Vehicle Platoon Trajectory Prediction Under Traffic Oscillation: A Causal Physics-Informed Deep Learning Approach

In the data file, NGSIM refers to the NGSIM trajectory data mentioned in the paper.In the data file, WHUT contains the TVD trajectory data.All code is stored in the “code” folder.The filenames in the “code” folder correspond to the comparison model code and my complete model code, respectively.Code Execution Order:1. First, set up a virtual environment according to the requirements in `requirements.txt`.2. Use the code in the “Data Preprocessing” section to build a standard training dataset. Note that you must manually adjust the prediction step size in the dataset to match that of each code file to ensure consistency.3. Place the constructed data in a location accessible to all scripts, then run each script individually. To facilitate execution by readers, each .py file includes the model architecture, training code, testing code, validation code, and code for visualizing and saving results.