Adaptive Estimation Technique Research Articles

Identification for nonlinear systems modelled by deep long short-term memory networks based Wiener model

This paper is concerned with modeling and identification methodology for practical nonlinear system via deep long short-term memory (DLSTM) networks-based Wiener model. To determine the unknown parameters and simplify parameters identification procedure for the Wiener model, a two-step identification scheme is implemented applying hybrid signals involving separable signal and random data. First, the separable signal to separate the two blocks in Wiener model is imported, then parameters of the linear block is estimated using correlation function-based least squares technique, which handles the issue that the intermediate variable information in Wiener model cannot be measured. Moreover, integrating the advantages of adaptive stochastic gradient descent algorithm and root mean square propagation, an adaptive momentum estimation technique is created to optimize the DLSTM networks parameters based on available random data, which improves the accuracy of identified Wiener model. Compared with the other existing schemes, the superiority of the proposed method in terms of predictive performance and control results are illustrated by permanent magnet synchronous motors.

An analytical approach for unsupervised learning rate estimation using rectified linear units.

Unsupervised learning based on restricted Boltzmann machine or autoencoders has become an important research domain in the area of neural networks. In this paper mathematical expressions to adaptive learning step calculation for RBM with ReLU transfer function are proposed. As a result, we can automatically estimate the step size that minimizes the loss function of the neural network and correspondingly update the learning step in every iteration. We give a theoretical justification for the proposed adaptive learning rate approach, which is based on the steepest descent method. The proposed technique for adaptive learning rate estimation is compared with the existing constant step and Adam methods in terms of generalization ability and loss function. We demonstrate that the proposed approach provides better performance.

Three‐dimensional path following control of underactuated autonomous underwater vehicles with nonzero roll dynamics: A novel line‐of‐sight‐guided approach

AbstractMost existing path following control approaches for underactuated Autonomous underwater vehicles (AUVs) assumed that the roll angle can be passively stabilized at zero, and simplified the problem to the 5‐degree‐of‐freedom model. However, in practice, the roll motion suffers from undesirable oscillations induced by environmental disturbances, unbalanced propeller torque, etc. To preserve the stability of the system under nonzero roll dynamics, this paper proposes a novel line‐of‐sight‐ (LOS) guided path following control approach. Firstly, the spatial path following error dynamics model is derived to comprehensively reveal the negative effect of roll dynamics. Secondly, an improved LOS guidance law is designed to compensate for the nonzero roll angle. A salient feature is that the proposed guidance law can be applied to the underwater vehicle no matter it is underactuated in the roll motion or not. Thirdly, by assigning the guidance angles to the desired attitude angles, a robust attitude tracking controller is developed, in which the adaptive estimation technique is employed to handle the lumped uncertainties. The tanh‐based saturation function is incorporated such that the resultant control signals are inherently bounded. The closed‐loop path following and attitude tracking errors are proved to be globally asymptotically stable at the origin. Comparative simulation results are provided to substantiate the effectiveness and superiority of the proposed method.

Application of empirical Bayes adaptive estimation technique for estimating winds from MST radar covering higher altitudes

Application of empirical Bayes adaptive estimation technique for estimating winds from MST radar covering higher altitudes

Adaptive Estimation of Quasi-Empirical Proton Exchange Membrane Fuel Cell Models Based on Coot Bird Optimizer and Data Accumulation

The ambitious spread of fuel cell usage is facing the aging problem, which has a significant impact on the cells’ output power. Therefore, it is necessary to develop reliable techniques that are capable of accurately characterizing the cell throughout its life. This paper proposes an adaptive parameter estimation technique to develop a robust proton exchange membrane fuel cell (PEMFC) model over its lifespan. This is useful for accurate monitoring, analysis, design, and control of the PEMFC and increasing its life. For this purpose, fair comparisons of nine recent optimization algorithms were made by implementing them for a typical quasi-empirical PEMFC model estimation problem. Investigating the best competitors relied on two conceptual factors, the solution accuracy and computational burden (as a novel assessment factor in this study). The computational burden plays a great role in accelerating the model parameters’ update process. The proposed techniques were applied to five commercial PEMFCs. Moreover, a necessary statistical analysis of the results was performed to make a solid comparison with the competitors. Among them, the proposed coot-bird-algorithm (CBO)-based technique achieved a superior and balanced performance. It surpassed the closest competitors by a difference of 16.01% and 62.53% in the accuracy and computational speed, respectively.

Composite adaptive online inverse optimal control approach to human behavior learning

There have been broad ranges of human-in-the-loop (HiTL) control systems, like fly-by-wire aircraft, share driving, and energy management. To develop advanced HiTL control systems with augmented intelligence, it is desirable that the machine is empowered to understand how a skilled human operator determines the control actions. It is common to model the human operator as an optimal controller with an unknown weighting matrix which depicts the tradeoff between multiple control objectives. Our aim is thus to learn the weighting matrix of the human objective function using only the system state measurement. Accordingly, for the HiTL system, a composite adaptive inverse optimal control (IOC) approach, which integrates concurrent learning (CL) based composite adaptive estimation and linear matrix inequality (LMI) optimization techniques, is proposed to learn human behavior online. The proposed composite adaptive IOC method consists of two steps: first, a composite adaptive update law is proposed to online estimate the human feedback gain matrix, which removes the persistent excitation condition required by traditional adaptive estimation methods, and then, an LMI optimization problem is solved to retrieve the weighting matrix of the human objective function with the learned feedback gain. The applicability and effectiveness of the proposed method are demonstrated with a steering control simulation.

Variational problem of adaptive optimal control. Theoretical and applied computer analysis

The problem of adaptive optimal control of a dynamical system, which belongs to the class of conditional variational problems with moving boundaries, is considered. A variational and computer study of the controlled adaptive motion of a material point is carried out for the problem of the energy quality functional minimizing with a moving, not predetermined right transboundary and in the case when the mass of the point changes depending on the unfixed final time. The problem is solved using the schemes and procedures of the classical calculus of variations, as well as adaptive estimation techniques, including the derivation of the variation of the auxiliary quality functional, the corresponding Euler equations, and the adaptive estimation algorithm. When solving a general conditional variational problem, the obtained closed system of differential equations was studied for the formation of an adaptive optimal control system for a dynamic plant with a given performance functional. The results of the unconditional formulation of the problem are generalized to the case of additional differential (nonholonomic) and holonomic constraints. In a variational adaptive optimal control problem, the transversality condition is formulated in terms of the local programming condition. The developed variational scheme of adaptive optimal synthesis can be used in the calculation and design of controlled dynamic systems. This optimization scheme is also promising for use in systems where operating time is non-fixed in advance. The results achieved in this paper concern obtaining specific equations, expressions, and formulas relative to the model example under study and finding graphs of the main time functions that determine the nature of the movement of the control object and the quality of the corresponding transients. The proposed adaptive optimal control algorithms for purposeful movement of the studied material point were tested in digital mode and showed their effectiveness which makes them promising for further use in more complex nonlinear adaptive systems of dynamic optimal control.

Adaptive Leader-Follower Consensus Control of Multiple Flexible Manipulators With Actuator Failures and Parameter Uncertainties

In this paper, the leader-follower consensus problem for a multiple flexible manipulator network with actuator failures, parameter uncertainties, and unknown time-varying boundary disturbances is addressed. The purpose of this study is to develop distributed controllers utilizing local interactive protocols that not only suppress the vibration of each flexible manipulator but also achieve consensus on joint angle position between actual followers and the virtual leader. Following the accomplishment of the reconstruction of the fault terms and parameter uncertainties, the adaptive neural network method and parameter estimation technique are employed to compensate for unknown items and bounded disturbances. Furthermore, the Lyapunov stability theory is used to demonstrate that followers' angle consensus errors and vibration deflections in closed-loop systems are uniformly ultimately bounded. Finally, the numerical simulation results confirm the efficacy of the proposed controllers.

Distributed synchronization control for multiple tailless flying‐wing UAVs with predefined tracking accuracy and input nonlinearities

SummaryThis article addresses the distributed adaptive synchronization tracking control problem for multiple flying‐wing UAVs with predefined accuracy and unknown input nonlinearities. Considering that the multiple flying‐wing unmanned aerial vehicles (UAVs) system is a large‐scale nonlinear and strongly coupled system, the control schemes in existing literature which only consider longitudinal, lateral or attitude control may not be applicable in practical engineering. Thus, a full‐state flying‐wing UAV model containing both translational and rotational motions is established, whose dynamics are six‐degree‐of‐freedom with twelve‐state‐variables. Furthermore, to handle the problem that the prior knowledge of the actuator nonlinearities of multi‐UAV system are difficult to obtain, a more general input nonlinear model is established and the adaptive boundary estimation technique is developed. Finally, a performance‐oriented controller is proposed by incorporating a series of smooth functions into adaptive neural control design and Lyapunov analysis. Theoretical analysis and simulation results show that the presented control scheme can guarantee the satisfactory transient tracking error performance, and the tracking error converges to a user‐defined interval ultimately.

Towards More Efficient Rényi Entropy Estimation

Estimation of Rényi entropy is of fundamental importance to many applications in cryptography, statistical inference, and machine learning. This paper aims to improve the existing estimators with regard to: (a) the sample size, (b) the estimator adaptiveness, and (c) the simplicity of the analyses. The contribution is a novel analysis of the generalized "birthday paradox" collision estimator. The analysis is simpler than in prior works, gives clear formulas, and strengthens existing bounds. The improved bounds are used to develop an adaptive estimation technique that outperforms previous methods, particularly in regimes of low or moderate entropy. Last but not least, to demonstrate that the developed techniques are of broader interest, a number of applications concerning theoretical and practical properties of "birthday estimators" are discussed.

An optimization for adaptive multi-filter estimation in medical images and EEG based signal denoising

Classical denoising techniques are efficient to extinguish the Gaussian noise but are unable to handle the impulse and additive noise. The blurring of edges in denoised data is critical in both images and signals. It has been found that edge blurring occurs due to limited spatial information. Adaptive multi-filtering methods can accommodate large spatial information. But the estimation of filter window sizes and their subsequent information fusion is a challenging task. To overcome this issue, an adaptive multi-filter estimation technique that fuses large information has been explored. The fusion function is formulated for multiple sizes of the filter as an objective function. The objective function takes the spatial information from the neighborhood of noisy data using multiple filters and fuses them while minimizing the objective function. The objective function is optimized for various sizes of filter windows with a stochastic flower pollination Algorithm (FPA). Impulse noise is efficiently removed using optimized filters. An empirical study is conducted with recent state-of-the-art using evaluation indices such as MSE, PSNR, MAXERR, L2RAT, SSIM, computational time, and complexity. For the proposed setup, the MSE, PSNR, MAXERR, L2RAT, and SSIM are evaluated as 1.33, 26.90, 201, 0.9964, 0.8922, and 0.9312 respectively that are outperforming the results obtained in the previous state-of-the-art denoising algorithms such as wavelet, compression, box averaging, tabu search and simulated annealing. Further, it is interesting to observe the improvement in complexity. The experimental study demonstrated the performance of the framework with previous qualitative indices in both Alzheimer’s images and EEG-based neural signal data.

Fully Distributed Optimal Position Control of Networked Uncertain Euler-Lagrange Systems Under Unbalanced Digraphs.

The distributed optimal position control problem, which aims to cooperatively drive the networked uncertain nonlinear Euler-Lagrange (EL) systems to an optimal position that minimizes a global cost function, is investigated in this article. In the case without constraints for the positions, a fully distributed optimal position control protocol is first presented by applying adaptive parameter estimation and gain tuning techniques. As the environmental constraints for the positions are considered, we further provide an enhanced optimal control scheme by applying the ϵ -exact penalty function method. Different from the existing optimal control schemes of networked EL systems, the proposed adaptive control schemes have two merits. First, they are fully distributed in the sense without requiring any global information. Second, the control schemes are designed under the general unbalanced directed communication graphs. The simulations are performed to verify the obtained results.

Online Learning Human Behavior for a Class of Human-in-the-Loop Systems via Adaptive Inverse Optimal Control

To enhance the machines’ intelligence, it is important for them to learn how humans perform tasks. In this article, the issue of online adaptive learning human behavior is addressed for a class of human-in-the-loop (HiTL) systems using the state measurement only. The hypothesis underlying our study is that human behavior can be described by a linear quadratic optimal control model with an unknown weighting matrix for the quadratic cost function. In this model, the weighting matrix depicts the human tradeoff of various objectives. Our aim is thus to only use the system state measurement for learning the weighting matrix under the condition that human feedback gain matrix is unknown. A novel adaptive inverse optimal control approach to online learning human behavior is proposed for the HiTL system, which integrates adaptive estimation and linear matrix inequality (LMI) optimization techniques. Our approach consists of two steps: First, an adaptive law is developed to learn the human feedback gain matrix online using the system state measurement only, and second, the weighting matrix of human cost function is retrieved by solving an LMI optimization problem with the learned feedback gain matrix. Finally, simulation and experiment results on a steering assist system of intelligent vehicles are presented to illustrate the effectiveness of the proposed method.

Adaptive estimation of PEMFC stack model parameters - An experimental verification

The growing popularity of using proton-exchange membrane fuel cells (PEMFCs) stacks in stationary, portable, and transportation applications is driving researchers to develop proper dynamic models for PEMFCs. These models are used to accurately capture the electrical characteristics and runtime performance. This work proposes a well-known equivalent circuit model of a battery, to be modified and used as a model for a PEMFC stacks voltage-current characteristics. This model is modified by finding suitable functions to model the open circuit voltage and the series resistance, required to model the electrical performance of a 200-W PEMFC stack. The paper also shows that the existing adaptive parameters estimation (APE) technique for Li-ion battery parameters estimation is also able to estimate parameters of the PEMFC stack's model. The model parameters are estimated using the APE technique that requires only five experiments. The model is validated experimentally under different load conditions for a 200-W PEMFC stack supplied from a hydrogen cylinder (voltage error −0.2 V to 0.5 V), and a 30-W PEMFC stack supplied from a fuel stick (voltage error −0.2 V–0.4 V). The results show that the parameters estimation methodology works well across PEMFC stacks of different sizes with different input fuel intake configurations, with a minimal terminal voltage estimation error in the order of millivolts. Open circuit voltage measurements (OCV) show that the OCV curve starts at a little lower than 31 V, declines slowly to around 30 V for a normalized hydrogen flow rate of 0.6, after which there is a sudden linear decline in OCV was observed. Most of the data has absolute estimation error less than 0.1 V. In fact, the terminal voltage estimation error across all tests, with different current discharge profiles, lies between −0.2 and 0.2 V only. Also, 95.84% of the error samples lie between ±0.1% error.

Adaptive state-constrained trajectory tracking control of unmanned surface vessel with actuator saturation based on RBFNN and tan-type barrier Lyapunov function

To promote the vigorous development of the marine technology, more and more scientific research institutes have begun to develop unmanned surface vehicles (USVs) with advanced control performance. This paper addresses trajectory tracking problem for state-constrained USV under the influence of actuator saturation and general disturbance uncertainties. An anti-windup system with auxiliary variables is employed to cope with actuator saturation. Besides, full-state constrains are not violated by combining the tan-type barrier Lyapunov function and backstepping method. Moreover, RBFNN is adopted to approximate some unknown uncertain function terms. By utilizing adaptive estimation technique, the unknown upper boundaries of disturbance uncertainties, input saturation difference, as well as RBFNN approximation error are compensated. The semi-global asymptotic convergence is guaranteed by rigorous theoretical analysis and Lyapunov proof. Finally, simulations results are given to verify the feasibility of the proposed algorithm.

A Distributed Fault Diagnosis and Cooperative Fault-Tolerant Control Design Framework for Distributed Interconnected Systems

This paper investigates a design framework for a class of distributed interconnected systems, where a fault diagnosis scheme and a cooperative fault-tolerant control scheme are included. First of all, fault detection observers are designed for the interconnected subsystems, and the detection results will be spread to all subsystems in the form of a broadcast. Then, to locate the faulty subsystem accurately, fault isolation observers are further designed for the alarming subsystems in turn with the aid of an adaptive fault estimation technique. Based on this, the fault estimation information is used to compensate for the residuals, and then isolation decision logic is conducted. Moreover, the cooperative fault-tolerant control unit, where state feedback and cooperative compensation are both utilized, is introduced to ensure the stability of the whole system. Finally, the simulation of intelligent unmanned vehicle platooning is adopted to demonstrate the applicability and effectiveness of the proposed design framework.

Modelling and forecasting based on recursive incomplete pseudoinverse matrices

Time series modelling has a wide spectrum of applications in several fields including engineering and finance. Most traditional modelling techniques rely on assumptions related to the input data and manual pre-processing, based on user observations, rendering them unsuitable for analysing time-series with varying characteristics automatically, while more general modelling techniques usually require increased computational work for application and tuning. Recently, a general modelling framework based on a recursive Schur complement technique, that utilizes an adaptively determined set of basis functions, has been proposed. Herewith, a novel modified approach based on recursive incomplete pseudoinverse matrices in conjunction with preconditioned iterative methods for large datasets, is proposed. This sparse approach greatly reduces storage requirements and the recursive nature of the procedure avoids recomputation of the preconditioner after addition of a new basis function. Moreover, update of the coefficients for a different window of data and predefined basis functions can be performed utilizing the incomplete pseudoinverse matrix as preconditioner. The case of sinusoidal basis functions is presented along with a novel adaptive frequency estimation technique. The stability of the resulting model is discussed with respect to the choice of basis functions. The case of basis derived from machine learning techniques is also discussed. Numerical results are given depicting the applicability, generality and effectiveness of the proposed technique. Comparative results with other methods show forecasting RMSE improvement between 7% to 80%, for the majority of the chosen time series.

Adaptive Attitude Control of a Quadrotor Using Fast Nonsingular Terminal Sliding Mode

As one type of unmanned aerial vehicles, the quadrotor typically suffers from payload variations, system uncertainties, and environmental wind disturbances, which significantly deteriorate its attitude control performance. To provide high-speed, accurate, and robust attitude tracking performance for the quadrotor, an adaptive fast nonsingular terminal sliding mode (AFNTSM) controller is proposed in this article. The proposed AFNTSM controller combines the advantages of fast nonsingular terminal sliding mode (FNTSM), integral sliding mode, and adaptive estimation techniques, which are effective to achieve the desired tracking performance and suppress control signal chattering. Furthermore, unlike conventional methods, the adaptive estimation removes the requirements for the upper bound information of the disturbances. It is proved that the proposed AFNTSM can guarantee finite-time convergence and zero tracking error for the quadrotor attitude control. Finally, comparative study with the FNTSM control only and conventional sliding mode control is conducted through experiments and the results demonstrate that the proposed AFNTSM can achieve faster convergence and stronger robustness in line with theoretical analysis.

Adaptive Integral Correction-Based State of Charge Estimation Strategy for Lithium-Ion Cells

Lithium-ion (Li-ion) battery systems are critical elements of future energy systems and electric vehicles. Accurate prediction of the state of charge (SoC) is necessary for the safe and reliable functioning of Li-ion battery systems. Achieving a precise SOC estimate is challenging due to the nonlinear characteristics and variations of model parameters caused over the cell lifetime. This paper introduces an adaptive estimation strategy that can compensate for the effect of cell degradation for achieving high accuracy SoC estimation. The proposed method uses an integral correction-based SoC estimation loop utilizing a Li-ion cell model. The effect of model parameter variation is corrected by introducing two additional correction factors, the cell model resistance, and capacity correction factor. These correction factors are employed to update the Li-ion cell model, resulting in an adaptive integral correction-based SoC estimation technique that can compensate for the influence of cyclic degradation-induced parameter change. The proposed method is validated through extensive simulations in the Matlab-Simulink environment, and its output is compared to the existing unscented Kalman filter-based SoC estimation method. The proposed estimation strategy can adapt to the cell circumstances and correct for model uncertainties. The results indicate that the proposed adaptive SoC estimation strategy provides more precise and accurate SoC estimates for the entire lifespan of the Li-ion cell.

A Novel Blind Deconvolution Method with Adaptive Period Estimation Technique and Its Application to Fault Feature Enhancement of Bearing

Minimum correlated generalized Lp/Lq deconvolution (MCG-Lp/Lq-D) is an important tool to detect periodic impulses in vibration mixture. It is proved to be a more stable technique than maximum correlated kurtosis deconvolution (MCKD) to recover the fault impulse under strong noise conditions. However, MCG-Lp/Lq-D still has limitations. One of the necessary conditions for the success of MCG-Lp/Lq-D is to provide a precise period of fault. An imprecise prior period will lead to performance degradation or even failure of the method. Therefore, in this paper, a MCG-Lp/Lq-D with adaptive fault period estimation capability is proposed, adaptive minimum correlated generalized Lp/Lq deconvolution (AMCG-Lp/Lq-D). The proposed method uses the autocorrelation function of envelope signal to estimate the fault period adaptively in each iteration and then takes the estimated period as the input parameter of MCG-Lp/Lq-D for the next iteration optimization. The proposed method does not require precise prior fault period input, which greatly improves the fault recovery accuracy and application range of MCG-Lp/Lq-D. Eventually, simulated and experimental data verify the effectiveness and superiority of AMCG-Lp/Lq-D.