Using Constraint Programming for Solving Distance CSP with Uncertainty

Abstract

Many problems in chemistry, robotics or molecular biology can be expressed as a Distance CSP (Constraint Satisfaction Problem). Sometimes, the parameters of this kind of problems are determined in an experimental way, and therefore they have an uncertainty degree. A classical approach for solving this class of problems is to solve the CSP without considering the uncertainties, and to obtain a set of solutions without knowing the real solution sub-spaces. A better approach is to apply a branch and prune algorithm to generate a set of disjoint boxes that include all the solution sub-spaces, but without information about independent solution sub-spaces or the different types of boxes.