Multi-Sensor Data Fusion Technology Based on Convex Optimization Adaptive

Abstract:

In research of SINS/GPS/SS integrated navigation system and multi sensor data fusion method, we proposed an adaptive federated Kalman filtering method based on convex optimization. The method uses real-time OLS-SVM obtained information distribution factor, so that the information distribution factor can change with the local filter performance. So that can make timely response to the performance and failure of local sensors and filter, which affects the whole system accuracy. The simulation results show that, the method has stronger adaptability of model and noise interference, can effectively restrain the divergence and improves the system precision and real time.

• Periodical: Advanced Materials Research (Volume 739)
• Pages: 580-585
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.739.580
• Online since: August 2013
• Authors: He Nian Wang, Guo Xing Yi, Chang Hong Wang, Yang Guang Xie
• Keywords: Adaptive Federated Filter, Convex Optimization, Support Vector Machine (SVM)

References

[1] Yager R R. Set Measure Directed Multi-Source Information Fusion[J]. Fuzzy Systems, IEEE Transactions on, 2011, 19(6): 1031-1039.

DOI: 10.1109/tfuzz.2011.2159725

[2] Yongmian Z, Qiang J. Efficient Sensor Selection for Active Information Fusion[J]. Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on, 2010, 40(3): 719-728.

DOI: 10.1109/tsmcb.2009.2021272

[3] Moallemi C C, Van Roy B. Convergence of Min-Sum Message-Passing for Convex Optimization[J]. Information Theory, IEEE Transactions on, 2010, 56(4): 2041-(2050).

DOI: 10.1109/tit.2010.2040863

[4] Ferras M, Cheung-Chi L, Barras C, et al. Comparison of Speaker Adaptation Methods as Feature Extraction for SVM-Based Speaker Recognition[J]. Audio, Speech, and Language Processing, IEEE Transactions on, 2010, 18(6): 1366-1378.

DOI: 10.1109/tasl.2009.2034187

[5] Keng-Pei L, Ming-Syan C. On the Design and Analysis of the Privacy-Preserving SVM Classifier[J]. Knowledge and Data Engineering, IEEE Transactions on, 2011, 23(11): 1704-1717.

DOI: 10.1109/tkde.2010.193

[6] Adankon M M, Cheriet M, Biem A. Semisupervised Learning Using Bayesian Interpretation: Application to LS-SVM[J]. Neural Networks, IEEE Transactions on, 2011, 22(4): 513-524.

DOI: 10.1109/tnn.2011.2105888

[7] Pal M, Foody G M. Evaluation of SVM, RVM and SMLR for Accurate Image Classification With Limited Ground Data[J]. Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of, 2012, 5(5): 1344-1355.

DOI: 10.1109/jstars.2012.2215310