Neuromorphic sensing and control: Applications to position, force, and impact control for robotic manipulators

Abstract

The authors present a neural network (NN)-based approach for sensing and control of a robotic manipulator. They corroborate the effectiveness of the proposed approach for impact control of a robotic manipulator. Collisions are very quick phenomena and have strong nonlinearity. Therefore, it is difficult to sense collisions and to control the robotic manipulator undergoing collisions. The proposed approach has robustness against the impact force. It also has effectiveness in sensing and recognition of collisions by using a proximity sensor. In this case, the NN-based controller can acquire desirable manipulation of its own accord, so as to avoid the impact.