Learning a robot controller using an adaptive hierarchical fuzzy rule-based system

Abstract

The majority of machine learning techniques applied to learning a robot controller generalise over either a uniform or pre-defined representation that is selected by a human designer. The approach taken in this paper is to reduce the reliance on the human designer by adapting the representation to improve the generalisation during the learning process. An extension of a Hierarchical Fuzzy Rule-Based System (HFRBS) is proposed that identifies and refines inaccurate regions of a fuzzy controller, while interacting with the environment, for both supervised and reinforcement learning problems. The paper shows that a controller using an adaptive HFRBS can learn a suitable control policy using a fewer number of fuzzy rules for both a supervised and reinforcement learning problem and is not sensitive to the layout as with a uniform representation. In supervised learning problems, a small number of extra trials are required to find an effective representation but for reinforcement learning problems, the process of adapting the representation is shown to significantly reduce the time taken to learn a suitable control policy and hence open the door to high-dimensional problems.

Advancing Fuzzy Logic: A Hierarchical Fuzzy System Approach

“…2) Advantages of Fuzzy Logic System: Additionally, FLSs have gained acceptance as a methodology for designing robust controllers that can effectively handle uncertainty and imprecision [13], [14]. They have been applied to various problems and have demonstrated their ability to generate more resilient and cost-effective solutions in the face of uncertainties.…”

A hierarchical fuzzy comprehensive evaluation algorithm for running states of an electromechanical system

“…Hierarchical fuzzy evaluation methods are widely used in learning robot control, construction goals and strategies, control systems, wind turbines, and power loads. The approach used in the paper was based on extending Hierarchical Fuzzy Rule-Based Systems, which learned a controller with a variable internal representation for both supervised and reinforcement learning problems while interacting with the environment (Waldock & Carse, 2016). Liu et al (2019) built the combination of multiobjective and optimization strategies and improved the comprehensive fuzzy evaluation to realize the ecological operation decision of the reservoir and the control system on the instance of Learning Management System.…”

Probabilistic fuzzy controller based robotics path decision theory

“…Problems like navigation from the narrow passages have effectively solved by Huq et al (2008). Fuzzy logic has been used with the combination with the sensor-based navigation technique (Gasos and Rosetti, 1999;Demirli and Turksen, 2000;Carelli and Freire, 2003;Waldock and Carse, 2015) to improve the incremental learning of the new environment. Recently, fuzzy triangulation method (Yaonan et al, 2011) and reinforced-based navigation (Jaradat et al, 2011) have been developed which helps to minimize the angular uncertainty and radial uncertainty present in the environment.…”