Skip to content

Commit

Permalink
added missing NVIDIA Runtime instructions for Docker
Browse files Browse the repository at this point in the history
  • Loading branch information
vamsikalagaturu committed Nov 15, 2024
1 parent 899b98a commit 1d1e912
Showing 1 changed file with 10 additions and 1 deletion.
11 changes: 10 additions & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -80,6 +80,10 @@ source ~/eddie_ws/install/setup.bash

## Docker

### Setup NVIDIA Runtime for Docker

Install the [nvidia-container-toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html) and configure it for the Docker.

### Build the Docker Image

```bash
Expand Down Expand Up @@ -109,6 +113,10 @@ sudo docker run -it --rm --runtime=nvidia --gpus all \
ros2 launch eddie_gazebo eddie_gazebo.launch.py
```

- To move the base using `Teleop`, select the `Teleop` from the `Tools` menu in the Gazebo GUI.
- Change the `topic` to `/model/robot/cmd_vel` and Enter.
- Then, you can move the robot using the selected controls.

1. By default, the `base_controller` for controlling the mobile base is a velocity controller and
the `arm_controller` for controlling the Kinova arms is a `trajectory_controller`. You can
customize the base_controller and arm_controller by providing launch arguments, for example:
Expand Down Expand Up @@ -153,7 +161,8 @@ Choose robot component to command using
```
---
> [!WARNING] The below instructions are not yet tested and may not work as expected.
> [!WARNING]
> The below instructions are not yet tested and may not work as expected.
The behaviour of this executable can be changed to match the controllers used when launching
`eddie_gazebo.launch.py`. This can be done by providing `arm_controller` and `base_controller`
Expand Down

0 comments on commit 1d1e912

Please sign in to comment.