33 results about "Social robot" patented technology

A social robot provides more believable, spontaneous, and understandable expressive communication via embodied communication capabilities by which a robot can express one or more of: paralinguistic audio expressions, sound effects or audio/vocal filters, expressive synthetic speech or pre-recorded speech, body movements and expressive gestures, body postures, lighting effects, aromas, and on-screen content, such as graphics, animations, photos, videos. These are coordinated with produced speech to enhance the expressiveness of the communication and non-verbal communication apart from speech communication.

Social robot formed by an artificial vision system composed of webcam cameras, a voice recognition system formed by three microphones arranged in a triangular configuration, an expression system composed of an LED matrix, formed by a plurality of LEDs and a status LED, and eyelids formed by half-moons connected to gearwheels which engage with respective servomotors via transmission wheels, a speech synthesis system composed of loudspeakers, a system for detecting obstacles which is formed by ultrasound sensors, and a movement system formed by two driving wheels.

The invention belongs to the technical field of robots and provides a learning type intelligent home social contact robot system and method. The learning type intelligent home social contact robot system comprises a man-machine interface module, a main controller, an environment monitoring module, a robot positioning module, a motor driving module, a vision module, a voice module, a data storage module, a wireless data transmitting module, a power management module and a touch type game platform, wherein all the modules except the main controller are connected with the main controller. The learning type intelligent home social contact robot system is characterized in that the voice module further comprises a voice input module, a specific person learning type voice recognition module, a specific person learning type semantic parsing module, a specific person learning type response statement generating module and a specific person learning type voice synthesis module, and the touch type game platform further comprises a specific person learning type card showing module. By means of the learning type intelligent home social contact robot system and method, recognition of the voice of all family members is achieved, and the learning type intelligent home social contact robot system and method have the advantages that the voice with the pitch consistent with that of imitated persons is given out through learning, and therefore chatting, card playing and interaction with people are achieved.

The invention belongs to the technical field of computers, particularly provides a social robot detection method and system capable of blending user behaviors with text information, and aims to solvethe problem in the prior art that characteristics are manually selected, logicality and a time sequence between social media posts are ignored and the user behavior information of a social platform isignored. The social robot detection method comprises the following steps that: obtaining the historical network data and the friend network data of a social media user to be detected; on the basis ofthe above data, obtaining a user text characteristic vector, a behavior characteristic vector and a friend network characteristic network, and blending all vectors to obtain the user characteristic vector of the social media user to be detected; and detecting the user characteristic vector to output a detection result. The method disclosed by the invention more conforms to the own characteristicsof the social media, the social media user to be detected is analyzed from multiple dimensions, and detection accuracy is improved. The system also has the above beneficial effects.

The present invention provides an interaction device adapted for setting own controller for maximizing expectation of information defined between a hypothesis about an interaction object and own input/output. Thus, the social robot can judge by using only simple input/output sensor whether or not the human being is present or absent at the outside world.

A computerized system comprising a social robot for interacting with a patient in order to improve the patient's adherence to a health or medical regime. A patient is instructed to perform a health activity (e.g., take medication, step on scale, measure glucose, play a game) at certain intervals in order to reach a specified goal or expected result. The results of the health activity are recorded in a wireless health device and transmitted to a computer. The computer interprets the result of health activity and communicates an action to the robot. If the patient performs the activity within a specified interval and results are consistent with the specified goal or expected result, the robot communicates relief or pleasure. If the patient does not perform the activity within a specified interval, or if the results are not consistent with the specified goal or expected result, the robot communicates anxiety, disappointment or exhibits symptoms of an effect such as hunger or illness.

The invention belongs to the field of human-machine interaction intelligence, and relates to a social robot which is used for detecting indoor personnel in real time, recording the form of current personnel by photographing, and sharing the form to a social network. The social robot comprises a chassis, wherein the chassis is provided with a power device; the top of the chassis is provided with a supporting device; the supporting device is provided with a user terminal; the power device can drive the chassis to move under the control of the user terminal; the upper end of the supporting device is provided with a binocular vision camera; the binocular vision camera is hinged to the upper end of the supporting device; the binocular vision camera can rotate to different angles relative to the supporting device; the binocular vision camera acquires depth path information of the front environment; the binocular vision camera photographs a human face under the control of the user terminal, and transmits pictures to the user terminal; the user terminal stores pictures photographed by the vision camera, and shares the pictures to the social network. The social robot has the advantages of low constructing price and cost, high flexibility, simplicity in operation, and diversity and sufficiency in human-machine interaction.

The invention discloses a microblog social robot detection method based on a deep neural network. The method comprises the steps of constructing an artificial mark data set, extracting classificationfeatures of user data, and constructing a deep neural network model based on a residual neural network, a bidirectional gating circulation unit and an attention mechanism to perform Sina microblog social robot detection. User data of Sina microblogs are obtained through web crawlers, behaviors of users and robots are analyzed, classification features of the user data are extracted from four different perspectives, and the social robots can be detected more comprehensively and accurately by means of the features; and the deep neural network model is constructed by combining the residual neuralnetwork, the bidirectional gating circulation unit and the attention mechanism, so that the accuracy and the stability of the detection model can be remarkably improved.

The invention discloses a social robot detection data set extension method and system based on active learning. The expansion method comprises the steps: collecting microblog user data, extracting a part of data samples for manual annotation, constructing an initial annotation data set, training a classifier by adopting an active learning method, annotating remaining unannotated data sets, and expanding the data sets according to an annotation result. According to the method, the problem that a reliable large sample data set is lacked in social robot detection research in microblogs is solved,full-automatic data annotation and classification are realized, unnecessary human resource consumption is reduced to a great extent, and the data annotation efficiency is improved to a great extent.

A method and system for improving the software programming of a robot system, comprising monitoring of plurality of human user-robot interactive pairs' (HURIP) interactions. System comprises each of said plurality of HURIPs as using ‘front-end’ semi-autonomous robot component linked by wireless two-way communications to a ‘back-end’ cloud-based computerized component. Monitoring comprises review of robot sensor-gathered data and data from camera and audio data from homes of users during user-robot interactions. Analysis of said monitoring by authorized observers such as psychologist, parent, teacher, system administrator, software programmer(s), enables identification of areas for software improvement. Improved software is tested, wherein testing comprises at least similar monitoring of HURIPs, and wherein said testing comprises social robots comprising said updated software. Cycles of such monitoring of HURIP interactions, analyzing data derived from said monitoring, focus for improvement derived thereof and followed-up in coding updates, testing of updates comprising use within monitored HURIP interactions, such cycles are applied in herein disclosed method to manufacture progressively improved code for and uses of social robot system.

A social robot detection method based on variational self-encoding and K-nearest neighbor combination belongs to the technical field of anomaly detection, and comprises the following steps: acquiring public data of a social robot through a network, extracting features through preprocessing, training by adopting the data, and encoding and decoding by using variational self-encoding. Normal sample features are more similar to initial features after being decoded, abnormal samples are greatly different from the initial features, the original features and the decoded features are fused, and then anomaly detection is performed by using an anomaly detection method K-nearest neighbor. In a social network environment, the number of abnormal user groups is smaller than that of normal user groups, so that collection of abnormal users is relatively troublesome in a data collection process. According to the method provided by the invention, the defects of high-cost labeling and unbalanced positive and negative samples in the existing social robot detection method are overcome, and efficient detection of social network machine users is realized by reducing abnormal sample participation model training.

The invention provides a social worker robot simulation method for loan investment network fraud, and belongs to the field of artificial intelligence/social engineering. The method specifically comprises the steps that firstly, a plurality of social work robots with different attributes are set according to actual requirements, and user attributes are configured for all the social work robots; filling a small loan application form in each loan platform according to the attribute value, and leaving personal information of the social robot to be leaked to the fraud; then, a behavior, language and voice database conforming to basic human settings of the social worker robot and a fraud process archive library are constructed; then, the social worker robot interacts with the fraud person, and all information is reserved and stored in a database; and when the fraud person induces the social worker robot to transfer accounts, the social worker robot sends an account transfer failure picture with the embedded Trojan horse program, traces the fraud person and feeds back the traced fraud person to the public security department. The method is easy to operate and high in practicability, and the implementation efficiency is greatly improved.

The invention discloses a robot detection method for multiple social network platforms, which comprises the following steps: acquiring user account data of the social network platforms, taking an account ID as a unique identifier of a user, and extracting user features, friend features, network features, content features, emotion features and time sequence features; constructing a high-dimensional original matrix, and obtaining a low-dimensional feature matrix through significance analysis; and a clustering algorithm or a classification algorithm is adopted to realize division, and a normal user account and a robot account are identified. The invention further discloses a robot detection device for the multiple social network platforms, electronic equipment and a storage medium. According to the method, account data of a plurality of domestic and overseas social network platforms are researched, social robot detection is performed through algorithms such as feature representation, feature significance analysis, clustering or classification and the like, and social robot accounts in the social network are identified, so that abnormal behaviors of large-scale social robots are early warned, and the security of the social network is maintained.

The invention relates to the field of machine learning and social media, and discloses a social media promotion method and device based on social action injection and a storage medium. The method comprises the following steps: registering a plurality of social media accounts, programming and developing application programs for executing social media actions, binding the application programs to the accounts, and executing the programs to generate a social robot; a public API provided by a web crawler or a social media is utilized to obtain part of user data including basic data of users and part of blog articles, user-level features x and blog article-level features xt are extracted from the data, subsets of a data set are obtained, and a simulator f (.) is trained by utilizing a small amount of obtained data to learn interaction in the social media; and calculating an influence vector based on the simulator f (.) obtained by training, and implementing popularization by forwarding blog articles and modifying basic data by using the social robot. The method is simple to operate, easy to implement and low in cost, and the acceptance degree of the target user to the specified content is effectively improved.

The present invention provides a method for quickly detecting social media robot groups, which is characterized in that it comprises the following steps: step 1, obtaining the basic attribute feature vector, behavior feature vector, text feature vector and relationship attribute feature vector of individual social accounts; step 2, Detect individual social robots through the machine learning model; Step 3, obtain the vector expression of the individual social account on the graph neural network as a graph vector; Step 4, use the graph vectors of all individual social accounts to calculate the similarity of the graph vectors , to determine whether the undetected individual social account is a social robot. The present invention first detects individual social robots, and then searches for users similar to the detected individual social robots through similar user judgment based on the graph neural network, so as to realize rapid detection of social media robot groups.