AUV Acceleration Prediction Using DVL and Deep Learning

TL;DR

Underwater navigation relies on accurate acceleration estimates to improve INS/DVL fusion. The authors present an end-to-end CNN-LSTM network that predicts the AUV acceleration vector from sequences of past DVL velocity measurements. Using real sea-trial data from the Snapir AUV near Haifa, they demonstrate up to 67.2% RMSE improvement over a model-based least-squares estimator. The results suggest that data-driven acceleration estimates can enhance navigation accuracy and enable reliable operation with lower-grade sensors, potentially improving INS convergence and fusion performance in challenging underwater environments.

Abstract

Autonomous underwater vehicles (AUVs) are essential for various applications, including oceanographic surveys, underwater mapping, and infrastructure inspections. Accurate and robust navigation are critical to completing these tasks. To this end, a Doppler velocity log (DVL) and inertial sensors are fused together. Recently, a model-based approach demonstrated the ability to extract the vehicle acceleration vector from DVL velocity measurements. Motivated by this advancement, in this paper we present an end-to-end deep learning approach to estimate the AUV acceleration vector based on past DVL velocity measurements. Based on recorded data from sea experiments, we demonstrate that the proposed method improves acceleration vector estimation by more than 65% compared to the model-based approach by using data-driven techniques. As a result of our data-driven approach, we can enhance navigation accuracy and reliability in AUV applications, contributing to more efficient and effective underwater missions through improved accuracy and reliability.

Figures (5)

• Figure 1: Our proposed AUV acceleration estimation approach.
• Figure 2: The CNN-LSTM network architecture for predicting AUV acceleration.
• Figure 3: The Snapir AUV during experimental trials.
• Figure 4: A representative training trajectory made by the Snapir AUV.
• Figure 5: The training and validation loss graph versus the number of epochs.