The faces are named U,D,L,R,F,B (for up, down, left, right, front, back).
Moves can be described in the following format:
- X (rotate face X clockwise, 90º)
- X2 (rotate face X 180º)
- X' (rotate face X anticlockwise, 90º)
The initial configuration of a cube can be described by a string, such as:
"DRLUUBFBRBLURRLRUBLRDDFDLFUFUFFDBRDUBRUFLLFDDBFLUBLRBD"
The idea is that each face of solved cube has a colour uniquely determined by the square in its center. If we have a shuffled cube, the center square of each face will remain as in the start. We read the squares in the order:
U1, U2, U3, U4, U5, U6, U7, U8, U9, R1, R2, R3, R4, R5, R6, R7, R8, R9, F1, F2, F3, F4, F5, F6, F7, F8, F9, D1, D2, D3, D4, D5, D6, D7, D8, D9, L1, L2, L3, L4, L5, L6, L7, L8, L9, B1, B2, B3, B4, B5, B6, B7, B8, B9
We look at square U1 and see what colour it has. Then the first element of the string will be labelled by the face which has said colour.
We look at square U2 and see what colour it has. Then the second element of the string will be labelled by the face which has said colour.
For example, using the string above, we know that square U1 has the colour corresponding with the down (D) face; square U2will have the colour corresponding to the right (R) face; etc ... .
We are using the Kociemba Algorithm, in particular this Python implementation. For more details on Kociemba, see this.
