Bredon cohomology and robot motion planning (Q2330954)

In [Discrete Comput. Geom. 29, No. 2, 211--221 (2003; Zbl 1038.68130)], \textit{M. Farber} defined the topological complexity \(\mathrm{TC}(X)\) of a space \(X\) as the Schwarz genus of the fibrational substitute of the diagonal \(d: X \to X^n\). Topological complexity has well-known applications to robotics. Furthermore, the issue is a relatively new subject, it is an area of intensive development, and any progress in this direction is very welcome. It is easy to see that \(\mathrm{TC}(X)\) is a homotopy invariant. Hence, given a discrete group \(\pi\), it makes sense to define (introduce) \(\mathrm{TC}(\pi)\) as \(\mathrm{TC}(K(\pi,1))\). So, we have the problem (posed by Farber) to describe \(\mathrm{TC}(\pi)\) in a purely algebraic way. For example, for the Lusternik-Schnirelman category \(\mathrm{cat}\) (a close relative of \(\mathrm{TC}\)) we know that \(\mathrm{cat}(\pi)\) coincides (with one possible exeption) with the cohomological dimension of \(\pi\). The situation with \(\mathrm{TC}\) is much more complicated, it is а really challenging problem. \textit{M. Farber} and \textit{S. Mescher} made some progress (by describing \(\mathrm{TC}(\pi)\) for an important class of groups), cf. [J. Topol. Anal. 12, No. 2, 293--319 (2020; Zbl 1455.55003)]. In the paper under review the authors find certain upper and lower bounds for \(\mathrm{TC}(\pi)\) for arbitary groups \(\pi\). The descriptions (definitions) of lower and upper bounds are quite cumbersome, but this is the life. Here the Bredon cohomology is the main tool for this research.