rosbot-navigation

Setting target point in RViz (running on your PC) for ROSbot XL driving autonomously thanks to Nav2 stack. Works both in the LAN network and over the Internet.

Quick Start

PC

Clone this repository:

git clone https://github.com/husarion/rosbot-xl-navigation.git

Create a map of the environment using therosbot-mapping project template.

Copy rosbot-xl-mapping/maps folder to rosbot-xl-navigation/maps (the same directory as this README).

Check your configs in .env file

LIDAR_SERIAL=/dev/ttyRPLIDAR

# for RPLIDAR A2M8:
# LIDAR_BAUDRATE=115200
# for RPLIDAR A2M12 and A3:
LIDAR_BAUDRATE=256000

DDS_CONFIG=DEFAULT
# DDS_CONFIG=HUSARNET_SIMPLE_AUTO

# RMW_IMPLEMENTATION=rmw_fastrtps_cpp
RMW_IMPLEMENTATION=rmw_cyclonedds_cpp

MECANUM=True
# MECANUM=False

# AMCL_PARAMS=amcl_params.yaml
AMCL_PARAMS=amcl_params_mecanum.yaml

# NAV2_PARAMS=nav2_params.yaml
# NAV2_PARAMS=nav2_params_mecanum.yaml
NAV2_PARAMS=nav2_params_pure_pursuit.yaml

Notes:

• Usually RPLIDAR is listed under /dev/ttyUSB0, but verify it with ls -la /dev/ttyUSB* command.
• If you have RPLIDAR A3 or A2M12 (with violet border around the lenses) set: LIDAR_BAUDRATE=256000. Otherwise (for older A2 LIDARs): LIDAR_BAUDRATE=115200.
• With DDS_CONFIG=DEFAULT your robot and laptop need to be in the same LAN network. If you want to use this demo over the Internet, set DDS_CONFIG=HUSARNET_SIMPLE_AUTO and enable Husarnet on ROSbot and you PC.

Sync workspace with ROSbot

./sync_with_rosbot.sh <ROSbot_ip>

Run Rviz and set initial 2D Pose Estimate by using the button in RViz UI:

xhost +local:docker && \
docker compose -f compose.pc.yaml up

ROSbot

Firmware version

Before running the project, make sure you have the correct version of a firmware flashed on your robot.

Firmware flashing command (run in the ROSbot's terminal)

docker run --rm -it --privileged \
husarion/rosbot-xl:humble \
flash-firmware.py /firmware.bin /dev/ttyUSB0

In the ROSbot's shell execute (in the /home/husarion/rosbot-xl-navigation directory)

docker compose -f compose.rosbot.yaml up

Alternative configurations

As you probably noticed, there are a few amcl and nav2 parameter config files. You can switch them by modifying the .env file.

Mecanum

Although mecanum wheels will work as well on default configs, you can use these ones to take advantage of an additional degree of freedom. It is necessary to change two env variables - AMCL_PARAMS (robot_model_type switched to omnidirectional) and NAV2_PARAMS. In this demo DWBLocalPlanner is configured to move only in the x and y position ignoring orientation.

Pure pursuit

This config uses a different controller - RegulatedPurePursuitController instead of DWBLocalPlanner used in the default config. When compared to DWB, movement should be smoother. To use it you have to set NAV2_PARAMS to nav2_params_pure_pursuit.yaml. It will work on both regular and mecanum wheels (amcl parameters can be set to either version).

Quick Start (Gazebo simulation)

Prerequisites

The compose.sim.gazebo.yaml file uses NVIDIA Container Runtime. Make sure you have NVIDIA GPU and the NVIDIA Container Toolkit installed.

First please copy rosbot-xl-mapping/maps folder (which was created during mapping) to rosbot-xl-navigation/maps (the same directory as this README). Then start the containers in a new terminal:

xhost +local:docker && \
docker compose -f compose.sim.gazebo.yaml up

Start by providing AMCL with an initial pose estimate using 2D Pose Estimate in RViz. Now you can use 2D Goal Pose to set destinations.

It is also possible to modify some options - MECANUM, AMCL_PARAMS, NAV2_PARAMS and RMW_IMPLEMENTATION. For details please refer to previous sections.