An upper bound for topological complexity (Q1725671): Difference between revisions

The topological complexity of a space \(X\) is defined as \(\text{TC}(X):=\text{secat}(\pi ),\) that is, the sectional category (or Schwarz genus) of the end-points evaluation fibration \[ \pi :X^I\rightarrow X\times X,\quad \pi (\alpha )=(\alpha (0),\alpha (1)) \] This is a numerical homotopy invariant introduced by \textit{M. Farber} [Discrete Comput. Geom. 29, No. 2, 211--221 (2003; Zbl 1038.68130)] in order to study the topological instabilities in any motion planning algorithm in robotics. Unfortunately, although its definition is simple to establish, the topological complexity of a space is quite hard to compute. In order to overcome this drawback a good strategy consists of considering upper or lower bounds for topological complexity. Following this spirit, in the paper [``Bredon Cohomology and robot motion planning'', Preprint, \url{arXiv:1711.10132}] the authors of the present paper defined and developed a new invariant, \(\text{TC}^{\mathcal{D}}\), called \(\mathcal{D}\)-topological complexity, that approximates topological complexity. In the paper under review the same authors first continue their previous work on \(\text{TC}^{\mathcal{D}}\) by establishing more interesting properties as well as giving connections with topological complexity and other invariants of Lusternik-Schnirelmann type. They also define a new invariant of TC-type. Indeed, consider the sequence \(\mathcal{F}\equiv E\overset{p}{\rightarrow }{\overline{X}}\overset{q}{\rightarrow }X\) where \(p\) is a fibration and \(q\) a covering map such that \(\overline{X}\) is connected. Then \(\widetilde{secat}(\mathcal{F})\) is defined as the minimal number \(k\) such that \(X\) admits an open cover \(X=U_0\cup U_1\cup \ldots\cup U_n\) such that for any \(i\) the fibration \(p\) admits a continuous section over \(q^{-1}(U_i)\subset \overline{X}.\) If \(P:\tilde{X}\rightarrow X\) denotes the universal cover of \(X\) then they define \[ \widetilde{\text{TC}}(X):=\widetilde{\text{secat}}(X^I\overset{p}{\rightarrow }\tilde{X}\times _{\pi }\tilde{X}\overset{q}{\rightarrow }X\times X) \] for certain maps \(p\) and \(q,\) and \(\pi \) being the fundamental group of \(X\). The main result of the paper is the following inequality, under mild conditions on the space: \[ \text{TC}(X)\leq \text{TC}^{\mathcal{D}}(X)+\widetilde{\text{TC}}(X) \] Such an inequality gives a connectivity-dimension estimate of topological complexity \[ \text{TC}(X)\leq \text{TC}^{\mathcal{D}}(X)+\left \lceil{\frac{2\text{dim}(X)-k}{k+1}}\right \rceil \] provided \(X\) is a finite dimensional simplicial complex with \(k\)-connected universal cover. As noted by the authors this is a refinement of an estimate given by \textit{A. Dranishnikov} [Topology Appl. 179, 74--80 (2015; Zbl 1304.55003)].