Bacterial Evolutionary Algorithm-Trained Interpolative Fuzzy System for Mobile Robot Navigation

Abstract

This paper describes the process of building a transport logic that enables a mobile robot to travel fast enough to reach a desired destination in time, but safe enough to prevent damage. This transport logic is based on fuzzy logic inference using fuzzy rule interpolation, which allows for accurate inferences even when using a smaller rule base. The construction of the fuzzy rule base can be conducted experimentally, but there are also solutions for automatic construction. One of them is the bacterial evolutionary algorithm, which is used in this application. This algorithm is based on the theory of bacterial evolution and is very well-suited to solving optimization problems. Successful transport is also facilitated by proper path planning, and for this purpose, the so-called neuro-activity-based path planning has been used. This path-planning algorithm is combined with interpolative fuzzy logic-based speed control of the mobile robot. By applying the described methods, an intelligent transport logic can be constructed. These methods are tested in a simulated environment and several results are investigated.