A base-point-free definition of the Lefschetz invariant (Q2491514)

Let \(f\colon X\to X\) be a self map on a compact polyhedron. Fix a base point \(x_0\) and a path \(w_0\) in \(X\) from \(x_0\) to \(f(x_0)\). Two elements \(\langle a \rangle\) and \(\langle b \rangle\) in \(\pi_1(X, x_0)\) are said to be in the same \(f^{w_0}\)-semi-conjugacy class if there is a loop \(c\) at \(x_0\) such that \(\langle b \rangle = \langle c^{-1} a w_0 f(c) w_0^{-1} \rangle\). Each fixed point class \(F\) of \(f\) has a well-defined \(f^{w_0}\)-semi-conjugacy class \(\Phi(F) = \langle uf(u^{-1})w^{-1}_0\rangle\) in \(\pi_1(X, x_0)\), where \(u\) is a path from \(x_0\) to a fixed point in \(F\). The so-called Reidemeister trace, due to K. Reidemeister and F. Wecken, was defined to be the sum \(\sum ind(f, F) \Phi(F)\), where \(F\) ranges over all fixed point classes of \(f\). The Reidemeister trace contains the information of the Lefschetz number \(L(f)\) and the Nielsen number \(N(f)\), and can be obtained algebraically. In any sense, this invariant depends on the choice of the base point \(x_0\) and the base path \(w_0\), although one-to-one correspondences can be written down between the Reidemeister traces of a map according to different base points and base paths. The author reformulates the Reidemeister trace, which is called Lefschetz-Nielsen invariant in this paper, by using the fundamental groupoid. This is what the term ``base-point free'' in the title refers to.