Multiple Identification Models Research Articles

Robust adaptive safety control of uncertain nonlinear systems and its application to differential robots

AbstractThis article focuses on the problem of robust adaptive safety control of a class of nonlinear systems with parametric uncertainties. Two novel definitions, that is, robust adaptive control Lyapunov function (RACLF) and robust adaptive control barrier function (RACBF) are introduced in this article. Also, an estimation of unknown parameters is obtained based on multiple identification models, state predictors and a logic switching mechanism. The switching mechanism plays a vital role in enhancing the transient properties of nonlinear systems. Meanwhile, a safe motion framework for differential robots is provided based on the existence of parametric uncertainties in RACLF and RACBF, and a safe and stable controller is obtained by solving quadratic programming. Finally, simulation results of differential robot capable of safe motion in space are presented to verify the validity of the design method.

Elasto-geometrical calibration of six-DOF serial robots using multiple identification models

Here, an elasto-geometrical calibration method for joint-dependent geometric errors and compliance parameters of six-DOF (degree of freedom) serial robot (six joints are all revolute where the second and third joints are parallel) based on multiple identification models is presented for a complete and accurate calibration. Simulations indicate that the unknown parameters from various sources have sensitivities with different scales. If the parameters are involved in a single model, then the identification Jacobian might become ill-conditioned, and identifiability is degraded. According to a set of evaluations on the observability index and condition number, excellent identifiability is achieved when the unknowns are lumped into different groups according to their sensitivities. Thus, the groups with higher sensitivities are identified in the submodels derived from the generally used single model, whereas the insensitive ones are measured directly using non-model-based methods. The use of different sets of measurement configurations is necessary for various parameter groups. The simulation results present more stable submodels, as well as the improved identifiability. Two equivalent strategies that implement the multiple identification models are designed. Compared with the residuals after using general methods, the results from the multiple identification models confirm the improvement of identification accuracy.

Identification Technology of Grid Monitoring Alarm Event Based on Natural Language Processing and Deep Learning in China

Power dispatching systems currently receive massive, complicated, and irregular monitoring alarms during their operation, which prevents the controllers from making accurate judgments on the alarm events that occur within a short period of time. In view of the current situation with the low efficiency of monitoring alarm information, this paper proposes a method based on natural language processing (NLP) and a hybrid model that combines long short-term memory (LSTM) and convolutional neural network (CNN) for the identification of grid monitoring alarm events. Firstly, the characteristics of the alarm information text were analyzed and induced and then preprocessed. Then, the monitoring alarm information was vectorized based on the Word2vec model. Finally, a monitoring alarm event identification model based on a combination of LSTM and CNN was established for the characteristics of the alarm information. The feasibility and effectiveness of the method in this paper were verified by comparison with multiple identification models.

Noise-robust speech triage.

A method is presented in which conventional speech algorithms are applied, with no modifications, to improve their performance in extremely noisy environments. It has been demonstrated that, for eigen-channel algorithms, pre-training multiple speaker identification (SID) models at a lattice of signal-to-noise-ratio (SNR) levels and then performing SID using the appropriate SNR dependent model was successful in mitigating noise at all SNR levels. In those tests, it was found that SID performance was optimized when the SNR of the testing and training data were close or identical. In this current effort multiple i-vector algorithms were used, greatly improving both processing throughput and equal error rate classification accuracy. Using identical approaches in the same noisy environment, performance of SID, language identification, gender identification, and diarization were significantly improved. A critical factor in this improvement is speech activity detection (SAD) that performs reliably in extremely noisy environments, where the speech itself is barely audible. To optimize SAD operation at all SNR levels, two algorithms were employed. The first maximized detection probability at low levels (-10 dB ≤ SNR < +10 dB) using just the voiced speech envelope, and the second exploited features extracted from the original speech to improve overall accuracy at higher quality levels (SNR ≥ +10 dB).

Multiple model adaptive control using second level adaptation for a class of nonlinear systems with linear parameterizations

In this research work, an adaptive controller using multiple models is proposed for a class of nonlinear systems with uncertain parameters. A second level adaptation scheme is introduced to combine all the models efficiently, unlike old technique of switching between models. The uncertain parameters are assumed to belong to a closed and bounded region. Multiple nonlinear identification models are evenly distributed in the region of uncertainty and the parameters of these identification models are adaptively tuned. The proposed method is applied to nonlinear plants with linear parameterization because of the high usecase for this class of systems. Closed loop stability and tracking error convergence of the overall system with second level adaptation is guaranteed using Lyapunov stability analysis. Simulation of a single link robotic manipulator and one theoretical example, shows superior performance of the proposed technique over some existing methods of adaptive control. The commonly known issues with adaptive control systems like oscillatory transient response, poor parameter convergence and sluggish performance are found to be improved considerably by using this proposed multiple model based adaptive controller.

Variable Elimination-Based Contribution for Accurate Fault Identification

We propose a new fault identification method, which can describe the contribution of each process variable to a detected fault and identify a faulty variable more accurately than conventional methods. In the proposed method, in addition to a fault detection model that describes normal operating condition (NOC), multiple fault identification models that describe the same NOC are also constructed by eliminating one variable from all monitored variables at a time. After a fault is detected with the fault detection model, the fault detection index, e.g. a combined index of the T2 and Q statistics, is calculated by using each of the fault identification models. When the faulty variable is eliminated, the index does not change before and after the fault occurs. On the other hand, when the normal variable is eliminated, the index is affected by the fault and increases after the fault occurs. Thus, the eliminated variable corresponding to the index that does not increase after the occurrence of the fault is identified as a faulty variable. In the proposed method, the ratio of the average index in NOC to the current index is used as a fault identification index or a contribution. To validate the proposed method, VEC was compared with the reconstruction-based contribution (RBC) through numerical examples. The results have demonstrated that VEC outperformed RBC in fault identification performance both in the linear case and in the nonlinear case.

A New Approach to Adaptive Control of Multi-Input Multi-Output Systems Using Multiple Models

This paper proposes a novel method as second level adaptation using multiple models to identify and control of a class of multi-input multi-output (MIMO) systems. Different uncertain environments change the system parameters and create multiple operating conditions. These conditions are designed as multiple identification models in a model bank using adaptive laws. These models are evaluated using some estimated weighting factors based on the errors between each of the models and the actual plant. The evaluated models are effectively used in identification and control process. Bounded signals, proper closed-loop tracking performance, and rapid and accurate parameter convergence to their real values are achieved through simulation results.

Dynamic adaptive trajectory tracking control of nonholonomic mobile robots using multiple models approach

This paper presents an approach for the trajectory tracking control of nonholonomic wheeled mobile robots (WMR) by combining one of the existing adaptive control methods and multiple identification models. The overall system includes two types of controllers in the control scheme. A kinematic controller developed by using kinematic model produces the required linear and angular velocities of the robot for tracking a reference trajectory. These required velocities are used to calculate the torques using an adaptive dynamic controller with multiple models. The proposed method uses the multiple models of the WMR for the identification of the dynamic parameters and performs switching between the given models. The models used in the identification are identical, except for the initial estimates of the parameters. By using an adaptive dynamic controller with multiple models of the WMR, enhancement in transient response is obtained. Stability analysis of the overall system is given, and simulation results are presented to demonstrate the effective performance of the adaptive control by using multiple models approach.

Dynamic Feedback Controller Based on Optimized Switching of Multiple Identification Models for Course Control of Unmanned Surface Vehicle

A dynamic feedback control algorithm based on multiple identification models switching is proposed to solve the problems in course control of the unmanned surface vehicle(USV),such as large external disturbances and unknown course control model parameters.Firstly,a transitional model set is constructed by identifying USV course model according to the least square method.Then,a temporal model set is selected from the transitional models by introducing mean fitting error,which can avoid huge computation cost caused by large amount of sub-models in model set.Finally,a dynamic feedback controller database is designed based on the temporal model set.Meanwhile,using control performance indexes as event driven factors,some switching methods for multiple identification models are taken to obtain optimal controllers from the controller database.Some lake trials show that the multi-model switching based dynamic feedback method can improve the control performances,and the USV can move without overshoots or static errors.

Fuzzy indirect adaptive control using SVM-based multiple models for a class of nonlinear systems

Adaptive control with multiple models can further improve the adaptation ability of controllers for the plant with wide-range uncertain parameters. Fuzzy modeling and control are introduced into the multiple-model adaptive control in this paper, which facilitates the intelligent behavior of a plant facing with uncertainty. Within the combination of fuzzy sets of state variables, the corresponding combined kernel functions of support vector machine are utilized to describe the unknown nonlinear dynamics. The coefficients of kernel functions are learned online through adaptive laws. The multiple identification models and indirect adaptive controllers are assigned to the plant through fuzzy inferences. The stability of adaptive law corresponding to the fuzzy identification model and the synthetic control input through fuzzy fusion has been proved for the proposed fuzzy multiple-model adaptive control (FMMAC). The simulation results demonstrate that the proposed FMMAC can achieve favorable control performance for a class of nonlinear systems.

A new approach to adaptive control using multiple models

SUMMARYThe demands of a rapidly growing technology for faster and more accurate controllers have always had a strong influence on advances in automatic control theory. In recent years, problems are arising in a variety of disciplines including medicine, computer vision, and financial markets where fast and accurate decisions have to be made in the presence of large uncertainties or rapidly time‐varying parameters. Classical adaptive control methods based on the use of a single identification model are found to be inadequate to cope with such problems. During the past 15years, efforts have been made to extend the general methodology of adaptive control by the use of multiple identification models. Among numerous methods that have been proposed, two approaches, referred to as “switching” and “switching and tuning”, have emerged over the years as the most successful ones. In this paper, a radically new way of using multiple models for the identification and control of an unknown linear time‐invariant plant is proposed. It uses the information generated by a finite number of conventional adaptive identifiers (referred to as first level) to re‐parameterize and identify rapidly the unknown plant (at a second level). The stability of the procedure, and the reasons for its resulting in faster convergence in the control of time‐invariant plants, are discussed and illustrated using simulation studies. The accepted philosophy among adaptive control theorists has been that if an adaptive system is fast and accurate in a time‐invariant environment, it will perform satisfactorily in a time‐varying environment. Simulation studies are included, at the end of the paper, of plants with rapidly time‐varying parameters, to compare the performance of the new approach with currently well‐established methods. Copyright © 2012 John Wiley &amp; Sons, Ltd.

Adaptive Control Using Collective Information Obtained from Multiple Models

AbstractA radically new approach to adaptive control using multiple models was recently proposed by the authors. Using parametric estimates and outputs generated by multiple adaptive identification models, an adaptive control procedure was proposed which resulted in significantly faster and more accurate response of the overall system, as compared to existing methods such as “switching” and “switching and tuning”, which also use multiple models. In this paper the ideas are extended to include fixed identification models. Further, many of the theoretical questions that have arisen concerning the stability of the overall system and the reasons for the improved performance are addressed in detail. Simulation results are included, towards the end of the paper, to indicate the effectiveness of the new methodology.

Indirect adaptive control of non-linear systems using multiple identification models and switching

Transient performance improvement of adaptive control of a class of single-input single-output (SISO) non-linear systems is considered in this study. The system under study is assumed to be minimum-phase and input-output linearisable. The non-linear dynamics is also assumed to be linearly parameterised in terms of the unknown parameters. The improvement in the transient performance under large parametric uncertainties is obtained with the use of multiple identification models and switching, and the closed loop system is shown to be stable with the switching mechanism. A new methodology is proposed for the quantitative evaluation of the transient performance. The study is verified by simulation of a non-linear physical system.

Nonlinear adaptive control using multiple identification models

A multiple model adaptive controller is proposed for nonlinear systems in parametric-strict-feedback form. By running in parallel multiple identification models and designing a suitable switching scheme, some models close to the real plant can be selected quickly, so that transient performance can be improved significantly. Global asymptotic stability of the closed-loop switching system is proved. A simulation example is given to demonstrate the effectiveness of the proposed multiple model adaptive controller.

Transient performance enhancement of direct adaptive control of nonlinear systems using multiple models and switching

Adaptive control of nonlinear systems with parametric uncertainties is considered. A novel approach is presented which makes use of multiple identification models and switching based on direct adaptive control scheme. The use of multiple fixed models and switching provides better transient response under sudden changes in the parameter values. The parameter estimates are reset to the fixed values whenever one of these fixed models yields smaller identification error. The restrictive matching conditions are not required, but the nonlinearities are assumed to be Lipschitz. This work is the generalisation of the methodology proposed by Ciliz and Cezayirli.

多重モデルとモデル選定機構をもつ適応制御系とその応用

An adaptive control methodology using multiple identification models is presented. In the proposed method, the real process is identified simultaneously by multiple models with different structure and an adaptive control is carried out based on the optimum model selected by some rules. The selection rules of the suitable model for an operating condition are decided by theoretical and heuristic knowledge about the parameter estimation and adaptive control. This method requires few a priori information about the process and can deal with the abrupt change of the process characteristics. In this paper, first, the principle of the adaptive control system with multiple models is presented. Next, the main parts of the proposd system: model following control loop, recursive identifier with multiple models, model selecting function block and higher level adaptation mechanism are discussed. In order to investigate the effectiveness of the proposed method, computer simulations are carried out for processes with time-varying characteristics. Finally, the proposed method can be successfully applied to the temperature control of an incinerator.

Adaptive Control of an Incinerator Using Multiple Models

A practical discrete-time adaptive contr ol system using multiple identification models is presented . This method is applicable to a real process which has variable characteristics depending on the operating conditions. In the proposed control system, the real process is identified simultaneously by several models with different structure and an adaptive control is carried out using a controller designed by an optimum model sel ected by some rules.In this paper, first, a short description of the controlled process (incinerator) and the control objective is given . Next, an adapt ive controller with multiple identification models and higher level adaptation mechanism is designed. Finally, on-line experiments are carrird out for a real process . It is shown that the proposed method can be successfully applied to the temperature control of an incinerator.