Xiaoyu006 / MATP-with-HEAT

This repo contains the code for our paper entitled "Multi-Agent Trajectory Prediction with Heterogeneous Edge-Enhanced Graph Attention Network".

MATP-with-HEAT

This repo contains the code for our paper entitled "Multi-Agent Trajectory Prediction with Heterogeneous Edge-Enhanced Graph Attention Network", IEEE T-ITS, 2022.

Install dependencies via pip.

pip install -r requirements.txt

Data preprocessing

The strucutre of the raw INTERACTION Dataset can be found in INTERACTION Dataset Tree.txt.

To obtain the sorted dataset, please refer to INTERPRET_challenge_regular_generalizability_track.

Run bash datapre_run.sh to process all the scenarios provided by the INTERACTION dataset.

Models

Base model -> Heat model -> HeatIR model.

Traning

Run python it_all_train.py -m Heat to train the one-channel HEAT-based trajectory predictor.

Validation

Citation

If you have found this work to be useful, please consider citing our paper:

@article{mo2022multi,
  title={Multi-agent trajectory prediction with heterogeneous edge-enhanced graph attention network},
  author={Mo, Xiaoyu and Huang, Zhiyu and Xing, Yang and Lv, Chen},
  journal={IEEE Transactions on Intelligent Transportation Systems},
  year={2022},
  publisher={IEEE}
}