Skip to content

About Electrode

Much like the name implies from the medical term, Electrode is the introspection and user interface to CogniPilot's Cranium and Cerebri.

Electrode is primarily a ROS 2 workspace that allows for controlling robotic platforms through either foxglove or RVIZ 2 backends.

Electrode uses a joystick whether (virtual or physical) to control platforms.

Foxglove-studio is the recommended backend for Electrode.

Do I have to use a ROS 2 workspace based Electrode to control platforms?

No, there is a way to use Electrode with foxglove and the CogniPilot extensions on windows and linux that only uses websockets and does not require ROS 2 to be present on the system.

Foxglove-studio backend for Electrode.

By default Electrode uses foxglove-studio as the backend, for foxglove-studio to connect to a physical platform or simulation foxglove_bridge must be running first on the target system.

Install foxglove-studio and the foxglove CogniPilot Joystick extension

Run the build_foxglove script after following initial development system setup and follow the prompts.

Run build_foxglove script:
build_foxglove

Proceed to using Electrode with foxglove-studio with the platforms specific layout:

RVIZ 2 backend for Electrode

Electrode can be optionally run with the RVIZ 2 backend, however, it requires a physical joystick device for input.

Electrode with RVIZ 2:
ros2 launch electrode electrode.launch.py rviz2:=true
By default RVIZ 2 uses B3RB as the vehicle platform but can be changed with vehicle:=<platform-name>

Joy Input

Any compatible Joy input can be used to control the platform.

Joy input message example for B3RB

The sensor_msgs/Joy message on topic /cerebri/in/joy controls the b3rb, inspect the message to see the field mappings.

ros2 topic echo /cerebri/in/joy
---
header:
  stamp:
    sec: 0
    nanosec: 0
  frame_id: ''
axes:
- 0.0
- 0.0      #(1)
- 0.0
- 0.0      #(2)
buttons:
- 0        #(3)
- 0        #(4)
- 0        #(5)
- 0        #(6)
- 0        #(7)
- 0        #(8)
- 0        #(9)
- 0        #(10)
---

  1. Throttle in  Manual Mode .
  2. Steering in  Manual Mode .
  3.  Manual Mode 
  4.  Auto Mode 
  5.  CMD_VEL Mode 
  6.  Calibration Mode 
  7. Front Lights On
  8. Front Lights Off
  9. Disarm
  10. Arm

Foxglove CogniPilot Joystick extension

Install foxglove-studio and build the foxglove CogniPilot Joystick extension

Make sure to have first run the build_foxglove script at some point and follow the prompts.

Run build_foxglove script:
build_foxglove

The foxglove CogniPilot Joystick extension allows users to control a platform without the need for a physical joystick.

Using the CogniPilot Joystick extension in foxglove.

Use is simple, click the corresponding buttons to perform their actions. An example on B3RB, click and drag the smaller circle around for throttle (up/down) and steering (left/right).

CogniPilot Joystick extension in foxglove.

Can I still use a physical josytick with foxglove?

A physical Joystick can still be used with foxglove but it requires closing the CogniPilot Joystick panel that is automatically opened when importing a layout file like the b3rb.json layout and running electrode with joy:=true

Run Electrode with foxglove and physical joystick
ros2 launch electrode electrode.launch.py joy:=true
Remove CogniPilot Joystick extension from b3rb.json layout in foxglove.

Example of using electrode with a physical joystick for B3RB.

While most physical joysticks on the market can be made to work with ROS 2, this example highlights a Logitech F310.

Logitech F310 button layout.

F310

Selecting a Mode

  •  A : Manual Mode
  •  X : CMD_VEL Mode (nav2)
  •  B : Auto Mode (other cmd_vel)
  •  Y : Calibration Mode

Arming

  • START: arm
  • BACK: disarm

Manual Mode

  • Left stick Up/Down: Throttle
  • Right stick Left/Right: Steering

Front Lights

  • LB: Front lights on
  • RB: Front lights off

Using Electrode without a ROS 2 workspace

To use Electrode without ROS 2 on a system use foxglove-studio.

  1. Download the latest foxglove-studio for your system (Linux/Windows) and install it.

    If using with linux only use debian installer.

    Do not use the snap version, only follow instructions for debian linux install.

  2. Download the latest foxglove-cognipilot-joy extension asset cognipilot.cognipilot-joystick-<version>.zip for Brave.

  3. Unzip compressed folder and put it in correct path for operating systems foxglove-studio.
    On Linux:

    Move and extract extension on Linux:
    mkdir -p ~/.foxglove-studio/extensions
    mv <path-to>/cognipilot.cognipilot-joystick-*.zip ~/.foxglove-studio/extensions
    cd ~/.foxglove-studio/extensions
    unzip cognipilot.cognipilot-joystick-*.zip
    
    On Windows:

    1. Go to Users/<User>/ and see if there is a .foxglove-studio/extensions folder, enable seeing hidden folders if not already enabled. If the folder path exists proceed to step d., if not continue with step b.
    2. Create a .foxglove-studio folder in Users/<User>/.

      Creating a .foxglove-studio folder Users/<User>/.

      Creating a .foxglove-studio folder in Users/<User>/.

    3. Create an extensions folder in Users/<User>/.foxglove-studio.

      Creating an extensions folder in Users/<User>/.foxglove-studio.

      Creating an extensions folder in Users/<User>/.foxglove-studio.

    4. Go to location where cognipilot.cognipilot-joystick-<version>.zip was downloaded right click the .zip and select Extract All...

      Select Extract All... on cognipilot.cognipilot-joystick-<version>.zip.

      Select Extract All... on cognipilot.cognipilot-joystick-<version>.zip.

    5. Provide extraction path to Users/<User>/.foxglove-studio/extensions and press extract.

      Provide extraction path to Users/<User>/.foxglove-studio/extensions.

      Provide extraction path to Users/<User>/.foxglove-studio/extensions.

  4. Download desired platform layout for foxglove by selecting the <robot>.json file on GitHub and pressing the download icon.

    Example of downloading b3rb.json layout for foxglove.

    Example of downloading b3rb.json layout for foxglove.

  5. Open the foxglove-studio application.

  6. Connect foxglove-studio to the correct foxglove_bridge websocket's ip address and port in the format ws://<ip-address>:<port>.

    Connecting foxglove-studio to the correct foxglove_bridge websocket's ip address and port.

    Connecting foxglove-studio to the correct foxglove_bridge websocket's ip address and port.