Design of parallel robots with reduced singularities and improved absolute accuracy

点击翻译按钮获取中文摘要

Parallel robots consist of a mobile platform connected to a base via several articulated legs. Due to their undisputable advantages of agility and rigidity, parallel robots have become a standard for applications where speed, precision, or payload capacity is a priority. Thousands of parallel robots have already been manufactured by dozens of companies, including Canadian firms CAE and Quanser. Unfortunately, parallel robots have certain disadvantages that are not negligible. They usually have a small workspace that is segmented by undesirable so-called parallel singularities. Nearly all commercially-available parallel robots are based on designs that avoid these singularities through a judicious choice of mechanical interferences and actuator limits. Obviously, this conventional approach is not necessarily optimal. The main objective of this research program proposal is to develop a singularity-oriented design methodology for parallel robots in which parallel singularities are eliminated or bypassed without compromising workspace and accuracy. The proposed research program consists of two intimately related topics to be addressed simultaneously: (1) minimizing the negative effect of singularities in some parallel robots through design modifications and (2) maximizing the accuracy of these novel designs. Three unorthodox approaches to reducing parallel singularities while enlarging workspace and improving accuracy will be adopted. Firstly, designs will be sought of redundant parallel robots with a minimal number of extra legs, in which parallel singularities are eliminated. Secondly, designs will be sought of non-redundant parallel robots in which the dimension of the parallel singularity manifold is reduced. Thirdly, designs will be sought of non-redundant parallel robots, in which access to a large workspace area is secured by crossing so-called serial singularities only. In all three approaches, systematic methods will be developed for finding these optimal kinematic designs. The mechanical design of each of the novel robots will also be performed and calibration methods will be proposed for two of the novel designs for which prototypes will be built.