Artintrek PRO AI and Neural Network Education Kit

Educational laboratories · Middle and high school · Neural networks · Computer vision · Artificial intelligence · Ministry of Industry and Trade register

The Artintrek PRO educational kit for artificial intelligence and neural networks processes video and audio streams with neural networks. It is a resource kit for the Robotrack Basic robotics kits.

Register

The kit is listed in the Russian Ministry of Industry and Trade register under No. 10299112.

Age

12 years and up

Description and specifications

The kit is listed as compliant with paragraph 2.22.27 of Russian Ministry of Education Order No. 838, dated 28 November 2024. See the Russian-language equipment list.

More information about resource kits is available on the Robotrack Wiki (in Russian).

Artintrek can run ready-made neural networks or networks developed in NNTrack. Ready-made scripts can send neural network results to a Trackduino controller to control robotic models. The kit can also operate independently and connect to external displays through HDMI. Unlike the Basic version, Artintrek PRO includes a multifunctional Trackduino controller, allowing learners to experiment with controlling the controller using neural networks.

13 built-in neural networks

  • handwritten digit recognition;
  • face detection;
  • face recognition;
  • emotion recognition;
  • coordinates of 5 facial key points;
  • coordinates of 35 facial key points;
  • gesture recognition;
  • gender and age recognition;
  • text recognition;
  • recognition of 1,000 objects in the surrounding environment;
  • head rotation angle detection;
  • Russian speech recognition;
  • English speech recognition.

Extending the kit

  • Integrate the kit with neural networks developed in NNTrack (in Russian), then use them to control robotic models built around a Trackduino controller.
  • With the Robotrack Basic robotics kit and the TFT 2.4 LCD touch display resource kit, learners explore neural network principles and their use in different tasks in the 21-lesson Introduction to Artificial Intelligence course (in Russian).
  • The same 21-lesson course can be studied with the Robotrack Basic Technology robotics kit and the TFT 2.4 LCD touch display resource kit.
  • Learners can also use the Robotrack Basic Expanded kit with the same 21-lesson course.

Kit contents

  • Artintrek video processing module, version 2;
  • two licences for NNTrack, a visual environment for modelling, training, testing, and exporting neural network architectures;
  • connection cable;
  • Trackduino controller;
  • USB camera;
  • power adapter;
  • USB cable;
  • micro-HDMI to HDMI cable;
  • microSDHC memory card;
  • microphone;
  • MP3 speaker;
  • 9 V battery holder.

Artintrek module features

  • microcomputer (64-bit, 1.8 GHz; RAM: 4 GB or 8 GB; flash storage: 16 GB or 32 GB);
  • built-in accelerator for video stream processing;
  • mounting connectors compatible with Robotrack kits;
  • UART or Wi-Fi connectivity;
  • two micro-HDMI ports;
  • two USB 2.0 ports;
  • cooling system;
  • built-in battery and charging connector;
  • power switch;
  • power indicator LED.

The kit includes a free digital teaching and learning package, provided after a network cooperation agreement is signed. Teacher training is available online, along with technical support.

Skills and competencies

Learners explore the fundamentals of artificial intelligence and neural networks, their types, and their applications. They learn about AI development prospects and how different neural networks work. They practise text recognition, gender and age estimation, gesture recognition, and other methods. Learners create robotics projects with Robotrack Basic kits, using both ready-made neural networks and networks trained in NNTrack. They also explore speech and sound recognition, emotion recognition in speech, and the basics of natural language processing (NLP).

Included software

  • Robotrack IDE is based on Arduino IDE and adds a visual block-programming environment. Learners can create programs for the Trackduino controller without writing or editing code, and can also work with Arduino C.
  • TrackyCode is a programming environment for controlling Trackduino-based robots and virtual sprites with block-based algorithms and artificial intelligence features. Learners can use ready-made sprites or create their own. TrackyCode product page (in Russian).

Robotrack IDE and TrackyCode are supported on Windows, Astra Linux, and Alt Linux.

Video material (external link)

Documents