2WD ROSduino

Differential-drive Arduino robot with ROS-based remote control and autonomous mode

Abstract

This project focuses on building a compact two-wheel differential-drive robot using standard electronic components and an Arduino-based control architecture.

The platform supports two operating modes:

• 🎮 Remote control via Xbox controller
• 🤖 Autonomous navigation using an ultrasonic distance sensor

The primary challenges involved designing a reliable electrical circuit, establishing a feasible ROS communication architecture, and implementing real-time control logic in C++.
The final result and real-world operation can be seen in the video section below.

Background Story

During the winter semester 2019/2020, I participated in — and won — a robotics challenge at TU Darmstadt. The competition involved autonomous pick-and-place of tennis balls using a TurtleBot3 equipped with a 5-DOF gripper arm.

A recording of the final run can be found here:
🎥 https://youtu.be/cxs0oeeQU-w

The prize was a standard Arduino-based 2WD car kit — which became the foundation for this project.

Initially, my goals were simple:

• Convert the kit into a remote-controlled vehicle
• Use an Xbox controller for intuitive control
• Integrate it with ROS for extensibility

This led to the introduction of a Raspberry Pi 4B as an intermediate control layer. While technically functional, it quickly revealed two limitations:

• The platform was overpowered from a computing perspective
• The mechanical drivetrain was underpowered

These insights eventually motivated the transition to the more ambitious Autonomous Driller project.

Assembly

The base kit consisted of a standard 2WD chassis, dual DC motors, and a basic Arduino controller.

The full system architecture with my custom upgrades included:

• Arduino Uno for low-level motor control
• Raspberry Pi 4B for ROS integration
• Motor driver module
• Battery pack
• Ultrasonic sensor for obstacle detection

The assembly process involved modifying the standard kit for better cable routing, stable motor mounting, sensor integration and micro-controller integration.

Controller-to-Arduino Communication

The control pipeline was structured as follows:

Xbox Controller → Laptop → ROS Node → Raspberry Pi → Arduino

Communication Roles

• Xbox Controller
  Provides user input via joystick and buttons.

• Laptop
  Maps controller input to velocity commands and transmits via ROS.

• Raspberry Pi
  Acts as the ROS master node, handling command processing and forwarding.

• Arduino
  Executes low-level motor control through PWM signals.

This architecture allowed seamless switching between:

• Manual teleoperation
• Autonomous obstacle-avoidance mode

Autonomous Mode

In autonomous mode, the ultrasonic sensor continuously measured distances ahead.
Simple logic ensured:

• Forward movement if clear
• Rotation if obstacle detected
• Adaptive correction depending on distance thresholds

This mode served as an introduction to sensor-based feedback control and robotic decision-making.

Video

First functional test with Xbox remote control and ROS-based communication:

Key Takeaways

This project marked my first end-to-end robotics system combining:

• Hardware assembly
• C++ embedded programming
• ROS architecture
• Real-world teleoperation

Although simple in hindsight, it laid the foundation for more complex systems — and ultimately inspired the transition toward high-performance autonomous platforms.

Small robot. First control loop. Big curiosity.