A Theory of Generalized Inverses Applied to Robotics

Abstract

Robokcs research has made extensive use of techniques based on me Moore-Penrose inverse, or generalized inverse, of matri als. Recently it has been pointed out how noninvariant results may, in general, be obtained by applying these techniques to other areas of robotics, namely hybrid control and inverse ve locity kinematics. Unfortunately, the problems are not restricted to just these particular areas in robotics but are connected with misleading definitions of the metric properties of the six- dimensional wrench and twist vector spaces used in robotics. The current definitions lead to inconsistent results (i.e., results that are not invariant with respect to changes in the reference frame andlor changes in the dimensional units used to express the problem. As a matter of fact, given a linear systemu = Ax, where the matrix A may be singular, the Moore-Penrose theory of generalized inverses may be properly and directly applied only when the vector space U of vectoru and the vector space X of vector x are inner product spaces. Arbitrary assignment of Euclidean inner products to the space U and X when the vectorsu andx have elements with different physical units can lead to inconsistent and noninvariant results. In this article the problem of inconsistent, noninvariant solutionsX s tou = Ax in robotics is briefly reviewed and a general theory for com puting consistent, gauge-invariant solutions to nonhomogeneous systems of the formu = Ax is developed. In addition, the dual relationship between rigid-body kinematics and statics is de fined formally as a particular, linear algebraic system whose solution system is also a dual system. Examples illustrate the theory.

A fractional approach for the motion planning of redundant and hyper-redundant manipulators

“…and that the relation between q and x is given by the consequence is that it is possible to generate internal motions that reconfigure the manipulator structure without changing the gripper position and orientation [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17]. Nakamura and Hanafusa [18] proposed a least squares formulation with a damping factor under the name of singularity-robust inverse of Jacobian matrix, to provide continuous and feasible solutions even at or in the neighborhood of singular points.…”

“…Many techniques for solving the kinematics of redundant manipulators that have been suggested control the end-effector indirectly, through the rates at which the joints are driven, using the pseudoinverse of the Jacobian (see, for instance, [1,2]). The pseudoinverse of the Jacobian matrix guarantees an optimal reconstruction of the desired end-effector velocity-in the least-squares sense-with the minimum-norm joint velocity.…”

“…In this paper, we proposed a modified algorithm to solve the inverse kinematics problem of redundant manipulator, which uses a fractional derivative approach where J # is the Moore-Penrose generalized inverse of the Jacobian J [2,16]. It can be easily shown that a more general solution to Eq.…”

Inertial parameter estimation of floating base humanoid systems using partial force sensing

“…The parameters φ also live on a non-euclidian space, but if we assume K is full column rank (see the following section), then (8) will be independent of any metric on φ. See [18] for a full discussion of these issues.…”

Controlling dynamic contact transition for nonholonomic mobile manipulators

“…It is usually denoted J(q) + . Detailed results and proofs regarding generalized inversion are presented in [6].…”