# I built a 3D-printed tracked robot using Arduino Mega, Raspberry Pi and OpenCV — looking for feedback

**URL:** <https://discourse.openrobotics.org/t/i-built-a-3d-printed-tracked-robot-using-arduino-mega-raspberry-pi-and-opencv-looking-for-feedback/57352>\
**Category:** Projects\
**Created:** [August 12, 2026, 3:40pm UTC](https://discourse.openrobotics.org/t/i-built-a-3d-printed-tracked-robot-using-arduino-mega-raspberry-pi-and-opencv-looking-for-feedback/57352 "2026-08-12T15:40:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Adilkhan](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/adilkhan/32/35002_2.png) [@Adilkhan](https://discourse.openrobotics.org/u/Adilkhan)\
**Post date:** [August 12, 2026, 3:40pm UTC](https://discourse.openrobotics.org/t/i-built-a-3d-printed-tracked-robot-using-arduino-mega-raspberry-pi-and-opencv-looking-for-feedback/57352/1 "2026-08-12T15:40:25Z")

</div>

Hi everyone!

I recently finished a robotic tracked platform that I developed as an engineering graduation project.

The goal was to build a functional mobile robot combining mechanical design, embedded control, radio communication, and basic computer vision.

The platform includes:

- 3D-printed tracked chassis designed in SolidWorks
- Arduino Mega 2560 for real-time motion control
- BTS7960 motor drivers
- FlySky FS-i6 / FS-iA6B radio control via iBus
- Differential track steering
- Two-speed transmission
- Raspberry Pi 4 for video processing
- Python + OpenCV for basic obstacle detection
- Camera-based real-time vision processing

The control architecture is split into two levels:

**FlySky → Arduino Mega → BTS7960 → DC Motors**

and

**Camera → Raspberry Pi → Python/OpenCV → Obstacle Detection**

I also documented the CAD design, hardware architecture, source code, real prototype, and demo video on GitHub.

GitHub:

> **[GitHub - adilkhann1/robotic-tracked-platform: Robotic tracked platform based on Arduino Mega...](https://github.com/adilkhann1/robotic-tracked-platform)**
>
> Robotic tracked platform based on Arduino Mega 2560 and Raspberry Pi 4 with remote control, camera streaming, computer vision, and 3D-printed components.

This was my first attempt at putting a complete robotics project together as a public engineering portfolio, so I’d really appreciate technical feedback.

In particular, I’m interested in suggestions on:

- improving the Raspberry Pi / OpenCV obstacle-detection pipeline;
- communication between Raspberry Pi and Arduino;
- sensors that would make sense for autonomous navigation;
- how you would approach turning this platform into a semi-autonomous robot.

Thanks!