A Fuzzy Cooperative Localisation Framework for Underwater Robotic Swarms

Area of Science:

• Robotics
• Artificial Intelligence
• Navigation Systems

Background:

• Underwater robotic swarms require robust localization, especially without acoustic sensors.
• Existing methods like Extended Kalman Filter have limitations in dynamic environments.
• Need for scalable and flexible navigation solutions for autonomous underwater vehicles (AUVs).

Purpose of the Study:

• To propose a holistic localization framework for underwater robotic swarms.
• To dynamically fuse multiple position estimates using a fuzzy decision support system.
• To provide navigation aids in the absence of traditional acoustic sensors.

Main Methods:

• Developed a fuzzy-based decision support system for dynamic data fusion.
• Integrated physics-based simulations including hydrodynamics and sensor characteristics.
• Validated the framework on a swarm of 150 AUVs.

Main Results:

• The fuzzy-based localization framework improved swarm mean localization error by 16.53%.
• Standard deviation of localization error was reduced by 35.17% compared to Extended Kalman Filter.
• Demonstrated superior performance over Extended Kalman Filter with round-robin scheduling.

Conclusions:

• The proposed framework offers simplicity, flexibility, and scalability for underwater swarm localization.
• Fuzzy logic provides an effective alternative for AUV navigation, enhancing accuracy and reliability.
• The framework can be extended to incorporate new localization methods by expanding the fuzzy rule base.