A problem on hinged dissections with colours (Q1889823)

The hinged dissection problem asks whether two planar polygonal regions \(A\) and \(B\) can be dissected into finitely many polygons, and the pieces fixed with hinges at some joints, with the collection of hinged pieces being connected, so that \(A\) can be transformed into \(B\) by moving the pieces and swinging them around their hinges. It is required that the moves be realizable continuously in the plane without overlappings occurring. (The idea can be illustrated by Dudeney's famous dissection of a regular triangle into four hinged pieces which can be moved to form a square.) The authors ask for the existence of a hinged dissection of a square into \(p^2\) congruent squares and a colouring of the edges of the smaller squares with \(k\) colours such that the original square can be transformed into another with its perimeter coloured with the \(i\)th colour, for all \(i\in\{1,\dots,k\}\). Surprisingly, they show that this is possible, if \(p\) is an even number at least \(2k+\sqrt{k^2-k}\).