人工情感及其应用

自然系统中,情感影响人的行为.目前,人工情感在拟人agent的研究得到了越来越多的重视.为此,本文从工程角度综述了人工情感研究的基本问题,简要介绍了几类情感模型,并着重讨论了基于人工情感的agent体系结构的典型应用.人工情感的研究不仅体现在情感辨识、情感表达等人机交互方面,同时情感的作用机理也影响到智能agent的控制体系结构和算法设计.基于人工情感的体系结构具有混合分层的特点,强调情感和其他过程的联系以增强agent在与动态环境交互中的自适应能力.在这个结构中人工情感的核心作用主要体现在两个方面:首先,人工情感是联系agent的内部状态和外部环境的基础,情感状态影响到agent信息处理的整个过程,包括紧急情况下的快速响应和规划任务时的复杂推理.其次,人工情感作为内部驱动机制对学习过程十分重要,人工情感作为强化刺激促使agent创建更复杂的行为功能.在设计中引入人工情感,对agent实现智能化和个性化至关重要.     

基于免疫网络的策略选择模型研究

研究免疫网络在一类路径选择和规划问题（羊群问题）中的应用。利用免疫网络动力学模型实现羊群问题中主动主体和被动主体之间相互作用,抗体和抗原分别对应选择的策略和动态变化的环境,建立基于免疫网络的主动主体行动策略选择模型。仿真结果表明,该方法中主动主体通过与被动主体之间相互作用,可以比传统方法自发形成概率的、较好的主动主体的行动策略,达到使羊归圈的目的。     

Architectural Methodology Based on Intentional Configuration of Behaviors

Intelligence has been an object of study for a long time. Different architectures try to capture and reproduce these aspects into artificial systems (or agents), but there is still no agreement on how to integrate them into a general framework. With this objective in mind, we propose an architectural methodology based on the idea of intentional configuration of behaviors. Behavior‐producing modules are used as basic control components that are selected and modified dynamically according to the intentions of the agent. These intentions are influenced by the situation perceived, knowledge about the world, and internal variables that monitor the state of the agent. The architectural methodology preserves the emergence of functionality associated with the behavior‐based paradigm in the more abstract levels involved in configuring the behaviors. Validation of this architecture is done using a simulated world for mobile robots, in which the agent must deal with various goals such as managing its energy and its well‐being, finding targets, and acquiring knowledge about its environment. Fuzzy logic, a topologic map learning algorithm, and activation variables with a propagation mechanism are used to implement the architecture for this agent.     

Aborting,suspending, and resuming goals and plans in BDI agents

Intelligent agents designed to work in complex, dynamic environments such as e-commerce must respond robustly and flexibly to environmental and circumstantial changes, including the actions of other agents. An agent must have the capability to deliberate about appropriate courses of action, which may include reprioritising tasks—whether goals or associated plans—aborting or suspending tasks, or scheduling tasks in a particular order. In this article we study mechanisms to enable principled suspend, resuming, and aborting of goals and plans within a Belief-Desire-Intention (BDI) agent architecture. We give a formal and combined operational semantics for these actions in an abstract agent language (CAN), thus providing a general mechanism that can be incorporated into several BDI-based agent platforms. The abilities enabled by our semantics provides an agent designer greater flexibility to direct agent operation, offering a generic means to manage the status of goals. We demonstrate the reasoning abilities enabled on a document workflow scenario.     

基于联邦结构Agent的柔性装配系统的研究

根据功能结构将柔性装配系统分成几个大的组成部分，每个部分用一个联邦结构agent进行建模，联邦结构agent内部由一个调度agent和若干个普通agent组成。这种结构具有功能结构层次化、任务分解执行以及智能化和鲁棒性等优点。对系统中的agent进行了构造，描述了不同agent在系统中的角色、任务和内容。系统中的各agent之间的分工合作是以调度agent为中心，通过“投标/仲裁”和“指派/执行”两种机制来配合实现的。构建了实验平台进行实验验证。     

Linking perception and action through motivation and affect

Research into cognitive architectures is described within a framework spanning major issues in artificial intelligence and cognitive science. Earlier work on motivation is extended with a cognitive model of reasoning which, together with an affective mechanism, enables consistent decision-making across a variety of cognitive and reactive processes. Cognition involves the control of behaviour within both external and internal environments. The control of behaviour is vital to an autonomous system as it acts to further its goals. Except in the most spartan of environments, the potential available information and associated combinatorics in a perception, cognition, and action sequence can tax even the most powerful agents. The affect magnitude concept solves some problems with BDI models, and allows for adaptive decision-making over a number of tasks in different domains. The cognitive and affective components are brought together using motivational constructs. The generic cognitive model can adapt to different environments and tasks as it makes use of motivational models to direct reactive and situated processes.     

Architectural mechanisms for dynamic changes of behavior selection strategies in behavior-based systems

Behavior selection is typically a "built-in" feature of behavior-based architectures and hence, not amenable to change. There are, however, circumstances where changing behavior selection strategies is useful and can lead to better performance. In this paper, we demonstrate that such dynamic changes of behavior selection mechanisms are beneficial in several circumstances. We first categorize existing behavior selection mechanisms along three dimensions and then discuss seven possible circumstances where dynamically switching among them can be beneficial. Using the agent architecture framework activation, priority, observer, and component (APOC), we show how instances of all (nonempty) categories can be captured and how additional architectural mechanisms can be added to allow for dynamic switching among them. In particular, we propose a generic architecture for dynamic behavior selection, which can integrate existing behavior selection mechanisms in a unified way. Based on this generic architecture, we then verify that dynamic behavior selection is beneficial in the seven cases by defining architectures for simulated and robotic agents and performing experiments with them. The quantitative and qualitative analyzes of the results obtained from extensive simulation studies and experimental runs with robots verify the utility of the proposed mechanisms.     

A multi-agent reinforcement learning approach to robot soccer

In this paper, a multi-agent reinforcement learning method based on action prediction of other agent is proposed. In a multi-agent system, action selection of the learning agent is unavoidably impacted by other agents’ actions. Therefore, joint-state and joint-action are involved in the multi-agent reinforcement learning system. A novel agent action prediction method based on the probabilistic neural network (PNN) is proposed. PNN is used to predict the actions of other agents. Furthermore, the sharing policy mechanism is used to exchange the learning policy of multiple agents, the aim of which is to speed up the learning. Finally, the application of presented method to robot soccer is studied. Through learning, robot players can master the mapping policy from the state information to the action space. Moreover, multiple robots coordination and cooperation are well realized.     

Can a rational agent afford to be affectless? a formal approach

In this article, we expose some of the issues raised by the critics of the neoclassical approach to rational agent modeling and we propose a formal approach for the design of artificial rational agents that includes some of the functions of emotions found in the human system. We suggest that emotions and rationality are closely linked in the human mind (and in the body, for that matter) and, therefore, need to be included in architectures for designing rational artificial agents, whether these agents are to interact with humans, to model humans' behaviors and actions, or both. We describe an Affective Knowledge Representation (AKR) scheme to represent emotion schemata, which we developed to guide the design of a variety of socially intelligent artificial agents. Our approach focuses on the notion of "social expertise" of socially intelligent agents in terms of their external behavior and internal motivational goal-based abilities. AKR, which uses probabilistic frames, is derived from combining multiple emotion theories into a hierarchical model of affective phenomena useful for artificial agent design. AKR includes a taxonomy of affect, mood, emotion, and personality, and a framework for emotional state dynamics using probabilistic Markov Models.     

Multi‐agent collaboration based on enhanced cognitive awareness: an architecture for agents' profiling on the semantic web

This paper applies cognitive models, inspired by cognitive science, with the aim to propose architectural and knowledge‐based requirements to structure ontological models for the cognitive profiling of agents. The proposed architecture aims to address the lack of flexibility that most agent‐based collaborations are affected by. The resulting agents, equipped with advanced cognitive profiling, have an increased cognitive awareness of themselves and are more capable of interacting with other agents in a multi‐agents based environment. In this research, cognitive awareness identifies the ability of the web agents to diagnose their processing limitations and to establish interactions with the external environment. The outcome is the enhanced flexibility, reusability and predictability of the agent behaviour; thus contributing towards minimizing human cognitive demands. The concept of cognitive profiling presented in this paper considers the semantic web as an action mediating space, where ontological models provide affordances for improving cognitive awareness through shared knowledge‐base. The conceptual model for the cognitive profile architecture is developed with Protégé Ontology editor to generate OWL Ontology and evaluated through a proof of concept. The results show that agents equipped with cognitive awareness can undertake complex tasks more dynamically.     

Improving Performance of Heterogeneous Agents

Agents provide services not only to humans users but also to agents in one or more multiagent systems. When agents are confronted with multiple tasks to perform (or requests to satisfy), the agent can reduce load on itself by attempting to take advantage of commonalities between the tasks that need to be performed. In this paper, we develop a logical theory by which such “heavily loaded” agents can merge commonalities amongst such tasks. In our framework, agents can be built on top of legacy codebases. We propose a logical formalism called invariants using which agent developers may specify known commonalities between tasks – after this, we propose a sound and complete mechanism to derive all possible derived commonalities. An obvious A ^*-based algorithm may be used to merge a set of tasks in a way that minimised expected execution cost. Unfortunately the execution time of this algorithm is prohibitive, even when only 10 tasks need to be merged, thus making it unusable in practice. We develop heuristic algorithms for this problem that take much less time to execute and produce almost as good ways of merging tasks.     

An extended JADE-S based framework for developing secure Multi-Agent Systems

Agent communities are self-organized virtual spaces consisting of a large number of agents and their dynamic environments. Within a community, agents group together offering special e-services for effective, reliable, and mutual benefits. Usually, an agent community is composed of specialized agents performing one or more tasks in a single domain/sub-domain, or in highly intersecting domains. However, secure Multi-Agent Systems require severe mechanisms in order to prevent malicious attacks. Several limits affect exiting secure agents platform, such as the lack of a strong authentication system, the lack of a flexible distributed mechanism for access control and the lack of a system for storing past behaviors of agent/user. Biometric owner agents authentication, agent/users policies to regulate agent's behavior and actions, and agent/users reputation level to select trusted agents can be used to overcome the above limits and enhance the level of security for these applications. In this paper an extended JADE-S based framework for developing secure Multi-Agent Systems is proposed. The framework functionalities are extended by self-contained FPGA biometric sensors providing secure and fast user authentication service. Each agent owner, by means of biometric authentication, acquires his/her own X.509v3 digital certificate. Policy files and a flexible, fast distributed Access Control Mechanism can regulate behavior and actions of any users/agent inside the platform. In addition, a mechanism based on the agent reputation is used: reputation is an attribute associated to each owner and/or agent on the basis of its past behavior and integrity. In order to prove the feasibility of the proposed framework, we have developed a multi-agent e-Banking system. System goal deals with e-Banking services such as bank account statements, account transactions and so on. In the paper, the experimental features of the biometric self-contained sensors are also outlined.     

Dialogue games that agents play within a society

Human societies have long used the capability of argumentation and dialogue to overcome and resolve conflicts that may arise within their communities. Today, there is an increasing level of interest in the application of such dialogue games within artificial agent societies. In particular, within the field of multi-agent systems, this theory of argumentation and dialogue games has become instrumental in designing rich interaction protocols and in providing agents with a means to manage and resolve conflicts. However, to date, much of the existing literature focuses on formulating theoretically sound and complete models for multi-agent systems. Nonetheless, in so doing, it has tended to overlook the computational implications of applying such models in agent societies, especially ones with complex social structures. Furthermore, the systemic impact of using argumentation in multi-agent societies and its interplay with other forms of social influences (such as those that emanate from the roles and relationships of a society) within such contexts has also received comparatively little attention. To this end, this paper presents a significant step towards bridging these gaps for one of the most important dialogue game types; namely argumentation-based negotiation (ABN). The contributions are three fold. First, we present a both theoretically grounded and computationally tractable ABN framework that allows agents to argue, negotiate, and resolve conflicts relating to their social influences within a multi-agent society. In particular, the model encapsulates four fundamental elements: (i) a scheme that captures the stereotypical pattern of reasoning about rights and obligations in an agent society, (ii) a mechanism to use this scheme to systematically identify social arguments to use in such contexts, (iii) a language and a protocol to govern the agent interactions, and (iv) a set of decision functions to enable agents to participate in such dialogues. Second, we use this framework to devise a series of concrete algorithms that give agents a set of ABN strategies to argue and resolve conflicts in a multi-agent task allocation scenario. In so doing, we exemplify the versatility of our framework and its ability to facilitate complex argumentation dialogues within artificial agent societies. Finally, we carry out a series of experiments to identify how and when argumentation can be useful for agent societies. In particular, our results show: a clear inverse correlation between the benefit of arguing and the resources available within the context; that when agents operate with imperfect knowledge, an arguing approach allows them to perform more effectively than a non-arguing one; that arguing earlier in an ABN interaction presents a more efficient method than arguing later in the interaction; and that allowing agents to negotiate their social influences presents both an effective and an efficient method that enhances their performance within a society.     

Behavior monitoring under uncertainty using Bayesian surprise and optimal action selection

The increasing trend towards delegating tasks to autonomous artificial agents in safety–critical socio-technical systems makes monitoring an action selection policy of paramount importance. Agent behavior monitoring may profit from a stochastic specification of an optimal policy under uncertainty. A probabilistic monitoring approach is proposed to assess if an agent behavior (or policy) respects its specification. The desired policy is modeled by a prior distribution for state transitions in an optimally-controlled stochastic process. Bayesian surprise is defined as the Kullback–Leibler divergence between the state transition distribution for the observed behavior and the distribution for optimal action selection. To provide a sensitive on-line estimation of Bayesian surprise with small samples twin Gaussian processes are used. Timely detection of a deviant behavior or anomaly in an artificial pancreas highlights the sensitivity of Bayesian surprise to a meaningful discrepancy regarding the stochastic optimal policy when there exist excessive glycemic variability, sensor errors, controller ill-tuning and infusion pump malfunctioning. To reject outliers and leave out redundant information, on-line sparsification of data streams is proposed.     

Toward a comparative theory of agents

The purpose of this paper is to address some of the questions on the notion of agent and agency in relation to property and personhood. I argue that following the Kantian criticism of Aristotelian metaphysics, contemporary biotechnology and information and communication technologies bring about a new challenge—this time, with regard to the Kantian moral subject understood in the subject’s unique metaphysical qualities of dignity and autonomy. The concept of human dignity underlies the foundation of many democratic systems, particularly in Europe as well as of international treaties, including the Universal Declaration of Human Rights. Digital agents, artificial organisms as well as new capabilities of the human agents related to their embeddedness in digital and biotechnological environments bring about an important transformation of the human self-appraisal. A critical comparative reflection of this transformation is important because of its ethical implications. I deal first with the concept of agent within the framework of Aristotelian philosophy, which is the basis for further theories in accordance with and/or in opposition to it, particularly since modernity. In the second part of this paper, I deal with the concept of personhood in Kantian philosophy, which supersedes the Aristotelian metaphysics of substance and builds the basis of a metaphysics of the moral human subject. In the third part, I discuss the question of artificial agents arising from modern biology and ICT. Blurring the difference between the human and the natural and/or artificial opens a “new space” for philosophical reflection as well as for debate in law and practical policy.     

AN AGENT-BASED SYSTEM FOR PRODUCTS PENETRATION STRATEGY SELECTION

This paper describes an agent-based system implementing an original consumer-based methodology for product penetration strategy selection in real-world situations. Agents are simultaneously considered according to two different levels: a functional and a structural level. In the functional level, we have three types of agents: task agents, information agents, and interface agents assuming task fulfillment through cooperation, information gathering tasks, and mediation between users and artificial agents, respectively. In the structural level, we have elementary agents based on a generic reusable architecture and complex agents considered as an agent organization created dynamically in an hierarchical way.     

Emotion in reinforcement learning agents and robots: a survey

This article provides the first survey of computational models of emotion in reinforcement learning (RL) agents. The survey focuses on agent/robot emotions, and mostly ignores human user emotions. Emotions are recognized as functional in decision-making by influencing motivation and action selection. Therefore, computational emotion models are usually grounded in the agent’s decision making architecture, of which RL is an important subclass. Studying emotions in RL-based agents is useful for three research fields. For machine learning (ML) researchers, emotion models may improve learning efficiency. For the interactive ML and human–robot interaction community, emotions can communicate state and enhance user investment. Lastly, it allows affective modelling researchers to investigate their emotion theories in a successful AI agent class. This survey provides background on emotion theory and RL. It systematically addresses (1) from what underlying dimensions (e.g. homeostasis, appraisal) emotions can be derived and how these can be modelled in RL-agents, (2) what types of emotions have been derived from these dimensions, and (3) how these emotions may either influence the learning efficiency of the agent or be useful as social signals. We also systematically compare evaluation criteria, and draw connections to important RL sub-domains like (intrinsic) motivation and model-based RL. In short, this survey provides both a practical overview for engineers wanting to implement emotions in their RL agents, and identifies challenges and directions for future emotion-RL research.     

战斗行为情绪机制

应用情绪行为选择机制,以Dom World机制为基础,建立基于情绪的竞争行为机制,揭示前期竞争结果以情绪方式影响后续行为。借用Dom World实验设计方法和数据,运用人工生命Swarm软件平台进行仿真实验。仿真结果以人工社会典型特征－自组织（涌现）过程的实现为核心,证明了所建立机制的正确性。应用ADI、Netto方法评估分析两种机制的仿真结果,证明情绪机制优于Dom World机制。进一步,将反馈和无反馈情绪机制的仿真结果进行比较分析,揭示反馈机制增加了战斗强度,使得达尔文关于情绪机制加速自然选择过程的论断得到证明。     

Reasoning about emotional agents

In this article we discuss the role of emotions in artificial agent design, and the use of logic in reasoning about the emotional or affective states an agent can reside in. We do so by extending the KARO framework for reasoning about rational agents appropriately. In particular, we formalize in this framework how emotions are related to the action monitoring capabilities of an agent. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 601–619, 2006.     

A framework for facilitating cooperation in multi-agent systems

This work introduces a multi-agent framework that facilitates cooperation in multi-agent robotic systems. It uses a layered approach based on Coloured Petri Nets for modelling complex, concurrent conversations among agents. In this approach each agent employs a Coloured Petri Net model that allows agents to follow a plan specifying their interactions. It also allows programmers to plan for the concurrent feature of the conversation and make sure that all possible states of the problem space are considered. The framework assists the agents to identify and adapt different strategies for teammates and task selection dynamically. The agents can change their strategies in the course of dynamic environments to improve their performance. We have examined the performance of the agents in this framework by developing some task selection and teammate selection strategies for agents in a disaster scenario.