Kinematically redundant manipulators

This chapter focuses on redundancy resolution schemes, i.e., the techniques for exploiting the redundant degrees of freedom in the solution of the inverse kinematics problem. This is obviously an issue of major relevance for motion planning and control purposes. In particular, task-oriented kinematics and the basic methods for its inversion at the velocity (first-order differential) level are first recalled, with a discussion of the main techniques for handling kinematic singularities. Next, different first-order methods to solve kinematic redundancy are arranged in two main categories, namely those based on the optimization of suitable performance criteria and those relying on the augmentation of the task space. Redundancy resolution methods at the acceleration (second-order differential) level are then considered in order to take into account dynamics issues, e.g., torque minimization. Conditions under which a cyclic task motion results in a cyclic joint motion are also discussed; this is a major issue, e.g., for industrial applications in which a redundant manipulator is used to execute a repetitive task. The special class of hyperredundant manipulators is analyzed in detail. Suggestions for further reading are given in a final section.