Abstract: This paper presents an AUV global path planning scheme using case-based learning algorithm. Case-based learning is relatively a new approach to path planning. Case-based learning is learning and reasoning from past episodic information about the environment. A new and suitable solution is generated by retrieving and adapting and old one, which approximately matches the current situation. In this research, the algorithm includes these parts: establishment of path attribute extracted from its features, matching and retrieving case from the case library, and updating the case library. The effectiveness of the algorithm is verified through simulations.