Accuracy Improvement of GPS Positioning Based on GA-Aided Particle Filter

Abstract:

Aiming at the weight degeneracy phenomena in particle filter algorithm, a resampling method improving the diversity based on GA-aided particle filter was presented. Taking the advantage of genetic algorithm ( GA ) in selection ,crossover and inheritance to make up for the shortcoming of resampling. Genetic operation on particles in real number domain is adapted to reduce the complex of the genetic algorithm. And the evolutionary idea of genetic algorithm was combined with particle filter, by using selection, and mutation to improve the weight degeneracy and diversity of particle filter. This GA-aided particle filter was applied in the established GPS system nonlinear dynamic state space model. The experimental results based on the collected real GPS data is compared with the tradition particle filter, and compared with the effective number of particles and particle distribution. The experimental results indicated that the GA-aided particle filter can increase the number of particle, and effectively solve the particle degradation phenomena, the estimation accuracy of GA-aided particle filter is better than that of particle filter (PF).

• Periodical: Applied Mechanics and Materials (Volumes 719-720)
• Pages: 737-743
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.719-720.737
• Online since: January 2015
• Authors: Er Shen Wang*, Tao Pang, Zhi Xian Zhang
• Keywords: Genetic Algorithm (GA), Global Positioning System (GPS), Particle Filter, Positioning Accuracy

References

[1] Mojtaba Bahrami and Marek Ziebart. A Kalman Filter based Dopper-smoothing of code Pseudoranges in GNSS Challenged Environments[C]. Proceedings of the ION GNSS 2011, Portland, 2011: 1-11.

[2] I. Sayim, B. Pervan, S. Pullen and P. Enge. Experimental and theoretical results on the LAAS sigma overbound, [C]. Proceedings of the ION GPS, Portland, Oregon, 2002: 29-38.

[3] GORDON N J, SALM OND D J, SMITH A F M. Nov el appro ach t o no nlinear/ no n- g aussian Bay esian state estimation[ C] . Minneapo : Pro c o f the Radar and Sig nal Processing , (1993).

[4] Junchuan Zhou, Yuhong Yang, Jieying Zhang, Ezzaldeen Edwan. Applying Quaternion-based Unscented Particle Filter on INS/GPS with Field Experiments[C]. Proceedings of the ION GNSS 2011, Portland, 2011: 1-14.

DOI: 10.2514/6.2011-6488

[5] Thanh-Trung Duong, Kai-Wei Chiang. Non-linear, non-Gaussian estimation for INS/GPS integration[C]. Proceedings of the ION GNSS 2011, Portland, 2011: 1-8.

[6] Christian Gentner, Jean-Marc Rawadi, Estefania Munoz, etc. Hybrid Positioning with 3GPP-LTE and GPS employing Particle Filters[C]. Proceedings of the ION GNSS 2012, 17-21 September, Nashville, 2012: 1-9.

[7] YANG Ning, QIAN Feng, ZHU Rui. Improved particle filter based on genetic algotrithm[J]. Journal of Shanghai Jiaotong University, 2011, 45(10): 1526-1530. (in Chinese).

[8] M. S. Arulampalam,S. Maskell,N. Gordon and T. Clapp. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J]. IEEE Transactions on Signal Processing, 2002, 50(2): 174–188.

DOI: 10.1109/78.978374

[9] YANG Yang, XUE Xiao zhong. Application of GA-aided particle filter intight compling GPS/SINS system[J]. Journal of PLA University of Science and Technology(Natural Science Edition), 2011, 12(4): 322-327. (in Chinese).

[10] Giovanni Giardini. Genetic algorithm for multi-agent space exploration[C]. Califo rnia: AIAA 2007 Conference and Exhibit , (2007).

DOI: 10.2514/6.2007-2824

[11] Fredrik Gustafsson, Fredrik Gunnarsson, etc. Particle filters for positioning, navigation, and tracking. IEEE Trans. on Signal Processing, 50(2), (2002).

DOI: 10.1109/78.978396

[12] E Kaplan, C Hegarty. Understanding GPS: Principles and Applications[M]. Second Edition. MA, US: Artech House, Inc, 2006: 239-264.