.png){kind=icon}
.png){kind=icon}
Return
Genetic Algorithm-Based Weighted Comprehensive Image Matching Algorithm for Underwater Gravity Gradient-Aided Navigation
.png){kind=icon}
PRE
AI
DOI：10.1109/JOE.2024.3379484.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Gravity gradient-aided inertial navigation is of great significance to the development of autonomous underwater vehicles (AUV). The matching algorithm is the core technology of gravity gradient-aided navigation. In this work, a weighted gravity gradient comprehensive image matching algorithm based on a genetic algorithm (GA) is proposed. This approach addresses the limitation of the comprehensive image matching algorithm in meeting real-time requirements. The characteristics of the gravity gradient reference map are analyzed, then the parallel feature matching methods of mathematical statistics, gray level concurrence matrix texture in the spatial domain and wavelet texture in the frequency domain are presented. In addition, the contribution rates of the independent components of the gravity gradient tensor are calculated, and dynamic weights are applied to synthesize the gravity gradient components. Finally, the longitude and latitude coordinates of the AUV to be estimated are integrated into a chromosome, and the fitness function is designed to realize the matching and positioning of the gravity gradient image based on the GA. The experimental results show that the positioning accuracy of the proposed algorithm in the straight trajectory segment and the entire U-shaped trajectory is 82.4% and 73.8% higher than that of the existing algorithms, respectively. It can be concluded that the proposed algorithm is real-time and has superior position correction performance for inertial navigation system.
Keywords:
Gravity
Genetic algorithms
Real-time systems
Image matching
Inertial navigation
Heuristic algorithms
Biological cells
Genetic algorithm (GA)
gravity gradient-aided navigation
image matching
inertial navigation system (INS)
Journal
IF:
5.3
Papers:
2.6K
Citations:
7.4K