Soft Computing Based Mobile Manipulator Controller Design

During the last decades, numerous papers have been written on how to apply neuronal networks, fuzzy (multi-valued) logic, genetic algorithms and related ideas of learning from data and embedding structured human knowledge. These concepts and associated algorithms form the field of soft computing. They have been recognized as attractive alternatives to the standard, well established hard computing (conventional) paradigms. Traditional hard computing methods are often too cumbersome for today’s problems. They always require a precisely stated analytical model and often a lot of computation time. Soft computing techniques which emphasize gains in understanding system behaviour in exchange for unnecessary accuracy have proved to be important practical tools for many real world problems. Because they are universal approximators of any multivariate function, the neuronal networks and fuzzy logic are of particular interest for modelling highly nonlinear, unknown or partial known complex systems. Due to their strong learning and cognitive ability and good tolerance to uncertainties and imprecision, soft computing techniques have found wide applications in robotic systems control. According to Zadeh (Zadeh, 1994), the basic premises of soft computing are