78 results about "Learning gain" patented technology

Dynamic systems and methods for gathering / tracking data, automatically adapting to an individual's pace of learning, selectively determining the type and difficulty of content provided to an individual, selectively providing an exposure frequency for the content, and / or enabling rapid design modifications within the educational environment. Educational content is dynamically designed / developed and customizably presented to an individual learner. An analysis is performed on the data to optimize learning. Modifications are selectively or automatically made to the educational content. The process of designing, implementing, analyzing, and selectively modifying creates a cycle that optimizes the learning process and adapts to groups and individual learners with the goal of improving learning outcomes and efficiency. Furthermore, the educational content is dynamically provided to the learner on an iterative basis according to the need of that learner in the learning process.

The invention provides a system and method for efficiently measuring and reporting performance relative to established learning outcomes for students, professors, classes, courses and programs. The invention provides an evidence-based approach to measuring student achievement and increasing educator accountability. Using associations between learning content, assessment content, and learning outcomes, the invention facilitates measurement of student achievement, content effectiveness, and teaching effectiveness, through reporting of student, class, teacher, course, and institution performance relative to set outcomes in an on-line educational environment.

Online course and learning material management system and methods are provided where students, instructors, publishers, administrators, publishers, software vendors, and industry partners utilize the disclosed systems' databases and networking features to interact, to network, to improve learning outcome, and to exchange information elements. The system furthermore allows for networking among different user categories based on users' roles, professional profiles, and professional interests.

A system for objective assessment of learning outcomes comprising a data repository comprising at least a hierarchical arrangement of a plurality of learning goals, a report generator coupled to the data repository, an analysis engine coupled to the data repository, a rules engine coupled to the data repository, and an application server adapted to receive application-specific requests from a plurality of client applications and coupled to the data repository. The application server is further adapted to provide an administrative interface for viewing, editing, or deleting a plurality of learning goals and relationships between them, learning assessment tools, learning outcome reports, and learning indexes, and the rules engine performs a plurality of consistency checks to ensure alignment between and among learning goals, learning assessment tools, learning outcomes, and learning indexes. The application server receives learning assessment data and the analysis engine performs analyses to generate a plurality of learning indexes.

A system and method for correcting repeatable runout errors during manufacture of a disc drive. The system includes a Kalman filter having a recursive learning gain input and includes a recursive learning Again-setting circuit coupled to the recursive learning gain input. The recursive learning gain is initially set based on an estimate of a ratio of non-repeatable run out error to an estimate of the repeatable run out error. On subsequent recursions, the recursive learning gain-setting is reduced.

A method for correcting track errors generated during iterative learning of an industrial robot comprises the following steps that firstly, a specific controlled object is determined, an electric current loop or a speed closed loop is adopted as the controlled object, and optimizing and setting of control parameters are carried out on a whole control loop; and secondly, according to the formula (please see the specification), the learning gain phi is changed, the position of the starting point of a N(z) Nyquist curve and the amplitude of the curve are adjusted, introduced offline lead compensation factors enable the N(z) Nyquist curve to achieve translation, more curves fall into a unit circle, gamma=1,2,3...n, and in the formula, q is the feedback gain, and gamma is the number of samplingperiods. According to the method, the design of a robot iterative learning controller is provided according to the characteristic that the industrial robot operates on the same track multiple times,the robot has the self correction capacity, the experience in the previous operating track process is gained to guide operating of subsequent tracks, the more the robot operates, the higher the accuracy is, following errors are reduced greatly, and the accuracy of track operating is improved.

The invention discloses an iterative learning-based subway train automatic running speed control method. The method comprises the following steps of: 1, establishing a train running dynamic model for an urban railway transit train automatic running speed control system; 2, automatically adjusting a learning gain through output errors and modified functions in an iteration process, and using the learning gain to update the input of a speed controller; and 3, in order to ensure the robustness, for the initial-state errors, of an algorithm, learning an iteration initial state while a controlled quantity is learnt to ensure that the system can restrain to an expected track under any initial condition without requiring the iteration initial state to be accurately located on an expected initial state, so as to finally realize the accurate tracking, for a target speed curve and a target displacement curve, of the train. According to the method, a learning gain initial value and a system state and tracing error of the last iteration are utilized to correct the system initial state of the current iteration, and an iteration initial state correction algorithm is given, so that the convergence, for any system initial state, of a law of learning is ensured.

An iterative learning control method for tracking consistency of a fractional-order multi-agent system comprises the following steps: a, transforming the control problem of coordinated tracking of fractional-order multi-agent systems with different orders into a stability control problem of a tracking error system in a fixed time interval; b, designing a distributed P-type iterative learning controller with initial state learning capability; and c, solving an initial state learning matrix and an iterative learning gain matrix to be determined in the iterative learning controller. The inventionsolves the problem of coordinated tracking in the case where the initial state offset and the model unknown exist simultaneously in the fractional-order multi-agent systems with different orders by using the iterative learning control method, and the proposed iterative learning controller is not only simple in design and solving, but also can resist the offset of the initial state, and can ensurethe consistency of fractional-order multi-agent systems with different orders during the whole moving process after certain number of iterations, thereby having high practicability.

The invention discloses a mechanical arm fractional order iterative learning control method and system with an initial state learning function. The mechanical arm fractional order iterative learning control method with the initial state learning function comprises the steps that firstly, a kinetic model of a mechanical arm system is established, and an expected movement track of the mechanical arm system is preset; secondly, the initial state of the state quantity of the mechanical arm system and system input of the mechanical arm system are initialized, and the actual movement track of the mechanical arm system is worked out according to the kinetic model of the mechanical arm system; thirdly, whether the tracking error between the actual movement track and the expected movement track is equal to zero or not is judged through calculation, if yes, the actual movement track overlaps with the expected movement track, and the process is ended, and if not, the next step is executed; and fourthly, the initial state of the state quantity of the mechanical arm system is corrected according to the initial state of the tracking error and a set initial state learning gain, system input of the mechanical arm system is corrected according to the tracking error, a set input learning gain, and a fractional order, the actual movement track of the mechanical arm system is worked out accordingly, and then the third step is executed.

The invention provides a system and method for efficiently measuring and reporting performance relative to established learning outcomes for students, professors, classes, courses and programs. The invention provides an evidence-based approach to measuring student achievement and increasing educator accountability. Using associations between learning content, assessment content, and learning outcomes, the invention facilitates measurement of student achievement, content effectiveness, and teaching effectiveness, through reporting of student, class, teacher, course, and institution performance relative to set outcomes in an on-line educational environment.

The invention relates to a method for controlling trajectory tracking of a mobile robot, which solves the problem of selection of gain parameters. The method comprises the following steps: (1) a state of an actual trajectory is participated in iterative operation each time on the basis of conventional proportion integration differentiation (PID) iterative learning control, so that the control effect is close to an expected value; (2) a form of a learning gain matrix in a model algorithm learning law is introduced into a learning gain matrix of a PID algorithm. The problem that the gain matrix is difficult to acquire is solved. According to the controller, the trajectory tracking of a wheeled robot can be well realized.

The invention discloses a network training system. The system comprises a student management module, a course management module, a test module, a cloud server, a teaching end, a student end and a parent end. The student management module manages personal information of students and sends the personal information of the students to the cloud server. The course management module is used for storing courses and course learning progresses of the students and sending the course learning progresses of the students to the cloud server. The cloud server pushes daily learning progresses of the students to the teaching end and the parent end, so that teachers and parents can know the learning progresses of the students in time conveniently. The test module is used for storing test questions, scoring test results of the students and sending scoring results to the cloud server. The cloud server pushes the test results to the teaching end, the student end and the parent end, so that the students can know their learning outcomes conveniently, and the teachers and the parents can know the learning effects of the students in time conveniently. Therefore, the learning of the students can be further supervised.

The invention discloses an interactive self-learning system for university English and belongs to the technical field of education and teaching. The system comprises a login module, a user management module is electrically connected with a network firewall and a registration unit in an output mode respectively, a teacher module is electrically connected with a teacher management module, a personal information management section, a student mentoring module, a communication module and a test question management module in an output mode respectively, and a manager module is electrically connected with a streaming media management module, the user management module and a safety management module in an input mode respectively. According to the interactive self-learning system for university English, according to the learning experience and characteristic data of a learner, through analysis, the personalized difference teaching service can be better provided for the learner, and the teaching achievement is improved; meanwhile, by using a testing module, the learning efficiency and learning achievement after the learner learns for a certain stage can be detected in real time, by using the streaming media technology, the learning enthusiasm of the learner is fully mobilized, and the learning efficiency is improved.

The invention discloses a method for suppressing the rotation speed fluctuation of a PMSM based on iterative learning. The method can suppress the periodic pulsating torque of the PMSM and improve servo precision. The method is a motor control algorithm that compensate for torque ripple by using a PI learning law and attenuates non-periodic agitation error accumulation by introducing a forgettingfactor. The method comprises steps of: firstly establishing a transfer function model of an iterative learning controller to analyze its stability; then determining the convergence condition of the algorithm; and designing two important parameters, namely the learning gain and the forgetting factor, in consideration of the steady-state performance and dynamic response performance of the system. The method is simple and easy to implement, and can track the desired trajectory with high accuracy by learning the periodic deviation signal.

The invention discloses an intelligent student score self-analysis method. The method includes steps: S1, recording each-time examination subjects and subject scores of a student from the beginning of school; S2, comprehensively analyzing the each-time examination subjects and the subject scores of the student in a preset time period to obtain a student score analysis result. According to the method, the examination subjects and the subject scores of the student can be analyzed in a preset time period to obtain contrast of learning interests and learning outcomes of different subjects in different time periods. The intelligent student score self-analysis method is not only beneficial to analysis of development tendency and interest preference of the student in different subjects but also beneficial to providing of comprehensive and accurate learning effect test standards for the student to make it convenient for the student to adjust own learning conditions and learning states according to the analysis result obtained according to the method, so that comprehensive and healthy development of student achievements can be guaranteed.

The invention provides a coordination control method for a system with multiple mobile robots. The coordination control method comprises the following steps of: obtaining a personnel current moment position, an input matrix, an output matrix and an experience control quantity of each mobile robot, and a current moment position of a neighbor robot; determining a first parameter of each mobile robot according to existence of predetermined expectation position information of each mobile robot; determining a learning parameter of each mobile robot according to the first parameter of each mobile robot and a weighting adjacent matrix of the system with the multiple mobile robots; determining a learning gain matrix of each individual mobile robot according to the input matrix and the output matrix of each mobile robot; and controlling each mobile robot according to the current moment position, the first parameter, the learning parameter, the experience control quantity and the learning gain matrix of each mobile robot, and the current moment position of the neighbor robot. According to the invention, the control precision is improved through adding a learning process, so that each mobile robot can complete a task in a limited time.

The invention provides a Block chain-based vocational education credit bank creating method, device and system. The vocational education credit bank creating method comprises the steps of building a unified reference system of learning achievement levels; obtaining all current learning achievements of a target person, and obtaining a learning achievement score of the target person according to theconversion weight of each current learning achievement; and based on the learning achievement score, benchmarking in the unified reference system to obtain the learning achievement grade of the target person in the unified reference system. The technical problems that in the prior art, methods and systems for integrating and benchmarking learning achievements obtained through various types of learning are lacked, so that the learning achievements obtained under different learning systems are administrated independently, barriers exist between education systems and between the education systems and a labor system, and effective butt joint between the systems cannot be achieved are solved.

The invention provides a social type language learning system. The social type language learning system comprises a plurality of user terminals, wherein the user terminals are in communication connection with one another through servers. The social type language learning system is characterized in that the social type language learning system further comprises a user registration module, a course announcement module, a course learning module, a voice quality grading module, a course evaluating module and an upgrading module. According to the social type language learning system, a cell phone or an IPAD or a computer serves as a carrier, an issuer uploads voice courses to the servers on line or uploads voice courses which are recorded in advanced to the servers, a learner searches and selects the interesting voice courses to learn in the servers on line and uploads learning achievements to the servers, the learner can know the mispronunciation on line and in real time after the issuer evaluates the learner and points out defects of the learner on line, and the learner can make right correction under the guidance of the issuer. According to the social type language learning system, social type interaction among people is achieved, a language can be learned through one-to-one communication and one-to-many communication, and the social type language learning system has great development value and use prospect.

A machine learning apparatus according to one embodiment of the present invention is a machine learning apparatus for learning a condition associated with adjustment of a current gain parameter in electrical machine control, and comprises: a state observing unit which acquires actual current as well as an integral gain function and a proportional gain function in a current control loop, and observes state variables which include the integral gain function, the proportional gain function, and at least one of an amount of overshoot, an amount of undershoot, and a rise time of the actual current occurring in response to a step-like torque command; and a learning unit which learns the condition associated with the adjustment of the current gain parameter in accordance with a training data set constructed from the state variables.

The invention discloses a video learning effect evaluation method based on artificial intelligence, and particularly relates to the field of digital learning, and the method comprises the steps: comprehensive testing before learning, wrong question arrangement, video adjustment, information recording playing, in-class testing, judgment model building, score judgment, and single-person learning result model building. According to the invention, process records of all users are collected and include data such as online visual evaluation watching time and playback times, after-learning test difficulty question answering accuracy and answering time, and the like, and the data are used as training data sources after a new teaching video is online, whether the teaching video is appropriate or not is predicted through the collected user data by taking the ability result of the user after the test as the classification basis, so that whether the video has the effect of improving the learning effect or not is evaluated.

The invention provides an express way access ramp control method and system based on switching system characteristics. The method includes acquiring a current error function according to an actual traffic flow density value and an expected traffic flow density value; setting an iterative learning control law and iterative learning gain according to the switching system characteristics according to the error function; applying the iterative learning control law based on the switching system characteristics to an express way system provided with a ramp access so as to enable the express way to reach the expected traffic flow density in a specific iteration number. Since parameters of an express way model system change along with influential factors including weather, road condition and the like in practical application, the express way system meets switching system characteristics. The method provided by the invention can solve a problem of control of the express way having the switching system characteristics and provided with the ramp access and meet practical requirements better.

The invention relates to a learning management method used in language learning machine, including the following steps: according to the condition selected and input by a user, language learning plan is set; taking the set learning plan as the standard, corresponding language contents are respectively executed to accumulate and manage learning outcome; when the learning outcome is displayed on a menu, any unexecuted language content is respectively executed according to the requirements of the user. When used in a digital machine, the invention stores language learning program of different contents in memories such as a hard disk and sets language learning plans suitable for children in a simpler way through a user interface menu; moreover, according to the language learning plans, related contents are respectively executed to ensure that children are more interested in language learning; meanwhile, learning outcome can be accumulated and managed through a learning recording part; in addition, the invention which can manage unexecuted learning plans not only increases language learning efficiency of children, but also contributes greatly to language learning for adolescents and parents.

The invention provides a learning feedforward controller and a vapor compression refrigeration control system. The learning feedforward control strategy comprises: t shows time, j shows the number of iteration loops, and j is a positive integer greater than or equal to 1; u<j>ff shows the learning feedforward term with jth iteration; u<j-1>ff shows the learning feedforward term with (j-1)th iteration; [gamma] is a learning gain, wherein 0<[gamma]<1; and u<j-1>fb shows an error feedback value in the learning feedforward term corresponding to the error between the output of the controlled system with the (j-1)th iteration and an expected output. The learning feedforward controller and the vapor compression refrigeration control system and method can comprehensively improve the performance and the robust stability of the feedforward controller, and the feedforward controller can also be applied to the vapor compression refrigeration control system to improve the real-time performance of the temperature control of the refrigeration system.

The invention relates to an ultrasonic motor speed control method and device based on variable gain iterative learning. that rotational speed of an ultrasonic motor based on improve P-type iterative learning controller or P-type iterative learning control is improved, The learning gain varies according to the actual speed of the ultrasonic motor. The larger the actual speed of the ultrasonic motor, the smaller the learning gain, and the smaller the actual speed of the ultrasonic motor, the greater the learning gain. The invention improves the convergence speed, but also prevents the response curve from oscillating, and effectively improves the ultrasonic iterative learning control performance.

The present invention disclose a learning test system. The system is used for students to perform learning and test through games, and mainly comprises a database system and an operation processing unit. The database system at least comprises a test game database, a test question bank database and a test record database. The operation processing unit at least comprises a grading and classification module, a game test module and a recording module. Corresponding test games and test question bank sets can be led out of the database system through the students' operation to allow the students to perform test in the game process, perform a lot of exercise through an interesting game mode and record each answer of the students to perform statistics of learning accomplishment of each part and perform diagnosis of weak directions to strengthen the review so as to reach the optimum learning effect.

The invention provides an iterative learning refrigeration control system and method based on fractional order. The iterative learning refrigeration control system based on the fractional order comprises a fractional order PID controller which is used for obtaining the feedback control quantity of the current operation period according to the refrigeration effect deviation of the refrigeration system of the current operation period and compensating the feedback control quantity to the feedforward control quantity to achieve fractional order PID feedback control over the refrigeration system; and a fractional order iterative learning controller which is used for estimating the feedforward control quantity of the next operation period according to the fractional order integral superpositionsum of the feedforward control quantity of the current operation period and the refrigeration effect deviation of the refrigeration system under the preset learning gain multiple so as to realize fractional order iterative learning control of the refrigeration system until the preset refrigeration effect is achieved, and the iterative control is ended.

A system for mitigation of a torque steer includes a base gain generation module that generates a base gain based on a transmission torque, a learning module that generates a learned gain based on the transmission torque and the estimated pinion torque, and a command generation module that generates a motor torque command from the learned gain, the transmission torque, and the base gain.

The invention discloses a police education and training system with learning centers. The system is provided with the learning centers and a server in remote communication connection with the learning centers, wherein each learning center is provided with a video playing unit for playing teaching and examination contents to a trainee and a brain wave area scanning unit for acquiring a brain area reflection signal when the trainee watches the teaching and examination contents and transmitting the reflection signal to the server; the server is used for comparing the received reflection signals with correct reflection signals corresponding to the currently played teaching and examination contents in a database to determine learning outcomes of the trainees. According to the police education and training system with the learning centers, brain wave detection devices for detecting excited areas of brains are arranged, so that whether the trainees correctly receive teaching contents or not can be directly and effectively judged, and necessary graduation tests at the end of a conventional teaching process are avoided; the scheme is novel and reliable, so that the system is suitable for large-scale popularization and application.

The invention discloses an online education platform student learning ability analysis feedback method based on user learning behavior deep analysis, and the method comprises the steps: obtaining the basic parameters of an education course on an online education platform; counting the mental state coefficient, the learning correlation coefficient, the learning interaction coefficient, the learning concentration coefficient and the learning result excellence coefficient of each student in the learning duration under each education course, calculating the comprehensive learning power coefficient of the student according to the counting result, and feeding back the comprehensive learning power coefficient to the student and the course teacher. The method achieves the comprehensive reliability evaluation of the online learning ability of the students according to the learning behaviors of the students, makes up for the steps that the evaluation indexes are too single and the evaluation results are too one-sided in the evaluation mode of the online learning ability of the students through an online education platform at present, perfects the evaluation index range, improves the reliability of the evaluation results, and meets the evaluation requirement of high comprehensive reliability of the online learning ability of the students at present.

The invention provides a speech speed-based course matching method, system and device, and a storage medium. The online education method comprises the steps of collecting preview audio of each studentin each to-be-allocated student group for reading a current course text before class; performing voice recognition on the preview audio to obtain a preview text, and obtaining a first parameter basedon a course text proficiency degree of the student before class according to a first voice speed and a first accuracy rate in the preview audio; according to the historical teaching video of each online teacher to be distributed, obtaining a second speech speed of the online teacher; sorting according to the first parameter average values of the to-be-distributed student groups and then matchingwith the online teachers, wherein the to-be-distributed student group with the lowest first parameter average value is preferentially matched with the teacher with the slowest second speech speed in the current to-be-distributed online teachers. According to the invention, teachers are flexibly distributed according to the student groups, the distribution scheme of the student groups and the teachers is optimized, the student learning achievements and the student user experience are improved, and the overall teaching quality of network courses is improved.