Sparse Neural Circuit Policies for Underwater Vehicle Trajectory Tracking in Complex Environments

Description

This dataset contains the real-world experimental data associated with the manuscript entitled “Sparse Neural Circuit Policies for Underwater Vehicle Trajectory Tracking in Complex Environments”. The dataset includes lake-experiment trajectory logs, underwater vehicle state data, sensor measurements, controller outputs used in this study. The provided data support the evaluation of trajectory-tracking performance, actuator smoothness, and closed-loop control behavior under disturbed underwater operating conditions. The dataset is organized to facilitate reproducibility and further research on sparse neural-network-based underwater vehicle control systems.

Steps to reproduce

The dataset can be reproduced by deploying the trained sparse NCP controller on the experimental underwater vehicle platform described in the manuscript. The recorded lake-experiment data include vehicle states, actuator commands, and trajectory-tracking results under disturbed operating conditions.

Institutions

• Hohai University

  Jiangsu, Nanjing

Categories

Licence