Design methodology for translational parallel manipulators exhibiting actuation redundancy

Abstract

In general, spatial manipulation of objects can be accomplished by parallel manipulators, whose number of actuators is equal to the demanded number of degrees of freedom. In order to improve, for example, positioning accuracy, stiffness characteristics, and transmission behavior, redundant drives can be added to the manipulator. Accordingly, this paper presents a methodology for the design of a translational parallel manipulator with redundant actuation. Based on the results of systematic structural syntheses and developed selection criteria, two valid configurations (i.e. 3-PŘŘŘ and 3-PUU) are analyzed. Since feasibility and performance of these configurations are dependent on the base geometry, five types of base geometries are introduced. First, the geometric parameters of each of the resulting 10 combinations of nonredundant configurations and base geometries are optimized by minimizing the maximal actuation force within a prescribed workspace. Second, the best combinations are used to generate redundant configurations with six legs. These redundant configurations are then analyzed with respect to the potential of improvement concerning homogenization of end-effector forces using force polytopes. It is shown that redundant actuation significantly improves the distribution of end-effector forces. This improvement has a positive influence on positioning accuracy and acceleration capabilities. In addition to these aspects, for further analysis it is planned to investigate the influence of homogenized end-effector forces on the dimensioning of actuators and finally on the energy efficiency of the entire configuration.