Learning Morphology and Control for a Microrobot

This is the repository for the paper "Learning Morphology and Control for a Microrobot". More information can be found on our website here. Included are demos for running the experiments laid out in the paper.

Installation

Most of the pip dependencies can be installed with pip install -r requirements.txt. First activate the venv with source venv/bin/activate, then install GPy with

git clone https://github.com/SheffieldML/GPy.git
pip install -e GPy

Due to the complexity of managing different python versions and dependencies between different modules, we recommend running this repository on the associated docker image.

Running Experiments

Experiments can be run either locally or via Docker.

Running locally

First, start V-REP running locally. Then use main.py to call the relevant experiment, for example:

python main.py hpcbbo --init_uc=5 --init_cn=5 --uc_runs_per_cn=5 \
                      --batch_size=5 --total=5 --obj_f=1 --contextual

Or

python main.py random --init_uc=5 --total=5 --obj_f=1

Call python main.py -h for additional help

Running on Docker

Instructions for setting up Docker are included in the related repository.

Simulator Setup

Before running any of the experiments, make sure V-REP is open (see the V-REP documentation for troubleshooting issues with installation/booting). Scenes are automatically loaded and can be found in scenes/. The default simulator settings should work fine, but check that the following settings are correct:

• Physics engine: Bullet 2.78
• Time step: 50 ms

Citation

Should you find this code useful, please support us by citing our paper:

T. Liao, G. Wang, B. Yang, R. Lee, S. Levine, K. Pister, R. Calandra. 
Data-efficient Learning of Morphology and Controller for a Microrobot.
In IEEE Int. Conf on Robotics and Automatation, ICRA '19, Montreal, Canada, May
2019. 

bibtex:

@inproceedings{liao2019,
  title                 = {Data-efficient Learning of Morphology and Controller for a Microrobot},
  author                = {Liao, Thomas and Wang, Grant and Yang, Brian and Lee, Rene and Pister, Kristofer and Levine, Sergey and Calandra, Roberto},
  booktitle             = {2019 IEEE International Conference on Robotics and Automation},
  year                  = {2019},
  url                   = {https://arxiv.org/abs/1905.01334}
}