the environment plays a significant role in an employee's mood, attention, and creativity. For example, excessively strong or weak light can affect an employee's visual comfort and mental agility, while noise interference can distract them, thus impacting their creativity. By combining this environmental data with the employee's physiological and behavioral data, the AI ​​system can more comprehensively analyze the relationship between the employee's innovative state and the environment, providing personalized suggestions for optimizing the innovation environment.

[0028] Optionally, the human-computer interaction unit includes a display screen, a speaker, and touch buttons. The display screen is used to display innovative suggestions, collaborative information, and terminal status information. The speaker is used to play voice prompts. The touch buttons are used for employee input.

[0029] Optionally, a location module may also be included to obtain employees’ location information in real time and send it to the collaborative innovation platform to facilitate location tracking and team collaboration management during the collaborative innovation process.

[0030] Optionally, the terminal also has an energy management system to optimize the terminal's energy consumption, extend the terminal's battery life, and ensure the terminal's continuous and stable operation during employee work.

[0031] The beneficial effects of this invention are as follows:

[0032] This invention combines an AI-based collaborative innovation system with a smart wearable terminal, enabling real-time collection of employees' physiological, motion, and voice data. This allows the collaborative innovation platform to fully understand employees' physical and mental state during the innovation process, providing a data foundation for personalized innovation support and services.

[0033] By leveraging AI analytics to assess employees' innovation status and generate innovation suggestions, their innovative potential can be effectively unlocked, and their creativity stimulated. AI can accurately analyze employees' strengths and weaknesses in the innovation process based on their historical data and real-time status, providing targeted improvement suggestions to help them overcome innovation bottlenecks and enhance their innovative thinking abilities.

[0034] The innovation support module provides employees with abundant innovation resources and convenient collaboration tools, enabling them to obtain necessary information and collaborate more efficiently during the innovation process. Employees no longer need to switch between multiple platforms to search for information and communicate, saving time and energy and allowing them to focus more on the innovation task itself, thereby improving innovation efficiency and quality.

[0035] The human-computer interaction unit of the smart wearable terminal can promptly display innovation suggestions and collaborative information sent from the collaborative innovation platform to employees, enabling real-time information transmission and feedback, and ensuring the closeness and continuity of team collaboration. Employees can stay informed about the team's innovation progress and their own task assignments, better collaborate with team members, and promote the achievement of the team's overall innovation goals.

[0036] For businesses, the application of this invention helps create a more positive and efficient innovation atmosphere, enhancing their innovation capabilities and competitiveness. By stimulating employee creativity and improving team collaboration efficiency, businesses can launch innovative products and services more quickly in fierce market competition, adapt to market changes, and achieve sustainable development. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating the overall system flow of the AI-based collaborative innovation system of this invention.

[0038] Figure 2 This is a flowchart of the data acquisition and preprocessing process in the AI-based collaborative innovation system of this invention;

[0039] Figure 3 This is a flowchart illustrating the AI ​​analysis and innovation assistance process in the AI-based collaborative innovation system of this invention.

[0040] Figure 4 This is a flowchart illustrating the interaction process of the smart wearable terminal of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention. Unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by those skilled in the art. The terms "comprising" and similar expressions used herein mean that the element or object preceding the word covers the element or object listed following the word and its equivalents, but do not exclude other elements or objects.

[0042] This invention relates to an AI-based collaborative innovation system and a smart wearable terminal. The system aims to collect multi-dimensional data on employees' physiological, behavioral, and vocal aspects through smart wearable devices, analyze this data using AI technology, and provide employees with personalized innovation support and collaborative tools. This, in turn, stimulates employee creativity and improves team collaboration and innovation efficiency. The specific implementation methods of this application will be described in detail below.

[0043] System architecture setup (see) Figure 1 ):

[0044] Hardware environment setup: Install and deploy servers as the operating platform for the collaborative innovation platform. Select servers with high-performance CPUs, large-capacity memory, and high-speed data storage devices to meet the needs of AI data processing and model computation. For example, use Intel Xeon series processors, equipped with 64GB or more of memory, and 1TB solid-state drives for data storage.

[0045] Select appropriate hardware chips and sensor components for smart wearable terminals. For example, use low-power, high-precision physiological sensor chips (such as the MAX30102 heart rate sensor chip) to collect physiological data such as heart rate and blood oxygen; select the MPU6050 nine-axis motion sensor for motion data acquisition; and incorporate a high-sensitivity silicon microphone for voice data acquisition. Simultaneously, equip the device with a low-power Bluetooth chip (such as the CSR1010) for data transmission, and a small lithium battery to provide power, along with a corresponding charging management circuit.

[0046] Software environment configuration: Install a Linux operating system (such as Ubuntu Server) on the server and deploy a Python runtime environment (Python 3.8 or later) for developing and running the backend services and AI algorithm models of the collaborative innovation platform. Simultaneously, install a database management system (such as MySQL or MongoDB) to store multi-dimensional employee data and innovation-related data.

[0047] Develop embedded software for smart wearable devices, using real-time operating systems such as RT-Thread or FreeRTOS, responsible for driving and controlling various sensors, data acquisition and preprocessing, and data communication with the collaborative innovation platform. Develop corresponding mobile applications (based on Android and iOS platforms) for employees to view innovation suggestions, collaborative information, and perform related operations.

[0048] Data acquisition and preprocessing (see Figure 2 ):

[0049] Smart wearable terminal data acquisition: When employees wear smart wearable terminals, physiological sensors collect heart rate, blood pressure, and body temperature physiological data at fixed time intervals (e.g., once per second) and convert them into digital signals. Motion sensors collect employees' limb movements and posture changes in real time, outputting raw data such as triaxial acceleration and angular velocity at a sampling frequency of 100Hz. When the voice sensor detects employee voice input (triggered via voice wake-up function), it begins recording voice signals and converts them into digital audio data.

[0050] The terminal device performs the following preprocessing operations on the various types of data collected:

[0051] Physiological data filtering: Low-pass filtering algorithms are used to remove high-frequency noise from physiological data. Taking heart rate data as an example, considering that the frequency range of heart rate signals is generally around 0.5-5Hz, a Butterworth low-pass filter with a cutoff frequency of 10Hz is selected. This is because the Butterworth filter has the characteristics of flat amplitude-frequency characteristics and good linearity of phase-frequency characteristics, which can effectively retain the main frequency components of the heart rate signal while effectively filtering out high-frequency noise interference above 10Hz, such as high-frequency noise generated by equipment electromagnetic interference, thereby improving the accuracy of heart rate data.

[0052] Motion data coordinate transformation and fusion calculation:

[0053] Data Transmission and Preliminary Processing: Coordinate Transformation Method: The raw motion data collected by motion sensors (such as the MPU6050 nine-axis motion sensor) is in its own coordinate system. To transform it to a unified global coordinate system for more accurate analysis of employee posture and trajectory, a rotation matrix is ​​used for coordinate transformation. Assuming the rotation relationship between the sensor coordinate system and the global coordinate system can be represented by the rotation matrix R, the transformed motion data vector is: global_data = R × sensor_data. The rotation matrix R can be determined by initial posture calibration of the sensor. The calibration process can employ a multi-position calibration method, where the sensor is placed in multiple known postures (e.g., horizontal, vertical, etc.), and its acceleration and geomagnetic data are measured in these postures. Then, the rotation matrix R is calculated using algorithms such as the least squares method.

[0054] Data fusion algorithm: The Extended Kalman Filter (EKF) algorithm is used to fuse multi-sensor data from the accelerometer, gyroscope, and magnetometer to obtain more accurate motion state information. Its observation equation is:

[0055]

[0056] Where, x k∣k-1 P is the predicted state value. k∣k-1 To predict the covariance matrix, y k h is the observed value. (xk∣k-1) For the observation model, v k To observe the noise, S k To observe the covariance matrix, R k To observe the noise covariance matrix, K k For Kalman gain, x k∣k Let Pk be the state estimate, and Pk|k be the estimated covariance matrix. Relevant parameters are optimized based on the actual data acquisition environment and sensor performance (such as sensor accuracy and sampling frequency).

[0057] Voice data processing: Audio data is converted and compressed to reduce data volume. Specifically, MP3 format is used for compression, and the compression ratio can be set between 10:1 and 20:1 according to actual needs, effectively reducing data storage space and transmission bandwidth while ensuring the intelligibility of the voice signal.

[0058] The pre-processed data is sent to employees' mobile devices (such as smartphones or tablets) via Bluetooth, and then the mobile devices upload the data to the collaborative innovation platform via Wi-Fi or mobile network. Corresponding data receiving and forwarding programs are developed on the mobile devices to ensure the stability and reliability of data transmission.

[0059] The collaborative innovation platform's data processing module further analyzes and organizes the received data, aligning different types of physiological, motor, and voice data according to timestamps to form a complete sequence of employee status data. For example, it associates physiological data such as heart rate and blood pressure with corresponding time-based motor and voice data, enabling the subsequent AI analysis module to accurately analyze changes in employees' physiological states under different behavioral and voice communication scenarios.

[0060] Preferably, an edge computing unit is integrated into the data processing module. Specifically, a small edge computing chip is integrated inside the smart wearable terminal to perform preliminary processing and filtering of the collected data. This reduces the pressure of transmitting large amounts of raw data to the collaborative innovation platform, improving data transmission efficiency and real-time performance. For example, the edge computing unit can perform simple feature extraction on physiological data, such as calculating basic indicators like the fluctuation range of heart rate and the stability of blood pressure, prioritizing the transmission of these key feature data. Redundant raw sampling data is stored locally or compressed before transmission, ensuring that the collaborative innovation platform can more quickly obtain key employee status information for innovation status assessment.

[0061] By leveraging edge computing to perform real-time pre-analysis of data, it's possible to preliminarily determine whether employees' innovation status is abnormal or undergoes noteworthy changes at the terminal side. For example, if the edge computing unit detects a sudden and significant fluctuation in an employee's physiological data (such as a sharp increase in heart rate or abnormal changes in body temperature), it may indicate that the employee is under excessive stress or experiencing emotional fluctuations. In this case, an early warning message can be promptly sent to the collaborative innovation platform, enabling the platform to respond quickly, such as adjusting innovation task assignments or providing psychological counseling suggestions.

[0062] AI analysis module training and application (see...) Figure 3 ):

[0063] Data annotation and model training data preparation: Collect a large amount of historical data on employees in different innovation states, including physiological, behavioral, and vocal data, as well as corresponding labels such as innovation achievements and task completion status. For example, extract data on employees during successful innovation and innovation bottleneck stages from internal R&D project records, and annotate and classify them. Preprocess this data, including data cleaning, normalization, and feature extraction, to meet the requirements of model training.

[0064] Based on the needs of innovation status assessment, corresponding feature vectors are designed to fuse key features from physiological, motor, and speech data. For example, features such as heart rate variability and blood pressure stability are extracted from physiological data; features such as activity intensity and movement pattern changes are extracted from motor data; and features such as speech rate, tone, and keyword frequency are extracted from speech data. These features are combined into a comprehensive feature vector, which serves as the input to the model.

[0065] AI Model Selection and Training: Choose a suitable deep learning model architecture, such as Long Short-Term Memory (LSTM) or Transformer, for modeling and analyzing employee innovation status. Taking the LSTM model as an example, construct a neural network structure containing multiple LSTM layers and fully connected layers, and optimize the model by adjusting its hyperparameters (such as the number of hidden layer neurons, learning rate, dropout rate, etc.).

[0066] The model is trained using prepared training data, and the backpropagation algorithm is used to update the model's weight parameters to minimize the error between the predicted results and the actual labels. During training, cross-validation is used to evaluate and validate the model's performance, ensuring that the model has good generalization ability and accuracy. For example, the training dataset is divided into an 80% training set and a 20% validation set, and through multiple iterations of training and validation, the model's accuracy on the validation set reaches over 90%.

[0067] Innovation Status Assessment and Recommendation Generation: When the collaborative innovation platform receives new employee status data, the AI ​​analysis module inputs this data into a trained model. The model outputs an assessment result of the employee's current innovation status, such as high innovation enthusiasm, innovation bottleneck period, or innovation fatigue. Based on the innovation status assessment result, corresponding suggestions are matched from a pre-designed innovation suggestion library, such as adjusting the difficulty of work tasks, providing creative stimulation activities, and arranging rest time. These suggestions are then sent to the employee's mobile device or smart wearable terminal through the collaborative innovation platform.

[0068] The innovation suggestion library is built upon extensive research in the psychology of innovation, corporate innovation management experience, and employee feedback data. For example, to address innovation bottlenecks, the library includes methods to guide employees in brainstorming, cross-departmental communication, and exposure to new fields of knowledge; to address innovation fatigue, it offers suggestions such as relaxation training, adjusting the work environment, and appropriately increasing recreational activities. Simultaneously, the platform dynamically updates and optimizes the innovation suggestion library based on actual employee feedback and changes in innovation status to improve the relevance and effectiveness of the suggestions.

[0069] Innovative auxiliary module function implementation (see) Figure 3 ):

[0070] Innovation Resource Management and Recommendation: Establish a rich innovation resource library on the collaborative innovation platform, including industry reports, technical documents, case studies, and creative materials. Classify and tag these resources, such as by industry sector, technology category, and innovation type. For example, categorize industry reports into different fields like artificial intelligence, biomedicine, and intelligent manufacturing, and classify technical documents according to technology categories such as algorithms, materials, and mechanics.

[0071] Based on employees' innovation status assessment results, personal interests, and historical resource usage records, corresponding resource recommendation algorithms are developed. For example, a collaborative filtering-based approach can be used to analyze the resources used by other employees with similar innovation statuses and recommend resources that the current employee might be interested in; or, based on the theme and keywords of the employee's current innovation task, relevant literature and case studies can be retrieved and recommended from the resource library. The recommended resource information is then displayed to employees via smart wearable devices or mobile devices, allowing them to directly access resources by clicking links or downloading them locally for viewing.

[0072] Collaboration Tool Integration and Application: Develop and integrate a series of collaboration tools into the collaborative innovation platform, including online document editing, real-time communication, and task assignment and tracking functions. The online document editing tool supports real-time collaborative editing of documents, drawings, etc., among employees, recording each employee's editing operations and version history, facilitating idea integration and output by team members. The real-time communication tool provides text chat, voice calls, and video conferencing functions, ensuring timely communication and exchange of innovative ideas and work progress among employees. The task assignment and tracking tool can break down and assign tasks to different employees according to the needs of the innovation project, setting task priorities, deadlines, and other attributes, and tracking task completion status in real time, providing effective task management support for team collaborative innovation.

[0073] Optimize the interface design and operation flow of collaboration tools to make their interaction with smart wearable terminals more convenient and efficient. For example, design simple communication buttons and task viewing shortcuts on smart wearable terminals, allowing employees to quickly send messages, answer calls, or check task progress via voice commands. Simultaneously, ensure data synchronization and sharing between collaboration tools across different devices, enabling employees to access and operate collaboration tools anytime on devices such as mobile phones, tablets, and computers, achieving seamless team collaboration.

[0074] User interaction design for smart wearable devices (see) Figure 4 ):

[0075] Interactive Interface Development and Customization: Design the interactive interface for the smart wearable terminal using development tools such as Qt or Android Studio. The interface layout adopts a simple and intuitive design style, displaying the innovation status information most relevant to employees (such as current innovation status evaluation results, innovation suggestion reminders, etc.) with prominent icons and text at the top of the main interface. Multiple functional module entry points are provided, such as data viewing, suggestion reception, collaboration tools, and resource access. Employees can switch between different modules via touch operations or gesture swipes.

[0076] The system offers a customizable interface, allowing employees to show or hide specific functional modules and adjust their order and size based on their personal habits and work needs. For example, employees can enlarge the innovation suggestion module and place it in the center of the main interface for easy access; or hide less frequently used data viewing modules to reduce interface complexity. It also supports multiple theme styles, allowing employees to choose different color schemes and icon styles to improve user satisfaction with the wearable device.

[0077] Multimodal Interaction Implementation and Optimization: Implement multimodal interaction functions, including touch operation, voice interaction, and gesture recognition. For touch operation, optimize the sensitivity and response speed of the touchscreen to ensure employees can smoothly perform clicks, swipes, zooms, and other operations. For voice interaction, employ advanced speech recognition algorithms (such as deep learning-based speech recognition models) to improve the accuracy of voice command recognition and support the recognition of multiple dialects and accents. For example, employees can quickly open the innovation suggestion module using the voice command "View my innovation suggestions," or directly save creative content to the collaborative innovation platform's note-taking function by voice inputting "Record my creative ideas."

[0078] For gesture recognition interaction, corresponding gesture recognition algorithms are developed. By analyzing and classifying gesture data collected by cameras or motion sensors, accurate recognition of employee gesture commands can be achieved. For example, employees can open or close a function module by waving their arms in the air, or confirm operation commands by clenching their fists. The accuracy and anti-interference ability of the gesture recognition algorithm are continuously optimized to ensure accurate recognition of employee gestures under different lighting conditions and postures, providing employees with a more natural and convenient interactive experience.

[0079] System integration and testing verification:

[0080] System Integration and Debugging: Integrate the various modules of the collaborative innovation platform (data processing module, AI analysis module, innovation assistance module, etc.) with the hardware and software system of the smart wearable terminal to ensure normal data communication and function calls between the parts. On the server side, configure the interfaces and data transmission protocols between the various modules, start the backend service program to enable it to receive data from the smart wearable terminal and send innovation suggestions and collaborative information to the terminal. On the smart wearable terminal side, install and run embedded software and mobile applications, establish a connection with the collaborative innovation platform, and complete the functions of data collection, transmission, reception, and display.

[0081] Conduct system integration testing to simulate employee usage in different scenarios and check the overall system operation. For example, simulate employees wearing smart wearable devices during creative design tasks, such as brainstorming and sketching, to observe whether the collaborative innovation platform can accurately collect employees' physiological, motor, and voice data; whether the AI ​​analysis module can correctly assess employees' innovative state and generate reasonable innovation suggestions; whether the innovation assistance module can provide timely support with relevant resources and collaborative tools; and whether the smart wearable device can accurately display this information and respond to employees' operation commands. Address and optimize any issues discovered during the integration testing process to ensure the system's stability and reliability.

[0082] Testing, Verification, and Optimization: A series of test cases were designed to conduct detailed testing and verification of the system's various functions and performance indicators. Test content included the accuracy of data collection, the stability of data transmission, the accuracy of AI analysis, the relevance of innovative suggestions, the ease of use of collaboration tools, and the battery life of the smart wearable terminal. For example, the accuracy of physiological data collection was verified by comparing the physiological data collected by the smart wearable terminal with the measurement results of professional medical equipment; packet loss rate and latency were tested by continuously transmitting data under different network environments; and feedback on system functionality and usability was collected by inviting employees with different professional backgrounds to participate in the testing.

[0083] Based on the test results, the system was optimized and improved. For example, to address data transmission stability issues, the communication protocol and data transmission algorithm were optimized, and a data retransmission mechanism and error correction algorithm were adopted to improve data transmission reliability. To address the insufficient accuracy of AI analysis, model parameters were further adjusted and the algorithm structure optimized, the amount and diversity of training data were increased, and the model's generalization ability was improved. To address the insufficient battery life of smart wearable terminals, hardware circuit design and software power management strategies were optimized to reduce device power consumption and extend battery life. Through continuous testing, verification, and optimization, the system was ensured to meet the performance and functional requirements of practical applications, providing effective support for employee collaborative innovation.

[0084] While embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations fall within the scope and spirit of the present invention. Furthermore, the present invention described herein may have other embodiments and can be implemented or carried out in various ways.

Claims

1. An AI-based collaborative innovation system, characterized in that, The system includes multiple smart wearable terminals and a collaborative innovation platform connected to the terminals. The smart wearable terminals are used to collect employees' physiological data, motion data, and voice data, and send them to the collaborative innovation platform. The collaborative innovation platform includes a data processing module, an AI analysis module, and an innovation assistance module. The data processing module is used to preprocess the data sent by the smart wearable terminals. The AI ​​analysis module is used to evaluate the employees' innovation status based on the preprocessed data and generate innovation suggestions. The innovation assistance module is used to provide employees with corresponding innovation resources and collaborative tools based on the innovation suggestions.

2. The AI-based collaborative innovation system according to claim 1, characterized in that, The smart wearable terminal includes a data acquisition unit, a data transmission unit, and a human-computer interaction unit. The data acquisition unit is used to collect employees' physiological data, motion data, and voice data. The data transmission unit is used to send the collected data to the collaborative innovation platform. The human-computer interaction unit is used to receive innovation suggestions and collaborative information sent by the collaborative innovation platform and display them to employees.

3. The AI-based collaborative innovation system according to claim 1, characterized in that, The collaborative innovation platform also includes a user management module for managing employee user accounts, including account creation, permission allocation, and information maintenance functions, to ensure data security and the orderly operation of the system.

4. The AI-based collaborative innovation system according to claim 1, characterized in that, The AI ​​analysis module uses deep learning algorithms to evaluate employees' innovation status. It trains a neural network model on a large amount of employees' historical innovation data to accurately judge the current innovation status of employees and generate personalized innovation suggestions.

5. The AI-based collaborative innovation system according to claim 1, characterized in that, The innovation support module provides innovation resources including industry reports, technical documents, and case studies. Collaboration tools include online document editing, real-time communication, task assignment, and tracking functions to meet the various needs of employees in the innovation process.

6. A smart wearable terminal, characterized in that, The AI-based collaborative innovation system, as described in any one of claims 1 to 5, further includes a data acquisition unit, a data processing unit, a data transmission unit, and a human-computer interaction unit. The data acquisition unit is used to collect employees' physiological data, motion data, and voice data. The data processing unit is used to perform preliminary processing on the collected data. The data transmission unit is used to send the processed data to the collaborative innovation platform. The human-computer interaction unit is used to receive innovation suggestions and collaborative information sent by the collaborative innovation platform and display them to employees.

7. The intelligent wearable terminal according to claim 6, characterized in that, The data acquisition unit includes a physiological sensor, a motion sensor, and a voice sensor. The physiological sensor is used to collect employees' heart rate, blood pressure, and body temperature physiological data. The motion sensor is used to collect employees' limb movements and posture changes. The voice sensor is used to collect employees' voice information.

8. The intelligent wearable terminal according to claim 6, characterized in that, The human-computer interaction unit includes a display screen, a speaker, and touch buttons. The display screen is used to display innovation suggestions, collaboration information, and terminal status information. The speaker is used to play voice prompts. The touch buttons are used for employees to input operations.

9. The intelligent wearable terminal according to claim 6, characterized in that, It also includes a positioning module, which is used to obtain employees' location information in real time and send it to the collaborative innovation platform to facilitate location tracking and team collaboration management during the collaborative innovation process.

10. The intelligent wearable terminal according to claim 6, characterized in that, The terminal also has an energy management system to optimize energy consumption, extend battery life, and ensure continuous and stable operation during employee work.

Images