Multibody System-Based Adaptive Formation Scheme for Multiple Under-Actuated AUVs.

Area of Science:

• Robotics
• Control Systems
• Underwater Acoustics

Background:

• Autonomous underwater vehicles (AUVs) coordination and formation are critical for complex missions.
• AUVs face challenges due to under-actuation and limited acoustic communication bandwidth.
• Existing frameworks struggle with robust formation control and accurate localization under communication constraints.

Purpose of the Study:

• To develop a novel framework for multiple AUV formation and communication link.
• To enhance cooperative localization accuracy and formation control robustness.
• To validate the proposed methods through extensive simulations and experiments.

Main Methods:

• A multibody system concept was used to establish the formation and communication framework.
• Adaptive radial basis function (RBF) strategies were employed for control and localization.
• Acoustic communication packet transmission schemes, topology, and protocols were investigated.
• Reinforcement learning RBF neural networks estimated cooperative localization errors caused by packet loss.
• An adaptive RBF formation scheme with potential energy functions was designed for formation cruising.

Main Results:

• The proposed framework effectively manages multiple AUV formation and communication links.
• Adaptive RBF strategies demonstrated robust control and accurate localization.
• The system showed resilience to packet loss, with errors estimated by RL-RBF networks.
• Formation cruising was successfully realized using the designed adaptive RBF scheme.

Conclusions:

• The developed framework and adaptive RBF strategies significantly improve AUV coordination and formation.
• The methods provide a viable solution for AUVs operating with limited communication.
• The study validates the effectiveness of the proposed approaches in complex underwater scenarios.