ResAlignNet aligns INS and DVL sensors to 0.8 degrees accuracy in 25 seconds using deep learning on synthetic-to-real data, reducing convergence time by 65% without prescribed motions or external positioning.
Adaptive identifica- tion on the group of rigid-body rotations and its appli- cation to underwater vehicle navigation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ResAlignNet: A Data-Driven Approach for INS/DVL Alignment
ResAlignNet aligns INS and DVL sensors to 0.8 degrees accuracy in 25 seconds using deep learning on synthetic-to-real data, reducing convergence time by 65% without prescribed motions or external positioning.