Trajectory Prediction of Buoy Drift based on Improved Complex Valued Neural Network

Liangkun Xu

《International Core Journal of Engineering》 8卷3期 (2022/03) Pp. 55-66

https://doi.org/10.6919/ICJE.202203_8(3).0011

摘要

This paper presents a buoy drift trajectory prediction algorithm based on improved complex valued neural network. Taking the longitude of the buoy as the real part of the input and the latitude as the imaginary part of the input, the complex input of the complex valued neural network is constructed. All longitudes and latitudes on the earth are perpendicular to each other. This satisfies that the real and imaginary parts of the complex form an orthogonal unit basis. The drift trajectories of buoys with different reporting intervals are estimated. The algorithm is tested by using the position data of buoys 11a and 2 in Meizhou Bay, Fujian Province. The effects of reporting interval, drift distance, adaptive factor and noise covariance on the estimated longitude are analyzed and compared. Experimental results and error analysis show that the new algorithm is superior to other algorithms in trajectory prediction. The longitude error and latitude error of the new method are 3.21e-04 and 6.36e-05 respectively, which is lower than the original algorithm. Therefore, this algorithm can be used to accurately predict the drift trajectory of dynamic time interval buoy.

關鍵字

Neural Network ； Buoy Drift ； Trajectory Prediction

參考文獻

Chen, Q. Bin, H. Huang, Z. Event-based master–slave synchronization of complex-valued neural networks via pinning impulsive control. Neural Networks 2021, Doi: 10.1016/j.neunet.2021.10.025.

Song, Q. Zhao, Z. Liu, Y. Alsaadi, F.E. Mean-square input-to-state stability for stochastic complex-valued neural networks with neutral delay. Neurocomputing 2021, Doi: 10.1016/j.neucom.2021.10.117.

Huang, Y, Wu, F. Finite-time passivity and synchronization of coupled complex-valued memristive neural networks. Information Sciences 2021, 580: 775-800.

Li, C. Liu, S. Wang, Z. Classifying interictal epileptiform activities in intracranial EEG using complex-valued convolutional neural network. International Journal of Psychophysiology 2021, 168: S104-S105.

Song, X. Sun, X. Man, J. Song, S. Wu, Q. Synchronization of fractional-order spatiotemporal complex-valued neural networks in finite-time interval and its application. Journal of the Franklin Institute 2021, 358(16): 8207-8225.