학술논문
AlexNet based Real-Time Detection and Segregation of Household Objects using Scorbot
Document Type
Conference
Source
2020 4th International Conference on Computational Intelligence and Networks (CINE) Computational Intelligence and Networks (CINE), 2020 4th International Conference on. :1-6 Feb, 2020
Subject
Language
Abstract
In this paper, the pick and place task of household objects is accomplished using Scorbot-ER 5 Plus robotic arm. A monocular camera is utilized to detect the image of the objects. AlexNet based deep learning technique is devoted to determine the class of the objects before grasping process. Thereafter, the grasping and segregation processes are executed based on the inverse kinematic solutions between predefined pick and place locations. These solutions are derived from the transformation matrix of the end effector with respect to the inertial frame, analytically. It has been observed from the experimental analysis that AlexNet based technique presents promising results with good accuracy in the detection and the classification of the grasped objects.