Genetic Algorithm and LSTM Based Approach for Diver Search and Path Prediction

Xiangyang Liu; Zhengyang Wang; Jintong Xiong; Zhaoyu Liu

doi:10.54097/7nby1w69

Authors

Xiangyang Liu
Zhengyang Wang
Jintong Xiong
Zhaoyu Liu

DOI:

https://doi.org/10.54097/7nby1w69

Keywords:

Kalman filtering algorithm; genetic algorithm; LSTM.

Abstract

With the rapid development of submersible technology, exploring the deep sea is no longer a distant dream. In this paper, a set of safety procedures for manned submersibles is designed to ensure that the main ship can respond in time and launch search and rescue in case of loss of connection. The digital elevation model and Kalman filter algorithm are used to process the deep-sea data, combined with the Monte Carlo method to estimate the position of the submersible, and the water depth of 2000m is taken as the working depth. The genetic algorithm is used to optimize the SAR strategy, and environmental factors are considered to affect the sailing speed. For the case of multi-diver lost connection, CNN and LSTM models are combined to predict the 3D path to avoid the risk of collision. These methods will significantly improve the safety of the submersible and the efficiency of search and rescue and promote the sustainable development of deep-sea exploration.

Downloads

Download data is not yet available.

References

Stel J H. Exploring and exploiting deep ocean space[J]. Ocean Literacy: Understanding the Ocean, 2021: 65-92.

Chen D, Qiao Y, Sun Y, et al. Human reliability assessment and risk prediction for deep submergence operating system of manned submersible under the influence of cognitive performance[J]. Ocean Engineering, 2022, 266: 112753.

Khodarahmi M, Maihami V. A review on Kalman filter models[J]. Archives of Computational Methods in Engineering, 2023, 30(1): 727-747.

Hemeida M G, Alkhalaf S, Senjyu T, et al. Optimal probabilistic location of DGs using Monte Carlo simulation based different bio-inspired algorithms[J]. Ain Shams Engineering Journal, 2021, 12(3): 2735-2762.

Behera N, Jeevitesh M S, Jose J, et al. Higher accuracy protein multiple sequence alignments by genetic algorithm[J]. Procedia Computer Science, 2017, 108: 1135-1144.

Anokye M, Cui X, Yang F, et al. CNN multibeam seabed sediment classification combined with a novel feature optimization method[J]. Mathematical Geosciences, 2024, 56(2): 279-302.

Song X, Liu Y, Xue L, et al. Time-series well performance prediction based on Long Short-Term Memory (LSTM) neural network model[J]. Journal of Petroleum Science and Engineering, 2020, 186: 106682.