DOI QR코드

DOI QR Code

Nonlinear System State Estimating Using Unscented Particle Filters

언센티드 파티클 필터를 이용한 비선형 시스템 상태 추정

  • Received : 2013.04.03
  • Accepted : 2013.05.13
  • Published : 2013.06.30

Abstract

The UKF algorithm for tracking moving objects has fast convergence speed and good tracking performance without the derivative computation. However, this algorithm has serious drawbacks which limit its use in conditions such as Gaussian noise distribution. Meanwhile, the particle filter(PF) is a state estimation method applied to nonlinear and non-Gaussian systems without these limitations. But this method also has some disadvantages such as computation increase as the number of particles rises. In this paper, we propose the Unscented Particle Filter (UPF) algorithm which combines Unscented Kalman Filter (UKF) and Particle Filter (PF) in order to overcome these drawbacks.The performance of the UPF algorithm was tested to compare with Particle Filter using a 2-DOF (Degree of Freedom) Pendulum System. The results show that the proposed algorithm is more suitable to the nonlinear and non-Gaussian state estimation compared with PF.

움직이는 물체를 추적함에 있어 언센티드 칼만 필터(UKF) 알고리즘은 미분 계산없는 빠른 수렴속도와 뛰어난 추정 성능을 지녔다. 그러나 이 방법은 가우시안 잡음 분포 하에서 적용해야 하는 등 제한적인 조건이 수반되는 문제점을 안고 있다. 반면에 파티클 필터(PF)는 제한적인 조건 없이 비선형/비가우시안 시스템에도 적용할 수 있는 상태 추정기법 이라 할 수 있겠다. 그러나 이 방법 또한 파티클의 갯수가 늘어나면 계산량이 크게 증가하는 등의 단점을 지니고 있다. 본 논문에서는 이러한 단점들을 극복하기 위하여 UKF와 PF를 결합한 언센티드 파티클 필터(UPF) 알고리즘을 제안하였다. 본 알고리즘의 성능을 확인하기 위하여 기존의 PF와 UPF 알고리즘을 2-자유도 펜듈럼 시스템을 이용하여 시뮬레이션 하였다. 결과적으로 본 논문에서 제안한 방법이 PF에 비하여 비선형/비가우시안 시스템의 상태 추정에 더욱 적합 함을 확인할 수 있었다.

Keywords

References

  1. M.S. Arulampalam, S. Maskell, N. Gordon, T. Clapp, "A Tutorial on Particle Filters for Online Nonlinear/ Non-Gaussian Bayesian Tracking, IEEE Trans. Signal Processing. vol. 50, no.2, pp 174-188, 2002. https://doi.org/10.1109/78.978374
  2. S.H. Lee, H.B. Yoon, "GPS/INS Data Fusion and Localization using Fuzzy Inference/UPF,"Jouranl of Fuzzy Logic & Intelligent Systems, vol. 19, No. 3, pp. 408-414, 2009. https://doi.org/10.5391/JKIIS.2009.19.3.408
  3. H.G. Oh, "Moving Object Tracking in UAV Video Using Particle Filter,"M.S. thesis Korea Aerospace Univ. Feb. 2008.
  4. W.J. Jun, "A Tightly Coupled GPS/INS Integration Based on the Unscented Particle Filter," M.S. thesis KonKuk Univ. Feb. 2005.
  5. T.Y. Kim, K.T.Chong, "Parameter Estimation of 2-DOF Dynamic System using Particle Filter," Journal of Electrical Engineers, vol. 49, SC. no.2, pp. 10-16, 2012. https://doi.org/10.5573/ieek.2012.49.12.010
  6. J. H. Seung, K.T. Chung, "Parameter Estimation of Dynamic System Based on UKF," Journal of Korea Academia-Industrial cooperation Society, vol. 13, no.2, pp 772-778, 2012. https://doi.org/10.5762/KAIS.2012.13.2.772
  7. S. Julier, J. Uhlmann, "A new extension of the Kalman filter to nonlinear systems," in: Proceedings of the 1997 SPIE AeroSense Symposium, SPIE, 21-24, April 1997.
  8. J.W. Park, C.K. Yang, D.S. Shim, "Particle Filter Performance for Ultra-tightly GPS/INS integration", Journal of Institute of Control, Robotics and Systems, Vol. 14, No. 8, 2008. https://doi.org/10.5302/J.ICROS.2008.14.8.785
  9. O.S. Kwon, "Parameter Estimation of Recurrent Neural Networks Using A Unscented Kalman Filter Training Algorithm and Its Applications to Nonlinear Channel Equalization", Journal of Korean Institute of Intelligent Systems, Vol. 15, No. 5, 552-559, 2005. https://doi.org/10.5391/JKIIS.2005.15.5.552
  10. P.M. Djuric, J.H. Kotecha, J. Zhang, Y. Huang, T. Ghirmai, M.F. Bugallo, J. Miguez, "Particle Filtering", IEEE Signal Processing Magazine, Setember, 2003

Cited by

  1. An Extended Finite Impulse Response Filter for Discrete-time Nonlinear Systems vol.21, pp.1, 2015, https://doi.org/10.5302/J.ICROS.2015.14.0110
  2. UUV의 DVL 항법을 위한 자세 추정 방법 비교 vol.9, pp.4, 2014, https://doi.org/10.7746/jkros.2014.9.4.216