Underwater Submarine Path Planning Based on Artificial Potential Field Ant Colony Algorithm and Velocity Obstacle...

Area of Science:

• Marine robotics
• Autonomous navigation
• Path planning algorithms

Background:

• Underwater navigation for submarines presents significant challenges due to environmental complexity.
• Effective path planning is crucial for safe and efficient submarine operations.
• Existing algorithms struggle with real-time dynamic obstacle avoidance in 3D marine environments.

Purpose of the Study:

• To develop an advanced algorithm for submarine global path planning and local dynamic obstacle avoidance.
• To enhance the speed and accuracy of underwater path planning.
• To improve the safety and stability of submarine navigation in complex marine settings.

Main Methods:

• Proposed an Artificial Potential Field Ant Colony Algorithm (APF-ACO) integrating improved potential field and ant colony methods.
• Implemented an inflection point optimization algorithm to reduce path complexity.
• Utilized a Clothoid curve fitting algorithm for path smoothing.
• Developed a 3D dynamic obstacle avoidance algorithm based on the velocity obstacle method.

Main Results:

• The APF-ACO algorithm demonstrated faster convergence and superior path planning results compared to existing methods.
• Inflection point optimization significantly reduced path length and the number of turns.
• Clothoid curve fitting resulted in smoother, more stable submarine paths.
• The velocity obstacle-based algorithm effectively identified and avoided dynamic obstacles, preventing collisions.

Conclusions:

• The proposed APF-ACO algorithm offers a robust solution for submarine global path planning.
• The integrated approach effectively addresses local dynamic obstacle avoidance in complex underwater environments.
• Semi-physical simulations validated the algorithm's effectiveness and reliability for real-world submarine navigation.