iConcept Press Logo
Email Password Remember me
Advances in Robotics: Modeling, Control and Applications
Advances in Robotics: Modeling, Control and Applications
Calin Ciufudean and Lino García

Chapter 14

Advances in Robotics: Modeling, Control and Applications

Autonomous Multi-Criteria Decision-Making in Ambient Intelligence

by Petr Tucnik

Viewed: 1650


The multi-criteria approach is typically used for complex (managerial) problems both for decision support or decision-making purposes. It is most useful when applied to difficult problems where many aspects are to be taken into consideration at the same time. But there is also another field of application of this approach – common everyday situations, requiring careful consideration of possible solutions and outcomes of our actions. Multi-criteria approach is, in fact, quite suitable model for this type of human decision-making processes and can be applied to a wide range of problems. Our effort is aimed to adaptation of this method for autonomous use in agents. Agent, which can be accepted as a universal implementation platform for many forms of artificial intelligence, can use multi-criteria technique as their control mechanism. To achieve rational behavior in such agent, many changes are to be made in standard multi-criteria mathematical conception. Control mechanism has to be able to function continuously, adapt to changes, work with input information in many formats, and act fast (etc.). It will be shown that modified multi-criteria approach can meet these requirements. Best results are achieved when it is used in higher-level reasoning when forming strategy and planning actions. Our formal model is intended to be used in domain of ambient intelligence; i.e. in agents distributed in environment helping user to achieve his/her goals. On higher-level of agent´s decision-making, discrete state representation will be used to represent its behavior. Proposed control system is a modular, scalable solution which can be easily modified or adapted to fit user´s needs.

Author Details

Petr Tucnik
University of Hradec Kralove, Czech Republic


Petr Tucnik. Autonomous Multi-Criteria Decision-Making in Ambient Intelligence. In Advances in Robotics: Modeling, Control and Applications. ISBN:978-1-461108-443. iConcept Press. 0000.