ROS2 Package for Freedom Rover Unit Host Development

Description

Fork of linorobot2. This repository holds FRU's modifications that are proprietary to the FRU-bot. The FRU-bot consists of an lds-01 2D lidar, MPU 6050, TT motors, dual L298N motor-driver, IR speed sensor, and ESP32-WROOM-DevKitC. Power is delivered via a custom pcb board consisting of 7.4V, 5V, and 3.3V headers for fascilitated attachment of hardware interfaces. The robot chassis was sourced from an external party on printables; although, the final chassis height was lowered for improved kinematics. Agent is intended to run on a remote docker container communicating wirelessly with esp32, but manual installation instructions are available.

Installation

1. Recommended installation via installation script.
   • Creates ros2 workspace in $HOME/FRU_ws by default.
   • Assumes docker in sudo usr group to pull micro-ROS agent docker image.

   source /opt/ros/<ros_distro>/setup.bash
   cd /tmp
   wget https://raw.githubusercontent.com/FreedomRoverUnits/FRU-bot/humble/install_FRU_bot.bash
   bash install_FRU_bot.bash <FRU_bot_ws>
   source ~/.bashrc
   

Quickstart

For Simulation steps skip to step 3.

1. Connect to robot wifi.

2. Start micro-ros agent via docker in another terminal.

   source /opt/ros/<your_ros_distro>/setup.bash
   
   docker run -it --rm --net=host microros/micro-ros-agent:humble udp4 --port 8888 -v6
   

3. Start EKF node and joint state publisher, set rviz flag to true to visualize robot description. Creates namespace "FRU_bot0" for all topics to be published under (except /tf topics, must set remap_tf to True).

   source install/setup.bash
   ros2 launch fru_bot_bringup bringup_default.launch.py rviz:=true use_loc:=True use_ns:=True idx:=0
   

   Or for simulation:

   ros2 launch fru_bot_bringup bringup_default.launch.py rviz:=true use_ns:=True idx:=0 sim:=True
   

4. Control the robot manually via teleop twist commands.

   python3 teleop_twist_keyboard_FRU.py --agents 1
   

Creating a map.

1. Run SLAM toolbox.

   ros2 launch fru_bot_navigation slam.launch.py rviz:=true
   

2. Move the robot to build the map, and save it.

   cd fru_bot/fru_bot_navigation/maps
   ros2 run nav2_map_server map_saver_cli -f <map_name> --ros-args -p save_map_timeout:=10000.
   

Autonomous Navigation.

1. First change MAP_NAME in navigation.launch.py to the name of the map you created. Or change it dynamically.

   cd <robot_ws>
   colcon build
   
   ros2 launch fru_bot_navigation navigation.launch.py map:=<path_to_map_file>/<map_name>.yaml
   

2. Run Nav2 package.

   ros2 launch fru_bot_navigation navigation.launch.py
   

Optional parameter for loading maps:

• map - Path to newly created map <map_name.yaml>.

Optional parameters for simulation on host machine:

• sim - Set to true for simulated robots on the host machine. Default value is false.
• rviz - Set to true to visualize the robot in RVIZ. Default value is false.

navigation.launch.py will continue to throw this error - 'Timed out waiting for transform from base_link to map to become available, tf error: Invalid frame ID "map" passed to canTransform argument target_frame - frame does not exist' until the robot's pose has been initialized.

• Refer to Nav2 tutorial for more info.

Troubleshooting Guide (Directly from linorobot2)

1. The changes I made on a file are not taking effect on the package configuration/robot's behavior.

• You need to build your workspace every time you modify a file:

  cd <ros2_ws>
  colcon build
  #continue what you're doing...
  

2. [slam_toolbox]: Message Filter dropping message: frame 'laser'

• Try to up transform_timeout by 0.1 in fru_bot_navigation/config/slam.yaml until the warning is gone.

3. target_frame - frame does not exist

• Check your <robot_type>.properties.urdf.xacro and ensure that there's no syntax errors or repeated decimal points.

4. Weird microROS agent behavior after updating the Linux/ROS

• Don't forget to update the microROS agent as well after your updates. Just run:

  bash update_microros.bash