Temporal Planning in Dynamic Environments for Mobile Agents Muhammad Adnan Hashmi Laboratoire d’Informatique de Paris 6 University Pierre and Marie Curie 75016 Paris, France Adnan.Hashmi@lip6.fr Amal El Fallah Seghrouchni Laboratoire d’Informatique de Paris 6 University Pierre and Marie Curie 75016 Paris, France Amal.Elfallah@lip6.fr ABSTRACT In this thesis we intend to propose and implement a planning based enhancement of the agent oriented programming lan- guage CLAIM, in which an agent is an intelligent and mobile entity. This new language which is called P-CLAIM, should have the following characteristics. Firstly, temporal plan- ning capability for the agents. Secondly, the agents should be able to work in a stable manner in the changing environ- ment. Thirdly, a balance should exist between reactivity and proactivity. Fourthly, the agents’ actions should not conflict with the actions of other agents in the same environment and the agents should avail the cooperative oppurtunities offered by other agents and they themselves should be able to help other agents where possible. General Terms Languages, Algorithms, Design Keywords Mobile Agents, Multi-Agent Systems, AOP Languages, Tem- poral Planning 1. INTRODUCTION In recent years, mobile agents have got much attention in the multi-agent systems community. A lot of agent ori- ented programming (AOP) languages and execution plat- forms have been proposed in the literature which make the programmer, able to create and deploy intelligent and mobile agents. Some of the AOP languages which support mobil- ity are Jinni[13] and CLAIM[5]. No programming language supporting mobility, gives agents the ability to plan ahead. Sometimes the execution of the actions without planning re- sults in the inability to achieve the goals. There has been some work to incorporate planning within some program- ming languages which does not support mobility [3], [8] but these systems do not take into account the duration of agent Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. FIT’09 December 16-18, 2009, CIIT, Abbottabad, Pakistan Copyright 2009 ACM 978-1-60558-642-7/09/12 ...$10.00. actions, neither do they consider the uncertainty of the en- vironment. These systems assume that the agents’ actions are instantaneous and that the effects produced on the en- vironment are only those which are produced by the agent’s actions. But these assumptions are unrealistic for the de- velopment of real world applications. In this thesis we are interested in the language CLAIM in which a multi-agent system is a set of distributed hierarchies of agents deployed on computers connected via a network. All the comput- ers have a global clock. With respect to the hierarchical representation, an agent is a node in a hierarchy. It is an autonomous, intelligent and mobile entity. It has a parent and contains (optional) sub-agents, running processes and cognitive elements (e.g. knowledge, goals, capabilities). Our goal is to propose and implement a planning based enhancement of CLAIM language which we shall call P- CLAIM. This new language should have the following char- acteristics. Firstly, temporal planning 1 capability for the agents. Secondly, the agents should be able to work in a sta- ble manner in the changing environment. Thirdly, a balance should exist between reactivity and proactivity. Fourthly, the agents’ actions should not conflict with the actions of other agents in the same environment and the agents should avail the cooperative oppurtunities offered by other agents and they themselves should be able to help other agents where possible. 2. LITERATURE SURVEY Some of the research related to ours is CYPRESS[9], RETS- INA[12] and the systems proposed in [3] and [8]. In our opinion, the system closest to our research is CY- PRESS system, which integrates a planning system SIPE- 2[15] with an execution system PRS[6]. It has the ability to react to the unanticipated changes in the environment by replanning. But CYPRESS system does not handle tempo- ral knowledge. Another aspect lacking in CYPRESS is the mobility of the agents. In mobile agents, context of an agent changes while moving from one machine to another and the planner component must be able to deal with the changing context which becomes more challenging when the planning is interleaved with execution. Another BDI like system incorporating planning is RETSINA. RETSINA system interleaves planning with execution and supports planning for dynamic and real environments. It also includes planning for information gathering tasks. But one major drawback of RETSINA system is that it plans 1 This type of planning takes into account the durations of agent’s actions.