ROS2 Graph (rqt_graph)

This graph visualizes the active ROS2 nodes and topic connections during teleoperation. The teleop_twist_keyboard node publishes velocity commands to /cmd_vel, which are consumed by the robot_serial_bridge node and translated into motor commands for the Arduino. This demonstrates real ROS middleware communication and node-based architecture.

ROS2 rqt_graph showing node and topic connections

System Architecture

High-level ROS2 control layer communicating with Arduino-based motor control via USB serial. The serial bridge node translates velocity commands into low-level motor instructions.

Keyboard teleoperation generates /cmd_vel messages, which flow through the ROS2 system into the serial bridge and down to embedded motor control, demonstrating end-to-end command propagation.

The robot supports multiple control paradigms: manual IR control, autonomous obstacle avoidance, and ROS2-based teleoperation, illustrating modular system evolution.

Documentation

Locomotion Technical Notes

Overview

This project implements a differential-drive robot that combines Arduino-based motor control with a ROS2 Jazzy teleoperation pipeline on Ubuntu. The system evolved through three control paradigms: manual IR remote control, autonomous obstacle avoidance using onboard sensors, and ROS2-based keyboard teleoperation using the /cmd_vel interface. The final implementation demonstrates middleware-based robot control, serial communication between high-level and low-level systems, and modular robotics architecture.

Hardware (BOM)

Arduino Uno with sensor shield
L298N motor driver
Differential-drive chassis with two DC motors and caster wheel
HC-SR04 ultrasonic sensor
Servo motor for sensor scanning
IR receiver and remote
Battery pack
USB connection to Ubuntu laptop for ROS2 teleoperation

Software Stack

Arduino IDE for firmware flashing and embedded motor logic
ROS2 Jazzy on Ubuntu for middleware and topic-based control
Python ROS2 node for serial bridging
teleop_twist_keyboard for keyboard-based teleoperation
Linux serial interface via /dev/ttyUSB0
GitHub for source control and project documentation

Bring-up / Calibration

Assemble the chassis, motor driver, ultrasonic sensor, and servo.
Verify Arduino firmware uploads successfully through Arduino IDE.
Confirm motor behavior using direct firmware tests.
Connect the robot to the Ubuntu laptop through USB.
Verify the Arduino serial device is available at /dev/ttyUSB0.
Launch the ROS2 serial bridge node.
Launch teleop_twist_keyboard.
Confirm /cmd_vel messages are published and translated into motion commands.
Validate movement directions and stop behavior before recording demos.

Demo Notes

The demo shows the robot being controlled through ROS2 using keyboard teleoperation. The terminal output includes the ROS topic list, /cmd_vel message values, and an rqt_graph screenshot showing the active ROS node graph. Additional demos demonstrate IR remote control and obstacle avoidance behavior implemented directly on the Arduino side.

Known limitations:

ROS control is currently tethered by USB to the Ubuntu laptop
No onboard ROS compute yet
No SLAM, Nav2, or vision-based autonomy in the current version
Obstacle avoidance and ROS teleoperation exist as separate operating modes rather than a unified decision layer

Roadmap

Move ROS execution from the laptop to onboard compute
Add onboard serial-independent compute using Raspberry Pi Compute Module 4
Integrate ultrasonic and IR sensing into a ROS-aware autonomy layer
Add camera-based perception and lightweight edge AI
Explore SLAM and navigation once onboard compute is stable

Locomotion Robot

Demo

Repository