Learning cooperative linguistic fuzzy rules using the best-worst ant system algorithm

Abstract

Within the field of linguistic fuzzy modeling with fuzzy rule-based systems, the automatic derivation of the linguistic fuzzy rules from numerical data is an important task. In the last few years, a large number of contributions based on techniques such as neural networks and genetic algorithms have been proposed to face this problem. In this article, we introduce a novel approach to the fuzzy rule learning problem with ant colony optimization (ACO) algorithms. To do so, this learning task is formulated as a combinatorial optimization problem. Our learning process is based on the COR methodology proposed in previous works, which provides a search space that allows us to obtain fuzzy models with a good interpretability-accuracy trade-off. A specific ACO-based algorithm, the Best-Worst Ant System, is used for this purpose due to the good performance shown when solving other optimization problems. We analyze the behavior of the proposed method and compare it to other learning methods and search techniques when solving two real-world applications. The obtained results lead us to remark the good performance of our proposal in terms of interpretability, accuracy, and efficiency.

Quick Design of Fuzzy Controllers With Good Interpretability in Mobile Robotics

“…It is a population search bioinspired technique that considers heuristic information to allow it to get good solutions quickly. Indeed, as was shown in [33] and [43], the use of ACO in COR, as opposed to other kinds of optimization techniques such as simulated annealing and genetic algorithms, performs a quick convergence obtaining accurate results. This section briefly describes the main components of the considered COR-based ACO algorithm [33].…”

“…The technique automates a number of tasks in the fuzzy controller design, thus leaving only a few components to be defined by the expert. With the aim of addressing some of the above-mentioned drawbacks by performing a quick learning and providing fuzzy controllers with good interpretability, we propose to use a simple but effective method: the cooperative rules (COR) methodology [31]- [33]. In this paper, we have adapted this methodology to our problem by considering several consequent variables and using a different fitness function to facilitate the rule base reduction.…”

Flow Discharge Modeling in Open Canals Using a New Fuzzy Modeling Technique (SMRGT)

“…A computational geometry approach was introduced in Wan et al to determine the minimum number of rules required in building a fuzzy model [78]. Casillas et al introduced a novel approach to the fuzzy rule learning problem with ant colony optimization (ACO) algorithms [79]. Finn presented an algorithm, FUZ-ZEX, for learning FRs from a corpus of data mapping input antecedents to output consequents [80].…”

Self-Organizing Multirobot Exploration through Counter-Ant Algorithm

“…The operating results led to a new field called swarm intelligence, covering ants' algorithms. These algorithms provide powerful methods for the design of algorithms and optimization of distributed problems involving a collaborative swarm behavior [2] [4] [5] [19]. The intrusion of these algorithms in the world of robotic improve the communication quality between robots but the convergence time remains problematic [1] [6] [7] [11] [12] [13] [18] [20] [21].…”