This library is a PyTorch implementation of dGPMP2 algorithm published in Differentiable Gaussian Process Motion Planning, ICRA 2020.
- Install Anaconda (Python 3.7)
- Clone the repository and run the following steps
conda create -n diff_gpmp2 python=3.7 conda env update -n diff_gpmp2 -f environment.yml conda activate diff_gpmp2
- Install OMPL with Python bindings (for generating data and expert trajectories)
- Dataset generation example
cd diff_gpmp2/datasets sh generate_2d_dataset.sh - Fully differentiable planning example
cd examples/ python diff_gpmp2_2d_example.py
Please use Github issue tracker to report bugs.
If you use this library in an academic context, please cite the following publication:
@article{bhardwaj2019differentiable,
title={Differentiable {G}aussian process motion planning},
author={Bhardwaj, Mohak and Boots, Byron and Mukadam, Mustafa},
journal={IEEE International Conference on Robotics and Automation (ICRA)},
year={2020}
}
dGPMP2 is released under the BSD license. See LICENSE file.