An Adaptive Control Algorithm to Improve Singularity Avoidance in 7-DOF Redundant Manipulators

摘要

In this article, we consider the singularity avoidance of a redundant manipulator. It is shown that small perturbations caused by singularities can destabilize the whole system. Its realization would be rigorous physical damages to the robot. The remedy for such structural sensitivity to singularities consists in using an adaptive controller by which the control signal power is dominant over the unstable configurations. In this note, an adaptive algorithm to control redundant manipulators is proposed and its singularity avoidance performance as the secondary task is studied. An outer-inner loop controller called the augmented hybrid impedance control scheme is used. The inner loop consists of a Cartesian-level potential difference controller, a redundancy resolution scheme at the acceleration level, and a joint-space inverse dynamic controller. The proposed scheme works on the redundancy resolution block and is comprised of a multi-state function based on the Jacobean matrix of robot. We applied the proposed algorithm to a Mitsubishi PA10 7-DOF manipulator and simulation results proved the effectiveness of the algorithm.