Skip to content

Latest commit

 

History

History
52 lines (35 loc) · 1.5 KB

README.md

File metadata and controls

52 lines (35 loc) · 1.5 KB

iBot-Hexapod_RL

Hexapod V3

  • Made it into a GYM Environment.

Hexapod V2

  • Added a Larger Plane
  • Correct the Motor Torque Outputs
  • Adjusted the Damping and Stiffness factor of the Bot
  • Added Transformation matrix for Easier SIM2REAL deployment.
  • General Stability of the bot improved and resembles much more closer to the real bot.

Hexapod V1

  • Modeled using STL files.
  • Hexapod Model with Basic Functionality and Python Interface

Installation Instructions

Prerequisites for Mujoco

  • Install following packages and edit bashrc
sudo apt-get install -y libglew-dev libosmesa6-dev patchelf
conda install -c anaconda patchelf
echo 'export LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libGLEW.so' >> ~/.bashrc
source ~/.bashrc

Install MuJoCo

  1. Download the MuJoCo version 2.0 binaries for Linux or OSX.

  2. Unzip the downloaded tar.gz file and copy the mujoco210 directory into ~/.mujoco

  3. Add following line to .bashrc: export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/$USER/.mujoco/mujoco210/bin

  4. Install gym version 0.21.0 and mujoco-py pip install 'mujoco-py<2.2,>=2.1' 'gym==0.21.0'

  5. Clone this repository and to test if everything works, run the following commands:

cd ~/Hexapod_V3
python hexapod_triple_gait.py

It should open up a window, with the Hexapod walking using Triple Gait... Check out the code for all the class implementations..