Navigation of Mobile Robots Using Potential Fields and Computational Intelligence Means

A number of various approaches for robot navigation have been designed.However, practical realizations strike on some serious problems like imprecision ofmeasured data, absence of complete knowledge about environment as well ascomputational complexity and resulting real-time bottlenecks. Each of proposed solutionshas some of mentioned drawbacks. Therefore one of possible solutions seems to be incombining several means of computational intelligence. In this paper a combination ofharmonic potential fields, neural networks and fuzzy controllers is presented. Also somesimulation experiments are done and evaluated.