Recently, planning based on answer set programming has been proposed as an approach towards realizing declarative planning systems. In this paper, we present the language K extends the declarative planning language K by action costs. K provides the notion of admissible and optimal plans, which are plans whose overall action costs are within a given limit resp.

Meta-heuristics support managers in decision-making with robust tools that provide high-quality solutions to important applications in business, engineering, economics and science in reasonable time horizons. In this paper we give some insight into the state of the art of meta-heuristics. This primarily focuses on the significant progress which general frames within the meta-heuristics field have implied for solving combinatorial optimization problems, mainly those for planning and scheduling.

This paper proposes a framework that enhances domain independent heuristic state-space planners with the ability to plan for multiple criteria simultaneously. A criteria hierarchy, together with preferences among them and boundaries over their values, is defined by the knowledge engineer. For the search process in the state-space, a weighted A* strategy is adopted. The heuristic function uses the criteria hierarchy, in order to measure the known accumulated value of the incomplete plan and to estimate the value of the remaining plan. The latter estimation is based on the assignment of multiple value-vectors to the facts of the problem, which estimate the value to achieve the facts from the current state, using alternative paths. A prototype system has been implemented and preliminary performance results have shown that the weights and the scales affect significantly both solution time and plan quality. The framework is general enough, so that it can be applied to all modern heuristic statespace planners.

Abstract. The Extended Behavior Network (EBN) is an architecture and action selection mechanism to design agents capable of selecting sets of concurrent actions in dynamic and continuous environments. It allows one to specify context-dependent motivations and build agents modularly, and has achieved good results in the Robocup and in the 3D action game Unreal Tournament. PHISH-Nets, another behavior network model capable of selecting just single actions, was applied to character modeling, with promising results. We investigate how EBNs fare on agent personality modeling via the design and analysis of 5 stereotypes in Unreal Tournament. We discuss three ways to build character personas and situate our work within other approaches. We conclude that EBNs provide a straightforward way to develop and experiment with different personalities, being interesting for building agents with simple personas and for character prototyping. 1

Resource management problems are problems in which the goal is to optimize the utilization of resources over time. Dependencies between resources make it necessary to plan the proper utilization of these resources. Such a planning must often be constructed in an environment that is non-deterministic, for example due to competing agents. Traditional state-based planning approaches are not suited for this type of planning problems.

Abstract—The task of pathfinding has received considerable attention in research as well as practice. Most of the approaches focus on minimizing the length of the path as well as optimizing the efficiency of the algorithm. Less attention has been paid to the problem of incorporating rich global conditions that require more than simply reaching a specific location. In this paper, we analyze elements of richer pathplanning problems, propose an enhanced specification format, and provide a first simple mechanism to solve such problems. I.