74 results about "Behavior learning" patented technology

The invention discloses a machine behavior learning method based on emotion driving and human expert feedback, and the method comprises the following steps: a) fusing environment image features extracted by BLIP-2 and text instruction semantic features analyzed by GPT-4, and forming cross-modal input; b) training a basic VLA model through supervised fine tuning (SFT) by using human expert remote control trajectory data and cross-modal input to obtain a basic behavior strategy; c) combining an emotion recognition module with a multi-head self-attention mechanism, fusing emotional interaction dependency into a basic strategy, and generating high-order emotion driven behavior representation; and d) inputting the high-order emotional behavior representation into a reinforcement learning module, storing a track by using a Replay Buffer, carrying out optimization through human expert feedback preference learning, and outputting a final behavior strategy. Compared with an existing method, the method has the advantages of being high in multi-modal feature extraction capacity, high in emotion fusion degree, sufficient in expert feedback utilization and the like, and the response accuracy and interaction experience of a machine to human instructions and emotions can be improved to a certain degree.

The invention relates to the technical field of voice recognition, in particular to a voice interaction system and method for customer service based on artificial intelligence, and the system comprises a voice input processing module, an intention classification and routing module, a context dynamic adjustment module, a user behavior learning module, a multi-level intention fusion module and a final result module. According to the method, a multi-dimensional feature system is constructed by extracting tone intensity, speech speed frequency and emotional fluctuation amplitude, intention categories, priority weights and confidence scores are generated to realize accurate acquisition of appeals, and dialogue history, context and emotional change dynamic reconstruction path nodes, switching rules and response time sequences are tracked during interaction. Historical behavior mining preference features, habit fusion intention relevance, emergency calculation of an optimal strategy, construction of service steps, resource allocation schemes and execution timelines, adjustment of an interactive interface, a service process and a feedback mechanism according to multi-dimensional analysis, guarantee of differentiated service experience, and improvement of response accuracy and user satisfaction.

The invention relates to the technical field of smart home and mobile terminals, and discloses a smart home anti-theft system based on a smart mobile terminal, which comprises a dynamic behavior learning module, a multi-modal sensor collaborative network, a dual-channel verification unit and a cross-platform emergency linkage engine. According to the smart home anti-theft system based on the smart mobile terminal, a false alarm scene is effectively eliminated and the false alarm rate is reduced through a multi-modal sensor collaborative network and a dual-channel verification unit, and a personalized security policy is dynamically generated by adopting a mixed learning algorithm and an edge-cloud collaborative computing architecture, so that the privacy data of a user is protected; through the cross-platform emergency linkage engine, hierarchical alarm and multi-dimensional response are supported, the intrusion event processing efficiency is improved, it is ensured that an evidence chain cannot be tampered, judicial evidence obtaining standards are met, and the problem that evidence of a traditional system is prone to being lost or tampered is solved.

The invention relates to the technical field of intelligent wheelchair control, in particular to an intelligent wheelchair control method and system. The method comprises the following steps: constructing a three-dimensional ground feature model by extracting historical environment data in a memory of electronic monitoring equipment on a wheelchair; thirdly, ground pit geometrical morphology analysis is carried out, pit data are obtained, wheel clamping stagnation situation simulation is carried out based on the pit data, and the wheel struggling behavior is simulated; and finally, constructing an optimization model by using the struggling behavior learning data, and performing optimization through training feature importance parameters to form a clamping stagnation struggling control optimization model. The model is then transmitted to the intelligent wheelchair control terminal. The intelligent wheelchair control technology is optimized, so that the intelligent wheelchair control technology is more perfect.

The invention discloses an emergency lane changing dangerous scene generation method and system based on behavior guidance and a medium. The method comprises the following steps: constructing a track data set of vehicle emergency lane changing behaviors; training a behavior learning module with short sequence learning ability according to the trajectory data set; training an active attack strategy model through guidance of a behavior learning module; outputting a reference track point of the active attack vehicle through the active attack strategy model, and obtaining an optimal control input in a future time domain according to the reference track point and a preset prediction control model; and controlling the active attack vehicle to interact with the simulation platform according to the optimal control input, and generating an emergency lane changing dangerous scene. The method improves the generation efficiency of the emergency lane changing dangerous scene, thereby improving the efficiency of the automatic driving test, and can be applied to the technical field of automatic driving.

The invention relates to the technical field of textbook construction and optimization, in particular to an automatic textbook construction and optimization method and device.The method comprises the following steps that original resources of multiple types of textbooks are collected, and core knowledge elements in the original resources are recognized; constructing a structured mapping relationship between the knowledge elements and the teaching resources based on the core knowledge elements; on the basis of the structured mapping relationship, constructing a knowledge graph covering multidisciplinary knowledge points and a preposition dependency relationship, and automatically identifying knowledge faults and complementing missing association so as to form a knowledge network with self-evolution ability; based on the knowledge network, utilizing a hybrid generation model to cooperatively work, and quickly generating a structured textbook framework containing knowledge veins, teaching cases and exercise resources; learning condition data including classroom behaviors, learning feedback and knowledge mastering conditions are obtained, and a quantity and chemical condition report is obtained after data processing; and updating data based on the quantitative chemical situation report and the knowledge graph.

The invention relates to the technical field of intelligent indoor illumination, and discloses a sunshade and illumination linkage intelligent adjustment method based on environment self-learning, and the method comprises the following steps: constructing an environment feature model representing external light environment features based on collected environment illumination data; based on the intervention behavior of the user, learning to obtain a user preference model reflecting the comfort preference of the user; in combination with the environment feature model and the user preference model, determining a linkage control strategy for the sunshade equipment and the lighting equipment; and according to the linkage control strategy, the sunshade equipment and the lighting equipment are controlled. By adopting a reverse modeling strategy, an environment characteristic model is reversely deduced from actually acquired illumination data, adaptive learning of an external light environment is realized, the implementation threshold and cost of the system are remarkably reduced, and the system can automatically adapt to long-term or temporary changes of a surrounding environment of a building, so that the system is more intelligent and more convenient to use. And the environmental adaptability and robustness of the system are improved.

The invention discloses a robot application secondary development method supporting user behavior pattern self-learning, and relates to the field of robot application development, and the method specifically comprises the following steps: S1, data collection and treatment; s2, performing behavior abstraction; s3, mode learning; s4, performing double-domain scheduling during operation; s5, performing mode capitalization; and S6, low code reuse. According to the robot application secondary development method supporting user behavior mode self-learning, during operation, a double-domain isolation framework physically isolates a safety domain from a learning domain, and the situation that key operation of a robot is disturbed in the learning process is avoided; the scheduling algorithm combining the SMP and the EDF and a real-time monitoring mechanism can accurately control core indexes such as thread scheduling delay and interrupt response time, ensure that key tasks such as robot motion control and emergency fault processing are executed preferentially, meanwhile, parallel operation of behavior learning and execution is achieved, the strict requirement for real-time performance of an industrial scene is met, and the real-time performance of the industrial scene is improved. And the operation safety of the robot is ensured.

The invention is suitable for the technical field of graphic processing, and provides a high-density element loading system and method applied to large-scale scene rendering, and the system comprises a scene data analysis module which is used for constructing a scene graph with a spatial index; the multi-level-of-detail management module is used for generating multi-level-of-detail (LOD) resources for the elements; the visual attention degree analysis module comprises a static feature analysis unit and a dynamic behavior learning unit and is used for generating and dynamically correcting an attention degree thermodynamic diagram; the rendering scheduling module is used for dynamically formulating a rendering strategy based on the thermodynamic diagram, the current window state and the LOD resources; and the synthesis module is used for synthesizing and outputting the cache bitmap and the dynamic element to the display canvas. Therefore, according to the method, the rendering performance, the memory efficiency, the interaction fluency, the convenience and other dimensions are all improved.

The invention relates to the technical field of automobile cabin environment control, and discloses an intelligent cabin environment adaptive adjustment system based on an AI large model, which comprises a sensing module, an AI large model decision module, an execution module and a feedback module. According to the intelligent cabin environment self-adaptive adjustment technology based on the AI large model, a mixed decision sub-model of an AI large model decision-making module adopts an MPC-TD3 fusion algorithm, and a user behavior learning sub-model of the AI large model decision-making module learns user physiological parameters and operation records collected by a sensing module; the multi-physics field prediction sub-model predicts a cabin environment and a power battery state, and corrects a decision result in combination with comfort parameters fed back by a PMV-PPD evaluation model and an SET evaluation model of the feedback module in real time, so that the problems that a single MPC model is insufficient in robustness under complex working conditions and a single reinforcement learning model is difficult to quantify the comfort feeling of a user are solved, and the user experience is improved. And the dynamic response speed and the adjustment precision are considered.

The invention relates to the technical field of intelligent household appliances and control, and discloses a temperature regulation and control system and method of an intelligent direct drinking machine. The system comprises a main control chip which is internally provided with a user behavior learning module, a diagnosis module and a prediction control module. The method comprises the following steps: collecting, clustering and analyzing historical water consumption data, and establishing a user portrait reflecting user habits; performing look-ahead diagnosis based on the portrait, and actively measuring and calculating the efficiency attenuation rate of the heating module caused by aging or incrustation; the future water demand is predicted in combination with the user portrait, and water is preheated to the optimal standby state in advance; when a user takes water, feedforward compensation is carried out by utilizing the efficiency attenuation rate, and compound control is carried out in combination with real-time feedback, so that hot water at a target temperature is accurately and quickly provided. The problems that an existing direct drinking machine is slow in response, high in energy consumption and inaccurate in temperature due to performance degradation are solved, and the intelligent level, the user experience and the energy-saving effect are remarkably improved.

The invention discloses an intelligent medicine chest management system based on multi-modal perception and adaptive learning, which obtains medicine visual information and millimeter wave feature codes through a rotary scanning platform, analyzes taking rules in combination with an edge side large model, and dynamically optimizes a reminding strategy based on user behavior learning. The system innovatively comprises a non-standard medicine multi-mode fusion recognition technology, a context-aware self-adaptive reminding engine and a hardware-level privacy protection architecture, can effectively improve family medicine taking compliance, obviously reduces the medicine taking error rate of a nursing home, and has great social value.

The invention discloses a scenic spot personnel distribution monitoring and early warning system and method, and relates to the technical field of scenic spot monitoring. The system comprises a memory for storing different security level early warning areas divided based on monitoring area boundaries and allowed moving paths; the real-time monitoring system is used for comparing the actual position of the personnel with the allowed path and generating an alarm signal when the path exceeds; the trajectory tracking system continuously tracks the subsequent path of the personnel until the personnel enter the allowed path again; and the behavior learning system updates the security rules according to feedback of the management terminal. The method comprises the steps of obtaining the real-time position of a person and carrying out trajectory compliance analysis, selecting an alarm level when deviation is detected, executing dynamic resynchronization through a trajectory tracking system, and updating safety rules of a geographic information model according to feedback. According to the system and the method, dynamic monitoring and early warning of the scenic spot personnel are realized, and the efficiency and the accuracy of scenic spot safety management are improved.

The invention discloses an automatic driving behavior simulation method and device, and the method achieves a high-precision simulation model which is consistent with the real world in a driving behavior distribution level through the unified vectorization representation of a driving environment, the behavior learning with a vehicle as a center, the introduction of routing guidance, and the construction of a trajectory prediction model. And it is ensured that large-area long-time simulation such as city-level simulation has high stability. Furthermore, illegal and unrealistic trajectory behaviors are avoided through a rule and security post-processing compensation mechanism, and a stable guarantee is provided for subsequent closed-loop simulation and real system deployment. Furthermore, through a closed-loop calling mechanism, the problem that the traditional single-step reasoning method cannot stably simulate for a long time is avoided, meanwhile, the complex behavior mode experienced by the vehicle in the urban road network is covered, and the authenticity and continuity of simulation are greatly improved.

This invention relates to the field of parcel delivery volume prediction technology, and discloses a method for predicting logistics parcel delivery volume based on a multi-level network architecture. The method includes: acquiring raw data and standardizing the data; constructing a multi-level network hierarchy and a multi-dimensional weighted ranking model; displaying the data in multiple dimensions; calculating the confidence interval based on the prediction model and attaching a confidence level label to each predicted value; introducing a user behavior learning model to record user click habits and query preferences, and optimizing the network display order and recommendation logic; using an integrated prediction model to fuse time series analysis, regression analysis, and machine learning prediction methods to form a combined prediction result; establishing a prediction error feedback mechanism to compare the actual delivery volume with the predicted delivery volume and optimize model parameters; and adopting a hierarchical prediction strategy. This invention provides reliable support for logistics delivery planning and resource allocation, reduces operating costs, and improves service quality.

The invention relates to the technical field of telecommunication, in particular to an eSIM card security encryption data transmission system which comprises a social simulation management module, a social behavior learning and strategy evolution module, a self-adaptive encryption strategy module, a trust collaborative verification module and an anomaly detection and self-healing module. According to the method, historical experience is precipitated through a digital twin memory bank, an optimization strategy is generated through reinforcement learning training, and a closed-loop mechanism of real-time calling and dynamic adjustment is adopted, so that autonomous collaborative optimization of the concealment and the transmission efficiency in a specific social situation is realized, and comprehensive risk quantitative evaluation based on multi-dimensional situation awareness and dynamic weighting is realized. And the dynamic matching is adaptive to encryption algorithms of different risk levels, so that the accurate balance among the security strength, the communication efficiency and the equipment resource consumption is realized.

The invention relates to a robot task level emotion feature extraction method and system based on multi-evaluator distribution modeling, and belongs to the technical field of artificial intelligence and robots. The method comprises the following steps: collecting emotion score data of multiple evaluators; calculating a plurality of emotion statistical characteristics; constructing an emotion prediction model and selecting an optimal feature subset based on the evaluation indexes; and outputting a structured emotion interface for a robot behavior learning system to call. According to the method, statistical modeling and feature selection mechanisms are introduced, so that the problem that an existing emotion recognition method lacks generalization and task-level coupling capability is solved, and the naturalness and decision-making flexibility of robot emotion interaction are remarkably improved.

The invention belongs to the technical field of vehicle control, and particularly relates to a vehicle energy-saving navigation method and system based on historical behavior learning, and the method comprises the steps: obtaining the travel data of a target vehicle in a historical time period, generating a historical path sequence composed of standard road segment identifiers through map matching, mining a high-frequency passing path of the target vehicle, and obtaining the high-frequency passing path of the target vehicle; according to the speed sequence, track points meeting a preset low-speed condition are identified, and the track points are clustered and verified as intersection waiting points; for each intersection waiting point, establishing a time-dependent waiting probability model according to historical passing records of the intersection waiting point under different time slices; when the target vehicle runs in real time and is matched with a certain high-frequency passing path, inquiring a waiting probability model of a waiting point of a next intersection in front according to the current position, the current time and the vehicle speed; and according to the queried waiting probability value, generating and outputting a hierarchical driving guide instruction to a vehicle end human-computer interaction interface. And a personalized intersection passing memory bank is established for vehicle owners, so that the mileage anxiety is relieved.

The application relates to the field of robots and discloses a skill simulation method, device and equipment of a robot and a storage medium. The method comprises the following steps: acquiring the gripper state when the robot completes an operation task and inputting the gripper state into a gripper simulation environment to obtain a first joint action position through simulation simulation; inputting the first joint action position into a joint simulation environment to obtain a second joint action position and a visual image through simulation simulation, taking the second joint action position and the visual image as input data, training and optimizing a preset initial prediction formula through a preset loss function, obtaining a prediction formula, inputting the initial joint action position and the initial visual image of the robot in the joint simulation environment into the prediction formula, obtaining a target joint action sequence, and driving the robot to perform corresponding actions. According to the scheme, the joint action information and the visual information are learned, the action sequence of the robot is predicted, the technical difficulty of the robot in operation behavior learning is reduced, and the intelligent degree and the generalization ability of skill simulation of the robot are improved.

The invention discloses an industrial control security isolation device and method suitable for an industrial control network, and relates to the technical field of industrial control system network security. The objective of the invention is to solve the problems of complex deployment, insufficient detection depth and lack of intelligent behavior learning ability of an existing general protection product at an industrial control terminal. The method is deployed between an operation host and a terminal controller, and a port sniffing and protocol identification module is used for automatically identifying a controller protocol and loading an analysis method; analyzing message characteristics through a communication learning and trust value modeling module, and establishing a dynamic trust value model; the risk identification agent module takes over communication in an agent mode, carries out grading processing on messages according to a trust value model, and carries out application layer deep protocol analysis and risk identification on low-trust-value messages. According to the technical scheme, targeted and intelligent safety isolation and protection of the industrial control terminal are realized.

The invention relates to the technical field of automobile artificial intelligence, and provides a vehicle-mounted interaction system and service prediction method based on cross-scene behavior learning, and the system comprises a data layer which collects behavior data of a user in a vehicle, constructs a four-dimensional behavior modeling engine, and compares the collected behavior data with behavior data stored in a cloud knowledge base; the learning layer runs a federal learning framework in a vehicle-mounted edge computing unit, generates a situation model of the current comprehensive state of the user and sends the situation model to the interaction layer; the interaction layer is used for establishing a service execution grading mechanism based on credibility, generating a prediction result, sending the prediction result to the execution layer, and determining a service execution mode according to prediction accuracy; and the execution layer is used for establishing a prediction service matrix and executing a corresponding service action according to the prediction result. According to the method, single-dimensional association is broken through; the dynamic situation is fused, so that the position prediction is more accurate; and a progressive trust mechanism is adopted.

Systems and methods for learning behavioral activity correlations. A method includes intercepting a plurality of requests, wherein each of the plurality of requests is directed to a respective destination entity of a plurality of destination entities; creating a request queue by queueing the plurality of requests; inspecting contents of the plurality of requests; separately forwarding each intercepted request to its respective destination entity based on the request queue; monitoring runtime output of each of the plurality of destination entities, wherein the runtime output includes behavioral activities of the plurality of destination entities; and training a machine learning model based on the contents of the plurality of requests the runtime output of each of the plurality of destination entities, wherein the machine learning model is trained to output request-output correlations between groups of requests and subsequent behavioral activities.

The invention discloses a multi-scene timer control method and system based on offline voice wake-up. The method comprises the following steps: S1, establishing a trigger starting point of natural interaction through offline voice wake-up; s2, determining the content and intention of the timed task through voice recognition and semantic understanding; s3, determining an applicable scene according to the voice content and the environmental condition, and generating a corresponding control strategy; s4, executing a timing process according to the control strategy and performing staged energy consumption management; s5, completing a task closed loop through diversified reminding and continuous task connection; and S6, the recognition effect and task response logic are optimized through voice behavior learning. According to the multi-scene timer control method and system based on off-line voice wake-up, the comprehensive control strategy table is generated in combination with environment characteristics after voice recognition, and a proper timing mode, a proper reminding mode and a proper feedback strategy are selected according to the use scene.

The invention discloses a KNX smart home adaptive control method and system based on user behavior learning, relates to the field of smart home control, and aims to solve the problem of insufficient flexibility of traditional KNX smart home control. The method comprises the following steps: collecting and storing historical behavior data of a user through a multi-source sensor, identifying and marking abnormal data through an anomaly detection algorithm, generating a training data set through time sequence interpolation correction, and training an LSTM model to obtain a dynamic trigger threshold value; and collecting real-time behavior data, processing the real-time behavior data by the same method, inputting the processed data into the model, comparing the real-time data with a threshold value, and generating a KNX standard control instruction to drive equipment linkage if a condition is met. The system supports the implementation of the method, can adaptively match user habits, and improves the intelligence and convenience of smart home control.

The present invention discloses a system and method for monitoring and early warning of personnel distribution in scenic areas, and relates to the technical field of scenic area monitoring. The system includes a memory for storing early warning areas of different safety levels and permitted movement paths divided based on the boundaries of the monitoring area; a real-time monitoring system for comparing the actual position of personnel with the permitted path, and generating an alarm signal when the path is exceeded; a trajectory tracking system for continuously tracking the subsequent path of the personnel until the permitted path is re-entered; and a behavior learning system for updating safety rules based on feedback from the management terminal. The method includes obtaining the real-time position of the personnel and performing trajectory compliance analysis, selecting an alarm level when a deviation is detected, performing dynamic resynchronization through the trajectory tracking system, and updating the safety rules of the geographic information model based on the feedback. The system and method realize dynamic monitoring and early warning of personnel in scenic areas, and improve the efficiency and accuracy of scenic area safety management.

The present application provides an application access control method based on system call restrictions, which relates to the field of network security technology and provides an application behavior learning module and an application permission control module; the application behavior learning module performs the following steps: A1, creating an application behavior list; A2, dynamic training monitoring, training and monitoring for each item in the application behavior list; A3, generating a mapping table, establishing a mapping relationship between each item in the application behavior list and the system call; the application permission control module performs the following steps: B1, user authorization application, initiating an application access request to the authorization center; B2, authorization approval, after receiving the application access request, the authorization center authorizes the application according to the requirements; B3, issuing permissions. The present application further refines the granularity of permission management, reduces the exposure of applications, protects enterprises, reduces the risk of malicious behavior, protects system security and data privacy, and improves overall security.

The invention discloses a file sending method, electronic equipment and a computer storage medium, and is applied to terminal equipment, the method comprises the following steps: obtaining a file operation of a user acting on a preset application, carrying out operation analysis on the file operation based on a preset user behavior learning program, and determining an operation identification result; if the operation identification result is that the file operation needs to be intercepted, a first file operation entry provided by a preset file sending application is called to replace a second file operation entry of the preset application so as to select files through the first file operation entry, and the first file operation entry comprises options of a predicted folder corresponding to the file operation; and the selected target file is transmitted to the preset application, so that the sending of the target file is executed through the preset application, and the file sending efficiency and experience of the user are improved.

The present invention discloses a thinking learning method and system based on AI convergent evolution, including inputting the historical designs of the current designer and the historical designs of other designers of the same type as the current designer, extracting design features in the historical designs, and generating multiple first behavior samples; limiting the behavior logic of the first behavior samples based on the process tree; expanding the first behavior samples based on the behavior logic to generate multiple second behavior samples; performing convergent evolution on the first behavior samples and the second behavior samples, and outputting a final design sample. The present invention generates multiple behavior samples by performing behavioral learning, logical restrictions, and behavioral expansion on the historical works of the designer and other designers of the same type, thereby generating a larger number of richer samples, and then performing AI convergent evolution on the generated behavior samples, thereby enhancing the iterative nature of AI design technology, and ultimately outputting diversified sample results to provide to designers for reference, thereby stimulating the designers' creativity.

The invention provides a step adjusting method and electronic equipment, and the method comprises the steps: an intelligent agent executes a first task, and triggers a behavior learning event; obtaining a first execution parameter of the target step, wherein the first execution parameter comprises first data representing a user operation behavior and second data representing a picture change; determining a matching result based on an association relationship between the first data and the second data; and adjusting the target step based on the matching result.

Some examples of a method and apparatus are described, including: for each of a plurality of time instants, receiving operation data, wherein the operation data comprises a current configuration of an operation and an output of the operation, where the operation is computer- implemented functionality for managing a communications network or for managing a service deployed in a communications network; determining a performance label associated with the output of the operation; generating behaviour data comprising the operation data and the performance label; and storing the behaviour data in a behaviour repository. The method comprises simulating the operation using the stored behaviour data; and triggering change to the configuration of the operation in response to output of the simulation.