Search

A novel approach to cooperative path-planning is presented. A SAT solver is used not to solve the whole instance but for optimizing the makespan of a sub-optimal solution. This approach is trying to exploit the ability of stateof- the-art SAT solvers to give a solution to relatively small instance quickly. A sub-optimal solution to the instance is obtained by some existent method first. It is then submitted to the optimization process which decomposes it into small subsequences for which optimal solutions are found by a SAT solver. The new shorter solution is subsequently obtained as concatenation of optimal subsolutions. The process is iterated until a fixed point is reached. This is the first method to produce near optimal solutions for densely populated environments; it can be also applied to domain-independent planning supposed that suboptimal planner is available.

We formulate a problem of goal satisfaction in mutex networks in this paper. The proposed problem is motivated by problems that arise in concurrent planning. For more efficient solving of goal satisfaction problem we design a novel global filtration technique. The filtration technique is based on exploiting graphical structures of the problem - clique decomposition of the problem is used. We evaluated the proposed filtration technique on a set of random goal satisfaction problems as well as a part of GraphPlan based planning algorithm. In both cases we obtained significant improvements in comparison with existing techniques.

Dynamic Constraint Satisfaction Problems play a very important role in modeling and solving real-life problems where the set of constraints is changing. The paper addresses a problem of maintaining arc consistency after removing a constraint from the constraint model. A new dynamic arc consistency algorithm is proposed that improves the practical time efficiency of the existing AC|DC algorithm by using additional data-structures. The algorithm achieves real time efficiency close to the so far fastest DynAC-6 algorithm while keeping the memory consumption low.