Adaptive and nonadaptive hybrid controllers for rheonomically constrained manipulators (Q1295146)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Adaptive and nonadaptive hybrid controllers for rheonomically constrained manipulators
scientific article

    Statements

    Adaptive and nonadaptive hybrid controllers for rheonomically constrained manipulators (English)
    0 references
    0 references
    0 references
    0 references
    6 June 2000
    0 references
    Denote the joint angles of a manipulator with \(N\) revolute joints by an \(N\)-vector \(q.\) When the end-effector of the manipulator is constrained on \(M\)-dimensional moving surfaces, the joint angle \(q\) must satisfy: \(\varphi(q, t) = 0\quad (1),\) where \(\varphi(q, t)\) is an \(M\)-vector function in which time \(t\) exists explicitly. Such constraints are said rheonomic. Assume that \(M < N.\) By differentiating the equation (1), one obtains \(J_\varphi(q, t)\dot q -\dot p(q, t) = 0,\) where \(J_\varphi(q, t)\) \((=\frac{\partial\varphi(q, t)}{\partial q})\) is an \(M\times N\) Jacobian matrix, and \(\dot p(q,t)\overset\triangle= -\frac{\partial}{\partial t}\varphi(q,t).\) An \(M\)-vector \(\dot p(q, t),\) called varying velocity of the constraints, depends on the change rates of the constraints. Under the constraints, the manipulator complies the following dynamics: (2) \(H(q)\ddot q + C(q,\dot q)\dot q + G(q) = \tau + J^T_\varphi(q, t)f\), where \(H(q)\) denotes the symmetric and positive definite inertial matrix, and \(C(q,\dot q)\dot q\) is the nonlinear force. Suppose that an exact expression of the constraints \(\varphi(q, t)\) is given, and a desired trajectory \(\{q_d,\dot q_d,\ddot q_d\}\) satisfying the motion constraints and a desired force \(f_d\) are specified. This paper deals with the problem of designing a proper joint torque \(\tau\) such that the real trajectory \(\{q_d,\dot q_d,\ddot q_d\}\) and force \(f\) governed by equation (2) follow the desired values as precisely as possible, respectively. A design method of hybrid controllers is proposed based on a modified joint-space orthogonalization scheme.
    0 references
    robotic dynamics
    0 references
    joint-space orthogonalization
    0 references
    hybrid position/force controller
    0 references
    manipulators
    0 references
    moving geometric surface
    0 references
    adaptive control
    0 references
    nonadaptive force control
    0 references
    exponential stability
    0 references
    tracking
    0 references
    regulation
    0 references

    Identifiers