Márcio Mendonça

Autonomous navigation system using Event Driven-Fuzzy Cognitive Maps

This study developed an autonomous navigation system using Fuzzy Cognitive Maps (FCM). Fuzzy Cognitive Map is a tool that can model qualitative knowledge in a structured way through concepts and causal relationships. Its mathematical representation is based on graph theory. A new variant of FCM, named Event Driven-Fuzzy Cognitive Maps (ED-FCM), is proposed to model decision tasks and/or make inferences in autonomous navigation. The FCM’s arcs are updated from the occurrence of special events as dynamic obstacle detection. As a result, the developed model is able to represent the robot’s dynamic behavior in presence of environment changes. This model skill is achieved by adapting the FCM relationships among concepts. A reinforcement learning algorithm is also used to finely adjust the robot behavior. Some simulation results are discussed highlighting the ability of the autonomous robot to navigate among obstacles (navigation at unknown environment). A fuzzy based navigation system is used as a reference to evaluate the proposed autonomous navigation system performance.

A dynamic fuzzy cognitive map applied to chemical process supervision

This work develops an intelligent tool based on fuzzy cognitive maps to supervisory process control. Fuzzy cognitive maps are a neuro-fuzzy methodology that can accurate model complexly system using a causal-effect fuzzy reasoning. In the proposed approach, new types of concept and relation, not restricted to cause-effect ones, are added to the model resulting in a dynamic fuzzy cognitive map (D-FCM). In this sense, a supervisory system is developed in order to control a fermentation process. This process has a non-linear behavior and presents several problems, such as non-minimum phase and large accommodation time. The supervisor goal is to operate the process in normal and critical conditions. The expert knowledge about the process behavior in both conditions is used to build the D-FCM supervisor. Simulation results are presented in order to validate the proposed intelligent supervisor.

A Combined FCM-GA Approach to Supervise Industrial Process

Abstract This paper presents a supervisory control strategy based on fuzzy cognitive map (FCM) and genetic algorithm (GA). Fuzzy cognitive maps are a neuro-fuzzy methodology that can model complexly system accurate. In the proposed methodology, the expert knowledge about the process behavior is used to build an initial FCM. This FCM is extended and refined to incorporate control strategies by means of a GA which runs with simulated process data. The resulting FCM is used to generate set points for the regulatory loops in the plant lower level. The developed supervisory control methodology is applied to an alcoholic fermentation process from chemical industry. Comparison of performance is made with another intelligent approach (Fuzzy-PD), and also with a predictive approach based on DMC (Dynamic Matrix Control).

Hybrid Dynamic Fuzzy Cognitive Maps and Hierarchical Fuzzy logic controllers for Autonomous Mobile Navigation

This work is an evolution of a previous work in which we used a knowledge-based system to autonomous navigation area. Two navigation systems are developed: one uses Fuzzy Cognitive Maps (FCM) and the other is based on a Fuzzy Logic Controller (FLC). For the first system, a variant of the classical FCM, named Hybrid-Dynamic Fuzzy Cognitive Maps (HDFCM), is used to model decision tasks and/or to make inference into a mobile navigation context. Fuzzy Cognitive Maps are a tool that model qualitative structured knowledge through concepts and causal relationships. The proposed model allows representing the dynamic behavior of a mobile robot in presence of changes in the environment. A Hierarchical Weighted Fuzzy Logic Controller (HW-FLC) composes the second navigation system. Simulation results are presented allowing a comparison among both systems and showing the ability of the mobile robot to navigate among obstacles in different scenarios (navigation environment). Finally, initial real results with Arduino micro-controller are showed.

Hierarchic Fuzzy Approach Applied in the Development of an Autonomous Architecture for Mobile Agents

The development of a controller architecture based on hierarchic fuzzy logic to create the trajectory of a mobile agent in a virtual environment is initially used. The objective of the low cost controller is to regulate an open-source autonomous explorer robot, which moves between two pre-established points. The system establishes a viable route in a scenario with obstacles. Simulations in virtual environment are used to ensure the autonomy of the developed controller. Initial results demonstrate the success of this proposal.

Intelligent Systems Applied to the Control of an Industrial Mixer

This paper presents the application of intelligent techniques to control an industrial mixer. Control design is based on hebbian evolution of fuzzy cognitive maps. In this context, this paper develops a dynamical fuzzy cognitive map (D-FCM) based on Hebbian Learning algorithms. Two strategies to update FCM weights are derived. Finally, the D-FCM is used to control an industrial mixer. Simulation results of this control are presented. Additionally, results are provided extending some of the algorithms into the Arduino platform in order to acknowledge the performance of the codes reported in this paper.

Hybrid Dynamic Fuzzy Cognitive Maps Evolution for autonomous navigation system

This work develops a knowledge-based system to autonomous navigation using Fuzzy Cognitive Maps (FCM). A new variant of FCM, named Hybrid-Dynamic Fuzzy Cognitive Maps (HD-FCM), is used to model decision tasks and/or to make inference into a mobile navigation context. Fuzzy Cognitive Maps are a tool that model qualitative structured knowledge through concepts and causal relationships. The proposed model allows representing the dynamic behavior of a mobile robot in presence of environment changes. A brief review of correlated works in the navigation area, using FCM evolutions, is presented. Some simulation results are discussed and compared with Weighted Fuzzy System for shows the ability of the mobile to navigate among obstacles in different scenarios (navigation environment).

Cooperative Autonomous Agents Based on Dynamical Fuzzy Cognitive Maps

This work presents an architecture for cooperative autonomous agents based on dynamic fuzzy cognitive maps (DFCM) that are an evolution of fuzzy cognitive maps. This architecture is used to build an autonomous navigation system for mobile robotics that presents learning capacity, on line tuning, self-adaptation abilities and behaviors management. The developed navigation system adopts a multi-agent approach, inspired on the Brooks’ subsumption architecture due to its hierarchical management functions, parallel processing and direct mapping from situation to action. In this paper, a DFCM is hierarchically developed, from low-level describing reactive actions to the highest level that comprises management actions. A multi-agent scheme to share experiences among robots is also implemented at the last hierarchy level based on pheromone exchange by ant colony algorithm. The proposed architecture is validated on a simple example of swarm robotics.

A comparative analysis of three metaheuristic methods applied to fuzzy cognitive maps learning

This work analyses the performance of three different population-based metaheuristic approaches applied to Fuzzy cognitive maps (FCM) learning in qualitative control of processes. Fuzzy cognitive maps permit to include the previous specialist knowledge in the control rule. Particularly, Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and an Ant Colony Optimization (ACO) are considered for obtaining appropriate weight matrices for learning the FCM. A statistical convergence analysis within 10000 simulations of each algorithm is presented. In order to validate the proposed approach, two industrial control process problems previously described in the literature are considered in this work.

Redes dinâmicas cognitivas aplicadas no controle supervisório de um fermentador

This paper uses dynamic cognitive networks (DCN) as an intelligent tool for supervisory control. The DCNs are an evolution of fuzzy cognitive maps (FCM). Intelligent systems and tools use expert knowledge to build models with inference and / or decision taking abilities. A supervisory control architecture for an alcoholic fermentation process is developed from the acquisition of empirical knowledge from an expert. The objective of the supervisor is to operate the process in normal and critical situations. For this, we propose the use of a DCN model with new types of concepts and relationships that not only represent cause-effect as in FCM models. Simulation results are presented to validate the architecture developed.