Redundant Robots

This chapter focuses on redundancy resolutionschemes, i. e., the techniques for exploiting theredundant degrees of freedom in the solution ofthe inverse kinematics problem. This is obviouslyan issue of major relevance for motion planningand control purposes.In particular, task-oriented kinematics andthe basic methods for its inversion at the veloc-ity (first-order differential) level are first recalled,with a discussion of the main techniques for han-dling kinematic singularities. Next, different first-order methods to solve kinematic redundancy arearranged in two main categories, namely thosebased on the optimization of suitable performancecriteria and those relying on the augmentation ofthe task space. Redundancy resolution methods atthe acceleration (second-order differential) levelare then considered in order to take into accountdynamics issues, e.g., torque minimization. Con-ditions under which a cyclic task motion resultsin a cyclic joint motion are also discussed; this isa major issue when a redundant manipulator isused to execute a repetitive task, e.g., in industrialapplications. The use of kinematic redundancy forfault tolerance is analyzed in detail. Suggestionsfor further reading are given in a final section.