Experiments on State and Unmeasured-Parameter Estimation of Two Degree-of-Freedom System for Precise Control Based on JAUKF

다국어 초록 (Multilingual Abstract)

We herein present system parameter estimation using the joint adaptive unscented Kalman filter and state estimation approach for a two degree-of-freedom (2-DOF) mechanical system. The unscented Kalman filter (UKF) is applied broadly in diverse engineering fields to estimate the state of the dynamic system and improve the control precision by reducing measurement noise. One aspect of parameter identification is that unmeasured parameter estimation is important in designing and maintaining the system performance with a suitable controller. This is because changes in the system parameters estimated will occur owing to external shock and deterioration in operation. State estimation has been studied thoroughly and developed widely, but not in terms of parameter estimation. Parameter estimation is important because system parameters can be altered owing to wear and tear, large disturbance, or exposure to extreme temperatures. It is difficult to disassemble and measure the parameter when it changes; hence, computational estimation is a solution. The proposed method simultaneously estimates the states as well as the parameters of custom-made 2-DOF mechanical system that is a nonlinear dynamical system. The adaptive rules in the estimation process are considered based on the moving average window method to address the effects of unexpected noise in the sensor measurements. The experimental results are analyzed to demonstrate the effectiveness of the proposed method for estimating the states and parameters. This method demonstrates better performance compared to using the joint-UKF in terms of convergence time, accuracy, and robustness to noise.

참고문헌 (Reference)

1 박수창, "Tracking Error Constrained Terminal Sliding Mode Control for Ball-Screw Driven Motion Systems with State Observer" 한국정밀공학회 19 (19): 359-366, 2018

2 Zhi, L., "State of charge estimation for Li-ion battery based on extended Kalman fi lter" 105 : 3515-3520, 2017

3 Tran, N. -T., "State of charge and state of health estimation of AGM VRLA batteries by employing a dual extended Kalman fi lter and an ARX model for online parameter estimation" 10 (10): 2017

4 Liu, X., "Robust state estimation for neural networks with discontinuous activations" 40 : 1425-1437, 2010

5 Wang, C., "Recursive least squares estimation algorithm applied to a class of linear-in-parameters output error moving average systems" 29 : 36-41, 2014

6 You, C., "Nonlinear driver parameter estimation and driver steering behavior analysis for ADAS using fi eld test data" 47 (47): 686-699, 2017

7 Min Ming, "Model Prediction Control Design for Inverse Multiplicative Structure Based Feedforward Hysteresis Compensation of a Piezo Nanopositioning Stage" 한국정밀공학회 19 (19): 1699-1708, 2018

8 Yu, Q., "Lithiumion battery parameters and state-of-charge joint estimation based on H-infi nity and unscented Kalman fi lters" 66 (66): 8693-8701, 2017

9 Davoodabadi, I., "Identifi cation of tire forces using dual unscented Kalman fi lter algorithm" 78 (78): 1907-1919, 2014

10 Manivannan, R., "Design of extended dissipativity state estimation for generalized neural networks with mixed time-varying delay signals" 424 : 175-203, 2018

11 Kazantzis, N., "Control-relevant discretization of nonlinear systems with timedelay using Taylor–Lie series" 1 : 149-154, 2003

12 Ma, J., "Combined state and parameter estimation for Hammerstein systems with time delay using the Kalman fi ltering" 31 (31): 1139-1151, 2017

13 Jianxing Zhang, "Co-State Variable Determination in Pontryagin’s Minimum Principle for Energy Management of Hybrid Vehicles" 한국정밀공학회 17 (17): 1215-1222, 2016

14 Cai, M., "Battery state-of-charge estimation based on a dual unscented Kalman fi lter and fractional variable-order model" 10 (10): 2017

15 Khodadadi, H., "Applying a dual extended Kalman fi lter for the nonlinear state and parameter estimations of a continuous stirred tank reactor" 35 (35): 2426-2436, 2011

16 Erazo, K., "An offl ine approach for output-only Bayesian identifi cation of stochastic nonlinear systems using unscented Kalman fi ltering" 397 : 222-240, 2017

17 Partovibakhsh, M., "An adaptive unscented Kalman fi ltering approach for online estimation of model parameters and state-of-charge of lithium-ion batteries for autonomous mobile robots" 23 (23): 357-363, 2015

18 Song, Q., "An adaptive UKF algorithm for the state and parameter estimations of a mobile robot" 34 (34): 72-79, 2008

19 Wei Sun, "An Approximate Solution Method of Dynamic Reliability for Wind Turbine Gear Transmission with Parameters of Uncertain Distribution Type" 한국정밀공학회 19 (19): 849-857, 2018

20 Sun, F., "Adaptive unscented Kalman fi ltering for state of charge estimation of a lithium-ion battery for electric vehicles" 36 (36): 3531-3540, 2011

21 de Marina, H. G., "Adaptive UAV attitude estimation employing unscented Kalman fi lter, FOAM and low-cost MEMS sensors" 12 (12): 9566-9585, 2012

22 To Xuan Dinh, "Adaptive Tracking Control of a Quadrotor Unmanned Vehicle" 한국정밀공학회 18 (18): 163-173, 2017

23 Hong, S., "A novel approach for vehicle inertial parameter identifi cation using a dual Kalman fi lter" 16 (16): 151-161, 2015

24 Panuska, V., "A new form of the extended Kalman fi lter for parameter estimation in linear systems with correlated noise" 25 (25): 229-235, 1980

25 Julier, S., "A new extension of the Kalman filter to nonlinear systems" (3068) : 182-193, 1997

26 Eftekhar Azam, S., "A dual Kalman fi lter approach for state estimation via output-only acceleration measurements" 60–61 : 866-886, 2015