A Real-Time Collision Avoidance Framework of MASS Based on B-Spline and Optimal Decoupling Control

Area of Science:

• Marine robotics
• Autonomous systems engineering
• Naval architecture

Background:

• Real-time collision avoidance is crucial for Maritime Autonomous Surface Ships (MASS) in diverse maritime operations.
• Existing algorithms struggle with multi-ship encounters in uncertain, limited-sensing environments.

Purpose of the Study:

• To propose a novel real-time collision avoidance framework for MASS.
• To ensure dynamic feasibility, high reliability, and safety in collision avoidance maneuvers.
• To address challenges posed by uncertain environments and limited sensor data.

Main Methods:

• Developed a B-spline-based collision avoidance trajectory search (BCATS) algorithm for effective trajectory generation.
• Implemented optimal decoupling control for waypoint-based trajectory optimization.
• Integrated collision risk assessment into the planning process.

Main Results:

• The BCATS algorithm effectively generates collision-free trajectories.
• Path-speed decoupling control reduces control costs and improves trajectory smoothness.
• Experimental validation using the Electronic Chart System (ECS) demonstrated system robustness.

Conclusions:

• The proposed framework provides robust and real-time collision avoidance capabilities for MASS.
• The system enhances navigational safety and efficiency in complex maritime scenarios.
• The B-spline and optimal decoupling control approach offers a promising solution for autonomous ship navigation.