A Subsumption Architecture to Develop Dynamic Cognitive Network-Based Models With Autonomous Navigation Application
- PDF / 1,587,940 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 32 Downloads / 160 Views
A Subsumption Architecture to Develop Dynamic Cognitive Network-Based Models With Autonomous Navigation Application Márcio Mendonça · Bruno Augusto Angélico · Lúcia Valéria Ramos de Arruda · Flávio Neves Jr
Received: 23 April 2012 / Revised: 20 August 2012 / Accepted: 22 September 2012 / Published online: 23 February 2013 © Sociedade Brasileira de Automatica–SBA 2013
Abstract This work proposes an original approach based on subsumption architecture to build dynamic cognitive networks (DCN). Dynamic cognitive network is a soft computing technique similar to fuzzy cognitive maps (FCM) which are able to easily model cause–effect behaviors. FCMs have been applied in several areas of knowledge; however, they present some restrictions for modeling dynamic systems, specifically temporal dependencies among events. Due to these restrictions, alternative approaches based on FCM and also fuzzy networks have appeared in the literature. Dynamic cognitive networks are one of these techniques. Hence, this study presents a new type of DCN which incorporates different types of concepts and causal relations able to circumvent the main drawbacks of FCM modeling. The new approach is based on the subsumption architecture, which allows to represent, model, and implement several behaviors of a dynamic systems through a composition of hierarchical DCNs. An application of the new DCN technique in autonomous navigation is also developed in order to validate the approach. Keywords Dynamic cognitive network · Fuzzy cognitive map · Subsumption architecture · Autonomous navigation
M. Mendonça · B. A. Angélico Universidade Tecnológica Federal do Paraná, Cornélio Procópio, Paraná, Brasil e-mail: [email protected] B. A. Angélico e-mail: [email protected] L. V. R. de Arruda (B)· F. Neves Jr Universidade Tecnológica Federal do Paraná, Curitiba, Paraná, Brasil e-mail: [email protected] F. Neves Jr e-mail: [email protected]
1 Introduction Cognitive maps (CMs) and fuzzy cognitive maps (FCMs) are some of the methods used for knowledge and inference modeling in intelligent systems. Cognitive maps (CMs) were originally proposed by (Axelrod 1976) as a formal model to the belief structure of a person or group. Axelrod (1976) also developed a mathematical treatment, by means of matrix operations and based on graph theory, for his cognitive maps. In the cognitive maps, the beliefs or statements about a limited area of knowledge are expressed through words or linguistic expressions with simple relationships of cause and effect (cause/non-cause). This mathematical model has been adapted by Kosko (1986) to consider uncertainties through fuzzy variables, generating the fuzzy cognitive maps (FCM). Like the original cognitive maps, the FCMs are signed and directed graphs (digraphs) in which the involved variables are fuzzy numbers. The graph’s nodes are linguistic concepts represented by fuzzy sets and each node is associated with others through connections. A fuzzy weight is assigned to each of these connections (graph’s edge) in order to represen
Data Loading...
