No description available.

Cette thèse de doctorat analyse les concepts impliqués dans la prise de décisions de groupes d'agents et applique ces concepts dans la création d'un cadre théorique et pratique qui permet la reconnaissance de comportements de groupes. Nous allons présenter une vue d'ensemble de la théorie de l'intention, étudiée dans le passé par quelques grands théoriciens comme Searle, Bratmann et Cohen, et nous allons montrer le lien avec des recherches plus récentes dans le domaine de la reconnaissance de comportements. Nous allons étudier les avantages et inconvénients des techniques les plus avancées dans ce domaine et nous allons créer un nouveau modèle qui représente et détecte les comportements de groupes. Ce nouveau modèle s'appelle Multiagent-Abstract Hidden Markov mEmory Model (M-AHMEM) et résulte de la fusion de modèles déjà existants, le but étant de créer une approche unifiée du problème. La plus grande partie de cette thèse est consacrée à la présentation détaillée du M-AHMEM et de l'algorithme responsable de la reconnaissance de comportements. Notre modèle sera testé sur deux applications différentes : l'analyse gesturale humaine et la fusion multimodale des données audio et vidéo. A travers ces deux applications, nous avançons l'argument qu'un ensemble de données constitué de plusieurs variables corrélées peut être analysé efficacement sous un cadre unifié de reconnaissance de comportements. Nous allons montrer que la corrélation entre les différentes variables peut être modélisée comme une coopération ayant lieu à l'intérieur d'une équipe et que la reconnaissance de comportements constitue une approche moderne de classification et de reconnaissance de patrons.

The ability to infer and mediate intentions has been recognized as a crucial task in recent robotics research, where it is agreed that robots are required to be equipped with intentional mechanisms in order to participate in collaborative tasks with humans. Reasoning about - or rather, perceiving - intentions enables robots to infer what other agents are doing, to communicate what are their plans, or to take proactive decisions. Intent recognition relates to several system requirements, such as the need of an enhanced collaboration mechanism in human-machine interactions, the need for adversarial technology in competitive scenarios, ambient intelligence, or predictive security systems. When attempting to describe what an intention is, agreement exists to represent it as a plan together with the goal it attempts to achieve. Being compatible with computer science concepts, this representation enables to handle intentions with methodologies based on planning, such as the Planning Domain Description Language or Hierarchical Task Networks. In this licentiate we describe how intentions can be processed using classical planning methods, with an eye also on newer technologies such as deep networks. Our goal is to study and define computational models that would allow robotic agents to infer, construct and mediate intentions. Additionally, we explore how intentions in the form of abstract plans can be grounded to sensorial data, and in particular we provide discussion on grounding over speech utterances and affordances, that correspond to the action possibilities offered by an environment.

<p>To incorporate new technical advances into military domain and make those processes more <i>efficient</i> in accuracy, time and cost, a new concept of Network Centric Warfare has been introduced in the US military forces. In Sweden a similar concept has been studied under the name Network Based Defence (NBD). Here we present one of the methodologies, called tactical plan recognition that is aimed to support NBD in future.</p><p>Advances in sensor technology and modelling produce large sets of data for decision makers. To achieve <i>decision superiority</i>, decision makers have to act agile with proper, adequate and relevant information (data aggregates) available. Information fusion is a process aimed to support decision makers’ situation awareness. This involves a process of combining data and information from disparate sources with <i>prior</i> information or knowledge to obtain an improved state estimate about an agent or phenomena. <i>Plan recognition</i> is the term given to the process of inferring an agent’s intentions from a set of actions and is intended to support decision making.</p><p>The aim of this work has been to introduce a methodology where prior (empirical) knowledge (e.g. behaviour, environment and organization) is represented and combined with sensor data to recognize plans/behaviours of an agent or group of agents. We call this methodology <i>multi-agent plan recognition</i>. It includes knowledge representation as well as imprecise and statistical inference issues.</p><p>Successful plan recognition in large scale systems is heavily dependent on the data that is supplied. Therefore we introduce a <i>bridge</i> between the plan recognition and sensor management where results of our plan recognition are reused to the control of, give <i>focus of attention</i> to, the sensors that are supposed to acquire most important/<i>relevant</i> information.</p><p>Here we combine different theoretical methods (Bayesian Networks, Unified Modeling Language and Plan Recognition) and apply them for tactical military situations for ground forces. The results achieved from several proof-ofconcept models show that it is possible to model and recognize behaviour of tank units.</p>

abstract: Humans and robots need to work together as a team to accomplish certain shared goals due to the limitations of current robot capabilities. Human assistance is required to accomplish the tasks as human capabilities are often better suited for certain tasks and they complement robot capabilities in many situations. Given the necessity of human-robot teams, it has been long assumed that for the robotic agent to be an effective team member, it must be equipped with automated planning technologies that helps in achieving the goals that have been delegated to it by their human teammates as well as in deducing its own goal to proactively support its human counterpart by inferring their goals. However there has not been any systematic evaluation on the accuracy of this claim. In my thesis, I perform human factors analysis on effectiveness of such automated planning technologies for remote human-robot teaming. In the first part of my study, I perform an investigation on effectiveness of automated planning in remote human-robot teaming scenarios. In the second part of my study, I perform an investigation on effectiveness of a proactive robot assistant in remote human-robot teaming scenarios. Both investigations are conducted in a simulated urban search and rescue (USAR) scenario where the human-robot teams are deployed during early phases of an emergency response to explore all areas of the disaster scene. I evaluate through both the studies, how effective is automated planning technology in helping the human-robot teams move closer to human-human teams. I utilize both objective measures (like accuracy and time spent on primary and secondary tasks, Robot Attention Demand, etc.) and a set of subjective Likert-scale questions (on situation awareness, immediacy etc.) to investigate the trade-offs between different types of remote human-robot teams. The results from both the studies seem to suggest that intelligent robots with automated planning capability and proactive support ability is welcomed in general. / Dissertation/Thesis / Masters Thesis Computer Science 2015

To incorporate new technical advances into military domain and make those processes more efficient in accuracy, time and cost, a new concept of Network Centric Warfare has been introduced in the US military forces. In Sweden a similar concept has been studied under the name Network Based Defence (NBD). Here we present one of the methodologies, called tactical plan recognition that is aimed to support NBD in future. Advances in sensor technology and modelling produce large sets of data for decision makers. To achieve decision superiority, decision makers have to act agile with proper, adequate and relevant information (data aggregates) available. Information fusion is a process aimed to support decision makers’ situation awareness. This involves a process of combining data and information from disparate sources with prior information or knowledge to obtain an improved state estimate about an agent or phenomena. Plan recognition is the term given to the process of inferring an agent’s intentions from a set of actions and is intended to support decision making. The aim of this work has been to introduce a methodology where prior (empirical) knowledge (e.g. behaviour, environment and organization) is represented and combined with sensor data to recognize plans/behaviours of an agent or group of agents. We call this methodology multi-agent plan recognition. It includes knowledge representation as well as imprecise and statistical inference issues. Successful plan recognition in large scale systems is heavily dependent on the data that is supplied. Therefore we introduce a bridge between the plan recognition and sensor management where results of our plan recognition are reused to the control of, give focus of attention to, the sensors that are supposed to acquire most important/relevant information. Here we combine different theoretical methods (Bayesian Networks, Unified Modeling Language and Plan Recognition) and apply them for tactical military situations for ground forces. The results achieved from several proof-ofconcept models show that it is possible to model and recognize behaviour of tank units. / QC 20101222

An area of study that has received much attention over the past few decades is simulations involving threat assessment in military scenarios. Recently, much research has emerged concerning the recognition of troop movements and formations in non-combat simulations. Additionally, there have been efforts towards the detection and assessment of various types of malicious intentions. One such work by Akridge addressed the issue of Strategic Intention Recognition, but fell short in the detection of tactics that it could not detect without somehow manipulating the environment. Therefore, the aim of this thesis is to address the problem of recognizing an opponent's intent in a strategic environment where the system can think ahead in time to see the agent's plan. To approach the problem, a structured form of knowledge called Template-Based Interpretation is borrowed from the work of others and enhanced to reason in a temporally dynamic simulation.

Several real world tasks involve data that is uncertain and relational in nature. Traditional approaches like first-order logic and probabilistic models either deal with structured data or uncertainty, but not both. To address these limitations, statistical relational learning (SRL), a new area in machine learning integrating both first-order logic and probabilistic graphical models, has emerged in the recent past. The advantage of SRL models is that they can handle both uncertainty and structured/relational data. As a result, they are widely used in domains like social network analysis, biological data analysis, and natural language processing. Bayesian Logic Programs (BLPs), which integrate both first-order logic and Bayesian net- works are a powerful SRL formalism developed in the recent past. In this dissertation, we develop approaches using BLPs to solve two real world tasks – plan recognition and machine reading. Plan recognition is the task of predicting an agent’s top-level plans based on its observed actions. It is an abductive reasoning task that involves inferring cause from effect. In the first part of the dissertation, we develop an approach to abductive plan recognition using BLPs. Since BLPs employ logical deduction to construct the networks, they cannot be used effectively for abductive plan recognition as is. Therefore, we extend BLPs to use logical abduction to construct Bayesian networks and call the resulting model Bayesian Abductive Logic Programs (BALPs). In the second part of the dissertation, we apply BLPs to the task of machine reading, which involves automatic extraction of knowledge from natural language text. Most information extraction (IE) systems identify facts that are explicitly stated in text. However, much of the information conveyed in text must be inferred from what is explicitly stated since easily inferable facts are rarely mentioned. Human readers naturally use common sense knowledge and “read between the lines” to infer such implicit information from the explicitly stated facts. Since IE systems do not have access to common sense knowledge, they cannot perform deeper reasoning to infer implicitly stated facts. Here, we first develop an approach using BLPs to infer implicitly stated facts from natural language text. It involves learning uncertain common sense knowledge in the form of probabilistic first-order rules by mining a large corpus of automatically extracted facts using an existing rule learner. These rules are then used to derive additional facts from extracted information using BLP inference. We then develop an online rule learner that handles the concise, incomplete nature of natural-language text and learns first-order rules from noisy IE extractions. Finally, we develop a novel approach to calculate the weights of the rules using a curated lexical ontology like WordNet. Both tasks described above involve inference and learning from partially observed or incomplete data. In plan recognition, the underlying cause or the top-level plan that resulted in the observed actions is not known or observed. Further, only a subset of the executed actions can be observed by the plan recognition system resulting in partially observed data. Similarly, in machine reading, since some information is implicitly stated, they are rarely observed in the data. In this dissertation, we demonstrate the efficacy of BLPs for inference and learning from incomplete data. Experimental comparison on various benchmark data sets on both tasks demonstrate the superior performance of BLPs over state-of-the-art methods. / text

[pt] Nesta tese propomos a aplicação de um paradigma de planejamento baseado em uma disciplina de modelagem conceitual para as tarefas de Mineração de Processos. Postulamos que a abordagem apresentada habilita as tarefas de descoberta de processos, checagem de conformidade e melhoria de modelos em domínios educacionais, que tem características de processos não-estruturados – dependências entre tarefas, múltiplas dependências, eventos concorrentes, atividades que falham, atividades repetidas, traços parciais e estruturas de nocaute. Relacionamos os conceitos em ambas as áreas de pesquisa e demonstramos a abordagem aplicada a um exemplo em um domínio acadêmico, implementando os algoritmos como parte de uma Biblioteca de Planos Típicos para Mineração de Processos que constrói sobre a extensa literatura prévia. / [en] In this thesis we propose the application of an automated planning paradigm based on a conceptual modeling discipline for the Process Mining tasks. We posit that the presented approach enables the process discovery, conformance checking and model enhancement tasks for educational domains, comprising characteristics of unstructured processes – with intertask dependencies, multiple dependencies, concurrent events, failing activities, repeated activities, partial traces and knock-out structures. We relate the concepts in both areas of research, and demonstrate the approach applied to an academic domain example, implementing the algorithms as part of a Library for Typical Plans for Process Mining that leverages the extensive prior art in the literature.

A Inteligência Artificial Distribuída traz uma série de novas perspectivas para a computação quando considera sistemas heterogêneos, adaptativos, evolutivos, continuamente em funcionamento e abertos. Estes sistemas, chamados de sociedades, apresentam tais características por permitirem que seus componentes, chamados de agentes, migrem entre sociedades, isto é, agentes podem sair e entrar em sociedades. Sociedades abertas permitem a migração dos agentes e coloca dois tipos de problemas para o agente que está migrando: problemas de linguagem e interação, que concernem ao uso de expressões usadas e à maneira como as interações são organizadas na nova sociedade; e, problemas de conhecimento e atuação, que se referem à como um agente irá se comportar a fim de realizar justamente aquilo que a sociedade espera dele. Este trabalho se atem aos problemas de conhecimento e atuação. Para que os agentes da sociedade possam cooperar e coordenar suas ações, é necessário que tenham conhecimento das capacidades, habilidades, desejos e planos dos outros agentes. Grande parte do conhecimento a respeito dos outros pode ser extraído dos papéis que estes podem assumir na sociedade. Assim sendo, o problema colocado para este trabalho é como os agentes da sociedade que receberam o agente imigrante e o próprio agente imigrante conhecerão/aprenderão os papéis uns dos outros. São desenvolvidos três mecanismos de identificação de papéis, bem como a comparação entre eles e sua adequação a tipos de migração. Os três mecanismos são os seguintes: i) Identificação de papéis por protocolo de apresentação: é proposta uma linguagem de descrição de protocolos (LDP) e uma especificação de protocolo de apresentação nesta LDP. Os agentes que utilizam este mecanismo conseguem se identificar com rapidez, porém necessitam conhecer várias informações “locais” da sociedade, o que pode ser muito restritivo para um agente migrante. ii) Identificação de papéis por observação e classificação: esta solução procura classificar o agente observado em um papel de um conjunto prédefinido de papéis. Neste conjunto, os papéis são descritos por meio de processos de interação (PI). Para isto, desenvolveu-se a noção de PI. Para isto, desenvolveu-se a noção de PI. Foram desenvolvidas duas formas de proceder a classificação: construir uma especificação do agente a partir da observação das suas ações e verificar se esta pertence ao conjunto pré-definido de papéis; e, verificar se o comportamento do agente confere com as execuções possíveis para algum dos papéis pré-definidos. Este mecanismo é mais adequado para sociedades abertas e tem boa precisão no resultado apresentado, porém, a identificação do papel de um agente pode ser demorada. iii) Identificação de papéis por reconhecimento de intenções em planos: este mecanismo baseia-se na existência de uma relação entre intenções e papéis. A partir das ações observadas para o agente, procura-se saber qual seu plano, sua intenção e, conseqüentemente, seu papel. Para isto foi implementado um procedimento de indução de planos. Este mecanismo também é adequado para sociedades abertas, no entanto, a identificação, embora satisfatória, nem sempre é completamente correta. Estes três mecanismos foram testados em simulações numa implementação da sociedade Produtor - Consumidor, onde puderam ser comprovadas as características de cada um. / Distributed Artificial Intelligence brings a number of new perspectives to Computing Science when heterogeneous, adaptative and evolutive systems, those under functioning and open, are taken into consideration. These systems, named societies, present these characteristics because they allow their components, named agents, to migrate within societies, that is, agents are allowed to enter and to leave societies. Agents' migration brings two kinds of problems to the migrating agent: language and interaction problems both related to the use of used expressions and to the way the interactions are organized in the new society; and, knowledge and performance problems refering to the way the agent will behave in order to accomplish exactly what society expects him to do. This work is limited to knowledge and performance problems. In order to cooperate and coordinate their actions, the society's agents need to learn about the capabilities, abilities wishes and plans of other agents. A great part of knowledge of others can be extracted from the social roles these agents can play. Therefore, the problem posed in this work is how social agents who has received an immigrating agent and the immigrating agent himself will know and learn one another's roles. Three role identification mechanisms, and the comparison between them and their adaptation to migration types as well are developed. The three mechanisms are the following: i) Role Identification by means of presentation protocol:a language of protocol description (LPD) and a specification of presentation protocol in this LPD are proposed. The agents who use this mechanism can rapidly identify each other, however they need know a number of 'local' social information, which can be very restrictive to the migrating agent. ii) Role identification by means of observation and classification: this solution tries to classify the observed agent as a role out of set of definite roles. In this set, the roles are described by means of interactional processes (IP). Therefore, the notion of IP was developed. Two ways to proceed the role classification were developed: to build the agent's especification departing from the observation of their actions and to check whether this especification belongs to a set of pre-defined roles; and to check whether the agent's behavior fits the possible executions to some predefined roles. This mechanism is more adequate to open societies and has good precision in the result presented, but, the agent's role identification can last longer. iii) Role Identification by means of intention and plans recognition: this mechanism is based on the existence of a relationship between intentions and roles. By departing from the agent's observed actions, his plan, intention, consequently, his role is recognized. Therefore an induced plan procedure was implemented. This mechanism is also adequate to open societies, however, the identification, though satisfactory, is not always totally correct. These three mechanisms were tested in simulated situations in a kind of Producer- Consumer Society implementation in which each one's characteristics could be verified.