DEPRICATION NOTE: As I'm currently restructuring the project, this README is not quite up-to-date with the changes. Working on it!

Primary Motor Cortex Documentation

This repository is part of the Roboost project. For more information visit the project website and the Roboost GitHub organization.

Description

This repository contains the code for the primary motor cortex of the robot. It is a PlatformIO project responsible for the control of a robot's motors. The micro-ROS framework is used to connect the robot to a ROS2 network. The robot can then be controlled using any ROS2 node publishing to /cmd_vel, such as the teleop_twist_keyboard.

As example application, here is a video of the Roboost robot being controlled using the teleop_twist_keyboard:

Configuration

For pin and hardware configuration (things like wheel dimensions, track width, etc.), changes can be made in the conf_hardware.h file. To change the hardware drivers, the core.cpp file needs to be modified. The conf_network_example.h file can be used to configure the network settings. Rename the file to conf_network.h and fill in the apropriate values to use it.

As mentioned in the Installation section, the micro-ROS agent can be configured to use either a wifi or serial connection. The default configuration is to use a wifi connection. To use a serial connection, the platformio.ini file needs to be modified. Remove board_microros_transport = wifi and adapt the core.cpp file to use the serial connection.

Supported Hardware

The primary motor cortex is designed to be modular. This means that the code can be easily adapted to different hardware configurations. The following components can be configured:

Motor Drivers

• L298N
• VESCs (Currently in development)

Motor Controllers

• Simple Controller
  • Directly controlling the motors
• PID Controller
  • Using half quad encoders as feedback
  • Different tuning methods (Currently in development)

Kinematics

• 4-Wheeled Meccanum Drive
• 3-Wheeled swerve drive (Currently in development)

Installation

The project can be installed as a standalone project or as part of the Roboost-Cerebrum repository. In case of the latter, the micro-ROS agent will be executed automatically as a Docker container. If you only want to use the primary motor cortex, you can install it as a standalone project. In this case, you will need to install the micro-ROS agent manually.

Standalone Installation

Prerequisites

• VSCode
• PlatformIO
• ROS2 Humble
• micro-ROS Agent

Building

To install the project, clone the repository and open it in VSCode. The project is written for the ESP32 microcontroller, so you will need to install the PlatformIO extension for VSCode. Once installed, you can build and upload the code to the microcontroller using the PlatformIO extension. All dependencies will be installed automatically.

Usage

To use the project, you will need to build the micro-ROS agent and run it on your host machine. The agent will communicate with the microcontroller via a wifi or serial connection. The agent will then publish and subscribe to ROS2 topics, which can be used to control the robot.

Micro-ROS Agent Installation

# Source the ROS 2 installation
source /opt/ros/$ROS_DISTRO/setup.bash

# Create a workspace and download the micro-ROS tools
mkdir microros_ws
cd microros_ws
git clone -b $ROS_DISTRO https://github.com/micro-ROS/micro_ros_setup.git src/micro_ros_setup

# Update dependencies using rosdep
sudo apt update && rosdep update
rosdep install --from-paths src --ignore-src -y

# Build micro-ROS tools and source them
colcon build
source install/local_setup.bash

# Download micro-ROS-Agent packages
ros2 run micro_ros_setup create_agent_ws.sh

# Build step
ros2 run micro_ros_setup build_agent.sh
source install/local_setup.bash 

Once the micro-ROS agent is built, run following command to make the robot accessible over the network:

cd microros_ws
source install/local_setup.bash
ros2 run micro_ros_agent micro_ros_agent udp4 -p 8888

Alternatively, you can use the following command to use a serial connection:

ros2 run micro_ros_agent micro_ros_agent serial --dev /dev/ttyUSB0

Note that depending on the communication method, you will need to modify the firmware of the microcontroller. Per default, the firmware is configured to use UDP over wifi.

Controlling the Robot

Once the micro-ROS agent is running, you can run the robot using the following command:

ros2 run teleop_twist_keyboard teleop_twist_keyboard

This will allow you to control the robot using the keyboard. You can also use any other ROS2 node to control the robot. Per default, the robot will subscribe to the /cmd_vel topic and publish to the /odom topic.

Roboost-Cerebrum Installation

When installing the project as part of the Roboost-Cerebrum repository, the micro-ROS agent will be executed automatically as a Docker container. This means that you will not need to install the agent manually. In this case, you only need to adapt the project according to your hardware configuration, upload the code to the microcontroller and run the Roboost-Cerebrum Docker container. For more information, visit the Roboost-Cerebrum repository.

Here are commands to get you started:

Start the micro-ROS agent:

docker run -it --net=host microros/micro-ros-agent:humble udp4 -p 8888

Or using a serial connection:

docker run -it --device=/dev/ttyUSB0 microros/micro-ros-agent:humble serial --dev /dev/ttyUSB0

Starting the TF broadcaster:

ros2 launch tf_broadcast_package mecanum_tf_broadcast_launch.py

Starting the joint state publisher:

ros2 run joint_state_publisher_gui joint_state_publisher_gui

Starting Rviz2:

ros2 launch rviz2 rviz2.launch.py

In Rviz2, set the fixed frame to odom and add a TF and RobotModel display. You should now see the robot in Rviz2. Note that you need to select the robot description in the RobotModel to see the robot.

TODO

• Add documentation of the motor shield PCB
• Add support for VESCs
• Add support for swerve drive
• Test accuracy of odometry
• Calculate odometry covariance
• Add parameter server for dynamic reconfiguration