Robot subtask performance with singularity robustness using optimal damped least-squares

The authors introduce a novel formulation of subtask performance for redundant manipulators. This formulation uses the singularity robust inverse (SRI), which is effective in yielding feasible joint motions in single regions, in conjunction with a homogeneous-like joint velocity component to perform an additional subtask such as obstacle avoidance. This novel inverse kinematic scheme enables the manipulator to overcome singularities (or, more accurately, avoid infeasible joint motions at or near singular configurations) and avoid obstacles in addition to the main motion task. The attractive aspect of this scheme is that it results in the satisfaction of two subtasks without requiring two explicit subtask criteria. This is because one of the subtasks, i.e., feasible joint motion, is implicitly incorporated in the particular solution in the form of the SRI. As depicted in simulations, the proposed formulation proves to be an effective inverse kinematic solution in singular regions of the manipulator workspace