Disjunctive and continuous constraint satisfaction problems

Abstract

In this work, we extend the class of Horn constraints to include disjunctions with an arbitrary number of linear inequalities, linear disequations and non-linear disequations. We propose a preprocess step in which two algorithms are carried out. The first algorithm called Constraint Selection Algorithm (CSA) translates the disjunctive non binary CSP into a non-disjunctive one. This algorithm selects the more appropriate set of atomic constraints that is more likely to be consistent.The second algorithm called Constraint Ordering Algorithm (COA)classifies the resultant constraints from the most restricted to the least restricted one. Then, a CSP solver carries out the search through the non-disjunctive and ordered problem.