Real-time collision avoidance planning for unmanned surface vessels based on field theory

Area of Science:

• Maritime technology
• Robotics
• Navigation systems

Background:

• Real-time collision avoidance is crucial for the increasing commercial application of unmanned surface vessels (USVs).
• Existing multi-vessel collision avoidance models suffer from structural complexity and high time consumption, limiting their efficiency.
• There is a need for simplified yet effective methods to enhance collision avoidance performance in complex maritime environments.

Purpose of the Study:

• To develop a simplified optimization model for collision avoidance in multi-vessel scenarios.
• To enhance the efficiency and performance of collision avoidance systems for USVs.
• To present a novel approach using field theory and an improved algorithm for generating optimal collision avoidance strategies.

Main Methods:

• Abstractly describing ship navigation using field theory to create a simplified optimization model.
• Developing a synthetic field combining virtual spatial electric and velocity fields to represent ship motion.
• Designing a simplified collision avoidance model integrated with an improved Electromagnetism-like mechanism (EM) algorithm.

Main Results:

• A simplified collision avoidance model based on field theory was successfully derived.
• The improved EM algorithm generated optimal collision avoidance strategies, considering energy loss.
• The algorithm demonstrated effectiveness and potential for advancing collision-free navigation in multi-vessel situations through validation under various layouts.

Conclusions:

• The proposed field theory approach significantly simplifies complex collision avoidance models for USVs.
• The integrated algorithm offers an efficient and effective solution for real-time collision avoidance in multi-vessel environments.
• This research paves the way for more advanced and reliable collision-free navigation systems in maritime autonomous operations.