After a lot of experiments with Raspberry, ESP32, STM32 and Arduino boards I’ve decided what will be a spine of the Zakhar project. ROS seems the most interesting and extendible. Inside it, I could combine Python, C, C++ programs (hence, libraries ) working inside a Linux environment and communicate with any other platforms.
Table of Contents
Say, I could build a system on a Raspberry, which will communicate with developer board like ESP-WROVER-KIT working on FreeRTOS. Even more, I could use my desktop as a part of the system for High-level computing (say, for image recognition with OpenCV).
I have bought some hardware for the new version of Zakhar, called Zakha_ros:
The new hardware
1. Platform. It is a simple platform which I used for my first experiments 3 years ago:

2. Manipulator. It based on DC motor and an aluminum chassis. Nice to grab something.

3. Camera. I’ve bought a standard Pi camera than mounted it onto a small cardboard panel

4. Raspberry Pi 3 B+. With a nice black case.

What’s Next
Next, I’m planning to update a Zakhar concept to follow ROS capabilities and using hardware. Stay tuned!