This is the code for NeurIPS'22 paper Learning-based Motion Planning in Dynamic Environments Using GNNs and Temporal Encoding
conda create -n Dynamic-GNN python=3.8
conda activate Dynamic-GNN
# install pytorch, modify the following line according to your environment
conda install pytorch torchvision cudatoolkit=11.3 -c pytorch
# install torch geometric, refer to https://github.com/pyg-team/pytorch_geometric
conda install pyg -c pyg
pip install pybullet transforms3d matplotlib
Please put those folders into /testcase into the root directory of this repo.
# generate test cases for 2arms using multi-processing
python oracle/generate_testcases.py