Maritime Autonomous Surface Ships (MASS) are deemed as the future of maritime transport. They can provide an effective solution, to reducing human errors and improving safety at sea. Although showing attractiveness, the applications of MASS have revealed some, problems in practice, among which MASS navigation safety presents a prioritised concern. To ensure maritime safety, rational route, planning for MASS is evident as the most critical step to avoiding any relevant collision accidents. This paper proposes a holistic, framework for the unsupervised route planning of MASS based on Automatic Identification System (AIS) and machine learning, methods, involving new feature measurement, pattern extraction, and route planning algorithms. Firstly, the features among trajectories, are captured by an automatic and adaptive Dynamic Time Warping (AADTW) approach to measure their similarities and provide the, criteria for trajectory clustering. Next, the trajectory patterns are mined based on the Automatic Spectral Clustering (ASC) algorithm, to discover all the hidden information. Finally, the representative trajectories are derived based on an optimal function, and the planned, routes for MASS are generated based on a feature extraction algorithm. The route planning method for MASS will provide insightful, solutions to aiding and realising autonomous and safe navigation at sea.