Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a hybrid A*–fuzzy path planning algorithm tailored for unmanned surface vehicles (USVs), integrating fuzzy logic for collision avoidance. By processing satellite imagery to create a grid-based navigable map and applying a Takagi-Sugeno fuzzy inference model for enhanced decision-making near obstacles, the method addresses real-world navigation challenges effectively. Simulation results indicate that this approach not only improves safety and redundancy in USV navigation but also optimizes computational efficiency and memory usage, making it a robust solution for autonomous maritime operations.
A novel A*–fuzzy hybrid algorithm enhances USV navigation safety and efficiency, balancing optimal pathfinding with real-world obstacle avoidance.
: This paper presents a global path planning algorithm for unmanned surface vehicle (USV) based on an A* – fuzzy hybrid approach that integrates fuzzy collision-avoidance reasoning. To address path planning challenges in real-world scenarios, the proposed method first processes actual satellite imagery through image preprocessing and binarization to generate a grid-based navigable map. The grid map is then scaled to physical dimensions, and obstacle boundaries are dilated according to the USV minimum safety radius to ensure navigational clearance. Subsequently, the A* algorithm is employed for initial path search, while a Takagi-Sugeno (T-S) fuzzy inference model is applied to refine node selection near obstacles, enhancing local decision-making under uncertainty. Finally, the generated trajectory is simplified by retaining only critical waypoints, significantly reducing data storage requirements without compromising path quality. Simulation results demonstrate that the proposed algorithm improves both redundancy and safety in USV navigation, maintains high computational efficiency, and offers a practical solution for autonomous maritime decision-making. The approach effectively balances path optimality, obstacle avoidance capability, and memory efficiency, providing valuable support for safe USV operations.