67 results about "Cognitive map" patented technology

The invention discloses an online learning cognitive map generation system and method for representing the knowledge learning mastering state of a leaner in a specific field. The system comprises an intelligent terminal and a server. The intelligent terminal is used for online learning interaction. The online learning cognitive map generation system runs in the server comprising a knowledge graph construction module, a learning interaction data acquisition module, a knowledge state calculation and prediction module and a visual output module. The online learning cognitive map generation system and method are suitable for a common online learning platform and can simulate the cognitive structure of the learner and diagnose the learning defect of the leaner to provide personalized learning resources and social network services for the learner, so that learning confusion is reduced, cognitive loads are reduced, and learning pertinency and accuracy are improved.

The invention provides a robot navigation map construction method based on rat brain hippocampus spatial cells. According to an information transfer loop of spatial navigation related cells in a hippocampus structure of a mammal, a robot acquires a current autonomic movement clue and color depth image information through environment exploration. The autonomic movement clue progressively forms coding to the spatial environment through path integration and characteristic extraction of the spatial cells in the hippocampus structure. The place field of place cells is gradually formed in an exploration process and furthermore covers the whole spatial environment, thereby forming a cognitive map. Meanwhile, Kinect acquires color depth image information of a view scene right in front of the current position as an absolute reference for performing closed-loop detection on the path and correcting a path integration error. At a closed-loop point, a system resets spatial cell discharging activity and performs correction on the path integration error. Nodes on a final navigation map comprise place cell mass coding information, corresponding visual cues and place topological relationships.

The invention belongs to the field of robot technologies and application, and relates to a robot situational cognitive map constructing and navigating method based on mouse brain location cells. In the process of exploring the environment, own motion information and image information in the front part are obtained through a robot, and the own motion information is transferred to a grid cell computing model, so that a corresponding location cell model is generated; by defining an event, location information in space and the obtained image information of the robot are encapsulated, and the eventis marked by state neurons; in the process of exploration, the state neurons are constantly generated to cover the entire environment and form a situational cognitive map of the environment; the generated situational cognitive map is utilized, and a global path planning algorithm is adopted to realized robot navigation based on a target; and through a local path planning algorithm, the global path of the robot is further optimized by a laser ranging sensor, and real-time obstacle avoidance and path local optimization of the robot are realized.

The present invention provides a method, system, and computer program for selecting a plurality of workers to accomplish a task. The method includes: identifying a task from a description; extracting at least one cognitive skill from the description of the task using machine learning methods; generating a group cognitive map which includes the at least one cognitive skill; and selecting a plurality of workers to accomplish the task based on at least the group cognitive map, wherein the plurality of workers comprises at least on artificial intelligent (AI) agent.

The invention discloses an intelligent vehicle positioning method and system for a cognitive map. The method comprises the following steps: treating, cognitive image data are collected and preprocessed; SSD landmark detection, continuously carrying out traffic sign image landmark detection on the preprocessed graphic data by utilizing an SSD detection algorithm; ORB feature extraction, feature tracking and feature screening, road sign positioning, intelligent vehicle self-positioning and other processes are carried out, an intelligent vehicle positioning mode based on a cognitive map is adopted, the cognitive map of an intelligent vehicle is mainly constructed based on vision, and a system that the intelligent vehicle can complete vehicle positioning only through a camera is achieved; According to the invention, the use of a laser radar is avoided, the hardware cost of the intelligent vehicle is reduced, the real-time processing data of the system is reduced, the calculation use cost is reduced, the intelligent vehicle positioning can be realized with small map data volume and small calculation amount, and the problems of large high-precision map data volume, high use cost and thelike in the unmanned field are solved.

The invention discloses a monitoring data query method and device, and is used for solving the problem of low query efficiency in the prior art. The method comprises the following steps that: on the basis of a first query condition and/ or cognitive map sent from a user, determining a second query condition, wherein the second query condition comprises a starting query node and a query area range;searching the cognitive map to obtain the path attribute information of at least one path which meets the second query condition in the cognitive map and the node attribute information of the node inat least one path, wherein the cognitive map comprises the path attribute information corresponding to the path formed by a plurality of nodes, wherein each node is the identification of camera equipment, and the node attribute information corresponding to each node comprises the physical position of the camera equipment; and according to the path attribute information, which is obtained by searching, of at least one path and the node attribute information of the node in at least one path, selecting M pieces of metadata from N pieces of metadata.

The invention discloses a learning cognitive path generation method based on a cognitive map. A cognitive learning path is recommended to a learner in combination with the cognitive dimension ontologyand the course knowledge graph; firstly, a cognitive map modeling and constructing method is provided; on the basis, for a learner and a target learning unit are given, firstly, the current cognitivestate of a learner is recognized according to a learning log; a cognitive learning path of the target knowledge unit is generated by combining the cognitive ability of the target knowledge unit; andrecommendation and generation of cognitive paths in personalized learning can be realized, wherein a cognitive dimension ontology diagram is designed, it is guaranteed that learners can analyze and diagnose the learning situation of each knowledge unit in a finer-grained mode, a learning target cognition path is generated based on the course knowledge graph, and the algorithm integrating the cognition dimension, the learning record and the learner cognition attribute can better provide more accurate personalized learning services for the learners.

The invention discloses an internet word frequency-based city cognitive map generation method. The method comprises the following steps: in a given territorial scope, determining the place names and road names in a study area, and establishing a place name table and a road name table; editing a status CAD drawing of the study area; capturing the word frequencies of the place names and road names in the study area, assigning the word frequencies of the place name in the place name table and assigning the word frequencies of the road names in the road name table; connecting a processed CAD file with the established tables, respectively calculating the word frequencies of intersections, roads and blocks, and generating a city cognitive map of the intersections, roads and blocks; and generating an optimum cognitive path between two points or among more points by using a traffic cost calculation method, and offering an optimization proposal of the study area according to the optimum cognitive path and the cognitive map of the intersections, roads and blocks. According to the internet word frequency-based city cognitive map generation method, the city cognitive measurement collected on the basis of network data can be used for providing new fundamental technical support for the planning and designing of city physical form.

The invention discloses a place cell bionic robot navigation algorithm which comprises the steps of: A1, firstly, capturing a panoramic photo by a robot wide angle camera, converting the panoramic photo into a gradient image, then filtering the gradient image by using a Gaussian difference filter, detecting characteristic points, carrying out log polar transformation on a local region near the characteristic points, wherein the characteristic points correspond to special road signs; A2, obtaining angle of each road sign relative to the true north direction as an azimuth angle, wherein the true north direction is provided by a compass and the azimuth angles and the road signs jointly determine a position cell of the current time; A3, jointly determining a transfer matrix by the position cell of the current time and a position cell of the former time, when a robot explores in an unknown environment, continuously forming transfer matrixes, forming a cognitive map by the transfer matrixes; and A4, jointly determining a movement transformation matrix by the transfer matrixes and the cognitive map, wherein the movement transformation matrix is in charge of sending a movement command. The invention can be applied to an intelligent ground-cleaning robot, a battlefield searching and rescuing robot, and the like.

The invention provides a bionic hippocampus cognitive map construction method based on a convolutional neural network, and the method can complete the construction of a cognitive map and guide the navigation of a robot. The method belongs to the technical field of robot environment cognition and motion navigation, and aims to solve the problem that a closed-loop detection module based on mouse brain hippocampus navigation is poor in robustness in a complex, changeable and high-repeatability environment at present. The specific process comprises the steps of firstly, in the environment exploration process of the robot, collecting a speed image, a direction image and an RGB image; inputting the speed and direction into a position sensing module to obtain real-time position information of therobot, and inputting the RGB image into a visual processing module to obtain image characteristics of the surrounding environment of the robot; and finally, associating the position information and the environment information of the robot and storing the position information and the environment information to cognitive map network nodes to complete construction of a cognitive map. And in the environment exploration process of the robot, when a closed loop is detected, position correction is carried out by using position information associated with the current image.

The invention relates to a document semantic automatic generating method, and particularly relates to a fuzzy cognitive map-based document semantic automatic generation method. Semantic representation is performed on paragraphs of a document through the fuzzy cognitive map, and the semantic representation of the whole document is realized by fuzzy cognitive map-base combination. The method comprises the following steps: first, preprocessing the document, including document segmenting, phrasing and word segmenting; second, converting each paragraph with atomic semantic into a corresponding atomic cognitive map; third, combining the paragraph atomic maps with the atomic semantic into a paragraph cognitive map; and fourth, combining the atomic cognitive map and the paragraph cognitive map of the same document into a document cognitive map so as to realize representation of document knowledge. In the method, the automation degree of document semantic representation can be improved, and the method is convenient for semantic representation applied to mass webpage text in a Web environment, so that the precision rate of Web searching can be improved.

The invention discloses an FCM-based waterlogging modeling and analysis method, which constructs an FCM model of waterlogging through modeling, and analyzes the occurrence of waterlogging based on the urban rainfall variation trend and waterlogging depth of waterlogging-prone points obtained therefrom. The method proposed by the present invention combines the strong fuzzy information expression and reasoning ability of the fuzzy cognitive map, and the gene-level operation of the DNA-encoded genetic algorithm technology to construct the waterlogging FCM model; the constructed waterlogging FCM model carries out the historical waterlogging data Integrate analysis and utilization, so that the ability of analysis is enhanced. Theoretically, the construction of this model is not restricted by the geographical location of waterlogging occurrence, and only needs to combine the parameter changes of a single waterlogging point under actual conditions to obtain the dynamic change trend of each parameter; therefore, it has great reference value.

The invention discloses a mechanical arm grabbing method based on cognitive maps. The method comprises the following steps that the outer layer cognitive map and the inner layer cognitive map are set up; according to collected target object image information, the current distance between a mechanical arm and an object is obtained, and the mechanical arm moves to the position of the outer layer cognitive map; according to the picture information of the position currently located and the proximate point in the outer layer cognitive map, outer layer matching is carried out; according to the result of outer layer matching, the track enabling the mechanical arm to move to the inner layer cognitive map can be planned in the outer layer map; according to the track, the mechanical arm moves to the area set up by the inner layer cognitive map; in the inner layer cognitive map, path planning is carried out according to the current position and the target position, and grabbing is completed. The method is good in environmental suitability and high in practicality.

The invention discloses a Mongolian multi-hop question and answer method based on a three-channel cognitive map and a map attention network. The method comprises the following steps: translating a Mongolian query question into a Chinese query question by machine translation; performing part-of-speech tagging on the Chinese query question after sentence segmentation and word segmentation; converting the Chinese question statement subjected to part-of-speech tagging into a Chinese query statement, and inputting the Chinese query statement into a cognitive map server; and allowing the cognitive map server to simulate a cognitive system of human beings in cognition, using three channels for parallel computing, finally, through normalization, query sorting and query selection, giving an answer with the maximum probability , and returning a result . Question and answer of different requirements are performed by means of respective advantages of the three channels, the accuracy and speed of a question and answer system in natural language processing are greatly improved, meanwhile, the channels 1, 2 and 3 can be reused in translation and question and answer stages, the speed is nearly doubled, and the question and answer quality is integrally improved.

The invention discloses a time series prediction method based on intuitionistic fuzzy cognitive map, the time series prediction model based on intuitionistic fuzzy cognitive map is constructed, the membership degree of the time series mapped to each cluster center at the current time is obtained by fuzzy C clustering, the hesitation degree and the non-membership degree are calculated according tothe membership degree, Each node of the model is obtained, and the parameters are learned according to the particle swarm optimization algorithm, and the weight matrix and weight hesitation degree after learning are obtained, and the hesitation degree is calculated according to the membership degree of the current time, and the time series of the next time is inferred according to the membership degree, the non-membership degree and the hesitation degree, and the prediction result is obtained, which has the advantage of high accuracy.

The invention provides a dynamic intuitionistic fuzzy cognitive map construction method, a time sequence prediction method and a time sequence prediction system. The dynamic intuitionistic fuzzy cognitive map construction method comprises the following steps: clustering standardized time sequence data into nodes of an intuitionistic fuzzy cognitive map by utilizing fuzzy C-means clustering; training a weight matrix of the intuitionistic fuzzy cognitive map to obtain an initial intuitionistic fuzzy cognitive map; receiving new standardized time sequence data and inputting the new standardized time sequence data into the initial intuitionistic fuzzy cognitive map; adjusting the position of a clustering center of the intuitionistic fuzzy cognitive map by utilizing dynamic fuzzy C-means clustering to obtain a dynamic intuitionistic fuzzy cognitive map; in the process of adjusting the position of the cluster center of the intuitionistic fuzzy cognitive map, judging whether the data at the current moment has concept drift or not by utilizing drift detection if the data at the current moment successfully fall into an existing cluster and an error between a real value solved by utilizing the adjusted intuitionistic fuzzy cognitive map and a predicted value is greater than a preset error threshold value and if so, adjusting a weight matrix of the intuitionistic fuzzy cognitive map; otherwise, not changing the intuitionistic fuzzy cognitive map.

The invention relates to a robot navigation method based on visual perception and a spatial cognitive neural mechanism. The collected visual images are transformed into visual nodes representing position and orientation angle information of the robot through a neural network to form a visual cell; a visual code of the visual cell is transformed into a spatial description of the environment, and acognitive map, similar to what is formed in the brain when the mammal freely moves, is constructed; and positioning and navigation of the robot are realized based on the cognitive map. According to aneural computation system of environment perception and spatial memory, the robot completes a series of tasks such as visual processing, spatial representation, self-positioning, and map update, thereby realizing robotic navigation with high bionics and strong autonomy in an unknown environment. Compared with the traditional simultaneous localization and mapping SLAM technology, the robot navigation method based on visual perception and the spatial cognitive neural mechanism in the invention avoid a series of complex calculations such as manual design visual features and feature point matchingand greatly improve the robustness of the system to factors such as illumination changes, viewing angle changes, object motion and the like in the natural environment.

The invention provides an online time sequence prediction method and system based on a granularity intuitionistic fuzzy cognitive map. The method comprises the following steps: converting a one-dimensional time sequence into a two-dimensional space form; clustering the preprocessed time series data into model nodes by using a fuzzy C-means clustering algorithm, and training and optimizing a numerical value weight matrix and granularity parameters of an intuitionistic fuzzy cognitive map by using a particle swarm optimization algorithm; expanding the numerical value weight matrix of the intuitionistic fuzzy cognitive map model into a granularity weight matrix by utilizing the optimized granularity parameters, and constructing a granularity intuitionistic fuzzy cognitive map prediction model; inputting to-be-predicted time sequence data into the trained granularity intuitionistic fuzzy cognitive map prediction model, and reasoning to obtain data at the next moment. According to the method and the device, the limitation of human factors is avoided, model establishment from original sequence data can be achieved, errors of a modeling process are considered, the prediction precision isimproved, results are expanded into intervals, and the results are covered as many as possible.

A system for insurance process management employing an advanced insurance management platform has been developed. A high speed data retrieval and storage module retrieves insurance related data from a plurality of sources. A data analysis module determines an activeness metric for an object, such as a physical asset, in order to categorize risk and also receives a plurality of individual, entity, and object data to create one or more cognitive maps which may analyzed to determine a propensity for risk. The data analysis module generates a cognitive density metric based on the activeness metric and cognitive map. A predictive analytics module performs predictive analytics functions on normalized insurance related data and using the cognitive density metric. A predictive simulation module performs predictive simulation functions on normalized insurance related data. As a result, the system can produce various models to determine risk and loss associated with an insured physical asset.

The invention discloses a man-machine cooperation natural language space navigation method and system, and the method comprises the steps of enabling a robot to receive a space navigation voice instruction transmitted by a user, and converting the space navigation voice instruction into a corresponding navigation instruction array according to a preset grammatical rule, wherein the space navigation voice instruction belongs to a natural language; enabling the robot to collect the position and attribute of a landmark object in an environment where the robot is located; determining a spatial cognition map of the robot according to the attributes of the landmark objects; enabling the robot to perform cognitive reasoning on the navigation instruction array based on the spatial cognitive map, determine the spatial three-dimensional coordinates of a destination to be reached by the robot required by the navigation instruction array in the spatial cognitive map, and determine the action to beexecuted when the robot reaches the destination; and enabling the robot to execute a related navigation action according to the spatial three-dimensional coordinates of the destination and the actionto be executed so as to reach the destination. According to the present invention, the understanding of the robot on the intention of the human natural language space navigation instruction is enhanced.

The invention relates to a cognitive map construction method based on space and motion joint coding, and belongs to the technical field of robot navigation. A sniffing layer-hippocampus neural circuitin an internal positioning system of the brain of a mammal is considered; a continuous attractor neural network model-spatial memory network is formed by utilizing spatial navigation coding characteristics of position cells, head orientation cells, grid cells and velocity cells, space and motion characteristics of animals are coded at the same time, and path integration is carried out on angularvelocity and linear velocity. In the network, the asymmetric connection weights among the neurons generate system inherent activity peaks capable of spontaneously moving, once speed input exists, theattractor network can steadily integrate the linear speed and the angular speed, and stable robot head direction and position coding is formed. And local view cells are formed through single-camera visual input, so that the accumulative error of path integration can be corrected. According to the invention, a consistent half-meter topological map can be stably constructed by using a single camera.

The invention discloses a multi-scale target-oriented navigation method based on cortical column network wavefront propagation, and belongs to the field of biomimetic navigation. The multi-scale target-oriented navigation method is characterized in that a cortical column network belongs to a cognitive map, and each cortical column unit comprises a plurality of neurons for receiving reward information, position information and the like. On the basis of the cortical column network, a wavefront propagation algorithm is used for path planning and navigation. Through a network weight STDP learninglaw, the system can calculate the shortest path leading to a target, and can memorize the change of the environment. Wavefront propagation adopts an integrated excitation type neuron, belongs to a non-attenuation navigation algorithm, and is suitable for navigation in various scale environments. In a navigation experiment, the multi-scale target-oriented navigation method can be used for successfully navigating in various scales and real-time changing environments, and has a relatively good practical effect.