A Comparison of Intelligent Models for Collision Avoidance Path Planning on Environmentally Propelled Unmanned Surface Vehicles

Abstract

Unmanned surface vehicles (USVs) are increasingly used for ocean missions and services aimed for safer, more efficient, and sustainable routine operations. Path planning is a key component of autonomy addressed to obstacle detection and avoidance. As a multi-optimization nonlinear problem, it should include computational time, optimal path, and maritime traffic standard procedures. This becomes even more challenging for USV technologies propelled by harvesting ocean energy from waves and wind. Sea current state and wind conditions significantly affect the USV energy consumption becoming the path planning approach key for navigation performance and endurance. To improve both aspects, an energy-efficient new path planning algorithm approach based on AI techniques for computing feasible paths in compliance with the Convention on the International Regulations for Preventing Collisions at Sea (COLREG) rules and taking energy consumption into account according to wind and sea current data is proposed.

Keywords

• unmanned surface vehicles (USV)
• obstacle avoidance (OA)
• path planning (PP)
• artificial neural network (ANN)
• random forest (RF)
• multiple logistic regressor (MLR)