Autonomous Rubik’s Cube Solver

Project

Purpose: Learning to solve a Rubik’s cube can be a daunting task and to do it efficiently can take a long time. We wanted to build a robot that could solve a Rubik’s cube without any input from the user. The main functions of the ARCS are:

• Identify the layout of the squares on each side of the cube.
• Figure out a series of moves to solve the cube.
• Use stepper motors to manipulate the cube into a solved state.

System

Methods

• Four stepper motors with four linear actuators control the cube.
• 3D printed cube grippers and rail system attach to motors.
• Single Board Computer (SBC) is a Raspberry Pi Zero W2. (picture)
• OpenCV C++ library assists in cube tile detection.
• PiGpio C++ library allows interfacing to LED, button, and motors via General Purpose IN/Out (GPIO). (picture of CV image)
• Off-the-shelf stepper motor drivers interface SBC to stepper motors. (picture)
• Relay board allows simple control of linear actuators. (picture)

Conclusion

The ARCS system is able to:

• Detect the state of the cube.
• Find a solution to solve the cube.
• Manipulate the cube into a solved state using motors.

Potential improvements:

• Use Pi’s Wi-Fi capability to make a guide on how to solve the Rubik’s cube.
• Optimize the algorithm to solve the cube.
• Improve the motor speed.

Lessons learned:

• Simple is often better for computer vision methods.
• Have details of design figured out before ordering parts.
• Plan adequate time for debugging and troubleshooting.