Learning of Multi-Context Models for Autonomous Underwater Vehicles

TL;DR

This paper presents a method using LSTM networks to identify multiple dynamic contexts of an autonomous underwater vehicle, improving robustness and scalability in marine robotics applications.

Contribution

It introduces a novel LSTM-based architecture for multi-context model learning in AUVs, capable of handling noisy data and large datasets.

Findings

High classification accuracy achieved

Robustness against noise demonstrated

Efficient scaling on large datasets

Abstract

Multi-context model learning is crucial for marine robotics where several factors can cause disturbances to the system's dynamics. This work addresses the problem of identifying multiple contexts of an AUV model. We build a simulation model of the robot from experimental data, and use it to fill in the missing data and generate different model contexts. We implement an architecture based on long-short-term-memory (LSTM) networks to learn the different contexts directly from the data. We show that the LSTM network can achieve high classification accuracy compared to baseline methods, showing robustness against noise and scaling efficiently on large datasets.