pypose 0.1.6

To connect classic robotics with modern learning methods.

pypose

To connect classic robotics with modern learning methods.

---

Current Features

LieTensor

• Lie Group: SO3, SE3, Sim3, RxSO3
• Lie Algebra: so3, se3, sim3, rxso3

Modules

• System
• IMUPreintegration
• ......

Second-order Optimizers

• GaussNewton
• LevenbergMarquardt
• ......

Efficiency-based design

• We support parallel computing for Jacobian of LieTensor.

Efficiency comparison of Lie group operations on CPU and GPU (we take Theseus performance as 1×).

More information about efficiency comparison goes to the paper.

Getting Started

Installing

Install from pypi

pip install pypose

From source

git clone https://github.com/pypose/pypose.git && cd pypose
python setup.py develop

For Early Users

1. Requirement:

On Ubuntu, MasOS, or Windows, install PyTorch, then run:

pip install -r requirements/main.txt

2. Install locally:

git clone  https://github.com/pypose/pypose.git
cd pypose && python setup.py develop

3. Run Test

pytest

For Contributors

1. Make sure the above installation is correct.

2. Go to CONTRIBUTING.md

Citing PyPose

If you use PyPose, please cite the paper below.

@article{wang2022pypose,
  title   = {{PyPose: A Library for Robot Learning with Physics-based Optimization}},
  author  = {Chen Wang, Dasong Gao, Kuan Xu, Junyi Geng1, Yaoyu Hu, Yuheng Qiu, Bowen Li, Fan Yang, Brady Moon, Abhinav Pandey, Aryan, Jiahe Xu, Tianhao Wu, Haonan He, Daning Huang, Zhongqiang Ren, Shibo Zhao, Taimeng Fu, Pranay Reddy, Xiao Lin, Wenshan Wang, Jingnan Shi, Rajat Talak, Han Wang, Huai Yu, Shanzhao Wang, Ananth Kashyap, Rohan Bandaru, Karthik Dantu, Jiajun Wu, Luca Carlone, Marco Hutter, Sebastian Scherer},
  journal = {arXiv},
  year    = {2022}
}