PROJECT 01 · SIMULATION CASE STUDY

ROS 2 AMR Mapping, Localization & Nav2 Navigation

A Gazebo differential-drive robot connects lidar, odometry, TF, online SLAM, AMCL, and Nav2 into an end-to-end mapping-to-navigation workflow.

ROS 2GazeboSLAM ToolboxAMCLNav2

Actual repository screenshot: Gazebo simulation on the left; lidar and the 2D map in RViz on the right. Original evidence

What this experiment validates

Bridge /scan and /odom into ROS 2 and establish map → odom → base_link → laser_link.
Build a map with asynchronous SLAM Toolbox and save map.pgm and map.yaml.
Localize with AMCL, send Nav2 goals, and inspect topics, TF, lifecycle nodes, and actions with check_nav2.sh.

Validation boundaries

An introductory simulation—not validated hardware safety settings or field navigation reliability.
Wheel odometry does not fuse an IMU. Slip, collisions, and prolonged wall contact can still produce map drift.

Environment and reproduction

Verified baseline: Ubuntu 26.04 / WSL 2, ROS 2 Lyrical, Gazebo ROS vendor 10.4.0, Cyclone DDS, and RViz 2 / WSLg. The test machine also has Jetty nightly packages; other versions have not been validated end to end.

Install ROS 2 Lyrical and configure its package repositories first. While mapping runs, use another terminal to execute bash scripts/save_map.sh. Exit mapping before running bash scripts/run_nav2.sh; SLAM and AMCL must not run together.

git clone https://github.com/zephastra/robotics.git
cd robotics
git checkout 8f2fbe53f616c905262efc589711b65510b1bdbd
cd projects/01-amr-slam
bash scripts/install.sh
bash scripts/run_mapping.sh

Commands are pinned to the reviewed revision. See the upstream README for full prerequisites and subsequent updates.

Evidence and source

The README records end-to-end mapping, map saving, localization, and goal navigation validation for v0.2.0. The Nav2 document provides operating steps and acceptance criteria—not a cross-hardware success-rate study.

GitHub / README ↗Validation record / criteria ↗Related: navigation vs. physical handoff →

Source revision: 8f2fbe53f616c905262efc589711b65510b1bdbd

Related engineering experiments

Mobile Manipulator: Visual Pick, Transport & Return Alignment

An independent ROS 2 and Gazebo experiment connects navigation, color-marker perception, SCARA manipulation, tray transport, and camera-based return alignment.

MuJoCo Humanoid Locomotion & Task Validation

Experiments with upstream pretrained Unitree G1 and EngineAI T800 gait policies focus on waypoint tasks, position holding, manual control, cancellation, and disturbance evidence.