ExoMy in Isaac Gym

Training ExoMy to navigate in an unstructured Mars environment

Install isaac-rover-mars-gym

Requirements

• Ubuntu 18.04, or 20.04.

• Python 3.6, 3.7, or 3.8

• Minimum recommended NVIDIA driver version: 470.74 (470 or above required)

Dependencies

# Docker
curl https://get.docker.com | sh \
&& sudo systemctl --now enable docker

# Setting up NVIDIA Container Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) \
 && curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - \
 && curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
# Install the nvidia-docker2 package (and dependencies) after updating the package listing:
sudo apt-get update
sudo apt-get install -y nvidia-docker2
# Restart the Docker daemon to complete the installation after setting the default runtime:
sudo systemctl restart docker

1. Download Isaac Gym

1. Download Isaac Gym from https://developer.nvidia.com/isaac-gym
2. Unzip Isaac Gym

3. nano isaacgym/docker/Dockerfile
4. Insert the follwing code at the bottom of the file and save.

RUN git clone https://github.com/AAU-RoboticsAutomationGroup/isaac_rover_mars_gym.git /home/gymuser/isaac_rover_mars_gym
RUN pip3 install -e /home/gymuser/isaac_rover_mars_gym/.
RUN git clone https://github.com/Toni-SM/skrl.git /home/gymuser/skrl
RUN pip3 install -e /home/gymuser/skrl/.
WORKDIR /home/gymuser/isaac_rover_mars_gym

5. bash docker/build.sh
6. docker run -it -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY —network=host —gpus=all —name=isaacgym_exomy_container isaacgym /bin/bash #11. bash docker/run.sh DISPLAYPORT
7. Enter container from different terminals --- sudo docker exec -it isaacgym_container bash

2. Run Isaac Gym

1. cd isaacgymenvs
2. python train.py

RL Architecture