https://github.com/idiap/ttgo
A PyTorch implementation of TTGO algorithm and the applications presented in the paper "Tensor Train for Global Optimization Problems in Robotics"
https://github.com/idiap/ttgo
Last synced: 9 months ago
JSON representation
A PyTorch implementation of TTGO algorithm and the applications presented in the paper "Tensor Train for Global Optimization Problems in Robotics"
- Host: GitHub
- URL: https://github.com/idiap/ttgo
- Owner: idiap
- License: gpl-3.0
- Created: 2022-06-21T08:30:04.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2024-11-05T08:23:27.000Z (over 1 year ago)
- Last Synced: 2025-03-06T03:04:23.721Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 16.5 MB
- Stars: 18
- Watchers: 5
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: COPYING
Awesome Lists containing this project
README
# TTGO: Tensor Train for Global Optimization Problems in Robotics
A PyTorch implementation of TTGO algorithm and the applications presented in the paper "Tensor Train for Global Optimization Problems in Robotics "
Website: https://sites.google.com/view/ttgo/home
Paper: https://arxiv.org/pdf/2206.05077.pdf
### Pre-requistes
- Install the tntorch library from: https://github.com/rballester/tntorch (pip install tntorch)
- Pybullet (only required for visualization of robotics applications): https://pypi.org/project/pybullet/
- RoMa (only required robotic applications; for quarternion calculations): https://naver.github.io/roma/
### Overview
- *./ttgo.py*: the TTGO algorithm is defined in this class
- *./function_optimization/*: includes the application of ttgo for optimization of several benchmark functions
- Recommendation: try these notebooks first to understand the approach
- *./toy_robots/*: application of ttgo for simple toy models of robotics problems (planar manipulator IK and reaching tasks)
- *./manipulator/*: application of ttgo for IK and reaching tasks with some standard manipulators
Note: All the implementations are fully compatible for use with GPU. For faster computation, it is highly recommended to use GPU
For any questions, contact the author Suhan Shetty