Real time visual control of two-axis robotic arm using motion estimation

Abstract

Visual servoing is the fusion of results from many elemental areas including high speed image processing, kinematics, dynamics, control system, and real-time computing. This paper presents the use of vision to control a two-axis robotic arm with active tracking of the probe. On specifying the target position in image coordinates, a suitable linear/non-linear mapping is used to transform it to real world coordinates. This is changed to joint level inputs and given to the actuators/drives. Once the probe starts moving the image acquisition is done continuously using NI-IMAQ 1407. The acquired image is processed real time using an accurate object-tracking algorithm, which has low computational cost and gives better performance. Predictor-corrector filter (Kalman filter) is used for real-time filtering of the noise. All the experimental results are presented.