O-Plan is an AI planner based on previous experience with the Nonlin planner and its derivatives. Nonlin and other similar planning systems had limited control architectures and were only partially successful at limiting their search spaces. O-Plan is a design and implementation of a more flexible system aimed at supporting planning research and development, opening up new planning methods and supporting strong search control heuristics. O-Plan takes an engineering approach to the construction of an efficient domain independent planning system which includes a mixture of AI and numerical techniques from Operations Research. The main contributions of the work are centred around the control of search within the OPlan planning framework, and this paper outlines the search control heuristics employed within the planner. These involve the use of condition typing, time and resource constraints and domain constraints to allow knowledge about an application domain to be used to prune the searc...

This paper describes the O-Plan2 agent oriented architecture and describes the communication which takes place between planning and execution monitoring agents built upon the architecture. Separate modules of such a system are identified along with internal and external interface specifications that form a part of the design. Time constraints, resource usage, object selection and condition/effect causal constraints are handled as an integral part of the overall system structure by treating specialised constraint management as supporting the core decision making components in the architecture. A close coupling of planning and time or resource scheduling is therefore possible within a system employing an activity based plan representation. 2 History and Technical Influences

Explicit causal structure representations have been widely used in classical planning systems to guide a variety of aspects of planning, including plan generation, modification and generalization. For the most part, these representations were limited to single-contributor causal structures. Although widely used, single-contributor causal structures have several limitations in handling partially ordered and partially instantiated plans. The foremost among these is that they are incapable of exploiting redundancy in the plan. In this paper, we explore multi-contributor causal structures as a way of overcoming these limitations. We will provide a general formulation for multi-contributor causal links, and explore the properties of several special classes of this formulation. We will also discuss their applications in plan generation, modification and generalization.

O-Plan2 is an AI planning architecture which supports research into a number of aspects of planning, scheduling and control. It is based on earlier work on the O-Plan System which was directed towards plan generation. The paper explores the different types of choice ordering decisions which need to be made in an architecture for command and control. The mechanisms for choice ordering and selection in the original O-Plan system were found to be too general for efficient use for all purposes. The paper describes a number of choice ordering mechanisms provided in the O-Plan2 Architecture which provide specialised mechanisms more suited to the range of different ordering problems that arise in planning, scheduling and control applications.

As realistic systems are built and commercial toolkits are created for planning and scheduling applications, it becomes increasingly important that modularisation of these "standard" components is attempted. This can lead to re-usability, embeddability and improved implementation provision. One early example of a standardised component is the Truth Criterion condition establishment algorithm which is now found at the heart of most activity based planning systems. The modularisation of this algorithm has led to an explosion of further development, empirical and theoretical study of this component. The provision of powerful constraint management languages and tools could lead to a rapid expansion of the benefits to be gained by identifying more standard components that can be combined and re-used in planning and scheduling systems. O-Plan2 is a command, planning and control architecture which has an open modular structure intended to allow experimentation on or replacement of various components. This paper describes the modular structure of the system along with the internal and external interface languages and protocols which are being developed on the O-Plan2 project. The research is seeking to isolate functionality that may be generally required in a number of applications and across a number of different planning, scheduling and control systems. This paper is intended to further discussions on the identification of suitable "standard" re-usable components in planning and scheduling systems.

This paper describes COLLAGE, a planner that utilizes a variety of non-traditional methods for plan construction within a partitioned or localized reasoning framework. The COLLAGE domain representation and plan construction methods are based on the use of action-based constraints. Not only do such constraints yield highly natural domain encodings, but they are also associated with cost-effective plan construction methods. The additional use of domain structure to partition the reasoning space and guide planning search likewise results in more efficient planning. COLLAGENS unconventional approach to planning is motivated by a specific target domain class: domains with parallel activities that require complex forms of coordination. We describe the design decisions underlying the COLLAGE approach in the context of a novel characterization of planning: a six-dimensional planning feature space. We also discuss how this six-dimensional space can serve a larger role in the planning community- as a framework for comparing planning techniques, enhancing communication between researchers, elucidating ternfinological difficulties, and clarifying the relationship between a domain's characteristics and its most suitable planning technique.