Partial-order Planning Research Articles

On-the-fly scheduling as a manifestation of partial-order planning and dynamic task values.

The aim of this study was to develop a computational account of the spontaneous task ordering that occurs within jobs as work unfolds ("on-the-fly task scheduling"). Air traffic control is an example of work in which operators have to schedule their tasks as a partially predictable work flow emerges. To date, little attention has been paid to such on-the-fly scheduling situations. We present a series of discrete-event models fit to conflict resolution decision data collected from experienced controllers operating in a high-fidelity simulation. Our simulations reveal air traffic controllers' scheduling decisions as examples of the partial-order planning approach of Hayes-Roth and Hayes-Roth. The most successful model uses opportunistic first-come-first-served scheduling to select tasks from a queue. Tasks with short deadlines are executed immediately. Tasks with long deadlines are evaluated to assess whether they need to be executed immediately or deferred. On-the-fly task scheduling is computationally tractable despite its surface complexity and understandable as an example of both the partial-order planning strategy and the dynamic-value approach to prioritization.

Optimally Relaxing Partial-Order Plans with MaxSAT

Partial-order plans (POPs) are attractive because of their least commitment nature, providing enhanced plan flexibility at execution time relative to sequential plans. Despite the appeal of POPs, most of the recent research on automated plan generation has focused on sequential plans. In this paper we examine the task of POP generation by relaxing or modifying the action orderings of a plan to optimize for plan criteria that promotes flexibility in the POP. Our approach relies on a novel partial weighted MaxSAT encoding of a plan that supports the minimization of deordering or reordering of actions. We further extend the classical least commitment criterion for a POP to consider the number of actions in a solution, and provide an encoding to achieve least commitment plans with respect to this criterion. We compare the effectiveness of our approach to a previous approach for POP generation via sequential-plan relaxation. Our results show that while the previous approach is proficient at heuristically finding the optimal deordering of a plan, our approach gains greater flexibility with the optimal reordering.

Composition method of folk-tales based on STRIPS-like framework

We propose a method for supporting story composition of folk-tales. Our idea is to derive events in a story and their order constraints from the outline specified by a storywriter. To achieve this, we introduce a technique of partial order planning, and represent the situations and the events in a story based on the framework of STRIPS. In our method, events and order constraints are derived according to the policy that respects the intentions of a storywriter specified in his/her outline. We investigated the effectiveness of the policy in a case study.

Narrative Planning: Balancing Plot and Character

Narrative, and in particular storytelling, is an important part of the human experience. Consequently, computational systems that can reason about narrative can be more effective communicators, entertainers, educators, and trainers. One of the central challenges in computational narrative reasoning is narrative generation, the automated creation of meaningful event sequences. There are many factors -- logical and aesthetic -- that contribute to the success of a narrative artifact. Central to this success is its understandability. We argue that the following two attributes of narratives are universal: (a) the logical causal progression of plot, and (b) character believability. Character believability is the perception by the audience that the actions performed by characters do not negatively impact the audience's suspension of disbelief. Specifically, characters must be perceived by the audience to be intentional agents. In this article, we explore the use of refinement search as a technique for solving the narrative generation problem -- to find a sound and believable sequence of character actions that transforms an initial world state into a world state in which goal propositions hold. We describe a novel refinement search planning algorithm -- the Intent-based Partial Order Causal Link (IPOCL) planner -- that, in addition to creating causally sound plot progression, reasons about character intentionality by identifying possible character goals that explain their actions and creating plan structures that explain why those characters commit to their goals. We present the results of an empirical evaluation that demonstrates that narrative plans generated by the IPOCL algorithm support audience comprehension of character intentions better than plans generated by conventional partial-order planners.

Solving Sequential Planning Problems via Constraint Satisfaction

Planning problems deal with finding a sequence of actions that transfer the initial state of the world into a desired state. Frequently such problems are solved by dedicated algorithms but there exist planners based on translating the planning problem into a different formalism such as constraint satisfaction or Boolean satisfiability and using a general solver for this formalism. The paper describes how to enhance existing constraintmodels of sequential planning problems by using techniques such as symmetry breaking (dominance rules), singleton consistency, nogoods, lifting, or techniques motivated by the partial-order planning.

On linear logic planning and concurrency

We present an approach to linear logic planning where an explicit correspondence between partial order plans and multiplicative exponential linear logic proofs is established. This is performed by extracting partial order plans from sound and complete encodings of planning problems in multiplicative exponential linear logic. These partial order plans exhibit a non-interleaving behavioural concurrency semantics, i.e., labelled event structures. Relying on this fact, we argue that this work is a crucial step for establishing a common language for concurrency and planning that will allow to carry techniques and methods between these two fields.

Evidence of an Inferior Total-Order Planning Strategy in Patients with Frontotemporal Dementia

We investigated planning abilities in patients with frontal temporal dementia (FTD) and normal controls (NC) using a chore location planning task. Planning tasks with increasing complexity could be either solved by constructing a linear total plan (total-order planning) or partially ordered sub-plans (partial-order planning). The NC group appeared to use partial-order planning, while the FTD group appeared to use total-order planning based on error pattern in performance, task-solution, and planning time. Our results are a reminder that besides social impairments, FTD patients also demonstrate higher-order cognitive impairment in domains such as planning and that these impairments may also have a profound effect upon the patients' day-to-day functioning.

Handling fuzzy temporal constraints in a planning environment

An interleaved integration of the planning and scheduling process is presented with the idea of including soft temporal constraints in a partial order planner that is being used as the core module of an intelligent decision support system for the design forest fire fighting plans. These soft temporal constraints have been defined through fuzzy sets. This representation allows us a flexible representation and handling of temporal information. The scheduler model consists of a fuzzy temporal constraints network whose main goal is the consistency checking of the network associated to each partial order plan. Moreover, we present a model of estimating this consistency, and show the monitoring and rescheduling capabilities of the system. The resulting approach is able to tackle problems with ill defined knowledge, to obtain plans that are approximately consistent and to adapt the execution of plans to unexpected delays.

A goal-oriented interface to consumer electronics using planning and commonsense reasoning

We are reaching a crisis with design of user interfaces for consumer electronics. Flashing 12:00 time indicators, push-and-hold buttons, and interminable modes and menus are all symptoms of trying to maintain a one-to-one correspondence between functions and physical controls, which becomes hopeless as the number of capabilities of devices grows. The root of this problem is that the devices do not have knowledge of (a) their own capabilities (b) the scenarios where they can be of use. By using commonsense knowledge and an AI partial-order planner we propose instead to orient interfaces around the goals that users have for the devices. We present Roadie, a user interface agent that provides intelligent context-sensitive help and assistance for a network of consumer devices. Roadie uses a commonsense knowledge base to map between user goals and functions of the devices, and an AI partial-order planner to provide mixed-initiative assistance with executing multi-step procedures and debugging help when things go wrong.

AN INTEGRATED LIFE CYCLE FOR WORKFLOW MANAGEMENT BASED ON LEARNING AND PLANNING

The ability to describe business processes as executable models has always been one of the fundamental premises of workflow management. Yet, the tacit nature of human knowledge is often an obstacle to eliciting accurate process models. On the other hand, the result of process modeling is a static plan of action, which is difficult to adapt to changing procedures or to different business goals. In this article, we attempt to address these problems by approaching workflow management with a combination of learning and planning techniques. Assuming that processes cannot be fully described at build-time, we make use of learning techniques, namely Inductive Logic Programming (ILP), in order to discover workflow activities and to describe them as planning operators. These operators will be subsequently fed to a partial-order planner in order to find the process model as a planning solution. The continuous interplay between learning, planning and execution aims at arriving at a feasible plan by successive refinement of the operators. The approach is illustrated in two simple scenarios. Following a discussion of related work, the paper concludes by presenting the main challenges that remain to be solved.

Combining Proof Plans with Partial Order Planning for Imperative Program Synthesis

The structured programming literature provides methods and a wealth of heuristic knowledge for guiding the construction of provably correct imperative programs. We investigate these methods and heuristics as a basis for mechanizing program synthesis. Our approach combines proof planning with conventional partial order planning. Proof planning is an automated theorem proving technique which uses high-level proof plans to guide the search for proofs. Proof plans are structured in terms of proof methods, which encapsulate heuristics for guiding proof search. We demonstrate that proof planning provides a local perspective on the synthesis task. In particular, we show that proof methods can be extended to represent heuristics for guiding program construction. Partial order planning complements proof planning by providing a global perspective on the synthesis task. This means that it allows us to reason about the order in which program fragments are composed. Our hybrid approach has been implemented in a semi-automatic system called Bertha. Bertha supports partial correctness and has been tested on a wide range of non-trivial programming examples.

Using Autonomous Planning Agents to Provide Model-based Decision-making Support

The purpose of this research is to investigate the potential of software agents in performing automated planning within Decision-Making Support Systems (DMSS). A planning agent that uses a partial-order planning mechanism to monitor and maintain corporate profit is built in Prolog and Java and embedded in an agent-integrated DMSS. The planning mechanism used by the agent supports the sequencing and interfacing of models within the DMSS for the purpose of generating alternatives. From this experience, an architecture for a general planning agent is constructed. Benefits of the agent-enhanced, planning DMSS are then discussed.

Event Calculus Planning Through Satisfiability

Previous work on event calculus planning has viewed it as an abductive task, whose implementation through abductive logic programming reduces to partial order planning. By contrast, this paper presents a formal framework for tackling event calculus planning through satisfiability, in the style of Kautz and Selman. A provably correct conjunctive normal form encoding for event calculus planning problems is supplied, rendering them soluble by an off-the-shelf SAT solver.

A New HTN Planning Framework for Agents in Dynamic Environments

In a dynamic environment, even if an agent makes a plan to obtain a goal, the environment might change while the agent is executing the plan. In that case, the plan, which was initially valid when it was made, might later become invalid. Furthermore, in the process of replanning, it is necessary to take into account the side effects of actions already executed. Nowadays, HTN planning is becoming popular among agent researchers because its task decomposition algorithm is efficient and suitable for joint planning in multi-agent systems. However, the dynamic replanning algorithm of HTN planning has not yet been established. In order to solve this problem, we have previously presented an agent life cycle that integrates HTN planning, action execution, knowledge updates, and plan modification. In that agent life cycle, the plans are always kept valid according to the most recent knowledge and situation. However, when one plan does not work and the agent starts using an alternative plan, it sometimes undoes some actions that do not have to be undone. The source of this problem lies in the need to have total-order plans. In other words, all the actions have to be sequentially ordered even when the order is not important in some parts of the plan. This paper extends our previous agent life cycle and presents new ways to avoid unnecessary action cancellation. This enables the agent to handle partial-order plans and to make use of two kinds of undoing actions: sequential undoing actions and concurrent undoing actions.

Temporal Reasoning for a Collaborative Planning Agent in a Dynamic Environment

We present a temporal reasoning mechanism for an individual agent situated in a dynamic environment such as the web and collaborating with other agents while interleaving planning and acting. Building a collaborative agent that can flexibly achieve its goals in changing environments requires a blending of real-time computing and AI technologies. Therefore, our mechanism consists of an Artificial Intelligence (AI) planning subsystem and a Real-Time (RT) scheduling subsystem. The AI planning subsystem is based on a model for collaborative planning. The AI planning subsystem generates a partial order plan dynamically. During the planning it sends the RT scheduling subsystem basic actions and time constraints. The RT scheduling subsystem receives the dynamic basic actions set with associated temporal constraints and inserts these actions into the agent's schedule of activities in such a way that the resulting schedule is feasible and satisfies the temporal constraints. Our mechanism allows the agent to construct its individual schedule independently. The mechanism handles various types of temporal constraints arising from individual activities and its collaborators. In contrast to other works on scheduling in planning systems which are either not appropriate for uncertain and dynamic environments or cannot be expanded for use in multi-agent systems, our mechanism enables the individual agent to determine the time of its activities in uncertain situations and to easily integrate its activities with the activities of other agents. We have proved that under certain conditions temporal reasoning mechanism of the AI planning subsystem is sound and complete. We show the results of several experiments on the system. The results demonstrate that interleave planning and acting in our environment is crucial.

On the Hybrid Propositional Encodings of Planning

Recently, casting planning as propositional satisfiability (SAT) has been shown to be an efficient technique of plan synthesis. This article is a response to the recently proposed challenge of developing novel propositional encodings that are based on a combination of different types of plan refinements and characterizing the tradeoffs. We refer to these encodings as hybrid encodings. An investigation of these encodings is important, because this can give insights into what kinds of planning problems can be solved faster with hybrid encodings.Encodings based on partial–order planning and state–space planning have been reported in previous research. We propose a new type of encoding called a unifying encoding that subsumes these two encodings. We also report on several other hybrid encodings. Next, we show how the satisfiability framework can be extended to incremental planning. State–space encoding is attractive because of its lower size and causal encoding is attractive because of its highest flexibility in reordering steps. We show that hybrid encodings have a higher size and a lower flexibility in step reordering and, thus, do not combine the best of these encodings. We discuss in detail several specific planning scenarios where hybrid encodings are likely to be superior to nonhybrid encodings.

Partial and total‐order planning: evidence from normal and prefrontally damaged populations

AbstractThis paper examines human planning abilities, using as its inspiration planning techniques developed in artificial intelligence. AI research has shown that in certain problems partial‐order planners, which manipulate partial plans while not committing to a particular ordering of those partial plans, are more efficient than total‐order planners, which represent all partial plans as totally ordered. This research asks whether total‐order planning and/or partial‐order planning are accurate descriptions of human planning, and if different populations use different planning techniques. Using a simple planning task modeled after tasks designed in artificial intelligence we tested 7–8 year‐old children, 11–13 year‐old children, adult controls, and adults with damage to the prefrontal cortex. We found that adults and older children exhibited performance on planning tasks of varying complexity which matched that of artificial partial‐order planners, and that this pattern of performance did not vary with multiple presentations of the planning task. In contrast, young children and adults with damage to the prefrontal cortex exhibited performance matching that of artificial total‐order planners. This pattern of performance did vary, however, with multiple presentations of the planning task, with the young children and adults with cortical damage displaying aspects of total‐order planning. In a further study we found that adolescents who had sustained damage to the prefrontal cortex as children displayed two different patterns of performance; when measures of reaction time were analyzed they revealed a pattern of performance suggestive of partial‐order plan representations. However, analyses of the adolescents' protocols revealed a pattern of performance suggestive of total‐order plan representations. The significance of these results to psychology, neuroscience, and artificial intelligence are discussed.

Partial-Order Planning with Concurrent Interacting Actions

In order to generate plans for agents with multiple actuators, agent teams, or distributed controllers, we must be able to represent and plan using concurrent actions with interacting effects. This has historically been considered a challenging task requiring a temporal planner with the ability to reason explicitly about time. We show that with simple modifications, the STRIPS action representation language can be used to represent interacting actions. Moreover, algorithms for partial-order planning require only small modifications in order to be applied in such multiagent domains. We demonstrate this fact by developing a sound and complete partial-order planner for planning with concurrent interacting actions, POMP, that extends existing partial-order planners in a straightforward way. These results open the way to the use of partial-order planners for the centralized control of cooperative multiagent systems.

Automatic generation of control sequences for manufacturing systems based on partial order planning techniques

This work presents an approach for the application of artificial intelligence planning techniques to the automatic generation of control sequences for manufacturing systems. These systems have some special features that must be considered in the planning process, but there are difficulties when the usual models of action are used to deal with these features. In this work, a specialized interval-based model of action is defined by extending the classic model of strips giving it more expressiveness so that it is able to deal with these features. In consequence, a specialized planning algorithm for this model of action, called machine, is defined based on a general partial order planning scheme, and it is able to obtain control sequences for manufacturing systems. These control sequences are actually the control program skeleton and may be easily translated into real control programs expressed as GRAFCET charts.

Hierarchically merging plans in decomposable domains

Recent works in domain-independent plan merging have shown that the optimal plan merging problem is NP-hard, and heuristic algorithms have been proposed to generate near-optimal solutions. We have developed a plan merging methodology that merges partial-order plans based on the notion of plan fragments. In contrast to previous works, mergeability classes no longer necessarily form a partition over the set of actions in the input plans. This methodology applies to a class of planning domains which are decomposable. We also investigate the previously unexplored notion of alternative actions in domain-independent plan merging, which can improve the quality of the resulting merged plan, and a novel approach is presented for removing cyclic dependencies that may result during the plan merging process. We provide theoretical analyses of several algorithms for computing the minimum cost cover of plan fragments, a central component of the methodology. We support the theoretical analysis of these algorithms with experimental data to show that a greedy approach is near-optimal and efficient.