Knaster's problem for almost \((\mathbb Z_p)^k\)-orbits (Q2268848)

In 1947, Knaster made the following conjecture: Let \(f: S^{d-1}\to \mathbb R^k\) be a map, let \(m=d-k+1\), and let \(x_1, \dots , x_m \in S^{d-1}\) be \(m\) points on \(S^{d-1}\). Then there exists a rotation \(\rho \in SO(d)\) such that \[ f(\rho (x_1))= \dots f(\rho (x_m)). \] There exist counterexamples to the Knaster conjecture. The authors prove some positive results. In particular, they show that if \(k=1\) or \(2\) and if \(X\subset S^{d-1}\) is an orbit of a \((\mathbb Z_p)^k\)-action with one point deleted, then \(X\) can be rotated so that \(f\) becomes constant on \(X\).