Parameter identification of hybrid-driven underwater glider based on differential evolution algorithm

With a view to the difficulty of the hybrid-driven underwater glider (HUG) hydrodynamic parameter identification, a method based on Differential Evolution (DE) algorithm which has designed adaptive factors is proposed. On the basis of dynamic modeling and analysis, this method is used to identify pa...

Full description

Saved in:
Bibliographic Details
Published in2021 International Conference on Artificial Intelligence and Electromechanical Automation (AIEA) pp. 338 - 342
Main Authors Zhang, Chao, Zhang, Hua, Zhang, Yong, Wang, Jian, Xiao, Donglin, Xu, Lingling, Chen, Wei, Cao, Yuanshan
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2021
Subjects
Adaptation models
Analytical models
differential evolution algorithm
Heuristic algorithms
hybrid-driven underwater gliders
hydrodynamic parameter
Hydrodynamics
Parameter estimation
parameter identification
Simulation
Unmanned underwater vehicles
Online AccessGet full text

Cover

More Information
Summary:With a view to the difficulty of the hybrid-driven underwater glider (HUG) hydrodynamic parameter identification, a method based on Differential Evolution (DE) algorithm which has designed adaptive factors is proposed. On the basis of dynamic modeling and analysis, this method is used to identify parameters from simulation observed data of the horizontal motions of HUG. After comparing with the actual value, the identification error is within the allowable range. The identification results are also compared with the results of Particle Swarm Optimization (PSO) algorithm and Least Square (LS) algorithm. The simulation results show that the parameter identification results using DE algorithm are significantly better than other comparison methods. The effectiveness of the algorithm is illustrated, and the method is not affected by the selection of initial parameter values, and has good robustness and global optimization characteristics. Finally, the model is verified by motion simulation, and the results show that the algorithm can effectively identify hydrodynamic parameters, which can provide a reference for engineering practice.
DOI:10.1109/AIEA53260.2021.00078