61 results about "Behavioral state" patented technology

A multimodal intelligent agent system for dynamic environmental monitoring and user-centered support, consisting of: a multimodal sensor module configured to continuously acquire environmental and behavioral data from multiple input modalities, including at least one visual sensor, at least one acoustic sensor, at least one environmental conditions sensor, and at least one proximity or motion detection sensor, each generating modality-specific data streams representing visual images, audio waveforms, physical environmental parameters, and motion signatures within a monitored environment; a data preprocessing and fusion subsystem that is operationally coupled with the multimodal sensor module and configured to normalize, temporally align, and transform the modality-specific data streams into high-dimensional feature embeddings using a variety of encoders, wherein the visual encoder uses convolutional or vision transformer architectures, the audio encoder uses a spectral-temporal feature extractor, and the sensor encoder transforms raw analog data into context vectors suitable for multimodal alignment; a multimodal processing unit consisting of a transformer-based large language model (LLM) trained on paired multimodal datasets and configured to perform semantic fusion, context abstraction, and inference across the aforementioned aligned multimodal feature embeddings to generate a contextual understanding of environmental and behavioral states; an adaptive agent controller coupled to the multimodal inference processing unit and configured to instantiate, manage, and terminate a variety of task-specific intelligent agents, each agent being a software unit configured to perform a specialized function selected from meeting summarization, behavioral analysis, misplaced object detection, or environmental anomaly identification, with the agents dynamically interacting with the inference engine to retrieve contextually relevant multimodal embeddings for task execution; a personalization and adaptive learning subsystem consisting of a user preference database and a neural memory structure configured to update and refine model parameters based on user-specific interaction history, thereby enabling personalized output generation, prioritization of recommendations, and long-term behavioral adaptation; and An output generation interface is operationally connected to the adaptive agent controller and configured to produce multimodal output in textual, visual, and auditory form. The interface is capable of displaying human-readable summaries, notifications, and visual reconstructions of identified entities or environmental states.

The invention relates to the field of breeding industry, and discloses a pig behavior-based pig health condition analysis method and system, and the method comprises the steps: carrying out the comprehensive monitoring of pig behaviors, capturing the gait, feeding mode, excretion behavior and activity range of a pig in real time based on a behavior feature extraction algorithm, and obtaining a behavior state video stream sequence; multi-dimensional time sequence correlation analysis is carried out on the behavior state video stream sequence, and historical behavior data, pig weight changes and physiological parameters are combined; based on the dynamic time sequence feature vector, dynamically identifying a change track of pig behaviors by applying a self-adaptive behavior identification algorithm; performing correlation analysis on the detected abnormal behavior pattern and the potential health risk of the pig, fusing the environmental factors, group behavior data and health history of the pig, and identifying a potential health problem; and based on a risk early warning result, automatically adjusting environmental parameters and feeding management strategies, and providing intervention measure suggestions. The pig health management system has the advantage of improving the efficiency and accuracy of pig health management.

The invention provides a penaeus vannamei behavior state acoustic intelligent identification method and a model building method, and belongs to the technical field of aquaculture acoustic monitoring. Comprising the following steps: acquiring audio data of penaeus vannamei in different behavior states; filtering and framing, threshold detection and feature extraction are carried out on the audio, and a standardized audio feature data set is constructed; a deep learning model ASP-SEResNet fusing attention statistical pooling and a channel attention mechanism is built, the deep learning model comprises a feature channel optimization branch based on a compression-excitation module and a time sequence feature expression branch adopting attention weighted statistic aggregation, and joint representation of acoustic features is achieved through a double-path feature fusion mechanism; training and optimizing the model to obtain a final model; experimental results show that the method shows excellent classification performance on a prawn audio feature data set, and high-precision identification of five behavior states of the penaeus vannamei can be realized.

The invention discloses an interaction method and device, electronic equipment and a readable storage medium, and belongs to the technical field of electronic equipment, and the method comprises the steps: obtaining a video stream of a user; determining a behavior state of the user according to the video stream; wherein the behavior state comprises at least one of the following items: an emotional state, an attention state, a facial expression, a body posture and an action; and outputting first interaction content according to the behavior state of the user.

The invention relates to the technical field of intelligent health monitoring, and particularly discloses a multi-modal AI-fused senior non-inductive assessment method and system and electronic device.The method comprises the following steps that in the daily living environment of the senior, original multi-modal data reflecting the physiological and behavior states of the senior are collected in a non-inductive mode through at least two non-contact sensors; and preprocessing the original multi-modal data, and extracting a single-modal health feature vector from the data of each modal by using a preset feature extraction model. According to the invention, non-inductive data acquisition is realized by deploying the non-contact multi-mode sensor in the daily environment of the elderly, the elderly does not need to cooperate deliberately, and the use convenience and acceptability are effectively improved; through preprocessing and feature extraction of multi-modal data, multi-dimensional health features can be mined from visual, audio and radar data, the limitation of single data dimension in the traditional technology is broken through, and comprehensive representation of physiological and behavior states of old people is realized.

The invention discloses a brain wave monitoring and intelligent feedback combined modularized head ring glasses device, and relates to the technical field of brain wave sensing and adjusting equipment.According to the device, brain electrical activities, eye movement behaviors, eye fatigue states, wearing fitness of the head ring glasses device, environmental noise interference and use situations of a user are monitored in real time; the brain wave fitting quality index is calculated through the pressure variance, the EEG signal-to-noise ratio and the historical template fitting degree, and self-adaptive compensation adjustment is carried out when the brain wave fitting quality index is unqualified; combining electroencephalogram and eye movement characteristics to calculate a brain-eye collaborative concentration index, and generating a comprehensive adjustment strategy when the brain-eye collaborative concentration index does not reach the standard; and calculating an emotional scene switching index according to the brain wave emotional characteristics, the situation parameters and the historical trend, and triggering personalized emotional regulation when the emotional scene switching index is not matched with the situation parameters. The device can quantify the brain wave and behavior state of the user in real time, realizes intelligent concentration and emotion management, and improves the reliability and individual suitability of electroencephalogram monitoring.

The invention relates to the technical field of animal behavior analysis, and discloses an animal behavior accurate recording and analyzing system based on multi-mode keyboard coding. A data acquisition and synchronization module of the system acquires original key event data and generates a time sequence behavior data stream with millisecond-level precision; the coding logic engine module is used for converting the physical key sequence into a hierarchical behavior semantic coding sequence through a dynamic mode switching and key mapping technology; the behavior state modeling module is used for constructing an animal behavior state transition probability model, performing state decoding and estimation by adopting a hidden Markov model, and outputting a behavior state sequence with confidence; the behavior analysis and prediction module is used for predicting a behavior evolution trend and identifying an abnormal behavior event; and the threshold value dynamic adjustment module is used for dynamically adjusting an early warning threshold value through interaction between the normal behavior generator and the real-time discriminator aiming at an abnormal behavior event so as to finish accurate recording and risk analysis of animal behaviors.

This invention discloses a remote discrimination method for giant panda behavior based on collar sensors, specifically relating to the field of animal behavior monitoring using collar sensors. The method includes collecting behavior monitoring data output by the collar during continuous monitoring, arranging the data chronologically according to sampling times, and calculating the corresponding activity and displacement values ​​at each sampling time to obtain a monitoring sequence. Based on the direction of activity and displacement changes at adjacent sampling times in the monitoring sequence, the connection point where the direction of activity change changes and the displacement change state simultaneously change is determined as the behavior transition point. This invention solves the problems in existing technologies where the correlation between behavior transitions is difficult to retain, and the correspondence and stability between remote discrimination results and actual behavior states are insufficient. It extracts continuous segments around the behavior transition point before and after the transition and determines the behavior state based on the corresponding changes in these segments.

This invention provides an intelligent delivery room care system integrating multimodal environmental control based on sound, light, and temperature, belonging to the field of medical care equipment technology. By integrating multimodal environmental sensing and control capabilities based on sound, light, and temperature, combined with real-time analysis of the mother's physiological and behavioral states, this invention achieves intelligent, personalized, and contextualized management of the delivery room environment. Its main advantages include: significantly reducing the mother's anxiety and pain perception during childbirth, increasing the success rate of natural childbirth; reducing the burden of environmental control operations on medical staff, improving nursing efficiency; supporting adaptive environmental switching across multiple stages and scenarios, meeting the needs of the entire cycle from labor to postpartum recovery; ensuring the dignity of the mother and data security through non-invasive monitoring and privacy protection design; and a modular system design that facilitates integration with existing hospital information systems (HIS) and intelligent ward platforms.

The invention discloses a pet emotion recognition method, device and equipment and a storage medium, and relates to the technical field of intelligent vehicles, and the method comprises the steps: obtaining the local state information and behavior state information of a pet leaving a vehicle; constructing a local feature vector according to the local state information and the local weight matrix, and constructing a behavior amplification coefficient according to the behavior state information and the amplification coefficient matrix; constructing a state analysis quantity based on the local feature vector and the behavior amplification coefficient; and determining whether the emotion of the car-leaving pet is in an abnormal state according to the state analysis quantity. According to the method and the device, whether the current emotion of the car-leaving pet is in the abnormal state or not can be determined by comprehensively identifying the body local state and the behavior state of the car-leaving pet, and the user is notified to pay attention in time when the current emotion is in the abnormal state, so that the user only needs to pay attention to the corresponding notification without worrying about anxiety all the time.

The invention provides an intelligent data analysis method for a pension information service system, relates to the technical field of physiological data processing, and aims to judge behavior stages according to the daily moving speed of the elderly and the continuity of the speed in time, select stable effective behavior stages through deterministic scores, and improve the accuracy of the behavior stages. Judging a corresponding behavior state according to the change characteristic of the heart rate in the effective behavior stage, identifying the body function expression of the old person in the corresponding state according to the change rate characteristic of the central rate of each behavior state, and judging the reaction intensity in the first behavior state and the flat intensity in the second behavior state; and a heart function change index is constructed by combining the long-term comprehensive expression of the reaction intensity, the flat intensity and the heart rate in the third behavior state, so that the overall body function change characteristics are reflected and quantified, and a guardian can conveniently know the overall body condition change of the old.

This application relates to a device control method, apparatus, device, and system based on behavior monitoring. The device control method based on behavior monitoring includes: extracting features from behavior monitoring data of a target object to obtain behavior features; performing a first behavior evaluation process on the behavior features to obtain a first evaluation result; performing a second behavior evaluation process on the behavior features to obtain a second evaluation result; identifying the target behavior state of the target object based on the first and second evaluation results; and adjusting the device in the space where the target object is located based on the target behavior state. This solution can improve the intelligence level of the device control process.

This invention discloses an antenna reverse compensation method and system using digital twin virtual commissioning. The method acquires continuous behavioral data streams of the antenna system, reconstructs the state using a pre-associated physical characteristic computational model of the digital twin, and performs temporal evolution prediction using a data-driven fitting model to generate a behavioral state stream. Within the digital twin, a multi-dimensional deviation evolution analysis is performed between the behavioral state stream and a healthy baseline behavioral template to extract early covert mode feature sets. Based on this, a strategy self-exploration domain is defined in the virtual commissioning space. A perturbation action generator generates a perturbation action self-combination sequence and injects it into the digital twin, monitoring the repair state stream to construct a repair state mapping table. Finally, based on the repair state mapping table and the healthy baseline behavioral template, effective trial strategy clusters are selected, and the optimal reverse compensation strategy is selected based on the strategy stability metric, generating physical antenna reverse compensation commands.

This invention provides a functional architecture and method for an intelligent human-machine adaptive collaborative system in an aircraft cockpit. It features advantages such as hierarchy, modularity, and reusability, supporting detailed design of the physical architecture of the intelligent human-machine system in the aircraft cockpit and providing a technical means to achieve efficient human-machine matching and dynamic natural interaction in the aircraft cockpit. The intelligent human-machine adaptive collaborative method provided by this invention, by adjusting the human-machine interaction level, the autonomous dimension of the intelligent agent, the information processing dimension, and the task allocation dimension, and based on the decision steps of the information flow graph, can obtain the optimal intelligent human-machine adaptive collaborative solution for the aircraft cockpit. The system functional architecture and method provided by this invention can realize intelligent human-machine interaction that adapts to the pilot's physiological, psychological, behavioral state and task requirements. It can provide technical support for solving problems such as pilot cognitive load imbalance, loss of situational awareness, and frequent human errors, enhancing the pilot's situational judgment and decision-making capabilities, and optimizing the overall effectiveness of the cockpit human-machine system.

The invention relates to the technical field of computers, and discloses a digital twin model management method and system based on behavior analysis. The method comprises the following steps: acquiring a historical behavior sequence and real-time sensing data of a target object, and respectively encoding the historical behavior sequence and the real-time sensing data into a first behavior representation vector and a second behavior representation vector; a behavior driving model composed of a behavior state observer, an environment disturbance compensator and an intention inference engine is constructed, and behavior state probability output and intention inference are achieved; generating a regulation and control instruction set of minimum intervention and risk avoidance and issuing and executing the regulation and control instruction set; feedback data increment is collected to update model parameters, and closed-loop optimization is achieved. According to the method, through multi-dimensional behavior feature fusion and an environment adaptive compensation mechanism, the prediction precision, the decision-making efficiency and the sustainable evolution ability of the digital twin model under complex working conditions are significantly improved.

This invention provides a postoperative patient monitoring system for neurointerventional therapy, relating to the fields of medical big data and artificial intelligence. It includes a multimodal data acquisition center, an interference feature decoupling unit, a trust capital quantification unit, a game strategy arbitration unit, and a monitoring execution unit. The multimodal data acquisition center is configured to retrieve time-series monitoring data of the monitored subjects. The trust capital quantification unit performs trust loss assessment and analysis on historical interaction feedback data. The game strategy arbitration unit is configured to compare and analyze the current clinical trust capital index with a preset trust threshold. This invention's system performs time-frequency domain matching between residual sequences and behavioral state labeling data, ensuring that the system only predicts the risk evolution of neurogenic hemodynamic changes. This significantly reduces the false alarm rate caused by external interference in complex postoperative monitoring environments and extracts pure pathological feature components.

This application discloses a music recommendation method, device, and vehicle. The music recommendation method includes: acquiring passenger images, with the number of passengers being no less than two; identifying the behavioral state characteristics of each passenger based on the acquired images, including the passenger's facial expression, whether the passenger is asleep, and the passenger's body posture; and analyzing the human body state data of each passenger based on the behavioral state characteristics, whereby the human body state data can characterize the user's cognitive and physiological state; recommending music that matches the human body state data of the same category when the number of passengers corresponding to the same category of human body state data reaches a first threshold; and recommending music that matches the human body state data of the category with the highest priority based on the priority of each category of human body state data when the number of passengers corresponding to the same category of human body state data does not reach the first threshold, thereby improving the user experience.

This invention discloses an energy-saving control method and device for a bird tracker based on sensor fusion. The method uses an accelerometer to collect triaxial acceleration data to identify the bird's behavioral state, activating the heart rate and blood oxygenation sensors only when the bird is stationary and deactivating them during movement. It combines illumination data, location information, and the device's real-time time to determine signal obstruction environments and dynamically adjusts the operating strategies of the positioning and communication modules. Before positioning calculation, it detects the number of satellites in a low-power mode, performing full positioning only when the target is met. Power management independently controls the on / off operation of the positioning, communication, and storage modules, with standby power supplied only to the real-time clock module. The corresponding device includes a main control unit, a sensor data acquisition and tracking unit, and an energy-saving control unit, integrating multiple types of sensors and a power management module. This invention significantly reduces the tracker's ineffective power consumption and extends its battery life in the field through multi-sensor fusion and dynamic energy-saving scheduling, adapting to the long-term monitoring needs of birds.

The system according to this embodiment aims to understand the individual learning and behavioral status and provide an optimal learning plan and behavioral modification program based on that understanding. [Solution] The system according to the embodiment comprises a shooting unit, an analysis unit, a generation unit, and a provision unit. The shooting unit captures the subject's daily activities in photographs and videos. The analysis unit analyzes the information captured by the shooting unit to understand the subject's learning and behavioral status. The generation unit generates an optimal learning plan and behavioral modification program based on the situation understood by the analysis unit. The provision unit provides the program generated by the generation unit.

A device and a method for detecting anomalies, including: acquiring a normal signal reflecting a normal state, determining on the basis of the normal signal a probability density ƒ which models a normal behavioral state, setting an information filter based on the probability density ƒ, the information filter being configured to converge toward a limit value L when it is applied to samples of a normal signal, while increasing its value in response to the detection of an anomaly, setting a threshold value S based on the convergence limit value, acquiring a current signal reflecting the current behavioral state, sampling the current signal in a current series of N samples, computing a result of applying the information filter to the current series of N samples, comparing the result with the threshold value S, an anomaly being detected if the result exceeds the threshold value.

The invention provides a wearable behavior monitoring method and device based on AI, relates to the field of artificial intelligence, and solves the technical problems that in the prior art, isolated analysis of physiological signals and external behavior information lacks effective fusion and joint reasoning, and the comprehensive behavior state of a user is difficult to accurately recognize. The method comprises the following steps: collecting video data and EEG signals of a visual field in front of a user through a head-mounted wearable device; obtaining a visual feature vector and an audio feature vector in the video data; identifying current scene information through the visual feature vector and the audio feature vector; distinguishing the current behavior state of the user based on the scene information; carrying out filtering processing on the EEG signals, and extracting time-frequency domain features from the filtered EEG signals; identifying the current emotional state of the user based on the time-frequency domain features; and inferring the current composite behavior state of the user based on the current behavior and emotional state of the user. The method and device are used in the behavior monitoring process.

The invention relates to a device and a method for detecting anomalies within an environment or system, comprising the following steps: - acquiring (E1) a normal signal (NS) reflecting the normal state of the environment or system, - determining from the normal signal (NS) a probability density that models a normal behavioral state of the environment or system, - establishing (E3) an information filter based on said probability density, said information filter being configured to converge to a limit value when applied to samples of a normal signal, while increasing its value in response to the detection of an anomaly, - establishing (E4) a threshold value based on said convergence limit value, - acquiring (E5) a current signal (CS) reflecting the current behavioral state of the environment or system, - sampling (E6) said current signal (CS) in a current series of N samples.- calculate (E7) a result of applying the information filter to the current series of N samples, - compare (E8) this result with the threshold value, an anomaly being detected if the result exceeds the threshold value. Figure for the abbreviation: Fig. 2.

To provide a determination system, determination program, determination device, and determination method that can appropriately determine the discrepancy between the actions a subject should take and the actions the subject actually takes. [Solution] The system comprises: a first acquisition means for acquiring ambient information indicating the surrounding circumstances of a subject; a first identification means for identifying a standard action, which is the action the subject should take in response to the surrounding circumstances, based on the ambient information acquired by the first acquisition means; a second acquisition means for acquiring behavioral state information to identify the state of the subject's actions; a second identification means for identifying the subject's actions, which is the actions of the subject, based on the behavioral state information acquired by the second acquisition means; and a determination means for performing a discrepancy-related determination, which is a determination of the discrepancy between the standard action identified by the first identification means and the subject's actions identified by the second identification means.

The application discloses a kind of bionic dynamic neural network, the neural network can form behavior state relation net by learning, and can be received when setting target task, state transfer and behavior activity control are carried out according to current state and the behavior state relation net formed, directly complete the target task.The bionic dynamic neural network provided in the embodiment of the application imitates the neural circuit topological structure and dynamic characteristics of biology, realizes the control to state transfer, so that state transfer and action control form the behavior state relation net with causal relationship, more in line with the way of human brain processing problem, improve the decomposition ability of complex task and the efficiency of reinforcement learning, so that reinforcement learning can be more realistically used in actual robot task.The bionic dynamic neural network of the embodiment of the application can be applied in reinforcement learning model, for building the reinforcement learning algorithm model with higher learning efficiency capable of decomposing task.

The invention relates to a method for operating at least one comfort system (16) of a motor vehicle (10), wherein an interior configuration device (22): receives an occupant observation signal (S1) that describes at least one behavior parameter relating to a current behavior of at least one user (12), and determines a current behavioral state of the user (12) (S2), receives a situation description signal (S3) that describes at least one situation parameter influencing the current journey, and categorizes the current journey based on a situation parameter (S4);Based on the combination of the determined behavior pattern and the specified driving category, a comfort fulfillment criterion is determined (S5) which describes the suitability of one or more settings of the at least one comfort system (16) for providing behavior-specific and situation-specific support to the user (12) during the journey, a configuration model is provided (S6) which describes a course of at least one configuration parameter for the at least one comfort system (16) over a specified driving period in order to fulfill the specified comfort fulfillment criterion, at least one configuration signal is generated (S8), and the at least one generated configuration signal is transmitted to a specified receiving device (16, 20) (S9).

The invention relates to the technical field of education technology and artificial intelligence processing, and discloses a psychological emotion and behavior analysis education management method, which comprises the following steps: S1, acquiring data through a multi-modal data acquisition device; s2, extracting features from the multi-modal data based on psychology; s3, generating high-dimensional hidden space features; s4, generating a prediction result of the emotional state and the behavior mode of the student; s5, generating a personalized analysis report of the current emotional state and behavior performance of the student according to a prediction result; and S6, outputting the emotional state in real time according to the prediction result of the emotional state, dynamically updating parameters in the steps S3 to S4 after new data is input, and adjusting an education management strategy in combination with the change trend of the emotion and behavior of the student. According to the invention, the technical scheme of combining multi-modal data acquisition and feature extraction is adopted, so that the emotion and behavior states of students can be comprehensively captured, and accurate multi-modal feature vectors are generated.

The invention discloses an enterprise employee risk early warning method and system based on artificial intelligence and a medium, and the method comprises the steps: obtaining a real-time work video, and determining the identity information and personalized evaluation information of each employee in the real-time work video; extracting an expression change feature vector and a behavior change feature vector of each employee; inputting the expression change feature vector and the behavior change feature vector into a pre-trained intelligent recognition model, determining an expression state anomaly degree and a behavior state anomaly degree, and performing quantization processing to obtain an expression anomaly quantized value and a behavior anomaly quantized value respectively; determining a first weight and a second weight according to the personality evaluation information of the corresponding employee; performing weighted summation according to the expression anomaly quantized value and the behavior anomaly quantized value to obtain a working state fluctuation score of the employee; and when the working state fluctuation score is greater than or equal to a preset early warning threshold value, generating an early warning signal. According to the invention, the timeliness and accuracy of risk judgment of the working state of the employee can be improved.

The invention discloses a control method and device of cleaning equipment, electronic equipment and a storage medium, the control method of the cleaning equipment is applied to a smart home management platform, and the control method of the cleaning equipment comprises the following steps: acquiring physiological state information of a user; the physiological state information indicates that the current behavior state of the user is a first preset state, and determining a first target working mode corresponding to the first preset state; and controlling the cleaning equipment to work based on the first target working mode. According to the control method of the cleaning equipment, the working mode of the cleaning equipment corresponds to the current behavior state of the user, the working mode of the cleaning equipment can be adaptively adjusted according to the behavior state of the user, and user requirements are better met.

The present application relates to the technical field of artificial intelligence monitoring, and more particularly to a multi-modal physiological parameter fusion continuous monitoring method and system, the method comprising: continuously collecting physiological electrical signals of a target individual under multiple behavior states; periodically dividing the multi-lead electrical signals according to the heartbeat rhythm, and standardizing them into feature segments with the same time length and channel number; performing two-stage weighted processing on each feature segment to obtain high-weight feature representation; performing feature alignment on the high-weight feature representation according to the corresponding behavior state, and performing fusion operation; inputting the fusion feature representation into a convolutional neural network to recognize the individual physiological state, and periodically determining whether the target individual currently has a physiological abnormality risk. Through the present application, the signal distortion problem caused by ignoring the multi-behavior state switching in the prior art is effectively solved, the feature adaptability and cross-state recognition robustness in the dynamic scene are significantly improved, and continuous and accurate physiological abnormality risk tracking is realized.