79 results about "Adaptive interaction" patented technology

The embodiment of the invention provides a vehicle-mounted emotion interaction method and device based on multi-dimensional recognition, and the method and device achieve the precise judgment of the emotion of a driver through innovatively constructing an emotion fusion recognition model and integrating the facial expression, voice emotion, driving behavior and physiological state features. And designing a scene-based self-adaptive interaction strategy, and establishing an interaction triggering threshold value for intelligent matching in combination with external environment data and a danger level. An interaction effect evaluation mechanism is introduced, an interaction strategy model is continuously optimized through an online learning module, and dynamic adjustment of personalized interaction content is achieved. According to the method, the defects of the traditional technology in the aspects of emotion recognition, interaction strategies, effect evaluation and the like are effectively overcome, and the intelligent level and the user experience of vehicle-mounted emotion interaction are remarkably improved.

The invention relates to the technical field related to automatic driving, in particular to a trajectory prediction method based on adaptive interaction and dynamic intention, which comprises the following steps: firstly, constructing a heterogeneous interaction map, dividing a traffic scene into a vehicle grid, an environment grid and a non-driving area grid, and embedding multi-dimensional dynamic features; then dynamically adjusting a region of interest based on the behavior intention of the target vehicle, and extracting a high-correlation interaction subnet; modeling an interaction relationship by adopting a heterogeneous graph convolutional network, and processing the motion characteristics of the target vehicle and the neighbor vehicle through a sub-channel coding strategy; further realizing dynamic intention perception through a double-branch parallel attention architecture, and fusing macroscopic intention and dynamic intention information; and finally, iteratively generating a future trajectory prediction result based on a decoding architecture of a message passing mechanism. The method can effectively improve the long-term prediction performance in a lane changing scene, adaptively captures a dynamic interaction relationship, and improves the adaptability of a prediction system to the behavior intention change of a driver.

The invention relates to the technical field of old people health management, and discloses a personalized health auxiliary management system based on artificial intelligence, which comprises a multi-mode sensing module, an intelligent preprocessing module, a dynamic knowledge graph module, a personalized inference engine module and a self-adaptive communication and interaction module. The multi-mode sensing module is used for acquiring physiological, environmental and behavioral data; the intelligent preprocessing module desensitizes and cleans data through federal learning, and extracts core features; the dynamic knowledge graph module constructs and updates an exclusive graph containing basic and dynamic nodes and associated edges; the personalized inference engine module generates a layered intervention scheme; and the adaptive interaction module adapts to the cognitive ability and synchronizes the scheme. According to the invention, health information integration and dynamic intervention are realized, the management accuracy and compliance are improved, and the method is suitable for home non-clinical scenes.

The invention provides an intelligent examination and approval method, system and device based on a multi-modal large model and a medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: analyzing a multi-modal examination and approval material through a visual large model to obtain multi-modal feature information; performing semantic understanding by using the language large model to generate structured semantic data; performing entity matching, relationship verification and logical reasoning based on the knowledge graph to generate a research and judgment report; intelligent decision support and risk early warning are realized; approval instruction conversion, service evaluation and self-adaptive interaction are realized through a multi-mode interaction technology; and continuously optimizing the model and the knowledge base by using the approval process data. According to the invention, by constructing a technical framework of deep fusion of vision, language and knowledge, comprehensive intelligentization of the approval process is realized, and the beneficial effects of improving the approval efficiency, guaranteeing decision fairness, enhancing the complex material processing capability, improving the user experience and realizing continuous evolution of the system are achieved.

The invention relates to the technical field of artificial intelligence and man-machine interaction, and discloses a picture book co-reading method, system and device based on multi-modal emotion recognition and a medium, and the method comprises the following steps: obtaining multi-modal time sequence data of a user in picture book co-reading; establishing and updating a longitudinal emotion file for recording historical interaction data for the user; based on the data and the archive, personalized multi-dimensional internal state evaluation is carried out, and the evaluation comprises calculation of cognitive load indexes representing difficult understanding, recognition of instant emotional states and analysis of contextual emotional deviations of the emotional states and current plot expectations; and finally, based on the multi-dimensional evaluation result, determining and executing dynamic interaction strategies such as simplification, guidance or excitation. According to the method, deep personalization is achieved by constructing the multi-dimensional state model and combining the longitudinal archives, the user state can be more accurately judged, prospective and self-adaptive interaction is achieved, and the interaction fineness and effectiveness are improved.

The invention relates to the technical field of artificial intelligence, in particular to a scene-based AI partner training self-adaptive interaction and multi-dimensional evaluation method. The method comprises the steps that S1, a system scene library and a role library are constructed to configure real service scene requirements, and a structured scene model is obtained and stored; s2, automatically matching and initializing AI agent roles according to scene nodes; s3, constructing an interaction entrance of a user and an AI agent role, and inputting a training demand and interaction content by the user; s4, performing real-time scene adaptation detection according to the interaction content; and S5, performing multi-dimensional data acquisition on the interactive content of the user to obtain a multi-dimensional feature data set. And S6, constructing an evaluation model based on the multi-dimensional feature data set and outputting a multi-dimensional evaluation report. According to the invention, the scene process can be flexibly adjusted along with the user performance through the nodal disassembly and jump rule; the AI role adjusts response logic and mood to improve interaction authenticity according to scene characteristics; and specific improvement suggestions are generated through multi-dimensional data acquisition and analysis, so that comprehensive evaluation and closed-loop promotion are realized.

The invention relates to the technical field of robot control, in particular to a robot control method and system for conference investigation based on artificial intelligence. The method comprises the following steps: acquiring audio data and monitoring data of a conference environment; analyzing the monitoring data and the audio data, and determining a speaking state and speaking content; according to the speaking state and the speaking content, determining a speaking distribution thermodynamic diagram; and determining a robot active interaction strategy according to the speaking content and the speaking distribution thermodynamic diagram, and sending the robot active interaction strategy to the robot to enable the robot to execute the active interaction strategy to guide conference discussion. The accuracy and real-time performance of intention understanding are improved, and structured input is provided for thermodynamic diagram generation and interaction strategies. Real-time visualization of conference discussion distribution is realized, the defect that a thermodynamic diagram cannot be automatically generated is eliminated, and the monitoring capability of participating in balance is improved. And a self-adaptive interaction mechanism is realized, delay depending on manual intervention is eliminated, and efficient and accurate proceeding of the conference is ensured.

The invention discloses an intelligent elevator operation box which comprises a user identification module used for obtaining and identifying identity information of a user in a multi-mode mode; the intelligent control module is connected with the user identification module and is used for generating a dynamically adjusted personalized control strategy according to the identified identity information and the environment parameters; the self-adaptive interaction module is connected with the intelligent control module and is used for providing a scenarized human-computer interaction interface and receiving user input; and the reinforcement learning module is connected with the intelligent control module and is used for continuously optimizing the personalized control strategy based on the user feedback and the environment change. According to the intelligent elevator operation box disclosed by the invention, high intelligence and individuation of elevator service are realized by fusing a multi-mode identification technology, a reinforcement learning algorithm and a multi-system collaborative architecture.

The invention relates to a perception, decision and communication integrated aircraft brain software system, which belongs to the field of artificial intelligence and comprises an intelligent master control scheduling module, a decision change module, a communication adaptive interaction module and a situation perception module. According to the method, a modular design idea is adopted, modules in the method are flexibly combined, functional modules are newly added, different tasks executed by an unmanned system can be achieved, and generalization of brain software is greatly improved. The intelligent capability and the online autonomous decision-making capability of the aircraft can be greatly improved, the method can be flexibly applied to various unmanned equipment, and the intelligent enabling capability of the unmanned equipment can be greatly improved.

The invention discloses a water conservancy flood control scene-oriented question and answer intention recognition and multistage classification model construction method and system and a computer readable storage medium. The method comprises the steps of data collection and knowledge fusion; data preprocessing and intention system construction; constructing training data; training an intention recognition model; an online application mechanism; and model optimization and feedback closed loop. By constructing a standardized data acquisition process, a specialized intention recognition model and a high-adaptability interaction mechanism, accurate understanding and structured expression of the natural language instruction of the user in the flood control scene are achieved, and then the intelligent level of key services such as scheduling and early warning is improved.

The embodiment of the application provides a kind of based on multi-dimension recognition's vehicle emotion interaction method and device, through innovatively constructing emotion fusion identification model, by integrating facial expression, speech emotion, driving behavior and physiological state characteristics, the accurate judgment of driver emotion is realized.Design scene-based adaptive interaction strategy, combined with external environment data and danger level, establish interaction trigger threshold for intelligent matching.Introduce interaction effect evaluation mechanism, through online learning module, continuously optimize interaction strategy model, realize the dynamic adjustment of personalized interaction content.The method effectively solves the deficiency of traditional technology in emotion recognition, interaction strategy and effect evaluation, significantly improves the intelligent level and user experience of vehicle emotion interaction.

This invention discloses a multimodal interactive interface design method for elderly care robots, relating to the field of intelligent elderly care. The method includes: collecting user facial images, voice signals, and touch operation information to form a raw multimodal data stream; processing the raw multimodal data stream to extract standardized user real-time state feature vectors; inputting the user real-time state feature vectors and historical interaction data into a support vector regression model; obtaining the real-time cognitive load level based on the user's real-time state feature vectors in the current session; and analyzing the rate of change of user touch operation accuracy within a time window using historical interaction data, calculating the acceleration value of capability change and the trend of touch operation accuracy, and fusing them to derive the interaction abstraction level. This invention constructs a joint decision matrix sensitive to the acceleration of capability change, dynamically triggering an adaptive interaction strategy when accelerated capability decline is detected, driving the interface engine to adjust interface rendering and multimodal guidance in real time.

The invention discloses an intelligent cabin emotion interaction method and system and a storage medium. The method comprises the steps that multi-modal feature data and current environment feature data of a user are collected in real time; carrying out fusion processing on the multi-modal feature data and the current environment feature data based on a preset feature fusion algorithm, and obtaining an emotion recognition result through an emotion database based on a fusion result; performing sentiment analysis by adopting a sentiment interaction model based on the sentiment recognition result to obtain a current interaction strategy; and generating a cooperative control instruction in cooperation with the intelligent cabin system based on the current interaction strategy so as to control the intelligent cabin system to execute a corresponding interaction operation. According to the method provided by the invention, the defects of user emotion recognition and lack of adaptive interaction strategies of an intelligent cockpit system in the prior art are effectively overcome, and the adaptive interaction strategies are selected by accurately recognizing the user emotion, so that the user emotion is effectively adjusted, and personalized interaction experience is provided.

The invention discloses a self-adaptive interaction method based on multi-modal fusion perception, and belongs to the technical field of man-machine interaction and artificial intelligence. According to the method, multi-modal data of vision, hearing, physiology and the like of a user and an environment are synchronously collected, depth feature alignment and semantic understanding are carried out by utilizing a fusion neural network, and a depth state vector representing a comprehensive state and context of the user is generated. Based on this, an adaptive decision module integrating a rule engine and deep reinforcement learning dynamically generates a structured interaction instruction including modality, content, time sequence and strength parameters. Finally, through a high-precision synchronous scheduling mechanism, multiple physical actuators are coordinated to output fusion and coherent interaction feedback. The method solves the problems of one-sided perception and rigid response of an existing interaction system, realizes accurate understanding and active and personalized adaptation of a dynamic scene and a user hidden state, and is particularly suitable for complex scenes with high requirements on safety and experience, such as an intelligent cabin and the like.

The invention relates to the technical field of intelligent blind guiding, and discloses an intelligent blind guiding cap system based on multi-modal perception, which comprises a multi-source data acquisition module, a data processing module, a multi-terminal interaction and prompt module and a collaboration and optimization module. Through multi-modal perception fusion, visual information of the obstacle is captured, sound source and temperature and humidity data are also included, and correlation reasoning is carried out, so that the problem of a single-modal perception blind area is solved, the accuracy of obstacle recognition in complex environments such as rainy days is improved, and the safety of the blind guiding process is greatly enhanced; meanwhile, through layered adaptive interaction and integration of bone conduction hearing, micro-vision display, tactile vibration and other multi-sensory prompt modes, users with residual vision and mild hearing impairment are covered, the prompt information receiving rate is increased, and adaptive coverage of different user groups is achieved.

This invention discloses an adaptive interaction method, device, equipment, and medium for intelligent vehicle cockpits. The method includes: collecting multi-source perception data, including visual perception data, voice perception data, physiological perception data, and vehicle state perception data from drivers and passengers; cleaning, time-synchronizing, and feature-level fusion processing of the collected multi-source perception data to obtain a feature vector to be identified; analyzing the feature vector based on a trained emotion recognition model to identify the behavioral and emotional state labels of drivers and passengers, generating a comprehensive risk label; matching a corresponding cockpit interaction strategy from a preset strategy library based on the comprehensive risk label, and adjusting the cockpit state based on the cockpit interaction strategy. This technical solution can achieve the goals of improving driving safety, alleviating negative emotions, and enhancing the driving experience.

This invention relates to the field of rehabilitation robot control technology, specifically an adaptive interactive control method for rehabilitation robots incorporating controllable damping. To simultaneously achieve joint position control within limits and adaptive variable-intensity human-machine interactive training for the rehabilitation robot, a hysteresis-fuzzy-PD (MD-Fuzzy-PD) inner-loop controller is designed by establishing a human upper limb resistance model. This enables the rehabilitation robot to accurately track the target position, ensuring response speed and stability while reducing position overshoot, avoiding dangerous postures, and ensuring the safety of exercise training. For variable-intensity training requirements, an adaptive trajectory generator based on Lyapunov functions is designed to correct the reference trajectory through interactive forces. By estimating model parameters online, the desired trajectory is calculated and corrected in real time, and then the inner-loop controller completes real-time tracking of the desired trajectory, ultimately achieving active and compliant human-machine interactive training.

The application belongs to the technical field of swarm intelligence control, and proposes an agent cluster pursuit and evasion game method based on hierarchical reinforcement learning. The method maps discrete instructions and continuous actions to target assignment and path planning respectively in view of the problem of mixed decision space in the cluster pursuit and evasion scene, and proposes a novel hierarchical reinforcement learning framework. In view of the uncertainty problem caused by the enemy strategy, moving obstacles and insufficient training, a robustness-enhanced adaptive method is proposed. In addition, the method embeds a probability integration network between the instruction network and the action network for quantifying uncertainty, and adjusts the interaction frequency of the two-layer network based on uncertainty. In view of the instability problem of the traditional training method, a fusion training method is proposed. The method independently pre-trains the instruction network and the action network, and cross-trains the two networks to promote the adaptive interaction mechanism.

The invention discloses a multi-mode AI interaction method and device, a medium and a product, and the method comprises the steps: obtaining user portrait data, real-time scene data and user memory data; a fusion feature vector is generated; determining a target style, and combining the target style with the fusion feature vector to obtain a plurality of candidate multi-modal interaction contents; carrying out sorting processing on the plurality of candidate multi-modal interaction contents, and determining target multi-modal interaction contents; outputting the target multi-modal interaction content, and obtaining a user voice stream, a user image stream, a user video stream and a user text stream; performing timestamp alignment processing on the user voice stream, the user image stream, the user video stream and the user text stream to obtain multi-modal data; performing emotion recognition processing on the multi-modal data to determine a current emotion state; and matching and determining an interaction strategy by using the current emotional state, generating an adaptive interaction response and outputting the adaptive interaction response. By implementing the scheme, the overall intelligent level of man-machine interaction and the user experience are enhanced.

The invention relates to the technical field of Internet of Things, and discloses a cloud platform general interaction method and device, computer equipment and a storage medium, a message rule group is configured on a cloud platform according to the category of an intermediate node, and matching between the intermediate node and a terminal device and message rules can be realized only by depending on the transmission direction of message data and a communication protocol. The message data can be subjected to adaptive interaction of the cloud platform, the intermediate node and the terminal equipment according to the matched message rule, codes or structures of the cloud platform, the intermediate node and the terminal equipment do not need to be transformed, zero-code adaptation is achieved, and adaptation time consumption and cost are greatly reduced; and the cloud platform updates the online state of the associated terminal equipment by monitoring the state of the intermediate node, so that the equipment difference is effectively dealt with, and the interaction efficiency and the system adaptability are improved.

An artificial-intelligence powered interactive toy system for adaptive engagement is disclosed, comprising an interactive toy device, an AI cloud, a supervisory dashboard, and a fallback subsystem. The toy device includes a plush body, a multi-modal interface, and an environmental multi-sensor array. The AI cloud comprises a state detection module, an intelligent processing module, a storytelling and learning module, a safety and risk detection module, and a content and response generator, enabling real-time analysis of emotional and contextual states. The system implements a closed-loop adaptive interaction, generating speech, motion, and visual outputs, adapting educational and storytelling content, and providing context-aware safety alerts. The supervisory dashboard and fallback subsystem support personalized, emotionally-aware, and developmentally appropriate engagement.

The application relates to the technical field of robot control, in particular to a conference investigation robot control method and system based on artificial intelligence. The method comprises the following steps: acquiring audio data and monitoring data of a conference environment; analyzing the monitoring data and the audio data to determine a speaking state and speaking content; determining a speaking distribution heat map according to the speaking state and the speaking content; determining a robot active interaction strategy according to the speaking content and the speaking distribution heat map, and sending the robot active interaction strategy to the robot to enable the robot to execute the active interaction strategy to guide conference discussion. The method improves the accuracy and real-time performance of intention understanding, provides structured input for heat map generation and interaction strategy, realizes real-time visualization of conference discussion distribution, eliminates the defect that a heat map cannot be automatically generated, improves the monitoring capability of participation balance, realizes an adaptive interaction mechanism, eliminates the delay of relying on manual intervention, and ensures efficient and accurate promotion of the conference.

The application discloses a wisdom scene self-adaptive interaction method and system based on a multi-modal large model, relates to the technical fields of artificial intelligence, natural language processing and the Internet of Things, and comprises the following steps: constructing a multi-modal entity understanding pre-training data set; encoding multi-modal data in the pre-training data set according to a pre-constructed multi-modal entity encoder to generate an entity input sequence with controllable length; performing multi-target joint optimization pre-training on a pre-constructed double-tower multi-modal large model according to the pre-training data set to obtain a general pre-training model; performing lightweight scene adaptation on the general pre-training model for a subdivided wisdom scene to obtain a scene adaptation model; inputting the entity input sequence and a current natural language query into the scene adaptation model to generate a thought chain reasoning conclusion; and generating an interaction instruction according to the thought chain reasoning conclusion, so that the wisdom scene self-adaptive interaction with high accuracy, low delay, low cost and feasibility is realized.

The invention provides a self-service certificate handling equipment service method and system based on robot guidance, and belongs to the technical field of intelligent government affairs. The method comprises the following steps: acquiring certificate handling demand information and associated feature data of a user, and combining a preset certificate handling knowledge graph to generate a self-adaptive guide path; a user is guided to target self-service equipment in an adaptive interaction mode, equipment operation and user behavior data are collected in real time, abnormity judgment is carried out, and a dynamic correction strategy is triggered; after the user completes the operation, the whole-process data is synchronized to the management system, the management system calls the reinforcement learning model to extract features, identifies node bottlenecks, generates an optimization scheme and updates a knowledge graph, and meanwhile, the service robot can push a subsequent service package. The system comprises a service robot unit, a self-service certificate handling equipment unit, a management system unit and a data interaction unit, realizes the intelligentization, self-adaption and continuous optimization of the self-service certificate handling service, and improves the certificate handling efficiency and the user experience.

The application discloses a game artificial intelligence character dynamic interaction method and system fusing multi-modal emotion feedback, and the method comprises the following steps: collecting player in-game interaction behavior data and terminal device sensor physiological data in real time; pre-processing and feature fusing are performed on multi-source heterogeneous data; the fused features are input into a pre-trained CNN-LSTM multi-modal emotion recognition model, and a high-confidence player real-time emotion state label is output; based on the emotion label, in combination with a game world view knowledge base composed of scene constraints, character settings and plot rules, a fuzzy logic behavior decision engine is used to dynamically generate a non-player character behavior adjustment parameter set; finally, the AINPC is driven to real-time adjust its confrontation strength, interaction frequency, guidance mode and behavior style, so that emotion-driven personalized and adaptive interaction is realized.

The application relates to the technical field of computers and discloses a self-adaptive interaction control method and device applied to a digital person, which comprises the following steps: when an initial interaction instruction is detected, determining an interaction personnel corresponding to the initial interaction instruction, collecting multi-dimensional interaction information of the interaction personnel in the process of executing an initial interaction operation according to the initial interaction instruction, determining an interaction demand and a personnel portrait of the interaction personnel according to the initial interaction instruction and the multi-dimensional interaction information of the interaction personnel, generating interaction feedback information according to the interaction demand and the personnel portrait, and driving the digital person to interact with the interaction personnel according to the interaction feedback information. It can be seen that the application can improve the intelligent degree and the individualization level of the digital person interaction, improve the feedback accuracy of the digital person interaction to the user demand, and significantly improve the interaction experience and the satisfaction of the user.

The application discloses a kind of personalized adaptive interaction methods, by collecting user voice, touch trajectory data and portrait information;Dialect enhancement and the text of recognition are carried out to voice data;Touch trajectory feature is extracted and is fused into joint feature vector with portrait;Analysis touch mode;Based on portrait and touch mode, adjust speech rate, UI font size and touch feedback strength using transfer learning;According to text, joint feature vector and the dynamic adjustment of instruction response of adjusted parameter interface;Evaluate feedback effect, trigger adaptive process to realize dynamic learning adjustment.The method improves dialect recognition accuracy by end-side distillation Whisper model combined with multi-modal dialect corpus, accurately identifies touch mode by touch trajectory processing combined with DBSCAN clustering, dynamically adjusts interaction parameters based on transfer learning and meta-learning, continuously optimizes through feedback evaluation mechanism, and realizes lightweight deployment by end-side distillation technology.

The invention discloses an AR (Augmented Reality) system cognitive security interaction method based on multi-modal large language model reasoning and bidirectional individuation, which utilizes a cross attention mechanism guided by an inertial measurement unit signal to explicitly model and eliminate electroencephalogram signal motion artifacts, and compared with the existing method of directly using electroencephalogram signals or simple filtering, the method has the advantages that the method is simple and convenient to implement, and the efficiency is high. The anti-interference capability of personalized attention perception is high, attention recognition is more accurate, the distraction moment of the user can be accurately captured, and the false alarm rate is reduced. Besides, personalized preference configuration information is generated through man-machine alignment and is embedded into the reasoning process of the multi-modal large language model, so that the subjective preference of the user to the risk is read, the problem of alarm fatigue is solved, and user trust is established. According to the method, a real agent closed loop is realized, and the method is a complete agent architecture with perception (physiological denoising)-cognition (LLM reasoning combined with preference)-action (self-adaptive interaction), and can adapt to physiological features and psychological preferences of different users.

The invention discloses a tree shrew work memory task training method based on an 8-shaped maze, and the method comprises the following specific operation steps: S01, preparing before task training: measuring the food intake of a tree shrew, carrying out adaptive interaction, carrying out adaptive grain limitation and setting an experimental environment; s02, a task training step: environment familiarity, shaping period and rule training are carried out in sequence, the rule training adopts a mode of shaping first and then free selection, and rewards are given only when the current arm entering selection of the tree shrew is opposite to that of the last time; and S03, delay time testing: after the tree shrews master the rules, increasing the waiting time of the starting area in a gradient manner, carrying out half-round shaping training which is the same as the testing time on the testing day, and then carrying out free selection testing. According to the tree shrew work memory task training method provided by the invention, through progressive training and personalized behavior preparation, the stress response of tree shrews is effectively reduced, the task motivation and behavior stability are improved, high accuracy can be kept within the delay time of 60 seconds, and the training efficiency is improved. And a reliable and efficient training and evaluation normal form is provided for the work memory research of the tree shrew.

The present application belongs to the technical field of bioelectric signal processing and control, and specifically relates to a non-stationary signal codec-free adaptive control method based on a neuromorphic device. The present application directly feeds the collected non-stationary bioelectric signal as a physical excitation into the neuromorphic hardware; by using the transient amplitude-frequency characteristics of the input signal, the dynamic evolution of the synaptic weight is induced through the bottom-layer pulse time-dependent plasticity, and the action pulse is triggered to drive the external physical actuator; the action of the external physical actuator causes natural sensory feedback, promotes the spontaneous remodeling of the power spectral density and phase-amplitude coupling index of the bioelectric signal, and then reversely guides the spontaneous attenuation and convergence of the hardware weight to the optimal steady state. The present application overturns the traditional closed-loop control paradigm of "analog-digital conversion-software algorithm decoding-feature matching", and can realize the adaptive closed-loop driving of complex signals without artificial algorithm code, thereby providing a new physical-level methodology for bionic control and bio-mechatronic adaptive interaction.