Nesrine Baklouti

Motion Planning in Dynamic and Unknown Environment Using an Interval Type-2 TSK Fuzzy Logic Controller

Motion planning of mobile robots in unknown and dynamic environments is faced with a large amount of uncertainties. Such those uncertainties we find; the measurement noise, the membership functions translations, data uncertainties... In fact, the known type of fuzzy logic (FL), Type-1, gave some solutions. But, in the last few years, new trends and theory in FL have been appeared, proposing by thus the Type-2 Fuzzy Logic Systems (Type-2 FLSs) which can handle and minimize the effects of the cited uncertainties with a better performance. This paper deals with the design of an Interval Type-2 fuzzy logic controller for the navigation of mobile robots in unknown and dynamic environments. The obtained results are presented and are compared with the navigation using the Type-1 Fuzzy Logic system. The Type-2 FLSs provide very good results and outperform the correspondent Type-1 FLS.

The geometric interval type-2 fuzzy logic approach in robotic mobile issue

Recently type-2 Fuzzy logic systems (FLSs) have demonstrated their competence in treating vagueness in real world dynamic systems. But, in the last few years, new trends and theory in Fuzzy Logic have been appeared, proposing the geometric type-2 Fuzzy logic approach. The main idea of this approach was to model fuzzy logic sets using computational geometry providing by this more accurate results and better performance in treating vagueness. Throughout this paper, we study the effect of the geometric approach in robotic mobile issue. We propose two controllers: a geometric interval type-2 fuzzy logic local avoiding obstacles controller and a geometric interval type-2 fuzzy logic wall following controller. The obtained results are presented and are discussed. The geometric type-2 FLSs provide good results…

Hybridization of Fuzzy PSO and Fuzzy ACO applied to TSP

Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) algorithms have attracted the interest of researchers due to their simplicity, effectiveness and efficiency in solving real world optimization problems. Swarm-inspired optimization has recently become very popular. Both ACO and PSO are successfully applied in the Traveling Salesman Problem (TSP). Our approach consists in combining Fuzzy Logic with ACO (FACO - Fuzzy Ant Colony Optimization) and PSO (FPSO - Fuzzy Particle Swarm Optimization) for solving the TSP. Experimental results and comparative studies illustrate the importance of Fuzzy logic in reducing the time and the best length for the TSP problems considered.

Real time PSO based adaptive learning type-2 fuzzy logic controller design for the iRobot Create robot

Recently, there has been a considerable interest on learning type-2 fuzy logic systems, essentially on how determining the footprint of uncertainties of linguistic variables. In fact, the complexity and difficulty in developing type-2 fuzzy systems can be located at the time of deciding which are the best parameters of membership functions (MFs). In real robot applications, the task of designing a type-2 fuzzy logic controller is complex enough essentially because the presence of many forms of noise and uncertainties, where the robot while navigating has to control many variables. In this paper we present a novel adaptive learning type-2 fuzzy logic controller (FLC) for robot motion planing task. The MFs are tuned instantanously using real time particle swarm optimization technique. The proposed architecture presented good results which were demonstrated on the “iRobot Create” robot.

Interval Type-2 Beta Fuzzy Basis Functions: Some Properties and their First-Order Derivatives

In this paper we introduce the Interval type-2 Beta fuzzy set as a membership function in a Fuzzy Logic System (FLS). First order derivatives of type-1 and type-2 Beta functions were developed for designing fuzzy logic systems based on given input-output pairs. Then, the steepest descent algorithm is used to train Beta fuzzy basis functions to obtain the final fuzzy system. The performance of the proposed model of Beta fuzzy logic system is evaluated using the benchmark of Forecasting of Time-Series and is compared to fuzzy systems using Gaussian membership functions as a popular example of shapes.

PSO based adaptive learning Fuzzy Logic Controller for the Irobot Create robot

In real robot applications, the task of designing a fuzzy logic controller is complex enough essentially because the presence of many forms of noise and uncertainties. The robot while navigating has to control many variables to get the best result at the end of the task: best smooth trajectory, the guarantee of arrive to goal, lowest time, etc. We present in this paper a novel Particle Swarm Optimization based adaptive learning Fuzzy Logic Controller design for a motion planning task. The membership functions of a fuzzy controller are tuned instantanously using particle swarm optimization technique. The proposed architecture presented good results which were demonstrated on the robot “iRobot Create”.

Interval type-2 beta fuzzy neural network for wheeled mobile robots obstacles avoidance

Navigation of mobile robots in dynamic and unknown environments is usually cluttered with noise and errors. In the literature, several solutions have been proposed. Recently, type-2 fuzzy logic have showed having the ability to handle uncertainties, imprecise and incomplete data. Since, it has been constituted a new hopeful and promising technique for further improve control of mobile robots in real time applications. In elaborating fuzzy controllers, the most used membership functions in existing works, are the Gaussian, trapezoidal or triangular. In this paper, we propose a new Interval Type-2 Beta Fuzzy Neural Network (IT2BFNN) for obstacles Avoidance task for wheeled mobile robots. The main and novel idea is to involve type-2 beta fuzzy sets in the design process of a fuzzy network for the navigation process. The proposed architecture controller is based on beta type-2 fuzzy sets in the antecedent part, while the consequent part performed the TSK (Takagi-Sugeno-Kang) fuzzy output strategy.

A Beta basis function Interval Type-2 Fuzzy Neural Network for time series applications

Abstract The huge complexity and uncertainty in real life requires the use of advanced automatic learning methods to find out better approximators and suitable relationship in real data behavior. Neuro fuzzy systems have been proved to be excellent universal approximators. In this paper we propose a new based function Interval Type-2 Fuzzy Neural Network denoted ”Beta basis function Interval Type-2 Fuzzy Neural Network”, the BIT2FNN. The main idea is to involve type-2 beta fuzzy sets in the design process of fuzzy networks. The proposed architecture is based on beta type-2 fuzzy sets in the antecedent part, while the consequent part achieves the TSK (Takagi–Sugeno–Kang) fuzzy output strategy. Thanks to the beta function flexibility, the network achieve a good performance and shows a good resistance to noisy data. First order derivatives of type-1 and type-2 Beta functions were developed for the first time for designing fuzzy logic systems based on given input–output pairs. The backpropagation algorithm was used for the learning process of antecedent fuzzy beta parameters and the consequent part. The performance of the proposed model of Beta fuzzy logic system is evaluated with mainly two problems of time series applications : the Mackey Glass Chaotic Time-Series prediction problem with different setting of parameters and levels of noise and the ECG heart-rate Time Series monitoring problem.

A beta-fuzzy-near-sets approach to research for visually similar content images

In the automated search system, similarity is a key concept for solving the human task. The human process is a natural categorization, which underlies many natural abilities such as image recovery, language comprehension, decision making or pattern recognition. In this paper, the focus is on the use of similarities in image retrieval search using near sets of similarity approaches. The results showed that a general framework for Near set is compatible with these foundations, and that similarity measurements can be involved in all steps of the image research process. We therefore focus on the fuzzy logic which provides interesting tools for data mining mainly because of its ability to represent imperfect information. We then introduce a new category of a fuzzy set : the Beta function. We finally illustrate our work with examples of similarities used in the real world of image retrieval problems.

A novel method for resemblance images using near fuzzy set

We introduce a new method Near-Fuzzy set for analysis image. Indeed, near sets are considered a generalization of the rough sets theory. A set X is close to another set Y insofar as the description of at least one of the object of X corresponds to the description of least one of objects of Y. Find the tolerance classes with objects of the same description is a major problem. Maximal Clique Enumeration Algorithm solves the same problem and improves field performance image resemblance. We propose an innovative technique that hybrids both near sets approach with Fuzzy sets approaches. In this paper we use the Near-Fuzzy sets method to obtain better results in the resemblance of facial images. The performance of use of near set approach has been proved throughout the Japanese Female Facial Expression (JAFFE) database.