/CNMP

Conditional Neural Movement Primitves

Primary LanguageJupyter Notebook

Conditional Neural Movement Primitives

This branch includes PyTorch implementation. For Keras/Tensorflow implementation please refer to Keras branch

CNMP.png

This repository includes the CNMP implementation and example material from the experiments of the paper:

M. Yunus Seker, Mert Imre, Justus Piater, Emre Ugur, Conditional Neural Movement Primitives, RSS 2019

What's New ?

  • PyTorch implementation of the framework with:
    • Multi-step/whole-trajectory prediction support
    • Easier data/model organization with a more basic batch-based approach
    • Recreated Experiment 1 from the paper

Requirements:

  • Python >=3.6

  • Jupyter Notebook

  • PyTorch >=1.6.0

Contact Information:

Cite CNMP:

You can use the BibTex below to cite CNMP

@INPROCEEDINGS{Ugur-RSS-19,
    AUTHOR    = {Muhammet Yunus Seker AND Mert Imre AND Justus Piater AND Emre Ugur}, 
    TITLE     = {Conditional Neural Movement Primitives}, 
    BOOKTITLE = {Proceedings of Robotics: Science and Systems}, 
    YEAR      = {2019}, 
    ADDRESS   = {FreiburgimBreisgau, Germany}, 
    MONTH     = {June}, 
    DOI       = {10.15607/RSS.2019.XV.071}}