Skip to main content
main-content

Über dieses Buch

This volume highlights new trends and challenges in research on agents and the new digital and knowledge economy, and includes 23 papers classified into the following categories: business process management, agent-based modeling and simulation, and anthropic-oriented computing. All papers were originally presented at the 11th International KES Conference on Agents and Multi-Agent Systems – Technologies and Applications (KES-AMSTA 2017) held June 21–23, 2017 in Vilamoura, Algarve, Portugal.

Today’s economy is driven by technologies and knowledge. Digital technologies can free, shift and multiply choices, and often intrude on the territory of other industries by providing new ways of conducting business operations and creating value for customers and companies. The topics covered in this volume include software agents, multi-agent systems, agent modeling, mobile and cloud computing, big data analysis, business intelligence, artificial intelligence, social systems, computer embedded systems and nature inspired manufacturing, etc., all of which contribute to the modern Digital Economy.

The results presented here will be of theoretical and practical value to researchers and industrial practitioners working in the fields of artificial intelligence, collective computational intelligence, innovative business models, the new digital and knowledge economy and, in particular, agent and multi-agent systems, technologies, tools and applications.

Inhaltsverzeichnis

Frontmatter

Agent and Multi-Agent Systems

Frontmatter

Personalized HealthCare and Agent Technologies

Remarkable gains in life expectancy and declines in fertility have led current society to an ageing global population. Different stakeholders, researcher communities and policy makers invest serious efforts to develop intelligent and smart environments that have to support as much as possible independent living of old population. As necessary prerequisite for these efforts rapid and fascinating development in ICT offers wide range of new technologies including wearable, 3D sensors and smart environments. These new technologies provide rich complex data from living environment and give the opportunity to learn and analyze them in order to discover the patient’s preferences, traits, and states. Further research efforts are oriented to personalized healthcare and development of sophisticated e-coaching facilities to obtain proper recommendations and advices to patients in order to increase their wellbeing.
Among different artificial intelligence methods and techniques agent technologies significantly influence and support different medical domains. The use of agents and multi agent systems in healthcare has also opened the ways to find out new applications like personalized and socialized healthcare platforms and systems with tailored recommendation capabilities. In this paper opportunities and challenges that agent technologies offer in personalized healthcare are discussed and presented.
Mirjana Ivanović, Srđan Ninković

Multiagent Environments for Dynamic Transportation Applications

Dynamic transportation applications have long been a domain of choice for the multiagent paradigm. Indeed, the presence of distributed entities, the highly dynamic character of these applications and the often presence of human actors in the system makes it very suitable for a multiagent design. This paper advocates for the primary consideration of multiagent environment design when dealing with such dynamic transportation applications. Transportation applications can greatly benefit from the use of the multiagent environment since most of them consider a dynamic geographical positioning of the system components. Indeed, the simultaneous consideration of the time and space dimensions makes the environment, which is shared and accessed by all the agents of the system, a candidate of choice to capture the dynamics of the application. The environment design can be envisioned at several levels of the system construction. It can be used as a medium for interaction between distributed entities. It could be used as a coordination entity of the system components. It can finally be designed as a mental model for the agents that they can use in their reasoning. We illustrate the possible uses of the environment with two transportation applications dealing with traveler information.
Mahdi Zargayouna

Microservices as Agents in IoT Systems

Developing robust monolith systems has achieved its limitations, since the implementation of changes in today’s large, complex, and fast evolving systems would be too slow and inefficient. As a response to these problems, microservice architecture emerged, and quickly became a widely used solution. Such modular architecture is appropriate for distributed environment of Internet of Things (IoT) solutions. In this paper we present a solution for service management on Machine-to-Machine (M2M) devices within IoT system by using collaborative microservices. Collaboration of distributed modules highly reminds of multi-agent systems where autonomous agents also cooperate to provide services to the end-user. Because of these similarities we consider microservices as modern agents that could improve systems in distributed environments, such as IoT.
Petar Krivic, Pavle Skocir, Mario Kusek, Gordan Jezic

Enhancing Tactical Information Assessment Using an Agent-Based Cognitive Architecture

The actualisation of an information-rich, expansive and complex battlespace have rendered current Tactical Information Assessment (TIA) within tactical systems as non-effective. Existing TIA techniques and algorithms perform limited cognitive processing because they lack fundamental characteristics of Situation Management (SM). In consequence, a dramatic reduction of situation awareness, information and decision superiority of the operators and the overall tactical system has emerged. Considerable attention in applying computational cognitive architectures and processing and SM within TIA has been gaining momentum.
This paper discusses the Cognitive Architecture for Tactical Information Assessment (CATIA), a proposed Multi-Agent System (MAS)-based cognitive architecture. CATIA employs cognitive architecture design principles and the Belief, Desire and Intention (BDI) framework to facilitate recognition and reasoning to deliberate in tactical situations and events. CATIA will be implemented within the Future Integrated Mission System (FIMS) to illustrate how superior TIA methodologies can dramatically improve information assessment, situation awareness and information and decision superiority within tactical systems.
Angela Consoli

Security and Trust on Mobile Agent Platforms: A Survey

Mobile agent technologies are known for their capacity to develop and construct distributed, heterogeneous and interoperable systems. Despite the presence of several platforms for the development of mobile agent applications, security issues act as a main deterrent against such trends. Based on this, we conducted a comparative study of the most promising and existing mobile agent platforms, showing the diverse security features employed to address various threats. We investigate also the trust models used by the platforms. The established study focuses on the contributions to verify security criteria by the used security mechanisms in every studied platform. This study is important not only to allow practitioners pick the most suitable platforms to meet their security requirements, but also to allow researchers address the voids by ameliorating the concluded limits and proposing possible improvements of new versions of these platforms.
Donies Samet, Farah Barika Ktata, Khaled Ghedira

A Self-adaptive System for Improving Autonomy and Public Spaces Accessibility for Elderly

Nowadays, there is an increasing need to provide a safe and independent living for cognitively deficient population. Notably, we have to improve seniors’ autonomy and their public spaces accessibility. Giving these observations, the aim of this paper is to provide a personalized adaptive assisting system for elderly. More precisely, this paper presents the specification and implementation of a self-organizing multi-agent system able to abstract the different distributed components involved in user’s environment. This system is able to detect different possible situations that a user could face in his daily outdoors activities and propose accordingly appropriate actions. This system not only learns user’s habits from its perceptions but also improves its recommendations thanks to feedbacks provided by stakeholders (family, doctors …) following a reinforcement learning reasoning. Finally, we present our system evaluation specially its learning capabilities through different scenarios that have been generated automatically.
Sameh Triki, Chihab Hanachi

Meaning Negotiation with Defeasible Logic

Agents negotiate the meaning of terms in numerous real-life situations. When they behave so, they can be used as the basis for providing an emulation paradigm for software agents, habilitating therefore socio-technical systems to perform meaning negotiation. In this paper we focus upon two methods for meaning negotiation in defeasible logic and provide room for an analysis of how the proposed approaches perform the aforementioned process. Finally we also provide a computational analysis of the process automation problem.
Matteo Cristani, Antonino Rotolo

Artificial Intelligence Techniques for the Puerto Rico Strategy Game

It was always a challenging task to create artificial opponents for strategy video games. It is usually quite easy to discover and exploit their weaknesses because their tactics usually do not adapt to changing conditions and to human opponent tactics. In this paper two artificial intelligence techniques for well known Puerto Rico strategy game are proposed. One of them does not rely on any precoded tactics, but tries to dynamically learn and adapt to the changing game environment. Both techniques were compared on the basis of results of games played against each other and also against human expert players.
Rafał Dreżewski, Maciej Klęczar

Simple Bounded MTLK Model Checking for Timed Interpreted Systems

We present a new translation of Metric Temporal Logic with knowledge operators (MTLK) to the Linear Temporal Logic with knowledge operators and with a new set of the atomic propositions (\({\mathrm{{\mathrm{{LTL}}}}_\mathrm{q}\mathrm {K}}\)). We investigate a SAT-based bounded model checking (BMC) method for MTLK. The semantics of MTLK is defined over timed interpreted systems (TIS). We show how to implement the bounded model checking technique for \({\mathrm{{\mathrm{{LTL}}}}_\mathrm{q}\mathrm {K}}\) and timed interpreted systems, and as a case study, we apply the technique in the analysis of the Timed Generic Pipeline Paradigm modelled by TIS. We also present the differences between the old translation of MTLK and the new one. The theoretical description is supported by the experimental results that demonstrate the efficiency of the method.
Agnieszka M. Zbrzezny, Andrzej Zbrzezny

An Algorithm for Allocating Structured Tasks in Multi-Robot Scenarios

Task allocation is an important aspect in dealing with coordination problems. However, there are challenges in developing appropriate strategies for multi-robot teams in such a way that robots perform their operations efficiently. Real-world scenarios usually require the use of heterogeneous robots and execution of tasks with different structures and constraints. In this paper we propose a dynamic, decentralised task allocation mechanism considering different types of tasks for heterogeneous robot teams playing different roles and carrying out tasks according to their own capabilities. We have run several simulations in order to evaluate the proposed mechanism. The results indicate that the proposed mechanism scales well and provides near-optimal allocations.
Tulio L. Basegio, Rafael H. Bordini

SAT-Versus SMT-Based BMC for TWIS and the Existential Fragment of WCTL with Knowledge

In this paper, we present the SAT-based bounded model checking method for Timed Weighted Interpreted Systems and for Weighted Existential Computation Tree Logic with epistemic operators. SAT-based bounded model checking consists in translating the existential model checking problem for a modal logic and for a model to the boolean satisfiability problem. We provide an implementation based on Cryptominisat and YicesSAT SAT-solvers and we present a comparison of the SAT-based BMC method and SMT-based BMC methods on common instances that can be scaled up to for performance evaluation.
Agnieszka M. Zbrzezny

Agent-Based Modeling and Simulation

Frontmatter

Communication and Autonomous Control of Multi-UAV System in Disaster Response Tasks

After disasters occurrence, advanced technologies always play significant contributions to various disaster response tasks. For instance, utilization of Unmanned Aerial Vehicle (UAV) for aerial imagery data helps to get a real-time monitoring system of the stricken areas. Recently, UAV is becoming a ubiquitous system and valuable technology in many civil applications, motivating researchers to endeavor to develop UAVs systems. Moreover, Internet of Things (IoT) has been employed to several study cases of disaster response. Using IoT to communicate, manage and control multiple UAVs after disaster occurrence is a real practical approach. However, this kind of integration wasn’t clearly presented especially in the technical sides like communication and autonomous control. Consequently, this paper aims to reveal the scientific scenes of communications and controls between multiple UAVs and Ground Control Stations (GCSs). Due to an unexpected failure within a single UAV system, this study presents a multi-UAV system for scanning and tracking missions. Authors turned UAV to an IoT device by using embedded LTE dongle on the UAV control board. In the beginning of the paper, authors raised the issues of existing systems. Then, presented the design and process of the proposed system. Lastly, they demonstrated some results through experiments.
Maher Aljehani, Masahiro Inoue

Decision Function Implementation in MAREA Simulations Influencing Financial Balance of Small-Sized Enterprise

The aim of this paper is to present the use of a decision function in the implementation of a multi-agent simulation model of a small-sized enterprise dealing with trading. The subject of the presented research are simulation experiments in MAREA software framework, which was designed to simulate trading behaviour of a trading company. Firstly, we present a multi-agent system and a mathematical description of a decision function, which is used to establish the price of traded goods. Secondly, we present MAREA software framework and lastly we discuss the simulation results of company dealing with retailing of fluorescence colours. The results obtained show that simulation experiments in MAREA could be used to support the decision-making process of a management of trading companies in the scope of predicting key performance indicators and changes of parameters and their impact on the company’s financial balance.
Roman Šperka, Dominik Musil

Application of I-Fuzzy Approach to Prediction of Blockability Values in Real-World Data

The main aim of this article is to discuss the setting and the construction of blockability values for decision-making agents, whose actions are modeled by transferable utility cooperative game, when possible coalitions of agents are vague – in this case expressed as I-fuzzy coalitions — and test the appropriateness of this approach on real data – in this case the data from voting in the Chamber of Deputies of the Czech Parliament during three subsequent Parliamentary periods. Results of calculations show improvement in values in ex-post I-fuzzy blockability values, as well as in tested predicted I-fuzzy blockability values based on results from preceding periods.
Elena Mielcová

Generalized Dynamic Model of Rating Alternatives by Agents with Interactions

Business process simulation models usually incorporate several essential components that reflect customer behavior for modeling system inputs and outputs and ranking and/or rating given alternatives. In this paper we deal with a general dynamic system of rating a number of alternatives given by pairwise comparison matrices on an alo-group. This system is based on a parametrized agent-based simulation with interactions among agents which is able to replicate various types of processes, e.g. financial market evaluation, evaluation of products’ demand and supply, evaluation of political parties in general elections, evaluation of universities etc. A simple simulation experiment is presented and discussed.
Radomír Perzina, Jaroslav Ramík

The Soft Tissue Implementation with Triangulated Mesh for Virtual Surgery System

In this paper, we describe specific issues arising during implementation of the virtual surgery simulator. Virtual surgery simulator is a software that provides realistic surgery experience using virtual reality technologies. We discuss the necessary requirements that a virtual surgery simulator shall meet, and suggest the possible solutions for its implementation, such as triangulated mesh for realistic rendering of soft tissue and haptic feedback. Among the supported operations with our solution for soft tissue there are cutting and stitching. Fluid dynamics are also mentioned.
Ruslan Akhmetsharipov, Murad Khafizov, Alexey Lushnikov, Shamil Zigantdinov

Anthropic-Oriented Computing

Frontmatter

“Thinking-Understanding” Approach in Spiking Reasoning System

In this position paper we propose the approach to use “Thinking-Understanding” architecture for the management of the real-time operated robotic system. Based on the “Robot dream” architecture, the robotic system digital input is been translated in form of “pseudo-spikes” and provided to a simulated spiking neural network, then elaborated and fed back to a robotic system as updated behavioural strategy rules. We present the reasoning rule-based system for intelligent spike processing translating spikes into software actions or hardware signals is thus specified. The reasoning is based on pattern matching mechanisms that activates critics that in their turn activates other critics or ways to think inherited from the work of Marvin Minsky “The emotion machine” [7].
Alexander Toschev, Max Talanov, Vitaliy Kurnosov

Pseudorehearsal in Value Function Approximation

Catastrophic forgetting is of special importance in reinforcement learning, as the data distribution is generally non-stationary over time. We study and compare several pseudorehearsal approaches for Q-learning with function approximation in a pole balancing task. We have found that pseudorehearsal seems to assist learning even in such very simple problems, given proper initialization of the rehearsal parameters.
Vladimir Marochko, Leonard Johard, Manuel Mazzara

Finding Correlations Between Driver Stress and Traffic Accidents: An Experimental Study

As the number of people getting injured or killed on the roads is constantly growing, it is crucial to identify and prevent potential factors causing traffic accidents. This paper focuses on one of such factors – namely, the drivers’ stress, which is known to be one of the main causes of traffic accidents, and timely detection of such situations becomes an important challenge. The paper aims to find a potential correlation between the driver stress when riding through a specific urban location and the recorded history of traffic accidents in that specific location. If proven, such a correlation can help to prevent traffic accidents and re-design urban spaces in a safer manner. To achieve this goal, the paper combines cross-disciplinary techniques from Computer Science and Physiology to measure drivers’ stress levels using physiological sensors during city rides, and match these experimental results against a map of previously recorded traffic accidents. As a result, the conducted study indicates that the correlation indeed exists, and measuring drivers’ stress levels using physiological sensors is a promising approach to minimise the amount of traffic accidents.
Margarita Pavlovskaya, Ruslan Gaisin, Rustem Dautov

Towards Robot Fall Detection and Management for Russian Humanoid AR-601

While interacting in a human environment, a fall is the main threat to safety and successful operation of humanoid robots, and thus it is critical to explore ways to detect and manage an unavoidable fall of humanoid robots. Even assuming perfect bipedal walking strategies and algorithms, there exist several unexpected factors, which can threaten existing balance of a humanoid robot. These include such issues as power failure, robot component failures, communication disruptions and failures, sudden forces applied to the robot externally as well as internally generated exceed torques etc. As progress in a humanoid robotics continues, robots attain more autonomy and enter realistic human environments, they will inevitably encounter such factors more frequently. Undesirable fall might cause serious physical damage to a human user, to a robot and to surrounding environment. In this paper, we present a brief review of strategies that include algorithms for fall prediction, avoidance, and damage control of small-size and human-size humanoids, which will be further implemented for Russian humanoid robot AR-601.
Evgeni Magid, Artur Sagitov

Business Process Management

Frontmatter

Modelling of the Logistic Supplier-Consumer Behavior

The paper highlights the problems of mathematical modelling in the delivery system. The system describes the suppliers who offer different types of products as well as the consumers who order different products. Products are ordered at stochastic times, however, manufacturers offer predictable demand. The problem becomes more complex when the number of orders grows. The structure of the system is shown, equations of state are introduced and control algorithms as well as criteria are proposed. Orders change their state which leads to modifying it at every decision stage. The same concerns the actual output of manufacturers which also has to be modified. Therefore, the problem consists of the design of such a delivery pattern which can minimise losses of the discussed company. The goal of the paper is to present the mathematical model of the logistic system taking into account the consumer-supplier relations. The model forms the basis for the subsequent information support tool.
Petr Suchánek, Robert Bucki

Conversion of Real Data from Production Process of Automotive Company for Process Mining Analysis

The aim of this paper is to convert the real data from the raw format from different information systems (log files) to the format, which is suitable for process mining analysis of a production process in a large automotive company. The conversion process will start with the import from several relational databases. The motivation is to use the DISCO tool for importing real pre-processed data and to conduct process mining analysis of a production process. DISCO generates process models from imported data in a comprehensive graphical form and provides different statistical features to analyse the process. This makes it possible to examine the production process in detail, identify bottlenecks, and streamline the process. The paper firstly presents a brief introduction of a manufacturing process in a company. Secondly, it provides a description of a conversion and pre-processing of chosen real data structures for the DISCO import. Then, it briefly describes the DISCO tool and proper format of pre-processed log file, which serves as desired input data. This data will be the main source for all consecutive operations in generated process map. Finally, it provides a sample analysis description with emphasis on one production process (process map and few statistics). To conclude, the results obtained show high demands on pre-processing of real data for suitable import format into DISCO tool and vital possibilities of process mining methods to optimize a production process in an automotive company.
Miroslav Dišek, Roman Šperka, Jan Kolesár

Multi-Agent BPMN Decision Footprint

Towards Decision Collaboration Along Distributed BI Process
Nowadays, we are confronted with increasingly complex information systems. Modelling these kinds of systems will only be controlled through appropriate tools, techniques and models. Work of the Open Management Group (OMG) in this area have resulted in the development of Business Process Model and Notation (BPMN) and Decision Model and Notation (DMN). Currently, these two standards are a pillar of various business architecture Frameworks to support Business-IT alignment and minimize the gap between the managers’ expectations and delivered technical solutions. Several research focus on the extension of these models especially BPMNDF which aims to harmonize decision-making throughout a single business process. The current challenge is to extend the BPMNDF in order to cover business process in a distributed and cooperative environment. In this paper, we propose the Multi-Agent BPMN Decision Footprint (MABPMNDF) which is a novel model based on both BPMNDF and MAS to support decision-making in distributed business process.
Riadh Ghlala, Zahra Kodia Aouina, Lamjed Ben Said

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise