Edmund H. Durfee

The distributed constraint satisfaction problem: formalization and algorithms

We develop a formalism called a distributed constraint satisfaction problem (distributed CSP) and algorithms for solving distributed CSPs. A distributed CSP is a constraint satisfaction problem in which variables and constraints are distributed among multiple agents. Various application problems in distributed artificial intelligence can be formalized as distributed CSPs. We present our newly developed technique called asynchronous backtracking that allows agents to act asynchronously and concurrently without any global control, while guaranteeing the completeness of the algorithm. Furthermore, we describe how the asynchronous backtracking algorithm can be modified into a more efficient algorithm called an asynchronous weak-commitment search, which can revise a bad decision without exhaustive search by changing the priority order of agents dynamically. The experimental results on various example problems show that the asynchronous weak-commitment search algorithm is, by far more, efficient than the asynchronous backtracking algorithm and can solve fairly large-scale problems.

Trends in cooperative distributed problem solving

The authors present an overview of cooperative distributed problem solving (CDPS), an emerging research area that combines aspects of AI (artificial intelligence) and distributed processing. CDPS can be used to study how a loosely coupled network of sophisticated problem-solving nodes can solve a complex problem which consists of a set of interdependent subproblems. Subproblems arise because of spatial, temporal, and functional distribution of data, knowledge, and processing capabilities. Application areas include distributed interpretation, distributed planning and control, cooperating expert systems, and computer-supported human cooperation. The authors survey the important approaches and empirical investigations that have been developed. The approaches covered include negotiation, functionally accurate cooperation, organizational structuring, multiagent planning, sophisticated local control, and theoretical frameworks. >

Distributed constraint satisfaction for formalizing distributed problem solving

Viewing cooperative distributed problem solving (CDPS) as distributed constraint satisfaction provides a useful formalism for characterizing CDPS techniques. This formalism and algorithms for solving distributed constraint satisfaction problems (DCSPs) are compared. A technique called asynchronous backtracking that allows agents to act asynchronously and concurrently, in contrast to the traditional sequential backtracking techniques used in constraint satisfaction problems, is presented. Experimental results show that solving DCSPs in a distributed fashion is worthwhile when the problems solved by individual agents are loosely coupled.<<ETX>>

Distributed problem solving and planning

Distributed problem solving involves tile collective effort of multiple problems solvers to combine their knowledge, information, and capabilities so as to develop solutions to problems that each could not have solved as well (if at all) alone. The challenge in distributed problem solving is thus in marshalling the distributed capabilities in the right ways so that the problem solving activities of each agent complement the activities of the others, so as to lead efficiently to effective solutions. Thus, while working together leads to distributed problem solving, there is also the distributed problem of how to work together that must be solved. We consider that problem to be a distributed planning problem, where each agent must formulate plans for what it will do that take into account (sufficiently well) the plans of other agents. In this paper, we characterize the variations of distributed problem solving and distributed planning, and summarize some of the basic techniques that have been developed to date.

Partial global planning: a coordination framework for distributed hypothesis formation

Partial global planning is used to provide a framework for coordinating multiple AI systems that are cooperating in a distributed sensor network. By combining a variety of coordination techniques into a single, unifying framework, partial global planning enables separate AI systems to reason about their roles and responsibilities as part of group problem solving, and to modify their planned processing and communication actions to act as a more coherent team. Partial global planning is uniquely suited for coordinating systems that are working in continuous, dynamic, and unpredictable domains because it interleaves coordination with action and allows systems to make effective decisions despite incomplete and possibly obsolete information about network activity. The authors implement and evaluate partial global planning in a simulated vehicle monitoring application and identifying promising extensions to the framework. >

CIRCA: a cooperative intelligent real-time control architecture

Most research into applying AI techniques to real-time control problems has limited the power of AI methods or embedded reactivity in an AI system. An alternative, cooperative architecture is presented. It uses separate AI and real-time subsystems to address the problems for which each is designed. A structured interface allows the subsystems to communicate without compromising their respective performance goals. By reasoning about its own bounded reactivity, cooperative intelligent real-time control architecture (CIRCA) can guarantee that it will meet hard deadlines while still using unpredictable AI methods. With its abilities to guarantee or trade off the timeliness, precision, confidence, and completeness of its output, CIRCA provides more flexible performance than previous systems. >

A Survey of Research in Distributed, Continual Planning

Complex, real-world domains require rethinking traditional approaches to AI planning. Planning and executing the resulting plans in a dynamic environment implies a continual approach in which planning and execution are interleaved, uncertainty in the current and projected world state is recognized and handled appropriately, and replanning can be performed when the situation changes or planned actions fail. Furthermore, complex planning and execution problems may require multiple computational agents and human planners to collaborate on a solution. In this article, we describe a new paradigm for planning in complex, dynamic environments, which we term distributed, continual planning (DCP). We argue that developing DCP systems will be necessary for planning applications to be successful in these environments. We give a historical overview of research leading to the current state of the art in DCP and describe research in distributed and continual planning.

Negotiating task decomposition and allocation using partial global planning

Abstract To coordinate as an effective team, cooperating problem solvers must negotiate over their use of local resources, information, and expertise. Sometimes they negotiate to decide which local problem-solving tasks to pursue, while at other times they negotiate over the decomposition and distribution of tasks. They might negotiate by sharing all of their information, or by exchanging proposals and counterproposals, or by working through an “arbitrator.” In general, negotiation is a complex process of improving agreement on common viewpoints or plans through the structured exchange of relevant information. In this paper, we describe how partial global planning provides a versatile framework for negotiating in different ways for different reasons, and we examine in detail its utility for negotiating over whether and how problem solvers should decompose and transfer tasks to improve group performance. Finally, we propose how our approach can be extended to capture even more fully the complexity, flexibility, and power of negotiation as a tool for coordinating distributed problem solvers.

Coordination as distributed search in a hierarchical behavior space

It is theorized that the process of coordination is a distributed search through a hierarchical space of agent behaviors. By specifying agent activities along multiple dimensions and at different levels of abstraction, the hierarchical behavior space provides a single, rich representation that agents can use to organize, plan, and schedule their collective actions. A computational instance of the evolving theory, which implements a particular choice of distributed protocol, local algorithm, metrics, and heuristics, as applied to resolving resource conflicts in an unstructured delivery domain, is described. In this domain, agents that initially do not know with whom they might interact exploit the hierarchical behavior representation to selectively exchange more details about themselves until they can resolve conflicting behaviors. It was experimentally demonstrated how the hierarchical protocol and multidimensional representation provide powerful and practical mechanisms for coordinating these agents, and important research issues to be addressed are highlighted. >

World modeling for the dynamic construction of real-time control plans

Abstract As intelligent, autonomous systems are embedded in critical real-world environments, it becomes increasingly important to rigorously characterize how these systems will perform. Research in real-time computing and control has developed ways of proving that a given control system will meet the demands of an environment, but has not addressed the dynamic planning of control actions. Building an agent that can flexibly achieve its goals in changing environments requires a blending of real-time computing and AI technologies. The Cooperative Intelligent Real-time Control Architecture (CIRCA) implements this blending by executing complex AI methods and guaranteed real-time control plans on separate subsystems. We describe the formal model of agent/environment interactions that CIRCA uses to build control plans, and we show how those control plans are guaranteed to meet domain requirements. CIRCAs world model provides the information required to make real-time performance guarantees, but avoids unnecessary complexity.