Design of kinematically-redundant two-module robots

当需要精确定位时（如在机床中，或在校准表中），并联机器人由于其高刚性而成为首选。然而，并联机器人的主要缺点之一是它们相对较小的工作空间，这进一步受到所谓的2型奇点的限制。当并联机器人接近2型奇异构型时，其末端执行器承受载荷的能力显著降低。由于这些奇异之处，很少有公司能够想出创新的设计，大多数公司都坚持使用流行的六足架，对于手头的应用来说，这种六足架通常设计过度（例如，太大）。

When precise positioning is needed (as in a machine tool, or in an alignment table), parallel robots are a choice of predilection due to their high rigidity. However, one of the major disadvantages of parallel robots is their relatively small workspace, which is further limited by the presence of so-called Type 2 singularities. When a parallel robot is close to a Type 2 singular configuration, its ability to withstand loads with its end-effector is significantly compromised. Because of these singularities, very few companies have managed to come up with innovative designs and most stick to the popular hexapod, which is often overdesigned (e.g., too big) for the application at hand. The divide-and-concur strategy is promising for achieving affordable high-accuracy robot systems. Instead of designing a single n-degree-of-freedom (n-DOF) parallel robot, the applicant would opt for the use of two separate robots, one with i DOFs aimed at manipulating the tool (possibly being a serial robot) and another with j DOFs for holding the workpiece. Furthermore, this two-module robot system will be kinematically redundant, which means that the sum of i and j will be greater than n. The main idea behind the proposed research program is to use a combination of two robots (at least one of which is parallel) with one or two degrees of kinematic redundancy, i.e., i + j = n + 1 or i + j = n + 2, in order to completely eliminate problems with singularities. The first objective of the proposed program is therefore to develop a systematic approach for analysing the geometric conditions leading to kinematically redundant designs with large workspace and singularity manifolds of reduced dimension. Particular attention will be paid to lower-mobility spatial parallel mechanisms with mixed DOFs, which have proven to be an excellent choice for machine tools. The second objective will be designing secondary modules for existing serial and parallel robots. Finally, the third objective will be the development of novel low-cost automated calibration techniques for these two-module robots.