BLENDS augments Bayesian smoothing with a transformer for learned covariance adaptation and state correction, achieving up to 25.6% better position RMSE on quadrotor GNSS outage data and compensating for GNSS bias versus RTK truth.
Bayesian Learning-Enhanced Navigation with Deep Smoothing for Inertial-Aided Navigation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
BLENDS: Bayesian Learning-Enhanced Deep Smoothing for GNSS-Denied Environments
BLENDS augments Bayesian smoothing with a transformer for learned covariance adaptation and state correction, achieving up to 25.6% better position RMSE on quadrotor GNSS outage data and compensating for GNSS bias versus RTK truth.