机器人学基础Robotics-Chap10.pdfVIP

CHAPTER 10 Nonlinear control of manipulators 10.1 INTRODUCTION Previous chapter: Linear controllers. Linear approximation results in nonuniform damping throughout the workspace. This chapter: nonlinear controllers • Computed-torque method • Lyapunovs method 10.2 NONLINEAR AND TIME-VARYING SYSTMES Fig. spring is not linear. Local linearization: approximation of the nonlinear equations in the neighborhood of an operating point. Manipulator-control problem not well-suited, no linearization valid for the wide region of a manipulator. Moving linearization: move the operating point with the manipulator as it moves. Result in a linear, but time-varying, system. quasi-static linearization. Linearizing control law: use a nonlinear control term to cancel the nonlinearity in the controlled system, so that the closed-loop system is linear. 3 EXAMPLE 10.1 nonlinear spring in Fig. described byf = qx . Construct a control law to keep the system critically damped with a stiffness of KCL. open-loop equation partitioned control law model-based portion servo portion error equation Select k and k to maintain poles in fixed locations v p EXAMPLE 10.2 Consider Coulomb friction, t (nonlinear) in a mass-spring system. Design a control system that uses a nonlinear model-based portion to damp the system critically at all times. Solution: open-loop equation partitioned control law where model-based portion servo control law motion of error linear, select kv and kp to damp it critically. EXAMPLE 10.3 Single-link manipulator. The mass is located at a point 2 at the end, so moment of inertia is ml . Coulomb and viscous friction act at the joint. Design a control system that damps the system critically at all times. Solution: open-loop equ