Error control in polytope computations (Q700748)

The paper assesses polytope computation and offers a fairly complete solution with logical consistency and conservative accuaracy bounds for the basic operations of convex hull determination, addition and subtraction of polytopes, and other operations [cf. \textit{J. P. Norton} and \textit{S. H. Mo}, Mathematics and Computers in Simulation 32, 527--534 (1990); \textit{E. Walter} and \textit{H. Piet-Lahanier}, IEEE Transactions in Automatic Control 34, 911--915 (1989; Zbl 0698.93017)]. A set of practically useful computational algorithms is developed in such a way that any compuation with real numbers is carried out through vertex fitting and plane fitting operations only. A numerical analysis of the foregoing operation is presented. It is also proposed that possible further research can look at adjustment procedures and polytope operations not covered in this paper such as computation with nonconvex polytopes.