Planning Domain Definition Language (PDDL)

We consider the challenges that arise when plans are based on an incorrect representation of the domain in which they are executed. We describe how information about plan formation, and how the way in which each plan step is related to the domain representation, can help identify problems in the underlying ontology when plan failure is encountered. We introduce the notion of a plan deconstructor, used to extract this information from the domain representation so that is it available when plan failure occurs. Thus a more accurate ontology can be developed and more robust plans can be formed.

Non-linear continuous change is common in real-world problems, especially those that model physical systems. We present an algorithm which builds upon existent temporal planning techniques based on linear programming to approximate non-linear continuous monotonic functions. These are integrated through a semantic attachment mechanism, allowing external libraries or functions that are difficult to model in native PDDL to be evaluated during the planning process. A new planning system implementing this algorithm was developed and evaluated. Results show that the addition of this algorithm to the planning process can enable it to solve a broader set of planning problems.

Real-world planning problems often feature complex temporal and numeric characteristics. These include concurrent activities and also effects that involve continuous change. This work presents the formalism behind reasoning with required concurrency that involves continuous change in temporal planning problems, together with a set of techniques to solve a class of tasks that to date are hard to solve with current state-of-the-art temporal planners. The motivation for this work is scenarios where planning actions have rich numeric effects on some physical system. One such example is automated demand dispatch for electricity provision, where actions that fulfil customer requirements have an effect on various metrics, such as wattage or costs, which could be subject to operational or commercial constraints. An algorithm that handles discrete interference of linear continuous effects, referred to as constants in context, is presented. This technique allows discrete actions to update the rate of change of a continuous effect taking place concurrently. This work builds on techniques used in current temporal planners that make use of linear programming, and also extends the heuristic to guide the search to a solution. This algorithm was implemented in a new temporal and numeric planner called DICE and evaluated with some benchmark domains. PDDL, the current de facto standard language for planning domains and corresponding planning tasks, was extended to support interactions with external class modules. The proposed extension, PDDLx, defines a generic planner-solver interface for both discrete and continuous effects. This enables planners that implement this interface to interact with external solvers and incorporate context-specific effects in a black-box fashion, enabling complex numeric behaviour to be encapsulated within such modules. Non-linear monotonic continuous effects, defined in the proposed PDDLx extension, are integrated within the planner using a non-linear iterative convergence algorithm. It searches for a linear approximation within an acceptable configurable error margin, which is then used within the linear program computed for each temporal state. This algorithm proves to be effective in various domains where non-linear continuous behaviour is prevalent. This technique was implemented as an extension to DICE, called uNICOrn, which performs non-linear iterative convergence for continuous effects whose duration needs to be determined by the planner. uNICOrn was also evaluated with some benchmark non-linear domains. A case study on the automated demand dispatch domain is presented to demonstrate the use of the planning algorithms proposed in this thesis. Linear and non-linear planning problems are evaluated and the performance of uNICOrn on these problem instances was analysed. This work builds on current techniques used for temporal planning with continuous numeric behaviour using linear programming, and enhances them to remove some of their intrinsic limitations. The result is a set of algorithms that are more effective in solving real-world applications that involve continuous change and rich numeric behaviour. This work would not have been possible without the academic, moral and financial support of various people and organisations. I am deeply grateful to my supervisors, Professor Maria Fox and Professor Derek Long, for their guidance, inspiration and continuous encouragement. Apart from their in-depth knowledge, they provided me with unique opportunities to get international exposure within the A.I. community, which has enriched my experience even further. This research was funded and supported by the UK Engineering and Physical Sciences Research Council (EPSRC) as part of the project entitled The Autonomic Power System (Grant Ref: EP/I031650/1). I would like to thank all the researchers participating in the project for their insight and interesting discussions during the various meetings and events. I would also like to thank my colleagues and friends at King's College London, whose company and friendship made the whole experience even more pleasant and enjoyable. The regular meetings of the King's A.I. Planning Research Group were also instrumental to generate ideas and gather invaluable feedback throughout my research. Last and foremost, I would like to show my utmost gratitude to my wife Thérèse, my parents, and the rest of our respective families for their immeasurable patience and support throughout my academic and professional endeavours.

This paper describes a preliminary work on Machine T , a temporal partial order planner based on the use of temporal constraint networks to represent and propagate temporal constraints over the actions of a plan. The use of this formalism provide several advantages like the possibility of including deadline goals, makespans and imprecise temporal references able to obtain a flexible timeline for the execution of temporal plans. The expressiveness of the approach also meet the main temporal specifications of PDDL 2.1 level 3 like durative actions and invariant conditions.

The Planning Domain Description Language (PDDL) was first devised and proposed by a committee led by Drew McDermott in preparation for what turned out to be the first in a series of international planning competitions, held in 1998. Its role at that stage was simply to support a uniform foundation for the competitors in order to simplify the problem of domain encoding. Prior to its development, planners used a mixture of similar, but slightly different, domain description languages. The differences made sharing domains significantly ...

In industrial domains such as manufacturing control, a trend away from centralized planning and scheduling towards more flexible distributed agent-based approaches could be observed over recent years. To be of practical relevance, the local control mechanisms of the autonomous agents must be able to dependably adhere and dynamically adjust to complex numeric goal systems like business key performance indicators in an economically beneficial way. However, planning with numeric state variables and objectives still poses a challenging task within the field of artificial intelligence (AI). In this article, a new general-purpose AI planning approach is presented that operates in two stages and extends existing domain-independent modeling formalisms like PDDL with continuous (i.e., infinite-domain) numeric action parameters, which are currently still unsupported by state-of-the-art AI planners. In doing so, it enables the solution of mathematical optimization problems at the action level of the planning tasks, which are inherent to many real-world control problems. To deal with certain difficulties concerning reliable and fast detection of action applicability that arise when planning with real-valued action parameters, the implemented planner allows resorting to an adjustable ''satisficing'' strategy by means of partial execution and subsequent repair of infeasible plans over the course of time. The functioning of the system is evaluated in a multi-agent simulation of a shop floor control scenario with focus on the effects the possible problem cases and different degrees of satisficing have on attained plan quality and total planning time. As the results demonstrate the basic practicability of the approach for the given setting, this contribution constitutes an important step towards the effective and dependable integration of complex numeric goal systems and non-linear multi-criteria optimization tasks into autonomous agent-controlled industrial processes in a reusable, domain-independent way.

The International Planning Competitions, run by Drew Mc-Dermott in 1998 and Fahiem Bacchus in 2000, have provided an important spur to the planning community, encouraging the development of planning technology and of a wide selection of planning bench mark domains. One of the most important contributions has been the introduction of a widely accepted standard for domain description, Drew McDermott's PDDL language (McDermott 2000), leading to the sharing of domains and planning systems. An equally important ...

Ce papier étudie l'importance de la complétude du marché en comparant le taux d'aversion au risque estimé dans le cas de marchés complets et incomplets. Dans le cas de marchés incomplets, nous utilisons des données de consommation de 50 états américains. Bien que l'utilisation de données agrégées au niveau des états est conceptuellement moins justifiée que l'utilisation de données individuelles, ce choix est préférable du fait que les données agrégées sont moins sujettes aux erreurs de mesures. Nous trouvons que le taux d'aversion au risque sous l'hypothèse de marchés incomplets est plus petit. De plus, la prise en compte du deuxième et du troisième moment de la distribution transversale de la croissance de la consommation dans le pricing kernel réduit le taux d'aversion au risque estimé. Ces résultats suggèrent que l'incomplétude du marché devrait être regardé comme une composante importante de la solution du puzzle de la prime d'équité. Mots clés : hétérogénéité, risque inhérent à la consommation, marchés incomplets, modèle d'évaluation d'actif basé sur la consommation, aversion au risque, puzzle de la prime d'équité This paper investigates the importance of market incompleteness by comparing the rates of risk aversion estimated from complete and incomplete markets environments. For the incomplete-markets case, we use consumption data for 50 U.S. states. While the use of state-level data is conceptually inferior to the use of data on individual consumption, it may be preferable because state-level data are less susceptible to measurement errors. We find that the rate of risk aversion under the incompletemarkets setup is much lower. Furthermore, including the second and third moments of the crosssectional distribution of consumption growth in the pricing kernel lowers the estimate of risk aversion. These findings suggest that market incompleteness ought to be seen as an important component of solutions to the equity premium puzzle.

In this paper, we argue that it would be fruitful to revisit foreign aid's potential as an insurance mechanism against macroeconomic shocks. In a simple model of aid flows between two endowment economies, we show that at least three fourths of the large welfare costs of macroeconomic fluctuations in poor countries could be alleviated by a simple reallocation of aid flows across time.

The work aims at extending Answer Set Programming (ASP) with the possibility of quickly introducing new predefined constructs and to deal with compound data structures. We show how ASP can be extended with 'template'predicate's definitions by introducing a wellsuited form of second order logics. We present language syntax and give its operational semantics. We show that the theory supporting our ASP extension is sound, and that program encodings are evaluated as efficiently as ASP programs. Examples show how ...

Le dispositif d'observation que nous avons adopté est simple. Nous avons repéré les sites à l'aide de moteurs de recherches (google et alta vista) et de plusieurs portails. Lorsque cela ne permettait pas de trouver le site, cette recherche était complétée par la visualisation du ...

are circulated to promote discussion and provoque comments. Any references to discussion papers should clearly state that the paper is preliminary. Materials published in this series may subject to further publication. The paper was selected for the 1 st Budapest Summer Workshop for young economists, organised by the KTK/IE on 10-11 July 2001. The Budapest Summer Workshops intend to bring together young economists with foreign PhD education, frequently still working or studying abroad.

In prior work, we examined the possibility of sound generation from colon tissue scan data using vocal synthesis models. In this work, we review key results and present extensions to the prior work. Sonification entails the mapping of data values to sound synthesis parameters such that informative sounds are produced by the chosen sound synthesis model. We review the physical equations and technical highlights of a vocal synthesis model developed by Cook. Next we present the colon tissue scan data gathered, and discuss processing steps applied to the data. Finally, we review preliminary results from a simple sonification map. New findings regarding perceptual distance of vowel sounds are presented 1 2. 1 Ryan J. Cassidy supported by the Natural Sciences and Engineering Research Council of Canada and The Banff Centre (Banff, Canada). 2 Supported by DARPA award F41624-03-1-7000.

Na Mata Atlântica, a conversão de áreas lorestadas em outro tipo de uso do solo ica condicionada, entre outros, à classiicação do estágio sucessional da loresta, sendo que diferentes estágios reletem em diferentes tamanhos e possibilidades de uso. Os parâmetros empregados na deinição dos estágios sucessionais deste Bioma encontram-se descritos em resoluções especíicas para cada estado, sendo que a resolução CONAMA 04/94 é própria para as lorestas de Santa Catarina. Este estudo objetivou avaliar a aplicação dos critérios quantitativos da Resolução CONAMA 04/94 na deinição dos estágios sucessionais da Floresta Ombróila Densa, tendo como estudo de caso os fragmentos lorestais da planície Norte do estado de Santa Catarina. Ao todo foram avaliados seis fragmentos localizados em diferentes unidades pedológicas, sendo avaliados o DAP e a altura média dos indivíduos, além da área basal e espécies indicadoras. Os resultados demonstram que a CONAMA 04/94 ignora características itoisionômicas particulares das planícies arenosas do norte do estado. Também é demonstrada a incoerência do texto da CONAMA 04/94 na deinição dos parâmetros estruturais e espécies indicadoras. Os resultados sugerem a necessidade urgente de revisão do texto da Resolução CONAMA 04/94, uma vez que os fragmentos lorestais do estado podem estar sendo sujeitos a usos de solo incompatíveis com os deinidos nas políticas públicas.

We propose a detailed Monte Carlo study of model selection criteria when the exact maximum likelihood (EML) method is used to estimate ARFIMA processes. More specifically, our object is to assess the performance of two automatic selection criteria in the presence of long−term memory: Akaike and Schwarz information criteria. Two special processes are considered: a pure fractional noise model (ARFIMA(0,d,0)) and an ARFIMA(1,d,0) process. For each criterion, we compute bias and root mean squared error for various d and AR(1) parameter values. Obtained results suggest that the Schwarz information criterion frequently selects the right model. Moreover, this criterion outperforms the other one in terms of bias and RMSE, for both pure fractional noise and ARFIMA processes. We are grateful to Emmanuel Dubois for helpful comments and invaluable computer assitance. We also thank an anonymous referee for his comments.

Vanbolnička potrošnja antibiotika u Republici Srpskoj u periodu od 2009. do 2014. godine SAŽETAK Uvod. Antimikrobna rezistencija predstavlja veoma ozbiljnu javnozdravstvenu prijetnju širom svijeta. Glavni uzrok nastanka antimikrobne rezistencije je neracionalna upotreba antimikrobnih lijekova. Cilj rada. Ciljevi rada su sljedeći: analizirati podatke o vanbolničkoj potrošnji antibiotika u Republici Srpskoj, u periodu od 2009. do 2014. godine, zatim prikazati vanbolničku potrošnju prvih šest antibiotika na petom nivou ATC klasifi kacije te uporediti potrošnju antibiotika u Republici Srpskoj u odnosu na potrošnju antibiotika u zemljama Evropske unije. Materijal i metode. Prikupljanje i analiza podataka o potrošnji lijekova se vrši ATC/DDD metodologijom koja je ustanovljena od strane SZO. Podaci su preuzeti iz Izvještaja o potrošnji i prometu gotovih lijekova u Republici Srpskoj, u periodu od 2009. do 2014. godine, koji se temelje na podacima dostavljenim iz apoteka otvorenog tipa registrovanih na području Republike Srpske. The goal of the article is to analyze data on outpatient antibiotic consumption in the Republic of Srpska during the period from 2009 to 2014, to present outpatient consumption of the fi rst six antibiotics (the fi fth level of the ATC classifi cation), and, fi nally, to compare the antibiotic use in Republic of Srpska in relation to the antibiotic use in European Union countries. Materials and Methods. Collection and analysis of data on drug consumption is done using ATC/DDD methodology, established by the WHO. Data are taken from the report on drug consumption and distribution in the Republic of Srpska during the period from 2009 to 2014, which are based on submitted data from pharmacies registered in the Republic of Srpska. Results. Results on outpatient antibiotic consumption in the Republic of Srpska, during the period from 2009 to 2014, indicate a total decrease by 3.41 DDDs/1000 inhabitants/day, with the exception of 2010 and 2013, when the antibiotic consumption was increased. In 2014, total outpatient antibiotic consumption in the Republic of Srpska was 31.2% lower than the average consumption in the European Union. During the period from 2010 to 2014, the European Union had a signifi cant increase in antibiotic consumption, while antibiotic consumption in the Republic of Srpska had a tendency of falling. Conclusion. Presented results show a quantitative analysis of antibiotic consumption and provide insight into the outpatient use of antibiotics in the Republic of Srpska, and can be used for further pharmacoepidemiological analysis of antibiotic consumption, which would give a better insight into the therapeutic practice, with the aim of improving the rational pharmacotherapy in the Republic of Srpska.

The BRS can be quantified as the slope between SBP and RR values identified in baroreflex events, estimated by ordinary least squares (OLS) minimization. Quantile regression (QR) is a more robust procedure than OLS and allows a more complete characterization of the data, by estimating conditional functions for different quantiles of interest. In this work, OLS and QR for BRS estimation are compared regarding slope estimates and dispersion. The EuroBaVar results indicate that OLS slope and QR slopes at different quantiles do not exhibit significant differences. Also, OLS and QR slopes require similar number of beats to achieve a given BRS precision in stationary recordings. Finally, BRS estimated with OLS exhibit relative dispersion lower than 10% and 5% when computed from stationary recordings of approximately 3 and 9 minutes length, respectively.

Computer games and the accompanying entertainment industry branch has become a major market factor. AI techniques are successfully applied to tasks like path planning or intelligent swarm behavior. On the decision-making level the state of the art are state machines with a xed set of behaviors. The perception of the computer player is perfect. They exactly know where the other players are located, even if they cannot see them. This approach seems to be limited for intelligent decision making. Therefore, we propose another approach. We use a variant of the logic-based action language GOLOG for implementing so-called game bots in the UNREAL TOURNAMENT 2004 environment. First results show that we can compete with the omniscience game bots in the Unreal domain.

Many planning domains have temporal features that can be expressed as durations associated with actions. Unfortunately, the conservative model of actions of most temporal planners is not appropriate for some domains which require more expressive models. Level 3 of PDDL2.1 introduces a model of durative actions which includes local conditions and effects to be satisfied at different times during the execution of the actions, thereby giving the planner freedom to plan concurrent actions. This paper presents a temporal planning system which combines the ideas of Graphplan and TGP to plan with such actions. Although that system becomes more complex, the experimental results demonstrate it remains feasible as a way to deal with durative actions. 1 INTRODUCTION Most temporal planners appeared in the recent literature, such as parcPLAN, TGP or TP4 [3, 8, 5] have yielded some success when dealing with temporality on actions. These planners have adopted the same conservative model of actions of non-temporal planners, which means that two actions cannot overlap in any way if they have conflicting preconditions or effects. This makes it possible to produce reasonable plans in most typical planning domains, but there exist some domains which require a richer model of actions in which better quality plans can be found. The new version of PDDL, called PDDL2.1 [4], provides a level 3 with a model of durative actions which subsumes the conservative model of actions. Level 3 allows actions to overlap even when their preconditions or effects refer to the same propositions. This paper presents a Temporal Planning SYStem (from now on TPSYS) to manage the model of durative actions proposed in level 3 of PDDL2.1. TPSYS is based on a three-stage process, which combines the ideas of Graphplan [1] and TGP [8] to deal with such durative actions. Hence, the main contributions of this paper are: An analysis of how durative actions can be managed in a Graphplan-based approach. An explanation of how a compact temporal graph can be generated, without ÒÓ ÓÔ actions and delete-edges. An extension of the mutual exclusion reasoning to manage PDDL2.1 durative actions, based on the work of TGP. A description of the plan extraction stage and the way it obtains the plan of optimal duration (in terms of the makespan). Some experimental results showing the importance of the mutex reasoning in richer models of actions, as indicated in [8].