Sensor integration for payload weight estimation in a robotic work cell

The objective of this paper is twofold - the calibration of a force/torque sensor and its integration into the robotic work cell, and the development of an exploratory procedure such as payload weight estimation that aids stable grasping of 'unknown' payloads. An artificial neural network is used to calibrate the force/torque sensor and its performance is compared with that of a conventional linear calibration procedure. The weight of the payload may not be specified a priori, and it may be required to estimate the weight in order to make a reasonable initial guess of the amount of gripper forces to be applied on the payload so that it neither slips out of the end-effector nor gets crushed.

Duke Scholars

Author Devendra P. Garg Thomas Lord Department of Mechanical Engineering and Materia ...