Abstract
This chapter presents a robotic object manipulation method and an adaptive force control method using fuzzy logic, neural network, and fuzzy neural network techniques. The fuzzy neural network technique provides effective adaptation ability to the controllers to deal with the uncertainty of both the object and the robot manipulators. An unknown object can be precisely manipulated by the learning ability of the main fuzzy neural controller. The sub-fuzzy-neural controllers precisely realize the required force for the desired object motion. The chapter presents two kinds of input adjustment methods for the instant adjustment of the sub-fuzzy-neural controllers according to the property of the unknown object: the input adjustment neural network (IANN) is effective when the kinds of materials used for the object are limited, and the fuzzy input adjustment method is effective when the kinds of material used for the object are not limited. The chapter also presents the fuzzy evaluation method for the effective learning of the sub-fuzzy-neural controllers.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call