Needle variation (Q1608119)

The paper investigates the notion of a needle variation and examines the behavior of the phase trajectory depending on the parameters of the variation for measurable and locally bounded controls following [\textit{L. S. Pontryagin}, \textit{V. G. Boltyanskiĭ}, \textit{R. V. Gamkrelidze} and \textit{E. F. Mishchenko}, ``Mathematical theory of optimal processes'' (Russian) (1961; Zbl 0102.31901)], [\textit{A. D. Ioffe} and \textit{V. M. Tikhomirov}, ``Theory of extremal problems'' (Russian) (1974; Zbl 0292.90042)], [\textit{V. M. Alekseev, V. M. Tikhomirov} and \textit{S. V. Fomin}, ``Optimal control'' (Russian) (1979; Zbl 0516.49002)]. The author attempts to present a modified exposition of the theory by introducing new notions. The exposition is not accurate. A new definition of a measurable and bounded control function taking values in a measurable set of a topological space is derived. He strives to show that his definition is equivalent to the classical one by referencing to the Lusin theorem, but this is not so because there is no suitable version of this theorem which will be true for an arbitrary topological space. A counter-example may be constructed.