Computing fundamental groups from point clouds (Q2352510)

The goal of this paper is to present and study an algorithm which computes the edge-path group (fundamental group) of a finite regular CW complex \(K\). In order to do this, the authors study a discrete vector field on the \(3\)-skeleton. Utilizing the Morse complex from discrete Morse theory, the authors show that in the case where \(K\) has a single critical \(0\)-cell, a maximal acyclic discrete vector field on the \(3\)-skeleton of \(K\) uniquely determines a presentation for the fundamental group. The algorithm given in the paper is thus a determination of such a maximal acyclic vector field on \(K\). After putting the paper into the more general context in the first section, Section 2 is devoted to giving much of the background needed to develop the authors' results. This includes the basics of CW complexes and other special kinds of complexes, as well as constructing the Čech complex from a lattice complex. Discrete Morse theory is developed insofar as it produces the Morse complex and relates to Whitehead's notion of simple homotopy type. This section ends with an algorithm (defined originally in a previous paper) which yields a maximal discrete vector field from a regular CW complex. As mentioned above, once we have a maximal acyclic discrete vector field on the \(3\)-skeleton, a presentation for the fundamental group is uniquely determined. Proving this is the goal of Section 3. Section 4 gives an explicit implementation in GAP for an application in computational knot theory. Specifically, it computes an induced homomorphism between the newly induced finite presentations of fundamental groups of certain complexes. The paper concludes with a (computationally proven) theorem which distinguishes certain isotopy classes of prime knots. This is a very readable paper with many understandable examples and pictures to help illustrate. It strikes a good balance between theory, computation, and examples.