43results about How to "Improve the ability of independent learning" patented technology

The invention relates to an on-line sequence limit learning machine method possessing an autonomous learning capability and belongs to the intelligent robot technology field. Nine portions are included and comprise an external state set, an external motion set, a rewarding signal, a value function, a state transition equation, a limit learning machine network hidden layer output set, an intermediate parameter transfer equation, a limit learning machine output set and a limit learning machine output weight transfer equation. In the on-line sequence limit learning machine method possessing the autonomous learning capability, an on-line sequence limit learning machine is taken as a framework and Q learning is combined and reinforced. The on-line sequence limit learning machine method possessing the autonomous learning capability is provided and a model is used in mobile robot path program research so that a robot realize autonomous learning navigation according to an external environment state and rewarding. An autonomous learning capability of the robot in an unknown environment is increased.

The invention discloses a parking garage entry method and device, and belongs to the field of automobile active safety. The method comprises the steps that a first action control parameter is determined according to the state information of an automobile; the position state of the automobile is adjusted according to the first action control parameter, and the adjusted position state of the automobile is a first position state; whether the first position state is the preset state or not is judged; if the first position state is the preset state, the position information of the automobile located in the parking environment when the automobile is in the first position state and the position information of a target stall in the parking environment serve as input parameters of a machine learning algorithm, and output parameters of the machine learning algorithm are obtained; a path corresponding to the output parameters serves as a target driving path; and the automobile is controlled to finish the parking garage entry action according to the target driving path. The parking garage entry method and device solve the problems that an existing parking garage entry method is poorer in stability, lower in reliability and poorer in flexibility. The effects of improving stability, reliability and flexibility are achieved, and the parking garage entry method and device are used for controlling the automobile parking garage entry.

The invention discloses a teaching management system based on a cloud platform. The system comprises a base layer which is used for providing hardware support and data support for the integral teaching management system and providing resource service for a user paid according to demands and comprises an infrastructure layer and a basic service layer, a platform service layer used for providing platform service for the integral teaching management system, and an education application layer used for providing application management for the integral teaching management system, and a structural mode of the teaching management system is an MVC design mode based on SOA configuration.

The invention discloses an obstacle avoidance method and device for a vehicle, and belongs to the active safety of vehicles. The method comprises the steps: detecting a first longitudinal distance from the vehicle to an obstacle vehicle; detecting the current distance from the vehicle to the obstacle vehicle when the first longitudinal distance is less than a first preset distance; inquiring the corresponding relation between preset lateral and longitudinal distances and a first enhanced signal value, and determining the first enhanced signal value corresponding to the current distance; Determining first motion control parameters through an enhancement learning method according to the current distance when the first enhanced signal value is greater than a first preset value, wherein the first motion control parameters comprises a reference force value of an accelerator or a brake and a rotation reference angle of a steering wheel; and controlling the vehicle to complete the obstacle avoidance operation according to the first motion control parameters. The method and device solve problems of poorer stability, lower reliability and poorer flexibility in an obstacle avoidance process, improve the stability, reliability and flexibility, and are used for vehicle control.

The invention discloses a semiconductor intelligent illumination control experimental teaching platform, which comprises a host machine managing system (computer), an integrated console and a single lamp control box. In the platform, a modern digital control technology, a sensor technology, a network transmission technology and a computer software and hardware technology are combined, and a modularization design principle is adopted, so that a plug-and-play effect is realized for various modules through a customized terminal and a bread board connector. Due to the adoption of the semiconductor intelligent illumination control experimental teaching platform, teaching experiments such as design of a semiconductor intelligent controller, designs of intelligent monitoring systems of different scales and the like can be finished. The semiconductor intelligent illumination control experimental teaching platform has the advantages of complete functions, safety, reliability, convenience for operating, capability of fully bringing the subjective initiatives of students into full play, and good teaching effect.

The invention relates to a work order classroom vocational education and teaching cloud platform. The teaching cloud platform comprises a work order classroom cloud service subsystem comprising a service server and a service database server, and is used for providing work order development, course development, classroom teaching, career development, internship and data analysis services for the teaching cloud platform. The Job Classroom Cloud Sales subsystem includes a marketing portal server, Platform management server and marketing database server, used for the construction of work orders, marketing, course construction, marketing, enterprise recruitment, part-time work, data analysis management, and responsible for the platform account management, authority management, work orders and course review, points management, column management, business management.

The invention discloses a method and device for parking a vehicle into a garage, and belongs to the field of vehicle active safety. The method includes the steps of determining a first movement control parameter according to state information of the vehicle, estimating the first movement control parameter to determine a second movement control parameter, adjusting the position state of the vehicle into the first position state according to the second movement control parameter, if the first position state is a preset state, making the position information of the vehicle in the parking environment in the first position state and the position information of a target vehicle spot in the parking environment serve as input parameters of a self-adaption dynamic scheduling algorithm to obtain output parameters, making a route corresponding to the output parameters serve as a target driving route, and controlling the vehicle to complete the park-into-garage movement according to the target driving route. The problems that a method for parking a vehicle into a garage is low in stability, reliability and flexibility are solved, the effects of improving stability, reliability and flexibility are achieved, and the method and device are used for controlling the vehicle to be parked into the garage.

The invention relates to an electric energy metering device abnormality detecting method based on a long and short term memory model. The method comprises the steps of (1) acquiring original collecteddata, (2) carrying out data preprocessing, (3) setting model hyperparameters, (4) training the model with a training sample, (5) testing the model with a test sample, and (6) output an electric energy metering device abnormality detecting result. According to the electric energy metering device abnormality detecting method based on the long and short term memory model, the deep learning theory isapplied to digital electric energy meter abnormality detection, and the change characteristics of each piece of collected data under various faults are automatically learned in case of a large amountof data and easy missing, at the same time, the method has good fault tolerance, and the improvement of the accuracy and timeliness of the abnormality detection of an electric energy metering deviceis helped.

The invention relates to a learning system and an assistant learning method thereof based on non-deletable user evaluation data management. The system comprises a system application manager, a user evaluation manager, an evaluation tool box, an evaluation data storage device, an evaluation certification manager, a learning optimization manager and the like, can effectively collect, store and utilize user evaluation opinions of learners for foreign language textbooks used, provides a learner-oriented (foreign language) learning system, help learners actively select foreign language learning textbooks fitting for basic conditions and individual characteristics of the learners according to the actual learning subjective needs, feeling and effect of the learners with the user evaluation opinions of the learners for foreign language textbooks used as the important basis of evaluation and selection of textbooks, and pertinently adjusts the learning content and learning process of the learners in learning foreign languages so as to improve the self learning ability and efficiency. Meanwhile, the invention can effectively analyze the applicability of foreign language to related user groups according to user evaluation information on learning effect of learners, and pertinently modifies and improves foreign language textbooks and continuously promotes the improvement in quality level of foreign language textbooks.

The invention relates to an informatization teaching system and method based on school-enterprise cooperation project courses, and belongs to the technical field of education. The informatization teaching system based on the school-enterprise cooperation project course comprises a micro-cloud server and a user terminal which can realize data communication with each other. The micro-cloud server isused for transmitting, storing, analyzing and processing the teaching whole-process data generated by all the user terminals, such as teaching organization, teaching management, teaching implementation, teaching evaluation and teaching feedback, and feeding back the optimized data to the corresponding user terminals in the form of reports. According to the system, all user terminals are connectedtogether by using a cloud technology; mutual communication among students, teachers in schools, enterprise teachers and practical training managers is promoted, the system timely and effectively feeds back teaching information to the teachers, guides the teachers to rectify teaching, timely and effectively feeds back personal learning information of the students to the students, guides the students to complete knowledge point review, and facilitates implementation of school-enterprise cooperation project courses.

The invention discloses an entity class online management system based on the Internet. By establishing online classes corresponding to entity classes and extending lecture space and time of the entity classes, the individualized teaching is realized, and the teaching efficiency is improved; by establishing an Internet system service end, support demands are provided for teaching and managing interacting, resource sharing, lecture scenarizing, teaching precision, family and school interacting and after-school tutoring, online teaching interaction experience is optimized, the learning interestand autonomous learning ability of students are improved, teachers can better know learning situations of the students, and high-quality teaching is realized.

The invention belongs to the field of aircraft control, particularly relates to an aircraft attitude control method, system and device based on reinforcement learning, and aims to solve the problem ofrealizing non-overshoot control under a low-precision model. The aircraft attitude control method based on reinforcement learning comprises the following steps of: acquiring aircraft state data at at-th moment; calculating an expected attitude angle and an attitude angle tracking error based on an expected flight path of an aircraft, and constructing a network input state at the t-th moment; based on the network input state at the t-th moment and the expected attitude angle, acquiring an optimal control action through utilizing a DDPG deep reinforcement learning network; and adjusting the attitude of the aircraft based on the optimal control action. Non-overshoot control can be realized under a low-precision model, the adjustment time is short, and the steady-state error is small.

The invention discloses a method and a device for intelligently obtaining learning resources, a printer and a computer storage medium, and the method comprises the steps: obtaining a picture comprising a question at the printer; identifying content information in the picture to obtain required knowledge points; sending the knowledge points to a server, so that the server searches learning resources related to the knowledge points from a pre-established knowledge point database; and receiving the learning resources sent by the server, and printing the learning resources. According to the invention, the user can quickly obtain the learning resources corresponding to the knowledge points according to wrong questions, so that the user can consolidate the unclear knowledge points in the learning process according to the learning resources and perform deep learning.

The invention discloses a child study type television set which comprises a study function module. Study content related to study is pre-stored in the study function module. Work of the study function module is set in the mode that when the operation state of the television set meets the operation start condition of the study function module, the television set is automatically switched to be in a study mode, in order words, the study content pre-stored in the study function module is automatically played through the television set, the television set can not be switched to a playing channel of a traditional television program until the television set detects that an operation stop condition of the study function module is met, the television set is automatically switched to be in an original playing state of the television set, and the study function module stops working. Thus, the requirement of a child for watching TV is met, knowledge study is facilitated, and the problem that the child becomes addictive after watching the television program for a long time is solved.

The invention relates to a single-transmission multi-frame infrared target detection method, which comprises the following steps of: starting from a feature pyramid network, describing the inequalityof contribution of each feature layer to fusion output on the basis of learnable weights, achieving bidirectional multi-scale feature weighted fusion between a low-resolution and strong-semantic feature layer and a high-resolution and weak-semantic feature layer, and constructing an auxiliary network; based on the cross-parallel ratio, considering the influence of an overlapping region and a non-overlapping region on a target function at the same time, constructing detector positioning loss keeping invariance to target scale change, constructing a target detector, and improving the sensitivityof a detection model to small target positioning errors. A VGG16 convolutional neural network is used as a feature extraction network and is integrated with an auxiliary network and a target detectorto form a single-shot multi-frame infrared target detection model fusing multi-scale feature weighted fusion and scale invariance positioning loss. The method has autonomous learning ability and highdetection rate, and is an effective way for solving the infrared imaging guidance target detection problem in a complex environment.

The invention discloses an intelligent health monitoring self-repairing system and method for a fiber grating sensor network, belonging to the field of engineering structure health monitoring technology. Aiming at the reliability of an intelligent health monitoring system of a fiber grating sensor in a large-scale engineering structure, the functions of information acquisition, extraction, analysis, evaluation and the like of a conventional distributed fiber grating intelligent health monitoring system are transformed into intelligent agents that can sense the own bodies and the environments and can communicate and mutually cooperate with other agents by adopting multi-agent cooperation and decision-making techniques, failure sensors are subjected to functional compensation based on the fiber grating sensor which is survived near a damaged area through the mutual cooperation among the agents, and thus the whole intelligent health monitoring system can be restored to normal work without changing the fiber grating sensor network in a monitored object, the functional compensation for local failure of the fiber grating sensor network can be achieved, and finally, the intelligent evaluation of the health state of the monitored object can be achieved with higher accuracy and reliability.

The invention discloses a learning resource acquisition method and device, a learning terminal and a computer readable storage medium. The method comprises the steps of receiving courseware sent by acurrently connected teacher terminal; displaying the courseware, and receiving annotations of a user on the courseware; obtaining courseware content, corresponding to the annotation, on the courseware; identifying the courseware content, and obtaining learning resources corresponding to the courseware content according to an identification result; displaying the learning resources. According to the invention, the corresponding learning resources can be obtained according to the annotated courseware content, so that the user can consolidate unclear knowledge points in the learning process and learn deeply according to the learning resources, and the autonomous learning ability of the user is improved.

The invention discloses an online education platform for training guidance. The online education platform comprises a task entering module, a concluding module, an uploading module, a user module, a display screen, a comparison module, a data analysis module, a controller, an information intercommunication module, a data processing module, a data distinguishing module and a storage module. According to the online education platform, the data analyzing modules receives score signals transmitted by the comparison module, starts analysis operations and respectively carries out comparison on acquired Zj and Y as well as acquired Hj and S so as to generate teaching signals, learning signals and problem solving signals, the teaching signals, and the learning signals and the problem solving signals are sequentially are sequentially transmitted to mobile phones of teachers, parents and students through the controller and the information intercommunication module, and the mistakes of the students in the learning process are guided in a targeted manner and are hierarchically answered in a targeted manner, so that the learning quality and effect of the students are improved.

The invention relates to a medicine self-service vending system and method based on face recognition. The system comprises a medicine vending machine, a plurality of collecting assemblies arranged onthe vending machine, a controller electrically connected with the collecting assemblies, and a database electrically connected with the controller. The system is characterized in that one side of themedicine selling machine is fixed to a wall, the vending machine is provided with screens besides the side attached to the wall, the screens can display medicine information and doctor guidance suggestions, and the screens are matched and can independently display the medicine information and the doctor guidance suggestions, wherein the acquisition assembly can acquire face information and analyzegender and age information of personnel, and the acquisition assembly acquires time information of viewing a screen by a drug purchaser in a fixed-eye manner and then inputs the information into a database for analysis; the screen can recommend matched amplified medicine information according to the eye-catching watching time of the medicine buyer acquired by the acquisition assembly; and the database can store case information of the medicine purchasing personnel.

The invention discloses an automatic car-following method and device, and belongs to the field of vehicle active safety. The automatic car-following method includes the steps that n strengthening signal values are obtained by equally dividing the car-following distance into n sections and determining strengthening signal values for sub-areas corresponding to all the sections of the distance, the car-following distance is a preset driving distance of a local car following a front car, the area corresponding to the car-following distance is a reliable area, and n is an integer larger than 2; n normalization values are obtained by carrying out normalization on the n strengthening signal values; whether the local car is in the reliable area or not is judged according to the n strengthening signal values, the n normalization values and the current driving position of the local car; when the local car is in the reliable area, the local car is controlled to finish the automatic car-following motion through a machine study method. The automatic car-following method resolves the problem of low reliability of automatic car-following, realizes the effect of improving the reliability of automatic car-following, and is used for automatic car-following.

The invention relates to a novel braking structure of an instructional car. The novel braking structure of the instructional car comprises a brake box, a reversing brake seat and a locking mechanism connected with a brake mechanism of a main cab of the instructional car; a telescopic mechanism is arranged inside the brake box; a processing unit of a radar sensing device on the instructional car controls the telescopic mechanism to stretch and contract; the telescopic end of the telescopic mechanism is connected with a steel wire rope; the telescopic mechanism stretches and contracts to drive the steel wire rope to stretch and contract synchronously; the externally stretching end of the steel wire rope penetrates through the reversing brake seat to be fixedly connected with the locking mechanism; and the position, between the brake box and the reversing brake seat, of the steel wire rope is sleeved with a protection sleeve. The novel braking structure of the instructional car is small, exquisite and simple in structure and convenient to install, and achieves intelligent mechanical operation, a coach does not need to sit on a co-driver seat all the time, the traditional structure that a brake device is arranged in the co-driver position is abandoned, the situation that the coach sits on the co-driver seat for a long time to cause fatigue teaching is avoided, potential safety hazards are eliminated, and the independent study capacity of trainees is improved.

The invention relates to a promotion network self-enhanced image voice deep learning model. According to the application, a promotion scheme and a neural network structure are fused to design a deep learning architecture which can realize autonomous learning of network parameters and modules and is compatible with linearity and nonlinearity, and the system stability is improved; linear and nonlinear promotion schemes are combined with a neural network classification model through loosening and compacting to construct a deep learning classification model, and the generalization ability is high; the convolution layer and the pooling layer are autonomously combined in the aspect of the network structure, a fully promoted network is realized, and grading processing has a better effect in classification and estimation of reasoning tasks; in the aspect of network operation, convolution operation and pooling operation are achieved through promotion, unification of linearity and nonlinearity is achieved, and the calculation complexity is low; in the network training process, hierarchical low entropy is adopted to accelerate network training, so that the network retains the learnability of convolution, a pooling layer becomes learnable, the model learning ability is strong, the error is smaller than that of the method in the prior art, and the accuracy and robustness of image speech recognition and classification are better.

The invention relates to an elevator medical information advertisement putting system and a method based on user behaviors. The elevator medical information advertisement putting system comprises a lift car, at least one collecting assembly, at least one putting assembly, a controller and a database, wherein the collecting assemblies and the putting assemblies are arranged on any inner wall of thelift car, the controller is electrically connected with the collecting assemblies and the putting assemblies, and the database is electrically connected with the controller. The collecting assembly and the throwing assembly are matched with the inner wall of the lift car through sliding ways. The collecting assembly and the throwing assembly can slide along the slide way. The acquisition assemblycan acquire face information and analyze sex and age information of passengers; the delivery assembly can deliver a corresponding advertisement according to the face information collected by the collection assembly; the acquisition assembly acquires time information of watching advertisements by passengers in a fixed-eye manner, and then inputs the information into the database for analysis and comparison. The database can store advertisements and advertisement preference information of passengers; and the controller is used for controlling information calling of the putting assembly, the collecting assembly and the database.

The invention relates to a work order creation unit for a work order classroom teaching cloud platform. The teaching cloud platform includes the work order classroom cloud business subsystem, a work order classroom cloud marketing subsystem and a work order classroom school-side business subsystem, the work order classroom cloud service subsystem comprises a service server and a service database server, the work order classroom cloud marketing subsystem comprises a marketing portal server, a platform management server and a marketing database server, and the a work order classroom school-sidebusiness subsystem comprises a college resource synchronization unit, a CACHE management unit and an Nginx proxy service unit, wherein the work order creation unit is arranged in the service server, and the work order used for the work order classroom teaching cloud platform comprises five parts of work order identification, work task book, learning resources, environment deployment and quality monitoring.

The invention discloses a teaching interaction method and electronic equipment based on gamification education positional relationship, the method includes: establishing a display interface in advance, the display interface includes a first display unit and a second display unit, and the first display unit It is a Jiugongge display unit; when the second display unit is 0, the first material is displayed in the first row and the first column of the Jiugongge display unit, and the seventh material is displayed in the third row and the first column; in the The second row and the first column of the Jiugongge display unit display the fourth material, and the sixth material is displayed in the second row and the third column; the fourth material or the sixth material is selected; when the fourth material is selected, Then update the second display unit to 1, and keep displaying 0 on the second display unit when the sixth material is selected. In this way, both teachers and students can improve teaching efficiency or learning efficiency through location-based learning software, which greatly saves time.

The invention discloses a method and device for automatically following a car, belonging to the field of active safety of automobiles. The method includes: obtaining n enhanced signal values, the n enhanced signal values ​​are obtained by dividing the vehicle following distance into n sections at equal intervals, and determining the enhanced signal value for each sub-area corresponding to the distance, and the following vehicle distance is preset The driving distance of the vehicle following the vehicle in front, the area corresponding to the following distance is the reliable area, and n is an integer greater than 2; n normalized values ​​are obtained, and n normalized values ​​are normalized to n enhanced signal values According to the n enhanced signal values, n normalized values ​​and the current driving position of the vehicle, it is judged whether the vehicle is in the reliable area; when the vehicle is in the reliable area, the machine learning method is used to control the vehicle The car completes the automatic follow-up action. The invention solves the problem of low reliability of automatic car following, realizes the effect of improving the reliability of automatic car following, and is used for automatic car following.

The invention relates to an online sequence extreme learning machine method with autonomous learning ability, which belongs to the field of intelligent robot technology and is divided into nine parts, namely external state set, external action set, reward signal, value function, and state transition equation. , extreme learning machine network hidden layer output set, intermediate parameter transfer equation, extreme learning machine output set, extreme learning machine output weight transfer equation. The online sequence extreme learning machine method with autonomous learning capability provided by the present invention uses the online sequence extreme learning machine as a framework and combined with reinforcement Q learning to propose an online sequence extreme learning machine method with autonomous learning capability, and uses the model Applied to mobile robot path planning research, the robot can realize autonomous learning and navigation according to the status and rewards of the external environment, and improve the robot's autonomous learning ability in unknown environments.

The invention discloses an English learning shorthand device comprising a learning device body, wherein the learning device body is rectangular; four corners of the learning device body are fillets; the four edges of the learning device body are each fixedly sleeved with an edge protection sleeve. One side of the learning device body is rotatably connected with a folding plate; the size of the folding plate corresponds to the size of the learning device body; four protection pads are fixedly connected to the bottom of the learning device body and located at the four corners of the bottom of the learning device body, and a word display screen, an entertainment display screen, a display panel button, a keyboard display screen and a keyboard are sequentially arranged on the front face of thelearning device body from top to bottom. According to the English learning shorthand device, through cooperative work of the English learning shorthand systems, the autonomous learning ability of students is improved, the English memory speed is increased, and the English application ability of college graduates can meet the social requirements correspondingly.

The invention belongs to the field of aircraft control, and in particular relates to an aircraft attitude control method, system, and device based on reinforcement learning, aiming at solving the problem of realizing no-overshoot control under a low-precision model. The aircraft attitude control method based on reinforcement learning of the present invention includes: obtaining the aircraft state data at the first moment; calculating the expected attitude angle and the attitude angle tracking error based on the expected flight trajectory of the aircraft, and constructing the network input state at the first moment; The network input state, the desired attitude angle, and the optimal control action are obtained through the DDPG deep reinforcement learning network; the attitude adjustment of the aircraft is performed based on the optimal control action. The invention can realize no overshoot control under the low-precision model, has short adjustment time and small steady-state error.

The invention relates to a scannable system and a scanning method thereof, which can be used for identifying different states including a static state, a working state and a post-working state and processing data so as to obtain relatively accurate data, and identifying different lesions in different states in combination with an established model. The system can continuously verify and optimize the model, can realize autonomous learning and updating, and improves the accuracy and reliability of lesion identification.