|
Online Black-box Model Identification and Output Prediction for Sampled-data Systems |
|
Asim Zaheer, Muhammad Salman(National University of Sciences & Technology, Pakistan) |
| Title: Online Black-box Model Identification and Output Prediction for Sampled-data Systems
Authors: Asim Zaheer and Muhammad Salman, National University of Sciences and Technology, Pakistan
Abstract: In this work, black-box model identification and output prediction for unknown sampled-data minimum phase system has been achieved. Feedforward neural network (multilayer perceptron) is used for system identification. Unscented Kalman Filter (UKF) online determine weights of neural network and predicts output in open-loop sampled-data configuration. |
|
|
An Aircraft's Parameter Identification Algorithm Based on Cloud Model Optimization |
|
Wei Zhang, Yi-Lei Liu(Northwestern Polytechnical University, China), Da-peng Guo(AVIC Aerodynamics Research Institute, China), Khayyam Masood, Jing Tian(Northwestern Polytechnical University, China) |
| The paper proposes an aircraft's parameter identification algorithm, which optimizes the ML function with the cloud model optimization theory in accordance with the ML estimation principle. The algorithm has no high requirements for initial values and is little affected by noise. Thus it is easy to apply, have rather fast convergence and nice global search capability. The Twin Otter airplane is used as a numerical example and the results show that the parameter identification algorithm has good precision and fast convergence, and does not reach locally optimal solutions. |
|
|
Identification of Continuous-time Hammerstein Models Using Simultaneous Perturbation Stochastic Approximation |
|
Mohd Ashraf Ahmad, Shun-ichi Azuma, Toshiharu Sugie(Kyoto University, Japan) |
| This paper performs an initial study on identification of continuous-time Hammerstein models based on Simultaneous Perturbation Stochastic Approximation (SPSA). For handling it, a piecewise-linear functions are used as a tool to approximate the unknown nonlinear functions. The SPSA based method is then used to estimate the parameters in both the linear and nonlinear parts based on the given input and output data. The simulation result shows that the SPSA based algorithm can give an accurate parameter estimation of the Hammerstein models with high probability. |
|
|
Multi-Target Tracking Algorithm Based on FIR filters |
|
Changjoo Lee, Kyung Min Min, Hyun Duck Choi, Choon Ki Ahn, Myo Taeg Lim(Korea University, Republic of Korea) |
| This research article proposes a multi-target tracking algorithm by using a finite impulse response (FIR) filter. The advantage of the FIR filter is to observe robust filtering to incorrect data such as model uncertainty and noise, etc. This paper demonstrates robust tracking performance for multi-target tracking using the properties of FIR filter. In order to use FIR structure, all possible paths are constructed and the estimate of each path is chosen as the target through Mahalanobis distance if it is appropriate target. |
|
|
Observer-based Admissible Control for Singular Delta Operator Systems |
|
Xin-zhuang Dong, MINGQING XIAO(Southern Illinois University Carbondale, United States), Yushun Wang, Wenxue He(Qingdao University, China) |
| This paper studies the problem of designing an observer-based admissible controller for singular delta operator systems. Sufficient conditions are provided for the existence of an asymptotical and physically realizable observer. Then an observer-based admissible controller is obtained in terms of strict linear matrix inequalities. The corresponding gain matrices appeared in our proposed approach can be constructed through the feasible solutions of a set of linear matrix inequalities. Some numerical examples are presented to demonstrate the theoretical outcomes of the paper. |
|
|
Sliding Mode Control of a Rotary Inverted Pendulum using Higher Order Differential Observer |
|
Philippe Faradja, Guoyuan Qi, Martial Tatchum(Tshwane University of Technology, South Africa) |
| The Inverted Pendulum is the benchmark system for control methods. Most linear controllers are not efficient under the situations of disturbances and other uncertainties. As high precision and robustness are required, a sliding mode controller is used in this work. Due to uncertainties, a model free based observer is used to estimate some states. The simulation and real experiment results demonstrate the robustness of the controller and the efficiency of a model free observer. |
|