LSTM-based Inverse Dynamics Learning for Franka Emika Panda

• Free software: MIT license

• Supplementary Information

• Repository

Features

• Project is based on Franka Emika Panda robot

• Motion parameter dataset acquiered with ROS and MoveIt

• LSTM-based neural network for inverse dynamic modelling

Repository Organisation

LSTM-based_inverse_dynamics_learning_for_franka_emika_panda
├── data: here you can find the datasets in different versions
│   ├── dataset_v1: first iteration
│   ├── dataset_v2: second iteration
│   ├── dataset_v3: third iteration
│   └── ...
├── src: here you find source code that is not directly related to data generation, preprocessing, analysis and neural network
│   ├── Data_Analysis: analysis scripts based on python
│   ├── Data_Preprocessing:
│   ├── Inverse_Dynamics: Deep Learning source code
│   │   ├── test results: figures and csv files
│   │   └── ...
│   ├── ROS: Data generation with ROS
│   └── ...
├── docs: everything related to the Supplementary Information pages
└── ...

Contents

• Data
  • Overview
  • Directory structure
  • How trajectories are generated
  • What is saved and how?
  • Robot joint limits
  • Contents
• Neural Network
  • Contents
• Credits
  • Development Lead
  • Contributors

Indices and tables

• Index

• Module Index

• Search Page