Clearo

A ROS2 package for controlling a floor-cleaning robot to follow a boustrophedon path, forked from Josh Newans' articubot_one project and modified for the Robotics Dojo 2024.

Overview

Clearo is a ROS2 package designed to control a floor-cleaning robot that navigates in a boustrophedon pattern (back-and-forth rows) for efficient cleaning. This project is based on Josh Newans' articubot_one, with modifications to support the Clearo robot's cleaning functionality. See Josh's YouTube playlist for foundational tutorials.

Quick Links

• Sources
• Hardware Configuration
• Software Dependencies
• Modified Parameters
  • Note
• Usage
  • Prerequisites
  • Simulation
  • Real Robot
  • Cleaning Path
• Misc.
• Future

Sources

• diffDriveArduino from RedstoneGithub's fork of Josh's diffdrive_arduino.
• sllidar and rplidar.
• joy_tester, serial, serial_motor_demo, and ros_arduino_bridge from Josh Newans.

Hardware Configuration

• Raspberry Pi running Ubuntu 22.04, dev machine running Ubuntu 22.04.
• RPLidar A1 M8 connected to Pi via USB.
• Arduino Mega running ROSArduinoBridge.ino, connected to Pi USB.
• 200 RPM motors with built-in encoders, connected as shown in encoder_driver.h.
• L298N motor driver, connected as shown in motor_driver.h.

Images of the robot: here, here.

Software Dependencies

• Ubuntu 22.04 with ROS2 Humble. Docker on Windows is not recommended due to network issues; native Ubuntu is preferred for performance.
• Required packages: twist-mux, ros-humble-teleop-twist-keyboard, ros-humble-teleop-twist-joy, slam-toolbox, ros-humble-navigation2, ros-humble-nav2-bringup, ros-humble-gazebo-ros-pkgs, ros-humble-xacro, ros2-control.
• tmux (or similar) is recommended for managing multiple terminals.

Modified Parameters

src/clearo/description

robot_core.xacro and robot.urdf.xacro

Adapted from Josh's code with some elements removed or modified.

gazebo_control.xacro

<wheel_separation>0.275</wheel_separation>
<wheel_diameter>0.085</wheel_diameter>

ros2_control.xacro

<!-- Arduino Mega -->
<param name="device">/dev/serial/by-id/usb-Arduino__www.arduino.cc__Arduino_Mega_2560_12148509806232150650-if00</param>
<param name="baud_rate">57600</param>
<param name="enc_counts_per_rev">988</param>

src/clearo/config/

mapper_params_online_async.yaml

resolution: 0.025 # (default 0.05)
loop_search_maximum_distance: 6.0 # (default 3.0, prevents "teleportation" during mapping)

nav2_params.yaml

• initial_pose: Added to save time during localization. Reset odom, launch online_async_launch.py, then set initial pose as (0, 0, 0) in RViz.
• controller_server: Uses MPPI controller (default: DWB).
• local_costmap & global_costmap: resolution: 0.025 (default 0.05), robot_radius: 0.025 (default 0.05).
• amcl: laser_max_range: 12.0, min_range: 0.3.
• velocity_smoother: feedback: "CLOSED_LOOP", velocities: [0.35, 0.0, -0], accelerations: [12.5, 0.0, 8.2].

my_controllers.yaml

wheel_separation: 0.275
wheel_radius: 0.0425

src/clearo/launch/

rplidar.launch.py

"serial_port": "/dev/serial/by-id/usb-Silicon_Labs_CP2102_USB_to_UART_Bridge_Controller_0001-if00-port0"

all files in src/sllidar_ros2/launch and src/rplidar_ros/launch

serial_port = LaunchConfiguration('serial_port', default='/dev/serial/by-id/usb-Silicon_Labs_CP2102_USB_to_UART_Bridge_Controller_0001-if00-port0') # rplidar

Note

• See Josh's [video](https://www.youtube.com/watch?v=4VVrTCnxvSw&list=PLunhqkrRNRhYAffV8JDi cambiato per il progetto Clearo.
• online_async_launch.py, localization_launch.py, navigation_launch.py, mapper_params_online_async.yaml, nav2_params.yaml were copied from slam-toolbox and nav2_bringup directories (/opt/ros/humble/share/) and modified (e.g., use get_package_share_directory('clearo') instead of nav2_bringup).
• Serial ports use /dev/serial/by-id for reliability.

Usage

Preliminary

1. Install ROS2 Humble.
2. Add source /opt/ros/humble/setup.bash to ~/.bashrc.
3. Install Nav2, SLAM Toolbox, colcon, and optionally set RMW_IMPLEMENTATION to cyclonedds:

   sudo apt install ros-humble-rmw-cyclonedds-cpp
   export RMW_IMPLEMENTATION=rmw_cyclonedds_cpp

4. Install additional packages:

   sudo apt install ros-humble-ros2-control ros-humble-ros2-controllers ros-humble-gazebo-ros2-control ros-humble-gazebo-ros-pkgs ros-humble-xacro ros-humble-twist-mux

5. Disable brltty service to free serial ports:

   systemctl stop brltty-udev.service
   sudo systemctl mask brltty-udev.service
   systemctl stop brltty.service
   systemctl disable brltty.service

6. Enable serial port access:

   sudo usermod -a -G dialout $USER

   Reboot after this step.
7. Clone the repository on both Pi and dev machine.
8. Edit src/clearo/description/ros2_control.xacro line 11 to match your microcontroller’s /dev/serial/by-id (check with ls /dev/serial/by-id).
9. Build the workspace:

   colcon build --symlink-install

   Ignore initial errors and run again if needed.
10. Source the workspace:

    source install/setup.bash

Simulation

Run on PC:

ros2 launch clearo launch_sim.launch.py use_sim_time:=true world:=./src/clearo/worlds/dojo2024

Set use_sim_time:=true in all relevant launch files.

Real Robot

1. On PC, run RViz:

   rviz2 -d ./src/comp.rviz

2. On Pi, launch the robot:

   ros2 launch clearo launch_robot.launch.py

3. On Pi, launch RPLidar:

   ros2 launch clearo rplidar.launch.py

   Alternatively, use:

   ros2 launch rplidar_ros rplidar_a1_launch.py

   or

   ros2 launch sllidar_ros2 sllidar_a1_launch.py

For Mapping Phase

1. On PC (preferred), run SLAM:

   ros2 launch clearo online_async_launch.py slam_params_file:=./src/clearo/config/mapper_params_online_async.yaml use_sim_time:=false

2. On PC or Pi, run teleop:

   ros2 run teleop_twist_keyboard teleop_twist_keyboard

   Or configure a gamepad using joy_tester and joystick.yaml (see Josh’s video), then:

   ros2 launch clearo joystick.launch.py

3. Drive the robot using u i o j k l m , . (keyboard) or gamepad until a satisfactory map appears in RViz.
4. In RViz, use the SLAM Toolbox panel to save the map (four files will be saved).

For Navigation Phase

1. For localization, edit mapper_params_online_async.yaml to set mode: localization and specify map_file_name (path to .data or .posegraph file). Then:

   ros2 launch clearo online_async_launch.py slam_params_file:=./src/clearo/config/mapper_params_online_async.yaml

   Or use:

   ros2 launch clearo localization_launch.py map:=path_to_map.yaml use_sim_time:=false

2. On PC (recommended), launch Nav2:

   ros2 launch clearo navigation_launch.py map_subscribe_transient_local:=true params_file:=./src/clearo/config/nav2_params.yaml use_sim_time:=false

3. Set waypoints in RViz for navigation.

Cleaning Path

To run the Clearo robot in a boustrophedon cleaning path:

ros2 launch clearo straight_line_waypoints.launch.py row_length:=3.0 num_rows:=8

Adjustable Parameters

• row_length: Length of each cleaning pass (default: 2.0m)
• lane_width: Width between passes (default: 0.3m, robot’s width)
• overlap: Overlap between passes (default: 0.05m or 5cm)
• num_rows: Number of rows to clean
• points_per_row: Waypoint density (more points = smoother path)

Example:

ros2 launch clearo straight_line_waypoints.launch.py row_length:=3.0 num_rows:=8

Misc.

• If RPLidar fails to launch, unplug/replug USB connections or try a different Type-C cable/power supply.
• Save RViz config with Ctrl+S.
• For simulations, use a powerful PC or split tasks across two PCs on the same network.
• Use floats in YAML files to avoid launch failures.
• Set ROS_DOMAIN_ID to avoid interference:

  export ROS_DOMAIN_ID=<unique-number>

• Use cyclonedds for better Nav2 compatibility:

  export RMW_IMPLEMENTATION=rmw_cyclonedds_cpp

Future

• Automate the boustrophedon path planning for dynamic environments.
• Optimize weaving in straight paths by adjusting angular velocity or controller settings.
• Reduce robot size (e.g., wheel separation) for better maneuverability.
• Improve power delivery to RPLidar motor (e.g., modify USB cable for 5V line).
• Explore high-resolution laser scans and costmaps.
• Test alternative controllers (RPP, TEB, Graceful) and planners (SMAC, ThetaStar).
• Incorporate IMU or depth camera for enhanced perception.
• Develop a simple BMS with MOSFET switch for battery safety.

License

This project is licensed under the same license as Josh Newans' articubot_one. See the LICENSE file for details.