Combination of fuzzy logic and neural networks for the intelligent control of micro robotic systems

Abstract

The authors present an advanced control concept for microrobotic systems, which is based on the combination of a neural network approach for the adaptation of manipulation parameters and a fuzzy logic approach for the correction of parameter values given to a conventional controller. This multilevel system architecture is suitable for the intelligent control of microrobots that can operate autonomously in changing environments. Typical tasks for these robots are exploration and fine manipulation, which demand intelligent task planning and motion/force control capabilities. The planning component deals with the successive determination of initial manipulation parameters, whereas the neural system performs during manipulation, computing suboptimal grasp forces and learning inference rules used for parameter adjustment.