Jinya Su

Continuous nonsingular terminal sliding mode control for systems with mismatched disturbances

A continuous nonsingular terminal sliding mode control approach is proposed for mismatched disturbance attenuation. A novel nonlinear dynamic sliding mode surface is designed based on a finite-time disturbance observer. The time taken to reach the desired setpoint from any initial states under mismatched disturbance is guaranteed to be finite time. In addition, the proposed method exhibits the fine properties of nominal performance recovery as well as chattering alleviation.

High-Order Mismatched Disturbance Compensation for Motion Control Systems Via a Continuous Dynamic Sliding-Mode Approach

A new continuous dynamic sliding-mode control (CDSMC) method is proposed for high-order mismatched disturbance attenuation in motion control systems using a high-order sliding-mode differentiator. First, a new dynamic sliding surface is developed by incorporating the information of the estimates of disturbances and their high-order derivatives. A CDSMC law is then designed for a general motion control system with both high-order matched and mismatched disturbances, which can attenuate the effects of disturbances from the system output. The proposed control method is finally applied for the airgap control of a MAGnetic LEViation (MAGLEV) suspension vehicle. Simulation results show that the proposed method exhibits promising control performance in the presence of high-order matched and mismatched disturbances.

On existence, optimality and asymptotic stability of the Kalman filter with partially observed inputs

For linear stochastic time-varying systems, we investigate the properties of the Kalman filter with partially observed inputs. We first establish the existence condition of a general linear filter when the unknown inputs are partially observed. Then we examine the optimality of the Kalman filter with partially observed inputs. Finally, on the basis of the established existence condition and optimality result, we investigate asymptotic stability of the filter for the corresponding time-invariant systems. It is shown that the results on existence and asymptotic stability obtained in this paper provide a unified approach to accommodating a variety of filtering scenarios as its special cases, including the classical Kalman filter and state estimation with unknown inputs.

Continuous finite-time anti-disturbance control for a class of uncertain nonlinear systems

The continuous finite-time anti-disturbance control problem for a class of nonlinear system under external disturbances and parameter uncertainties is investigated in this article. First, a continuous terminal sliding mode control (CTSMC) is introduced to stabilize the nominal system dynamics. Concerning the unsatisfying performance of the CTSMC in the presence of severe external disturbances and parameter uncertainties, a finite-time disturbance observer (FTDO) is employed to estimate the uncertainties to its nominal dynamics. By integrating the CTSMC method with the FTDO technique, a composite controller is presented for such kind of nonlinear system under external disturbances and parameter uncertainties. The composite controller obtains finite-time convergence property in the presence of disturbances and also nominal control performance recovery in the absence of disturbances. Moreover, compared with conventional sliding mode control, the proposed control law is continuous and no chattering phenomenon exists. The property of stability and the finite-time convergence of the closed-loop system under the proposed controller is guaranteed by means of Lyapunov stability criteria. The proposed control method is finally applied for the tracking control problem of robotic manipulators. Simulation results show that the proposed method exhibits promising control performance in the presence of severe external disturbances and parameter uncertainties.

On Relationship Between Time-Domain and Frequency-Domain Disturbance Observers and Its Applications

This paper provides a generic analysis of the relationship between time/frequency-domain DOB design methodology. It is discovered that the traditional frequency-domain DOBs using a low pass filter with unity gain can only handle disturbances satisfying matching condition, while the traditional time-domain DOBs always generate an observer with a high order. A Functional Disturbance OBserver (FDOB) is proposed to improve the existing results together with its design guideline, frequency analysis and existence condition. Compared with the existing frequency-domain DOBs, the proposed FDOB can handle more classes of disturbances, while compared with the existing time-domain DOBs the proposed FDOB can generate an observer with a lower order. Numerical examples are presented to illustrate the main findings of this paper including a rotary mechanical system of nonminimum phase.

Fault diagnosis for vehicle lateral dynamics with robust threshold

This paper investigates the robust fault diagnosis problem for vehicle lateral dynamics, which play a key role in vehicle stability and driving safety. The proposed fault diagnosis system consists of two sub-systems: fault diagnosis observer and robust threshold. By treating faults as disturbances, Disturbance/Uncertainty Estimation technique is used as fault diagnosis observer to generate residuals. Considering that residuals of model-based fault diagnosis are subject to the effect of uncertainties and consequently large false alarm rate may be resulted in, a novel robust threshold is then proposed based on reachability analysis technique for uncertain systems. The proposed fault diagnosis system is finally applied to the accelerometer and gyrometer sensor fault diagnosis problem of vehicle lateral dynamics, where initial states and velocity are considered to be uncertain. Simulation study verifies the effectiveness of the proposed fault diagnosis system.

Disturbance Observer Based Control with Anti-Windup Applied to a Small Fixed Wing UAV for Disturbance Rejection

Small Unmanned Aerial Vehicles (UAVs) are attracting increasing interest due to their favourable features; small size, low weight and cost. These features also present different challenges in control design and aircraft operation. An accurate mathematical model is unlikely to be available meaning optimal control methods become difficult to apply. Furthermore, their reduced weight and inertia mean they are significantly more vulnerable to environmental disturbances such as wind gusts. Larger disturbances require more control actuation, meaning small UAVs are far more susceptible to actuator saturation. Failure to account for this can lead to controller windup and subsequent performance degradation. In this work, numerical simulations are conducted comparing a baseline Linear Quadratic Regulator (LQR) controller to integral augmentation and Disturbance Observer Based Control (DOBC). An anti-windup scheme is added to the DOBC to attenuate windup effects due to actuator saturation. A range of external disturbances are applied to demonstrate performance. The simulations conduct manoeuvres which would occur during landing, statistically the most dangerous flight phase, where fast disturbance rejection is critical. Validation simulations are then conducted using commercial X-Plane simulation software. This demonstrates that DOBC with anti-windup provides faster disturbance rejection of both modelling errors and external disturbances.

Simultaneous state and input estimation with partial information on the inputs

This paper investigates the problem of simultaneous state and input estimation (SSIE) for discrete-time linear stochastic systems when the information on the inputs is partially available. To incorporate the partial information on the inputs, matrix manipulation is used to obtain an equivalent system with reduced-order inputs. Then Bayesian inference is drawn to obtain a recursive filter for both state and input variables. The proposed filter is an extension of the recently developed state filter with partially observed inputs to the case where the input filter is also of interest, and an extension of the SSIE to the case where the information on the inputs is partially available. A numerical example is given to illustrate the proposed method. It is shown that, due to the additional information on the inputs being incorporated in the filter design, the performances of both state and input estimation are substantially improved in comparison with the conventional SSIE without partial input information.

Reduced-order disturbance observer design for discrete-time linear stochastic systems

Conventional disturbance observers for discrete-time linear stochastic systems assume that the system states are fully estimable and the disturbance estimate is dependent on the estimated system states, hereafter termed full-order disturbance observers (FODOs). This paper investigates the design of reduced-order disturbance observers (RODOs) when the system state variables are not fully estimable. An existence condition of RODOs is established, which is shown to be more easily satisfied than that of conventional FODOs and consequently it has substantially extended the scope of applications of disturbance observer theory. Then a set of recursive formulae for the RODO is developed for online applications. Finally, it is further shown that the conventional FODOs are a special case of the proposed RODO in the sense that the former reduces to the RODO when the states become fully estimable in the presence of disturbances. Examples are given to demonstrate the effectiveness and advantages of the proposed approach.

Dimension Reduction Aided Hyperspectral Image Classification with a Small-sized Training Dataset: Experimental Comparisons

Hyperspectral images (HSI) provide rich information which may not be captured by other sensing technologies and therefore gradually find a wide range of applications. However, they also generate a large amount of irrelevant or redundant data for a specific task. This causes a number of issues including significantly increased computation time, complexity and scale of prediction models mapping the data to semantics (e.g., classification), and the need of a large amount of labelled data for training. Particularly, it is generally difficult and expensive for experts to acquire sufficient training samples in many applications. This paper addresses these issues by exploring a number of classical dimension reduction algorithms in machine learning communities for HSI classification. To reduce the size of training dataset, feature selection (e.g., mutual information, minimal redundancy maximal relevance) and feature extraction (e.g., Principal Component Analysis (PCA), Kernel PCA) are adopted to augment a baseline classification method, Support Vector Machine (SVM). The proposed algorithms are evaluated using a real HSI dataset. It is shown that PCA yields the most promising performance in reducing the number of features or spectral bands. It is observed that while significantly reducing the computational complexity, the proposed method can achieve better classification results over the classic SVM on a small training dataset, which makes it suitable for real-time applications or when only limited training data are available. Furthermore, it can also achieve performances similar to the classic SVM on large datasets but with much less computing time.