Development Of Autonomous Agents Research Articles

Higher Education in the Age of Artificial Intelligence

The breakneck expansion of artificial intelligence (AI) technologies in recent years has attracted the attention of higher education researchers. However, their curiosity often comes down to using specific AI tools, such as text and image generators, translators, and personal assistants in the educational process. This paper considers a broader question: what fundamental problems and challenges does the penetration of AI technologies into human lives originate for higher education? The authors offer a working definition of what it means to give/receive higher education in the age of AI. This definition identifies five critical characteristics of higher education: 1) the professor and the student continue to be in a subject-subject relationship and learn from each other; 2) higher education should prepare for life in conditions of “human-machine” interdependence; 3) these conditions imply choice in situations of uncertainty; 4) the spread of AI technologies brings enormous opportunities and 5) almost unpredictable dangers, risks and threats to humans. The authors consider general principles and specific problems associated with entering AI tools into society and higher education. They discuss the inevitability of the evolvement of “machine-machine interdependence” related to the development of autonomous agents. In conclusion, several theses and counter-theses summarize the article’s argumentation.

From Pluralistic Normative Principles to Autonomous-Agent Rules

With recent advancements in systems engineering and artificial intelligence, autonomous agents are increasingly being called upon to execute tasks that have normative relevance. These are tasks that directly—and potentially adversely—affect human well-being and demand of the agent a degree of normative-sensitivity and -compliance. Such norms and normative principles are typically of a social, legal, ethical, empathetic, or cultural (‘SLEEC’) nature. Whereas norms of this type are often framed in the abstract, or as high-level principles, addressing normative concerns in concrete applications of autonomous agents requires the refinement of normative principles into explicitly formulated practical rules. This paper develops a process for deriving specification rules from a set of high-level norms, thereby bridging the gap between normative principles and operational practice. This enables autonomous agents to select and execute the most normatively favourable action in the intended context premised on a range of underlying relevant normative principles. In the translation and reduction of normative principles to SLEEC rules, we present an iterative process that uncovers normative principles, addresses SLEEC concerns, identifies and resolves SLEEC conflicts, and generates both preliminary and complex normatively-relevant rules, thereby guiding the development of autonomous agents and better positioning them as normatively SLEEC-sensitive or SLEEC-compliant.

Deep reinforcement learning for multi-agent interaction

The development of autonomous agents which can interact with other agents to accomplish a given task is a core area of research in artificial intelligence and machine learning. Towards this goal, the Autonomous Agents Research Group develops novel machine learning algorithms for autonomous systems control, with a specific focus on deep reinforcement learning and multi-agent reinforcement learning. Research problems include scalable learning of coordinated agent policies and inter-agent communication; reasoning about the behaviours, goals, and composition of other agents from limited observations; and sample-efficient learning based on intrinsic motivation, curriculum learning, causal inference, and representation learning. This article provides a broad overview of the ongoing research portfolio of the group and discusses open problems for future directions.

A Design Framework of Exploration, Segmentation, Navigation, and Instruction (ESNI) for the Lifecycle of Intelligent Mobile Agents as a Method for Mapping an Unknown Built Environment.

Recent work on intelligent agents is a popular topic among the artificial intelligence community and robotic system design. The complexity of designing a framework as a guide for intelligent agents in an unknown built environment suggests a pressing need for the development of autonomous agents. However, most of the existing intelligent mobile agent design focus on the achievement of agent’s specific practicality and ignore the systematic integration. Furthermore, there are only few studies focus on how the agent can utilize the information collected in unknown build environment to produce a learning pipeline for fundamental task prototype. The hierarchical framework is a combination of different individual modules that support a type of functionality by applying algorithms and each module is sequentially connected as a prerequisite for the next module. The proposed framework proved the effectiveness of ESNI system integration in the experiment section by evaluating the results in the testing environment. By a series of comparative simulations, the agent can quickly build the knowledge representation of the unknown environment, plan the actions accordingly, and perform some basic tasks sequentially. In addition, we discussed some common failures and limitations of the proposed framework.

Qualitative case-based reasoning and learning

The development of autonomous agents that perform tasks with the same dexterity as performed by humans is one of the challenges of artificial intelligence and robotics. This motivates the research on intelligent agents, since the agent must choose the best action in a dynamic environment in order to maximise the final score. In this context, the present paper introduces a novel algorithm for Qualitative Case-Based Reasoning and Learning (QCBRL), which is a case-based reasoning system that uses qualitative spatial representations to retrieve and reuse cases by means of relations between objects in the environment. Combined with reinforcement learning, QCBRL allows the agent to learn new qualitative cases at runtime, without assuming a pre-processing step. In order to avoid cases that do not lead to the maximum performance, QCBRL executes case-base maintenance, excluding these cases and obtaining new (more suitable) ones. Experimental evaluation of QCBRL was conducted in a simulated robot-soccer environment, in a real humanoid-robot environment and on simple tasks in two distinct gridworld domains. Results show that QCBRL outperforms traditional RL methods. As a result of running QCBRL in autonomous soccer matches, the robots performed a higher average number of goals than those obtained when using pure numerical models. In the gridworlds considered, the agent was able to learn optimal and safety policies.

Survey on Computer Vision for UAVs: Current Developments and Trends

During last decade the scientific research on Unmanned Aerial Vehicless (UAVs) increased spectacularly and led to the design of multiple types of aerial platforms. The major challenge today is the development of autonomously operating aerial agents capable of completing missions independently of human interaction. To this extent, visual sensing techniques have been integrated in the control pipeline of the UAVs in order to enhance their navigation and guidance skills. The aim of this article is to present a comprehensive literature review on vision based applications for UAVs focusing mainly on current developments and trends. These applications are sorted in different categories according to the research topics among various research groups. More specifically vision based position-attitude control, pose estimation and mapping, obstacle detection as well as target tracking are the identified components towards autonomous agents. Aerial platforms could reach greater level of autonomy by integrating all these technologies onboard. Additionally, throughout this article the concept of fusion multiple sensors is highlighted, while an overview on the challenges addressed and future trends in autonomous agent development will be also provided.

Modeling Workplace Evacuation Behaviors Using Intelligent Agents

A major challenge in artificial intelligence has been the development of autonomous agents (AAs) capable of displaying believable behaviours. To achieve such objective, the underlying architectures of these intelligent systems have been designed to incorporate Learning Classifier System that provides an adaptation naturally with the environment. It is expected that through the interaction of this type of components, AAs can implement more intelligent and believable behaviour. Although the literature reports several computational models of behaviors, attention, and emotions developed to be included in cognitive agent architectures, these have been implemented as separated processes, disregarding essential interactions between these behaviors whose modeling and computational implementation may increase the believability of actions developed by AAs. In this paper, we propose an evolutive computational model. This model is designed to provide AAs with adequate mechanisms to attend and react to conditions and changes in the environment.

Design patterns for multi-agent programming

Various agent-based programming languages and frameworks have been proposed to support the development of autonomous agents and multi-agent systems. They have provided a valuable contribution to the identification and operationalisation of agent concepts and abstractions by proposing specific programming constructs. Unfortunately, these contributions have not yet been widely adopted by industry. In this paper, we follow the argument that multi-agent programming technology can find its way to industry by providing a methodology that guides the development of autonomous agents and multi-agent systems in standard programming technology. The proposed methodology explains how some characteristic concepts and abstractions related to autonomous agents and multi-agent systems can be implemented in object-oriented technology. This is done by initiating a Java library of object-oriented design patterns for some characteristic but established programming constructs that have been developed in some agent-based programming languages.

Design patterns for multi-agent programming

Various agent-based programming languages and frameworks have been proposed to support the development of autonomous agents and multi-agent systems. They have provided a valuable contribution to th...

Development of Computational Models of Emotions for Autonomous Agents: A Review

It has been recognized that human behavior is an observable consequence of the interactions between cognitive and affective functions. This perception has motivated the study of human emotions in disciplines such as psychology and neuroscience and led to the formulation of a number of theories and models that attempt to explain the mechanisms underlying this human process. In the field of artificial intelligence, these theoretical findings have posed a series of challenges in the development of autonomous agents (AAs) capable of exhibiting very believable and human-like behaviors. One of these challenges is the design and implementation of computational models of emotions (CMEs), which are software systems designed to provide AAs with proper mechanisms for the processing of emotional information, elicitation of synthetic emotions, and generation of emotional behaviors. In this paper, we review this type of computational model from the perspective of their development. Particularly, we investigate five design aspects that influence their development process: theoretical foundations, operating cycle, interaction between cognition and emotion, architectural design, and role in cognitive agent architectures. We provide discussions about key issues and challenges in the development of CMEs and suggest research that may lead to more robust and flexible designs for this type of computational model.

Robo-CAMAL: A BDI Motivational Robot

Abstract Motivation is a central concept in the development of autonomous agents and robots. This paper describes an architecture that uses a psychological BDI model of reasoning, combined with a distributed multi-level model of motivation. The robot controlling architecture makes use of a generic set of deliberative components plus an environment task-centred set of reactive components that reflect the architecture's embodiment. The architecture has been used in a number of simulated environments and here is used to control a mobile robot. A theoretical framework for motivation and affect is given, and related to the nature of autonomy and embodiment. A BDI model, based on a psychological model of reasoning in a 5 year old child, is described in terms of the nature of motivation and affect within the architecture. Finally, criteria for judging the nature of an agent's motivation are introduced, and used to validate the motivational constructs implemented within the architecture. Experimental results lead to a comparative discussion.

Task-technology fit and user acceptance of online auction

Word Wide Web intelligent agent technology has provided researchers and practitioners, such as those involved in information technology, innovation, knowledge management, and technical collaboration with the ability to examine the design principles and performance characteristics of the various approaches to intelligent agent technology, and to increase the cross fertilization of ideas on the development of autonomous agents and multi-agent systems among different domains. This study investigates the employment of intelligent agents in a web-based auction process, with particular reference to the appropriateness of the agent software for the online auction task, consumers’ value perception of the agent, the effect of this consumer perception on their intention to use the tool, and a measure of consumer acceptance. In the initial case study, both consumers and web operators thought the use of software agents enhanced online auction efficiency and timeliness. The second phase of the investigation established that consumer familiarity with the agent functionality was positively associated with seven dimensions: online auction site's task, agent's technology, task-technology fit, perceived ease of use, perceived usefulness, perceived playfulness, intention to use tool, and negatively associated with perceived risk. Intelligent agents have the potential to release skilled operator time for the use of value-adding tasks in the planning and expansion of online auctions.

A formal mathematical framework for modeling probabilistic hybrid systems

The development of autonomous agents, such as mobile robots and software agents, has generated considerable research in recent years. Robotic systems, which are usually built from a mixture of continuous (analog) and discrete (digital) components, are often referred to as hybrid dynamical systems. Traditional approaches to real-time hybrid systems usually define behaviors purely in terms of determinism or sometimes non-determinism. However, this is insufficient as real-time dynamical systems very often exhibit uncertain behavior. To address this issue, we develop a semantic model, Probabilistic Constraint Nets (PCN), for probabilistic hybrid systems. PCN captures the most general structure of dynamic systems, allowing systems with discrete and continuous time/variables, synchronous as well as asynchronous event structures and uncertain dynamics to be modeled in a unitary framework. Based on a formal mathematical paradigm exploiting abstract algebra, topology and measure theory, PCN provides a rigorous formal programming semantics for the design of hybrid real-time embedded systems exhibiting uncertainty.

An autonomous agent development environment for engineering applications

Agent technology is playing an increasingly important role in developing distributed, intelligent, and collaborative engineering applications. The innate difficulties of interoperation between heterogeneous agent communities and rapid development of multi-agent systems have motivated the emergence of FIPA specifications and the proliferation of multi-agent system platforms or toolkits that implement FIPA specifications. This paper presents a FIPA compliant multi-agent framework called Autonomous Agent Development Environment (AADE). Originating from the engineering fields, AADE can facilitate the rapid development of collaborative engineering applications (especially in engineering design and manufacturing fields) through the provision of reusable packages of agent level components and programming tools. An agent oriented engineering project on the development of a collaborative product design environment is implemented based on the facilities provided by AADE.

Robot shaping: developing autonomous agents through learning

Learning plays a vital role in the development of autonomous agents. In this paper, we explore the use of reinforcement learning to “shape” a robot to perform a predefined target behavior. We connect both simulated and real robots to Alecsys, a parallel implementation of a learning classifier system with an extended genetic algorithm. After classifying different kinds of Animat-like behaviors, we explore the effects on learning of different types of agent's architecture and training strategies. We show that the best results are achieved when both the agent's architecture and the training strategy match the structure of the behavior pattern to be learned. We report the results of a number of experiments carried out both in simulated and in real environments, and show that the results of simulations carry smoothly to physical robots. While most of our experiments deal with simple reactive behavior, in one of them we demonstrate the use of a simple and general memory mechanism. As a whole, our experimental activity demonstrates that classifier systems with genetic algorithms can be practically employed to develop autonomous agents.

The emergence of high level structure in classifier systems - a proposal

While Artificial Intelligence approaches to the representation of world knowledge has taken an unashamedly structured approach, Artificial Life workers have shunned such forms (Maes, 1992). Recent problems in the production of symbolic reasoning within classifier systems (Forrest, 1991), the debate between the ‘flat’ and ‘hierarchical’ schools of thought within autonomous agent development (Tyrell, 1992a,b). and developments in coevolving classifier systems (Dorigo & Schnepf, 1992) have fuelled new interest in higher level structures linking multiple classifier systems. It is shown that further work is required in order to understand what form such structures might take. and proposals are presented for the development of these structures