Content area
Full Text
Abstract
In this paper, dynamic equations of motion of a rigid-link non-redundant n-DOF robot manipulator consisting of mechanical arms, all with revolute joints and electrical actuators are considered and application of passivity-based and integrator backstepping techniques for trajectory tracking of the robot in presence of disturbance, friction and uncertainty is studied.
One of the main advantages of the backstepping control techniques is that they impose desired properties of stability, initially by fixing the candidate Lyapunov functions, then by calculating the storage functions, stabilizing functions and feedback control laws in a recursive way.
Simulation results of applying these schemes show the ability of backstepping methods in trajectory tracking in presence of disturbance, friction and parameter uncertainty. Comparison of simulation results of the two backstepping methods shows that passivity based with respect to integrator based method has more smooth control commands and better trajectory tracking by using passivity properties of the robot and by using some weighting matrices in its algorithm.
Key Words
Passivity-based backstepping, integrator backstepping, actuators dynamics, RLED robot manipulators, robot control
(ProQuest: ... denotes formulae omitted.)
1. Introduction
In robotic systems, there are actuators used to drive the links, wheels, legs, etc. Usually, electrical actuators are utilized. Electrically driven robot manipulators can be controlled in two different levels: torque level and voltage level. When one uses a torque-level controller, actuator dynamics is disregarded which implies a simplified (typically linear) relation between torque and voltage vectors is considered. Recently, actuator dynamics has been explicitly included in robot manipulator control schemes [1-11]. This dynamics becomes extremely important during situations in which motor currents are rapidly changing. However, this inclusion increases the order of equations of system dynamics and complicates the controller structure and its stability analysis. To avoid the increase of order of system dynamics, the mechanical arms and electrical actuators dynamics can be considered as two cascade loops [4].
Most of model-based control schemes proposed for rigid-link electrically driven (RLED) robot manipulators are different versions of adaptive and robust controllers. One of the attractive features of the adaptive controllers is that the control implementation does not require a priori knowledge of unknown system parameters. Several direct adaptive algorithms have been presented for trajectory tracking of RLED robots in the literature. In some of them...