Learning of Multi-Context Models for Autonomous Underwater Vehicles
This work addresses model learning for marine robotics, but it is incremental as it applies an existing LSTM method to a specific domain problem.
The paper tackles the problem of identifying multiple contexts for autonomous underwater vehicle models by using LSTM networks to learn from data, achieving high classification accuracy compared to baselines with robustness against noise and efficient scaling on large datasets.
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.