Fuzzy modeling. Part I: A spatial model approach for robot navigation via fuzzy potential energy

Abstract

This two-part series discusses a novel design and application of fuzzy modeling. There are two types of fuzzy modeling: (1) irritating expert experience or implementing engineering knowledge and (2) modeling a complex or unknown system. In Part I of this paper, the second type of fuzzy modeling is employed to define Fuzzy Potential Energy (FPE). Different FPE values are then assigned into a workspace to construct a spatial model for robot navigation. How the FPE guides a robot is revealed by its gradient directions. A workspace that contains a U-shaped obstacle is used as an example to illustrate the FPE applications. Two situations, one with the start position outside the U-shaped obstacle and the other with the start position inside, are discussed. The discussions show that FPE provides a way to merge both global and local path-planning strategies. The proposed FPE concretely improves traditional path planning strategy. This two-part series pioneers a novel design and application of fuzzy modeling for path planing and motion control.