A New Particle Filter Inspired by Biological Evolution: Genetic Filter

In this paper, we consider a new particle filter inspired by biological evolution. In the standard particle filter, a resampling scheme is used to decrease the degeneracy phenomenon and improve estimation performance. Unfortunately, however, it could cause the undesired the particle deprivation problem, as well. In order to overcome this problem of the particle filter, we propose a novel filtering method called the genetic filter. In the proposed filter, we embed the genetic algorithm into the particle filter and overcome the problems of the standard particle filter. The validity of the proposed method is demonstrated by computer simulation.

Authors:

• S. Park
• J. Hwang
• K. Rou
• E. Kim

Keywords:

• Particle filter
• genetic algorithm
• evolutionary algorithm.

References:

[1] Kalman, R.E, "A New Approach to Linear Filtering and Prediction
Problems," Trans, ASME, J. Basic Engineering, vol.82, pp34-45, Mar.
1960.
[2] B. Ristic, S. Arulampalam, and N. Gordon, Beyond the Kalman Filter:
Particle Filters for Tracking Applications, Boston, London: Artech
House., 2004.
[3] S. Thrun, W. Burgard, and D. Fox, Probabilistic Robotics, Cambridge,
Massachisetts, London, England: The MIT Press, 2005.
[4] J.E. Handschin and D.Q. Mayne, "Monte Carlo techniques to estimate the
conditional expectation in multi-stage non-linear filtering," Intern.
Journal of Control, vol.9, no. 5, pp.547-559, 1969.
[5] K. Uosaki, Y. Kimura and T. Hatanaka, "Nonlinear State Estimation by
Evolution Strategies based Particle Filters," Proc. Congress on
Evolutionary Computation Vol. 1, pp884 - 890, 19-23-th June 2004.
[6] Z. Michalewicz. Genetic Algorithm + Data Structure = Evolution
Programs. Berling Heidelberg, New York: Springer-Verlag, third ed.,
1996.