Models of Manipulators and Manipulation
Book, Wayne John
MetadataShow full item record
As a complex and expensive mechanical system, a manipulator, whether commanded by man or computer, is often better studied by and developed from a mathematical model of the system than by direct observation. Furthermore if we can build such a model of a system we have demonstrated our understanding of that system. The better our understanding, the more accurate and simple the model tends to be. This circular process of "understand a system to build a model to understand the system" is a basis, of progress in the natural sciences. In dynamic models of manipulators a fundamental feature is the compliance and mass of the components, modeled adequately in the most basic sense by Newton's laws of motion and Hooke's law for elastic deformation. Yet the understanding to build simple models, models simple enough to understand, does not exist. More critically the model of how the performance of manipulation is related to the mechanical performance of the manipulator does not exist, not at the basic level of Newton’s laws, nor at an empirical level based on more than twenty-five years of direct observation. The state of the art, its direction, and its preferred direction must be described, from an individual's perspective. The perspective here is of one who has been involved for several years in seeking understanding for modeling (MIT Department of Mechanical Engineering [1, 2, 3]) and in building models for understanding (research fellow NASA Johnson Space Center  and a current project at the Georgia Institute of Technology to design and build a planar experimental manipulator).