In this paper, two inverse-kinematic solutions with singularity robustness for a fully parallel robot manipulator are proposed. First, two inverse-kinematic solutions are presented based on the damped leastsquares method, where the first solution (OSVC) is derived using a velocity constraint, and the other (OSFC) uses a force constraint. It can be shown that both solutions are special cases of a general damped least-squares method (GIKS). Next, under the assumption that direct position measurement is available, a feedback-correction term is introduced into the general damped least-squares method to eliminate positioning error in resolved-rate control. It can be shown that both solutions are stable with bounded errors, but each has its own relative merits and shortcomings. Illustrative examples are also given to show their effectiveness in practical applications.

Two inverse kinematic solutions with singularity robustness for parallel robot manipulators are proposed in this paper using, respectively, velocity constraint and force constraint. They are unified in a general damped least-squares method with feedback corrections are introduced to eliminate positioning error in resolved-rate control (ORRCFC). Both solutions are shown to be stable.