An Intelligent Control Method for Redundant Robotic Manipulators with Output Constraints

Abstract

The kinematic redundancy of the robot leads to many solutions to the inverse kinematic problem. Thanks to this feature, redundant robots can be flexible in working environment with desired constraints. In this paper, a control approach via an intelligent two-layer method is presented for high-accuracy problems of redundant robots under output constraints. In the high-level control layer, the main control objective is to derive virtual velocities to minimize the output control errors that are formulated based on the system kinematics. To avoid the physical collision of the joint movement, proper constraints are synthesized as a second objective utilizing nonlinear functions. Furthermore, the second constraint is associated with a soft-gain function to effectively emphasize its effect in the general control structure. The overall control objective is then realized by a proportional-integral controller in the low-level layer. A 3 degree of freedom (DOF) planar arm was chosen to verify the performance of the proposed controller. The comparative simulation and experimental results confirmed the advantages and practical applicability of the proposed algorithms.