Model Mismatch Problem Research Articles

Improved IMM Algorithm based on RNNs

The Interactive Multi-Model (IMM) algorithm uses multiple motion models to simultaneously track the target, which effectively solves the problem of model mismatch when a single model tracks the maneuvering target, and is widely used in maneuvering target tracking tasks. However, the Interactive Multi-Model recognition motion model is not accurate enough, and there is a certain delay in the maneuver recognition of the target, which leads to a reduction in tracking accuracy. To solve this problem, considering that deep neural networks are very good at processing classification tasks, we introduce it into target tracking tasks, combining the respective of deep neural networks and traditional tracking filtering methods for maneuvering target tracking. we use the Recurrent Neural Networks to identify the motion model of the target and propose an improved LSTM-IMM model algorithm based on the interactive multi-model algorithm. Finally, we compare the traditional interactive multi-model algorithm and verify the algorithm using Monte Carlo simulation. The results show that the proposed algorithm has improved the recognition accuracy and recognition speed of the model, and the tracking accuracy has been improved.

Design of Disturbance Extended State Observer (D-ESO)-Based Constrained Full-State Model Predictive Controller for the Integrated Turbo-Shaft Engine/Rotor System

In the modern helicopter design and development process, constrained full-state control technology for turbo-shaft engine/rotor systems has always been a research hotspot in academia and industry. However, relevant references have pointed out that the traditional design method with an overly complex structure (Min-Max structure and schedule-based transient controller, i.e., M-M-STC) may not be able to meet the protection requirements of engine control systems under certain circumstances and can be too conservative under other conditions. In order to address the engine limit protection problem more efficiently, a constrained full-state model predictive controller (MPC) has been designed in this paper by incorporating a linear parameter varying (LPV) predictive model. Meanwhile, disturbance extended state observer (D-ESO) (which a sufficient convergence condition is deduced for) has also been proposed as the compensator of the LPV model to alleviate the MPC model mismatch problem. Finally, we run a group of comparison simulations with the traditional M-M-STC method to verify the effectiveness of this controller by taking compressor surge prevention problems as a case study, and the results indicate the validity of the proposed method.

Accurate State of Charge Estimation With Model Mismatch for Li-Ion Batteries: A Joint Moving Horizon Estimation Approach

The accurate state of charge (SOC) estimation plays a significant role in charge/discharge control, balance control, and safe management of lithium-ion batteries (LIBs). However, due to the model mismatch issues, either from battery inconsistency or battery dynamic characteristics difference, the accuracy of the model-based SOC estimation method is usually unsatisfactory. To solve this problem, a joint moving horizon estimation (joint-MHE) approach that can simultaneously estimate the model parameter and state is proposed here. In this paper, the circuit-equivalent battery model is first constructed by parameterizing the circuit parameters as polynomial function of SOC. Then, by the sensitivity analysis, the update parameters are selected and added to the state-space model as additional states. Finally, the joint-MHE strategy is conducted for the simultaneous parameter and SOC estimation. To investigate the performance of the proposed method thoroughly, three model mismatch conditions are considered, including battery inconsistency, battery dynamic characteristics difference, and the combination of both. The results demonstrate that the joint-MHE approach is an effective way to solve the model mismatch problem. Moreover, compared to joint extended Kalman filtering, the proposed approach can offer a more reliable, robust, and accurate SOC estimation of LIBs under various model mismatch conditions.

MRI and PET image fusion using the nonparametric density model and the theory of variable-weight

Medical image fusion is important in the field of clinical diagnosis because it can improve the availability of information contained in images. Magnetic Resonance Imaging (MRI) provides excellent anatomical details as well as functional information on regional changes in physiology, hemodynamics, and tissue composition. In contrast, although the spatial resolution of Positron Emission Tomography (PET) provides is lower than that an MRI, PET is capable of depicting the tissue's molecular and pathological activities that are not available from MRI. Fusion of MRI and PET may allow us to combine the advantages of both imaging modalities and achieve more precise localization and characterization of abnormalities. Previous image fusion algorithms, based on the estimation theory, assume that all distortions follow Gaussian distribution and are therefore susceptible to the model mismatch problem. To overcome this mismatch problem, we propose a new image fusion method with multi-resolution and nonparametric density models (MRNDM). The RGB space registered from the source multi-modal medical images is first transformed into a generalized intensity-hue-saturation space (GIHS), and then is decomposed into the low- and high-frequency components using the non-subsampled contourlet transform (NSCT). Two different fusion rules, which are based on the nonparametric density model and the theory of variable-weight, are developed and used to fuse low- and high-frequency coefficients. The fused images are constructed by performing the inverse of the NSCT operation with all composite coefficients. Our experimental results demonstrate that the quality of images fused from PET and MRI brain images using our proposed method MRNDM is higher than that of those fused using six previous fusion methods.

An improved general type-2 fuzzy sets type reduction and its application in general type-2 fuzzy controller design

The Karnik–Mendel (KM) or the enhanced Karnik–Mendel (EKM) algorithm is widely used for interval type-2 fuzzy sets type reduction in many applications. Compared with iterative procedures of KM/EKM, an iterative algorithm with a stop condition or an enhanced iterative algorithm with a stop condition based on the KM algorithm that converges monotonically is more efficient. In this article, a new iterative algorithm with stop condition type reduction for interval type-2 fuzzy sets is proposed, in which switch points are initialized and unidirectional search is performed based on monotone properties of the KM algorithm. Furthermore, the proposed algorithm is applied to general type-2 fuzzy sets type reduction based on α-plane representation. The experimental results of a triangular and gaussian secondary membership function show practicality and efficiency of this method. In accordance with the conventional PI, type-1 or interval type-2 fuzzy controller is difficult to achieve a desired control effect for steam temperature at collector outlet of trough solar thermal power generation system with large time delay, strong inertia and parameter time-variation, and a general type-2 fuzzy controller with more adjustable controller parameters is proposed in this article. In different working conditions, the proposed controller can reduce system overshoot and ensure system stability. Moreover, when the working condition changes, the controller can solve a model mismatch problem under same controller parameters and has faster response.

Strong Tracking Tobit Kalman Filter with Model Uncertainties

The Tobit Kalman filter (TKF) is a good choice for applications of discrete time-varying systems with censored measurements. The traditional TKF is usually designed under the assumption of exactly knowing system state and measurement functions, which can not guarantee the good performance and is even unsuitable for system with model mismatch. To ensure the performance, this paper proposes a novel TKF algorithm in the presence of both censored measurements and model uncertainties. Our proposed algorithm, called strong tracking TKF (STTKF), adopts the orthogonal principle as an additional criterion to overcome model mismatch problem. By introducing fading factor into a priori error covariance, STTKF adaptively adjust gain matrix according to different model mismatch degree. Firstly the recursive fading factor formulation is deduced based on orthogonal principle. Then the designed STTKF process is given in a recursive manner. Finally the effectiveness of STTKF is verified by two examples of oscillator system and unmanned aerial vehicle (UAV) transmitter.

Altitude measurement based on characteristics reversal by deep neural network for VHF radar

A novel direction of arrival (DOA) estimation method is proposed for very high-frequency (VHF) radar by the deep neural network (DNN) under strong multipath effect and complex terrain environment. The classical methods are all based on the classical multipath signal model, hence, it often causes the problem of model mismatch and results in poor performance in estimation. It is generally considered that the serious multipath effect reduces the precision of elevation estimation. However, the characteristics of the multipath signal are exploited and used to improve the precision in this study. This is the highlight of the proposed method. The approach of the deep neural network is applied to learn the received data's characteristics from a different elevation. A new characteristic space is constructed in the training procedure. In the test procedure, the characteristic of data is extracted by the well-trained network and projected into the constructed characteristic space. Reversing the DOA is finished at last. The results of simulation data verify the validity of the proposed method. The results of the practical data show the practicability of the proposed method for the low-elevation target under the serious multipath effect. Combining the DNN approach with altitude measurement of a low-elevation target for VHF radar is meaningful to explore.

Total variation-based method for radar coincidence imaging with model mismatch for extended target

Originating from traditional optical coincidence imaging, radar coincidence imaging (RCI) is a staring/forward-looking imaging technique. In RCI, the reference matrix must be computed precisely to reconstruct the image as preferred; unfortunately, such precision is almost impossible due to the existence of model mismatch in practical applications. Although some conventional sparse recovery algorithms are proposed to solve the model-mismatch problem, they are inapplicable to nonsparse targets. We therefore sought to derive the signal model of RCI with model mismatch by replacing the sparsity constraint item with total variation (TV) regularization in the sparse total least squares optimization problem; in this manner, we obtain the objective function of RCI with model mismatch for an extended target. A more robust and efficient algorithm called TV-TLS is proposed, in which the objective function is divided into two parts and the perturbation matrix and scattering coefficients are updated alternately. Moreover, due to the ability of TV regularization to recover sparse signal or image with sparse gradient, TV-TLS method is also applicable to sparse recovering. Results of numerical experiments demonstrate that, for uniform extended targets, sparse targets, and real extended targets, the algorithm can achieve preferred imaging performance both in suppressing noise and in adapting to model mismatch.

A multiple model approach for predictive control of indoor thermal environment with high resolution

Realtime and precise control strategies for indoor micro-climate are increasingly needed due to the requirements of thermal comfort and energy consumption saving in recent years. In this paper, a precise control scheme with high resolution is proposed for indoor thermal regulation. Finite volume method discretization/linearization are employed to construct the state-space model of the thermal process. Proper orthogonal decomposition method is further applied for model order reduction. On this basis, a multiple model approach is used to treat the thermodynamic coupling effect and a multi-model switching model predictive control (MPC) scheme is proposed for temperature tracking. Two cases including cooling and warming scenarios are designed, respectively, for performance validation. Results show that compared with the classic MPC method, the developed method can alleviate the model mismatch problem, and facilitate an optimal control of the spatial temperature in the considered ventilation room.

New framework for unsupervised universal steganalysis via SRISP-aided outlier detection

Formulating steganalysis as a binary classification problem has been highly successful. However, the existing detection algorithms are difficult to obtain high detection accuracy when applied in real-world circumstances. Because so-called model mismatch problem often occurs owing to unknown cover source and embedding parameters. To avoid the mess of model mismatch, we propose a new unsupervised universal steganalysis framework to detect individual stego images. First, cover images with statistical properties similar to those of the given test image are searched from a retrieval cover database to establish an aided cover sample set. Second, unsupervised outlier detection is performed on a test set composed of the given test image and its aided cover sample set to determine the type (cover or stego) of the given test image. Our proposed framework, called Similarity Retrieval of Image Statistical Properties (SRISP)-aided unsupervised outlier detection, requires no training, and thus it does not suffer from model mismatch. The framework employs standard steganalysis features and detects each test image individually. Experimental results illustrate that the framework substantially outperforms one-class support vector machine and the traditional unsupervised outlier detectors without considering SRISP; its detection performance is independent of the proportion of stego images in the test samples.

Noised image segmentation based on rough set and orthogonal polynomial density model

In order to segment a noised image, a method is proposed based on the rough set and orthogonal polynomial density model, in which the nonparametric mixture model can accurately fit the image gray distribution and the rough set can deal with the inaccuracy and uncertainty problems. First, the nonparametric mixture density model is constructed based on the upper and lower approximations of the rough set which can address the problem of over-relying on the prior presumption. Second, the nonparametric expectation-maximization is used to estimate the mixture model parameters. Finally, image pixels are classified according to Bayesian criterion. Experiments on different datasets show that our method is effective in solving the problem of model mismatch, restraining the noise, and preserving the boundary for the noised image segmentation.

Study of Predictive Control in Industrial Networked Control System

In order to improve the performance of networked control system which has the characteristic of uncertain time delay, predictive control is adopted. The model of networked control system is analyzed. Take into account uncertain network time delay caused by various factors and problem of model mismatch, predictive controller is designed and simulation model of system is built using software of Matlab. A method of reducing model mismatch is proposed. Study results show that predictive control can make networked control system with good performance.

Medical image segmentation based on non-parametric mixture models with spatial information

Because of too much dependence on prior assumptions, parametric estimation methods using finite mixture models are sensitive to noise in image segmentation. In this study, we developed a new medical image segmentation method based on non-parametric mixture models with spatial information. First, we designed the non-parametric image mixture models based on the cosine orthogonal sequence and defined the spatial information functions to obtain the spatial neighborhood information. Second, we calculated the orthogonal polynomial coefficients and the mixing ratio of the models using expectation-maximization (EM) algorithm, to classify the images by Bayesian Principle. This method can effectively overcome the problem of model mismatch, restrain noise, and keep the edge property well. In comparison with other methods, our method appears to have a better performance in the segmentation of simulated brain images and computed tomography (CT) images.

Multi-loop Internal Model Controller Design Based on a Dynamic PLS Framework

In this paper, a multi-loop internal model control (IMC) scheme in conjunction with feed-forward strategy based on the dynamic partial least squares (DyPLS) framework is proposed. Unlike the traditional methods to decouple multi-input multi-output (MIMO) systems, the DyPLS framework automatically decomposes the MIMO process into a multi-loop system in the PLS subspace in the modeling stage. The dynamic filters with identical structure are used to build the dynamic PLS model, which retains the orthogonality among the latent variables. To address the model mismatch problem, an off-line least squares method is applied to obtain a set of optimal filter parameters in each latent space. Without losing the merits of model-based control, a simple and easy-tuned IMC structure is readily carried over to the dynamic PLS control framework. In addition, by projecting the measurable disturbance into the latent subspace, a multi-loop feed-forward control is yielded to achieve better performance for disturbance rejection. Simulation results of a distillation column are used to further demonstrate this new strategy outperforms conventional control schemes in servo behavior and disturbance rejection.

Real-time Optimization with Estimation of Experimental Gradients

For good performance in practice, real-time optimization schemes need to be able to deal with the inevitable model-mismatch problem. Unlike the two-step schemes combining parameter estimation and optimization, the modifier-adaptation approach uses experimental gradient information and does not require the model parameters to be estimated on-line. The dual modifier-adaptation approach presented in this paper drives the process towards optimality, while paying attention to the accuracy of the estimated gradients. The gradients are estimated from the successive operating points generated by the optimization algorithm. The novelty lies in the development of an upper bound on the gradient estimation error, which is used as a constraint for locating the next operating point. The proposed approach is demonstrated in simulation by the real-time optimization of a continuous reactor.

Safe Adaptive Switching Control: Stability and Convergence

A formal theoretical explanation of the model-mismatch instability problem associated with certain adaptive control design schemes is proposed, and a solution is provided. To address the model-mismatch problem, a primary task of adaptive control is formulated as finding an asymptotically optimal, stabilizing controller, given the feasibility of adaptive control problem. A class of data-driven cost functions called cost-detectable is introduced that detect evidence of instability without reference to prior plant models or plant assumptions. The problem of designing adaptive systems that are robustly immune to mismatch instability problems is thus placed in a setting of a standard optimization problem. We call the result safe adaptive control because it robustly achieves adaptive stabilization goals whenever feasible, without prior assumptions on the plant model and, hence, without the risk of model-mismatch instability. The result improves the robustness of previous results in hysteresis switching control, both for discrete and for continuously-parameterized candidate controller sets. Examples are provided.

Model Enforcement: A Unified Feature Transformation Framework for Classification and Recognition

Bayesian classifiers rely on models of the a priori and class-conditional feature distributions; the classifier is trained by optimizing these models to best represent features observed in a training corpus according to certain criterion. In many problems of interest, the true class-conditional feature probability density function (PDF) is not a member of the set of PDFs the classifier can represent. Previous research has shown that the effect of this problem may be reduced either by improving the models or by transforming the features used in the classifier. This paper addresses this model mismatch problem in statistical identification, classification, and recognition systems. We formulate the problem as the problem of minimizing the relative entropy, which is also known as the Kullback-Leibler distance, between the true conditional PDF and the hypothesized probabilistic model. Based on this formulation, we provide a computationally efficient solution to the problem based on volume-preserving maps; existing linear transform designs are shown to be special cases of the proposed solution. Using this result, we propose the symplectic maximum likelihood transform (SMLT), which is a nonlinear volume-preserving extension of the maximum likelihood linear transform (MLLT). This approach has many applications in statistical modeling, classification, and recognition. We apply it to the maximum likelihood estimation (MLE) of the joint PDF of order statistics and show a significant increase in the likelihood for the same number of parameters. We provide also phoneme recognition experiments that show recognition accuracy improvement compared with using the baseline Mel-Frequency Cepstrum Coefficient (MFCC) features or using MLLT. We present an iterative algorithm to jointly estimate the parameters of the symplectic map and the probabilistic model for both applications.