Make A Robot Kit - for hands on AI learning

Features 

• Make a self-driving car
• Interactive & engaging through AI recognition 
• On-device model training
• Hands-on from construction to coding
• Supports Micropython
• Customizable, extendable & open-source

Description

MARK, your hands-on AI learning experience

Make A Robot Kit aka MARK, as seen on Kickstarter & in the Press

Media Links:

CNX Software MARK AI Robot Kit Aims to Teach AI & Robotics to 12+
hackster.io the Make A Robot Kit — Aims to Bring Artificial Intelligence to STEM Education
fabcross (JP) 自動運転ロボットカーを作ろう——AI搭載ロボットキット「Make A Robot Kit」
Notebook check (DE) Entwicklerkit wird zum Roboter mit AI-Funktionen

Make A Self Driving Car

Implement your self-driving, advanced map & story-based projects or use in robotics competitions.

MARK is designed in the form of a car, so we ll preload code which allows you to implement basic self-driving car functionality once everything is assembled and connected. Scroll down for a project video of the self-driving function.

Interactive & Engaging through AI Recognition

Mark comes pre-loaded with functionality to recognize colors, shapes, and objects within these categories. But of course, you should train your own custom models! We provide a colab notebook to guide you through this process. Scroll down to check out a video of MARK interacting with Pet animals.

On Device Model Training

Simple custom classifier models can be trained on MARK through data collection via the onboard camera, automatic data processing & model training by the CyberEye (brains of the MARK), and model deployment directly on MARK which can be saved for reuse on SD card.

Interactive with the external world

Aside from the obvious camera sensor, there is an ultrasonic sensor included for collision avoidance, a button which you can program to change modes and an electromagnet which gives MARK a "pick up" function. Simply attach metallic stickers (included in pack) on the back of an image card.

Hands-On from Construction to Coding

For a Hands-on construction experience, MARK will come disassembled in parts with the tools you need to make your robot.

Interface board - Markduino

Included are stepper motor & spare servo ports which you could utilize to extend MARK as a robot with a grabber or conveyor functionality. The Markduino also includes spare Digital, Analog, and I2C ports, into which you can easily plug & play some compatible Grove series sensors. Read more about grove here. Those familiar with Arduinos will notice that the markduino pinout fits, so projects can be implemented with this controller too.

We ve developed a mini-grabber and rubber slingshot attachment to extend the fun you can have with your MARK. This will be available in the Robotics Extension Pack but you are also welcome to make your own designs to extend MARK s applications.

Supported with Coursework and Graphical Coding Software

make2Learn is our project and lesson sharing platform where you can access course content directly related to MARK and our other support projects. There are courses for Machine Learning and Autonomous Driving. The above is an extract from Autonomous Driving. Find Lessons on make2learn.tinkergen.com

Accessible Yet Intelligent

We want AI learning to be accessible to those from a variety of backgrounds so you ll be able to code MARK graphically by dragging and dropping code blocks.

Codecraft is a graphical programming software based on Scratch 3.0, developed by us which is available on desktop & in-browser. Codecraft will support MARK AI functions in drag and drop code, as well as enabling you to see what the MicroPython code would look like. Simply connect and then click upload to start using your MARK with it s edited program. If you re confident with your coding skills you can also go ahead and edit text-based Python code in Codecraft. You can even use your own custom models through Codecraft.

Supports Micropython

MaixPyIDE from SIPEED - More advanced users can choose to directly write code for MARK using MaixPyIDE. The cyberEYE mainboard s K210 chip has Microython firmware which features openMV support.

Customizable, Extendable & Open Source

Extend with our extension packs (available soon), or DIY. 3D print your own cover to customize MARK into whatever color you like. Add structural modifications onto the Chassis which includes multiple attachment points.

*note grabber, slingshot attachment shown in images above are not included in Make A Robot Kit.

**multiple color choices are not available for purchase, but STL files for 3D FDM printing with pla colors will be provided

This kit is suitable for:

• Maker teachers
• High school students
• Homeschooling educators
• STEM camps
• AI experimenters
• AI & robotics learners

Specifications

• Power Supply ：6 x AA batteries
• Data connection: USB C
• Display: 2.4" color LCD
• Storage: micro SD slot
• Connectivity via onboard ESP32: Wifi 2.4G, Bluetooth 4.2
• CPU: RISC-V Dual Core 64 bit with FPU
• KPU: CNN Hardware accelerator 0.25 TOPS @ 0.3W, 400MHz; overclock to 800MHz, 0.5 TOPS
• Static RAM 8MB Total: 6MB SRAM, 2MB AI SRAM
• Image Recognition: QVGA@60fps VGA@30fps or overclocked 60fps

MARK Project Videos 

Part List

1 x Project Map 

1 x Quick Start Guide 

16 x Recognition cards 

3 x Card Support 

1 x Chassis Cover 

1 x M.A.R.Kduino 

1 x Headlight Board 

1 x Speaker 

1 x Ultrasonic Sensor 

1 x Electromagnet 

1 x Line Follower 

1 x Button 

1 x Chassis 

1 x Pan-Tilt Camera 

2 x Tyre 

1 x Universal Wheel 

1 x Line Follower Stand 

1 x cyberEye & 2.4" Color LCD

1 x Philips Head 

1 x Flat Head Screwdriver

14 x Magnetic Sticker

3 x 20cm Grove Cable 

3 x 10cm Grove Cable 

1 x 1m Type-C USB cable

6 x AA Battery 

1 x Hardware Box 

20 x M3x 8mm Screws

10 x Spacer