22 results about "Adaptive behavior" patented technology

The invention discloses a poultry behavior abnormity real-time monitoring system based on multi-modal image fusion, particularly relates to the technical field of intelligent breeding behavior recognition, and is used for solving the problem of poor behavior monitoring accuracy under feather shielding. The method comprises the following steps: firstly, through combined perception of a visible light image and an infrared image, extracting a claw track interruption point and an anus temperature gradient direction, and realizing analysis of a motion state of a sheltered area; then, in combination with the heat conduction delay characteristic and the group movement direction, the flexion and extension angle of the covered leg joint is inverted, and a complete gait sequence is generated; thirdly, multi-source features such as gaits, temperature differences and body postures are fused, and a dynamic deviation model of the individuals relative to the mass center of the group is constructed; and finally, generating a stress behavior threshold curve according to the ground temperature and the ammonia gas concentration, outputting an abnormal behavior type and confidence, and realizing intelligent distinguishing of mechanical obstacles and adaptive behaviors.

The invention belongs to the technical field of artificial intelligence, and discloses an operation behavior characteristic-driven rehabilitation training effect dynamic evaluation method, which comprises the following steps of: extracting compensatory behavior characteristics and recessive behavior mode characteristics by acquiring multi-dimensional operation behavior data and physiological sensor data of a patient on a simulated driving device; and constructing a behavior pattern migration feature set. And calculating a behavior pattern migration index based on the compensatory behavior characteristics in the characteristic set, and accurately identifying the progressive transformation process of the patient from the compensatory behavior to the normal behavior. Further constructing a rehabilitation capability dynamic evaluation model, generating a real-time rehabilitation capability score, dynamically adjusting the complexity of a simulated driving scene, collecting adaptive behavior data, constructing a behavior mode migration trend curve and a rehabilitation effect prediction model, and realizing closed-loop optimization of a personalized training scheme. According to the method, subjectivity and static limitation of traditional rehabilitation evaluation are broken through, and the accuracy and effectiveness of neural rehabilitation are remarkably improved.

The invention discloses an education robot teaching strategy generation system based on learning behavior big data analysis, and particularly relates to the technical field of intelligent education. The method comprises the following steps: acquiring learning behavior data of students in an education robot teaching process; identifying the active adaptation behavior of the student, and generating an active adaptation behavior identification result; based on the active adaptation behavior recognition result, analyzing a response relationship between the existing teaching strategy and the student learning behavior, and outputting a teaching strategy adaptability analysis index; according to the teaching strategy adaptability analysis index, the deviation degree of the existing teaching strategy deviating from the real learning requirement is evaluated, and a teaching strategy deviation evaluation result is obtained; an adaptive feedback suppression constraint and a strategy depolarization weight are constructed to generate a teaching strategy candidate set, screening and confirmation are carried out according to a consistency verification rule and a robustness threshold, an education robot teaching strategy is updated, and an education robot teaching execution instruction is formed. And the scientificity and individuation level of the teaching strategy are improved.

According to an embodiment, disclosed is a system comprising, a sensor and a processor; wherein the processor storing instructions in a non-transitory memory that, when executed, cause the processor to detect, via the sensor of a host vehicle, a target vehicle; capture, via the sensor, a first region comprising a first characteristic of the target vehicle; analyze, the first region using an image processing module; determine, an identity of the target vehicle; capture, via the sensor, a second region comprising a second characteristic of the target vehicle; analyze, the second region using the image processing module; and determine, a safety feature that is enabled in the target vehicle.

An adaptive behavior recognition method and apparatus, a device, a medium, and a product. The method comprises: acquiring behavior feature data respectively corresponding to at least two individuals to be recognized (S110), wherein the behavior feature data represents statistical behavior information exhibited by said individuals in response to repetitive threat stimulus signals during repeated threat stimulus experiments; on the basis of at least two pieces of behavior feature data, clustering said at least two individuals to obtain at least two cluster sets (S120); and on the basis of cluster index values respectively corresponding to the at least two cluster sets, determining adaptive behavior results respectively corresponding to said individuals in each cluster set (S130).

The invention discloses a personality self-adaption-based personalized communication scheme generation method and device, and the method comprises the steps: receiving a communication scene inputted by a user, and extracting the personality features of the user and the personality features of an interaction object; according to the personality characteristics of the user and the personality characteristics of the interaction object, a behavior disorder diagnosis report is generated in combination with a personality disorder rule base; generating a scene simulation interaction task according to the behavior disorder diagnosis report, and sending the scene simulation interaction task to the user for interaction; collecting interaction data of finishing the scene simulation interaction task by the user, and comparing the interaction data with risk points in the behavior disorder diagnosis report to obtain a comparison result; and an intervention instruction is generated according to a comparison result, and the non-adaptive behavior or psychological state glide of the user is intervened and corrected in real time according to the intervention instruction, so that a final communication scheme is generated, and the behavior change efficiency and continuous motivation of the user in multiple fields such as interpersonal communication, sales, fitness training, learning and growth and the like are improved.

The invention discloses a refined behavior recognition method and system for a virtual teaching scene, and solves the technical problem that the precision of a final behavior recognition result is poor due to an existing refined behavior recognition method for the virtual teaching scene. The method comprises the steps that a virtual teaching scene behavior image is acquired, the virtual teaching scene behavior image is input into a behavior measurement model based on multilayer perception self-adaption, and the behavior measurement model based on multilayer perception self-adaption comprises a feature extraction network, a feature enhancement network and a posture decoding network; performing multi-scale feature extraction on the virtual teaching scene behavior image through a feature extraction network, and outputting multi-scale fusion features; performing feature enhancement on the multi-scale fusion features by adopting a feature enhancement network, and outputting target enhancement features; and inputting the target enhanced feature into a posture decoding network for posture decoding, and generating a behavior recognition result.

The invention relates to a hierarchical reinforcement learning method for crowd adaptive behavior simulation, and the method comprises the steps: constructing a static planning optimal path of an intelligent agent in a static closed environment based on a current position, a target point and an environment structure of the intelligent agent in a crowd, and generating a target trend speed of the intelligent agent moving from the current position point to the target point; establishing nonlinear mapping from laser radar observation data to a collision avoidance speed, providing dynamic obstacle avoidance guidance based on the surrounding environment of the intelligent agent, and dynamically adjusting a target trend speed weight and a collision avoidance speed weight according to distribution information of personnel around the intelligent agent; and performing speed weighted coupling according to the current target trend speed of the intelligent agent, the adjusted target trend speed weight value, the target trend speed weight adjustment value and the adjusted collision avoidance speed weight value to obtain the adaptive walking speed of the intelligent agent. Therefore, global path optimization is realized under the condition that the intelligent agent does not cause local collision, and the reinforcement learning efficiency is improved.

Methods, systems, devices and computer software / program code products enable the generation of synthetic images based on actual images captured by one or more physical cameras. The generated synthetic images provide visual spatial awareness to enable adaptive behavior in a dynamic environment, including robotics and simulation.

The invention discloses a student psychological assessment early warning method and system based on group behavior clustering analysis, and the method comprises the steps: obtaining multi-source campus behavior time series data, extracting group behavior feature vectors, and dynamically dividing a plurality of behavior normal form groups; calculating a multi-dimensional behavior baseline of each behavior normal form group; calculating a group relative deviation degree between the current behavior feature vector of the target student and the multi-dimensional behavior baseline of the behavior normal form group to which the target student belongs; analyzing the continuous increasing trend of the relative deviation degree of the group based on a sliding time window to judge adaptive behavior deviation; and performing attribution analysis on the target student with adaptive behavior deviation, generating a semantic risk early warning signal, and outputting targeted assistance suggestions according to the association intervention strategy library. According to the invention, through dynamic group division and relative deviation analysis, accurate and explainable early warning and effective intervention guidance of student behavior risks are realized.

The application belongs to the technical field of artificial intelligence, and discloses a rehabilitation training effect dynamic evaluation method driven by work behavior characteristics, multi-dimensional operation behavior data and physiological sensor data of a patient on a simulated driving device are acquired, compensatory behavior characteristics and implicit behavior mode characteristics are extracted, and a behavior mode migration characteristic set is constructed. Based on the compensatory behavior characteristics in the characteristic set, a behavior mode migration index is calculated, and the gradual conversion process of the patient from compensatory behavior to normal behavior is accurately identified. Further, a rehabilitation ability dynamic evaluation model is constructed, real-time rehabilitation ability scores are generated, the complexity of the simulated driving scene is dynamically adjusted, adaptive behavior data are collected, a behavior mode migration trend curve and a rehabilitation effect prediction model are constructed, and closed-loop optimization of a personalized training scheme is realized. The method breaks through the subjectivity and staticity limitation of traditional rehabilitation evaluation, and significantly improves the accuracy and effectiveness of neural rehabilitation.

An adaptive multi-agent system (100) for disaster simulation and crisis response, comprising: a. a data acquisition module configured to receive unstructured disaster-related inputs from one or more external data sources, wherein the data acquisition module includes a large language model-based submodule for semantic extraction and is configured to transform the unstructured disaster-related inputs into a structured semantic representation; b. a knowledge base storing geospatial data, infrastructure data, and historical disaster data; c. a retrieval-based generation module communicatively coupled to the data acquisition module and the knowledge base, and designed to retrieve relevant data from the knowledge base and merge the retrieved data with the structured semantic representation to generate a context-aware scenario descriptor;d. a role-based multi-LLM orchestration module configured to assign different-sized language models to respective roles, including data extraction, agent behavior generation, and simulation state evaluation; e. an agent generation module configured to instantiate a multitude of heterogeneous agents from the context-aware scenario describer, each agent being equipped with demographic attributes, mobility restrictions, a vulnerability level, and one or more adaptive behavior parameters; f. a disaster propagation engine configured to model hazard propagation as a probabilistic, graph-based model; g. a simulation engine configured to coordinate the time-graded interaction between the multitude of heterogeneous agents and a modeled environment;and i.e. a geospatial visualization module and a decision support module coupled with the simulation engine and configured to provide real-time monitoring, scenario-based what-if analyses, and predictive decision support results;characterized in that the system (100) further comprises a closed-loop feedback adjustment module configured to receive a current simulation state from the simulation engine, forward the current simulation state together with retrieval-based reference material obtained via the retrieval-based generation module to a language model determined by the role-based multi-LLM orchestration module for the role of simulation state evaluation, and dynamically adjust one or more of the decision guidelines of one or more of the plurality of heterogeneous agents, propagation parameters of the disaster propagation engine, and global parameters of the system (100) depending on an output of said assigned language model.

Embodiments of the present disclosure provide a method and apparatus for determining a PDCCH monitoring adaptive behavior. The method comprises: receiving PDCCH monitoring adaptive indication information sent by a network side device, the PDCCH monitoring adaptive indication information comprising first indication information; and determining a PDCCH monitoring adaptive behavior on the basis of the PDCCH monitoring adaptive indication information. In the method for determining the PDCCH monitoring adaptive behavior provided in the embodiments of the present disclosure, different configurations of the PDCCH monitoring adaptive indication information, the PDCCH monitoring adaptive behavior is determined, such that different PDCCH monitoring behaviors of a terminal device can be flexibly indicated, or a PDCCH monitoring behavior can be switched, thereby saving the monitoring power consumption of the terminal device and ensuring the timely transmission of data.

The application discloses a human-computer interaction method and system based on character emotion coupling and adaptive behavior rendering, relates to the technical field of robots, and comprises the following steps: receiving and simultaneously inputting signals, performing priority arbitration after fusion, and converting into an interactive dose vector IDV; performing character-emotion double-layer coupling calculation on the IDV, obtaining an original increment, and performing emotion mutual exclusion and attenuation constraint to obtain a final emotion state and character parameter; acquiring a basic action template, correcting the atomic action frame of the basic action template by using a four-dimensional dynamic correction coefficient to obtain final execution parameters, and delivering the final execution parameters to an execution mechanism. The application solves the response conflict problem of multi-sensor concurrent input, realizes that the character determines the emotion sensitivity and reflects individual differences, introduces a mutual exclusion constraint mechanism to avoid unreasonable emotion state superposition, and maps and penetrates the internal character and emotion parameters into the physical execution of each action frame in real time to solve the action mechanical problem.

The application discloses a human-computer interaction method and system based on character emotion coupling and adaptive behavior rendering, relates to the technical field of robots, and comprises the following steps: receiving and simultaneously inputting signals, performing priority arbitration after fusion, and converting into an interactive dose vector IDV; performing character-emotion double-layer coupling calculation on the IDV, obtaining an original increment, and performing emotion mutual exclusion and attenuation constraint to obtain a final emotion state and character parameter; acquiring a basic action template, correcting the atomic action frame of the basic action template by using a four-dimensional dynamic correction coefficient to obtain final execution parameters, and delivering the final execution parameters to an execution mechanism. The application solves the response conflict problem of multi-sensor concurrent input, realizes that the character determines the emotion sensitivity and reflects individual differences, introduces a mutual exclusion constraint mechanism to avoid unreasonable emotion state superposition, and maps and penetrates the internal character and emotion parameters into the physical execution of each action frame in real time to solve the action mechanical problem.

The invention relates to the technical field of large models, in particular to a user preference dynamic tracking model training method suitable for an electronic signature system, which specifically comprises the following steps of: acquiring signature micro-behavior data, document security element data and enterprise compliance strategy data of a user in the electronic signature system; performing degree-of-freedom analysis on a signing process by integrating user signing micro-behavior data and enterprise compliance strategy data, and quantifying the locking degree of a strategy on a user preference expression opportunity; switching sequence information of an enterprise compliance strategy is obtained, user adaptability behaviors and real preference drift conditions caused by strategy change are comprehensively evaluated, and adaptability behavior analysis is carried out; obtaining a degree-of-freedom analysis result and an adaptive behavior analysis result of the signing process, and carrying out preference dominance potential evaluation; according to the user preference dynamic tracking method and device, the problem that in the prior art, a reliable model capable of stably tracking the real preference dynamic state of the user is difficult to construct is solved.

The application provides an adaptive behavior planning system for a perception robot, and relates to the technical field of perception robots, and is based on a perception robot extracting a multi-sensor, performing environment perception, obtaining an environment perception dataset, preprocessing the environment perception dataset, dimensionally reducing the environment perception dataset according to a preprocessing result, generating an environment perception dimensionally reduced data array, performing behavior analysis on the perception robot according to the environment perception dimensionally reduced data array, combining multiple behavior information, constructing a behavior tree, performing strategy training through the behavior tree combined with the environment perception dimensionally reduced data array, and executing a behavior strategy to plan adaptive behaviors for the perception robot. The technical problem of low behavior planning efficiency and poor adaptability of the perception robot in a complex environment in the prior art is solved. The technical effects of improving the behavior planning efficiency of the perception robot, enhancing the adaptability to the complex environment and realizing intelligent behavior decision are achieved.

The invention belongs to the technical field of data processing, and particularly relates to an abnormal behavior recognition data processing method based on a skeleton key point topological graph, which comprises the following steps of: S1, accessing a real-time video stream of a field high-position camera, performing attitude estimation on each frame of image in the video stream, extracting two-dimensional skeleton key points of an operator in a scene, and extracting the two-dimensional skeleton key points of the operator; the adjacent key points are connected according to a human anatomy structure to construct a skeleton topological structure; s2, obtaining the observation length of each limb in the skeleton topological structure on the current frame image, and determining the standard reference length corresponding to each limb in combination with a preset standard human body prior library. According to the method, the problem that human body behaviors are difficult to recognize under high-angle monitoring is solved from the mathematical principle, meanwhile, zero-sample self-adaption is achieved, and the method has the advantages of being high in sensitivity and low in false alarm rate balance.

The invention relates to the technical field of robots, and discloses a multifunctional four-wheel-foot composite structure intelligent mechanical dog and a control method thereof, and the control method comprises the steps: configuring a main control unit, executing an adaptive behavior tree, distributed edge calculation, visual-infrared feature fusion tracking, dynamic topology path planning and wheel-foot mode switching; configuring a sensing module; configuring a driving module; configuring a communication module; executing the adaptive behavior tree; executing distributed edge calculation, and distributing calculation tasks to the main control unit and the auxiliary nodes; executing vision-infrared feature fusion tracking to generate a target trajectory; dynamic topology path planning is executed, and an environment adaptation path is generated; and executing wheel foot mode switching. The dynamic coordination management of target tracking, path planning, obstacle avoidance, attitude control and interaction tasks is realized through the adaptive behavior tree, the limitation of a fixed decision framework is broken through through dynamic priority calculation and node reconstruction, and the adaptability of a multi-task scene is remarkably improved.

Methods, systems, devices and computer software / program code products enable the generation of synthetic images based on actual images captured by one or more physical cameras. The generated synthetic images provide visual spatial awareness to enable adaptive behavior in a dynamic environment, including robotics and simulation.

The invention relates to the technical field of artificial intelligence, and discloses a law enforcement video intelligent analysis method based on deep learning, and the method comprises the following steps: S01, carrying out the time period division of a law enforcement video, and obtaining an abnormal time period video, S02, extracting a to-be-analyzed time period video, carrying out the adaptive enhancement processing of a video frame, and obtaining a video frame; s03, performing target detection on the enhanced video frame in combination with an edge processing technology, S04, matching a self-adaptive behavior tracking technology based on a target detection result, and S05, comprehensively analyzing a self-adaptive behavior tracking result to perform abnormal behavior responsibility judgment, and performing rapid detection on the video frame in combination with an edge end to identify an abnormal type. And carrying out fine-grained detection to obtain the number of target objects, and carrying out comprehensive analysis on a single target behavior tracking result and a plurality of target behavior tracking results by matching an adaptive behavior tracking technology, so that abnormal behaviors in a video are accurately judged, and a responsibility subject of the abnormal behaviors is determined.

This invention discloses a refined behavior recognition method and system for virtual teaching scenarios, solving the technical problem that existing refined behavior recognition methods for virtual teaching scenarios result in poor accuracy of the final behavior recognition results. The method includes acquiring behavior images of a virtual teaching scenario and inputting these images into a multi-layered perception adaptive behavior measurement model, which includes a feature extraction network, a feature enhancement network, and a pose decoding network. The feature extraction network extracts multi-scale features from the behavior images of the virtual teaching scenario, outputting multi-scale fused features. The feature enhancement network enhances the multi-scale fused features, outputting target enhanced features. The target enhanced features are then input into the pose decoding network for pose decoding to generate the behavior recognition result.