This paper introduces an innovative method for predicting ship trajectories using advanced machine learning techniques, specifically combining data encoding representation, the Attribute Correlation Attention (ACoAtt) module, and Long Short-Term Memory (LSTM) networks. The primary goal is to enhance navigational safety, prevent collisions, and improve vessel management efficiency by accurately forecasting ship trajectories. The method addresses the challenges of complex marine environments and data quality issues through the following steps: 1. **Data Encoding Representation**: AIS data is processed using binary encoding to improve representation efficiency and reduce complexity, while preserving key information. 2. **Attribute Correlation Attention Module**: The ACoAtt module captures complex relationships between dynamic ship attributes such as speed and direction using a multi-head attention mechanism, enhancing the model's understanding of implicit time series patterns. 3. **Long Short-Term Memory Network**: LSTM networks are utilized to model temporal dependencies in AIS data, effectively predicting future ship trajectories. The proposed method was evaluated using a historical AIS dataset, demonstrating superior performance in trajectory prediction accuracy and stability compared to other classic intelligent models and advanced models with attention mechanisms. The study also discusses the limitations of traditional methods and highlights the advantages of the proposed approach in handling complex maritime environments and variable sailing conditions.This paper introduces an innovative method for predicting ship trajectories using advanced machine learning techniques, specifically combining data encoding representation, the Attribute Correlation Attention (ACoAtt) module, and Long Short-Term Memory (LSTM) networks. The primary goal is to enhance navigational safety, prevent collisions, and improve vessel management efficiency by accurately forecasting ship trajectories. The method addresses the challenges of complex marine environments and data quality issues through the following steps: 1. **Data Encoding Representation**: AIS data is processed using binary encoding to improve representation efficiency and reduce complexity, while preserving key information. 2. **Attribute Correlation Attention Module**: The ACoAtt module captures complex relationships between dynamic ship attributes such as speed and direction using a multi-head attention mechanism, enhancing the model's understanding of implicit time series patterns. 3. **Long Short-Term Memory Network**: LSTM networks are utilized to model temporal dependencies in AIS data, effectively predicting future ship trajectories. The proposed method was evaluated using a historical AIS dataset, demonstrating superior performance in trajectory prediction accuracy and stability compared to other classic intelligent models and advanced models with attention mechanisms. The study also discusses the limitations of traditional methods and highlights the advantages of the proposed approach in handling complex maritime environments and variable sailing conditions.