Krzysztof Cetnarowicz

A formal framework for multi-agent systems analysis and design

Abstract Agent based computing is generally intended for modeling and implementation of distributed complex problems. Despite the existence of many applications, the problem of rational engineering of multi-agent systems remains complex and difficult. The purpose of this paper can be summarized within two claims. First, we aim at providing an approach that gives some guidelines for specifying and designing multi-agent systems. Secondly, we focus on the formalisms as a language for describing the models produced in each development process phases. These seem to be straightforward, while the development of multi-agent systems is still done, in most cases, without using methods and formal modeling such as those generally used in object oriented software. We illustrate this approach by specifying an example based upon a specific agent architecture.

Models of social groups in blogosphere based on information about comment addressees and sentiments

This work concerns the analysis of number, sizes and other characteristics of groups identified in the blogosphere using a set of models identifying social relations. These models differ regarding identification of social relations, influenced by methods of classifying the addressee of the comments (they are either the post author or the author of a comment on which this comment is directly addressing) and by a sentiment calculated for comments considering the statistics of words present and connotation. The state of a selected blog portal was analyzed in sequential, partly overlapping time intervals. Groups in each interval were identified using a version of the CPM algorithm, on the basis of them, stable groups, existing for at least a minimal assumed duration of time, were identified.

Algorithm of behavior evaluation in multi-agent system

Behavior based detection of unfavorable activities in multi-agent systems (presented in [3,4,5]) is an approach to the problem of detection of intruders. This approach refers to evaluation of behavior of every agent which exists in a multi-agent system. Process of behavior evaluation is distributed – every agent makes autonomous behavior evaluation of other agents. This means that an agent is evaluated separately by all agents in the environment of the secured system. That separate results of behavior evaluations have to be collected and an algorithm should be used in order to elect the worst agents which should be eliminated. Collecting and data processing of results of distributed behavior evaluations is the main topic of this article.

Software Agent Systems for Improving Performance of Multi-Robot Groups

Agent paradigm is becoming more and more popular in design of complex multi-robot systems. Similarities between features attributed to robots and agents lead to defining an agent as a single robot control program or even calling a robot an “embodied agents”. This approach can cause serious issues concerning scalability, extensibility and performance of large-scale multi-robot systems. In this paper a different approach to design and implementation of software systems managing mobile robots is presented. It is based on the logical separation of a hardware robot and a software agent. The paper focuses on improvements in system performance achieved by using the approach. Three exemplary applications are described and experimental results are provided.

Behavior Based Detection of Unfavorable Resources

This article considers a problem of security in a computer systems in context of some mechanisms which act in societies and which mechanisms are seen as being useful to assure security in a computer system. A new approach to the security problem is discussed, which refers to identifying computer resource (e.g. agent, program) on the basis of this resource behaviour (actions which this resource undertakes). Mechanisms are presented, which may enable to recognize and dismiss resources undesirable or harmful in the computer system on the basis of behavior observation. Proposed mechanisms were tested in simulations of the computer system under DoS attack. The results of tests are presented and discussed.

Robot Task Allocation Using Signal Propagation Model

Abstract The problem of Multi-Robot Task Allocation has received significant attention over last years. The solutions with decentralized decision making have proven better durability than those using centralized planning. In this paper a method which does not use any explicit communication is presented. It is especially suitable for very high dynamics of tasks priorities, when other methods must often perform time-consuming replanning. The method uses a simple model of signal emission and propagation in the environment. Simulation experiments are provided to demonstrate usability and interesting features of the method.

On the concept of agent in multi-robot environment

Although, the software agent paradigm has been widely accepted, there is still a problem with situating (implementing) concrete agents in real environments. Agent is an abstract notion so that there is no straightforward mapping of real entities like robots, spacecrafts or any other autonomous systems to agents. An interesting discussion on this subject was done within the framework of LOGOS and ACT Agent Architecture in the context of ground and space systems, see [10,11]. In the paper we discuss the problem of situating agents in open systems consisting of cognitive heterogeneous robots that are supposed to perform jointly complex tasks. The starting point of the discussion is the archtecture of the M-agent.

Agent Oriented Technology of Decentralized Systems Based on the M-Agent Architecture

Abstract An idea of autonomous agents that arises as an extension of the object and process concepts may be applied to distributed and decentralized systems development. In the paper the authors have undertaken an attempt to describe formally the architecture of multi-agent systems (called M-agent architecture) that may be considered as a starting point to develop a decentralized multiagent, system technology such as Agent Oriented Analysis, Design and Programming.

Towards Credible Driver Behavior Modeling

In order to predict the drivers behaviour, to adapt different mechanisms of the car or to tune up the traffic controls, one has to create a reliable drivers model. Such approaches have already been made, based on the assumption, that certain similarity must be observed for one driver doing the same maneuver (e.g. braking) or even in certain situations for different drivers. The research presented in this paper shows clearly that such assumptions are wrong in general, based on a number of real-life experiments using a real car (equipped with necessary sensors) in real traffic conditions. The experiments show that the same driver performs his/hers maneuvers differently, even though they could have been assumed to be the same or at least very similar. Thus necessity of further looking for reliable models of drivers behaviour is proven.

Extensible Platform for Studying the Behavior of Drivers in Urban Traffic

Research on traffic management and automation focuses on macro-scale problems involving large number of vehicles. It seems that the influence of individual decisions of particular drivers has not received sufficient attention over last decades. Understanding the characteristics of particular drivers is crucial for traffic optimization and for decisions making process performed by automated vehicles. In this paper we would like to present a prototype platform for studying characteristics of individual human drivers. The platform collects data during driving in urban environment and performs analysis and classification of observed behaviors. It is extensible in terms of measured parameters and classification methods. We also present results of first experiment using the created platform - classification of drivers during the left turn maneuver.