Object Recognition and User Interface Design for Vision-based Autonomous Robotic Grasping Point Determination

Abstract

The integration of Robot Operating System (ROS) with Human-Machine Collaboration (HMC) currently represents the future tendency toward Autonomous Robotic In-Situ Assembly on Construction Sites. In comparison with the industrial environment, construction sites nowadays are extremely complex and unpredictable, due to the different building components and customized design.This paper presents a visual-based object recognition method and user interface enabling on-site robot arms to autonomously handle building components, to build specific designs without the influence of material, shape, and environment. The implementation is an object recognition approach that serves with KUKA industrial robotic manipulator along with an RGB-depth stereo camera in an eye-in-hand configuration to grasp and manipulate found elements to build the desired structure. Opportunities for using vision-based autonomous robotic in-situ assembly on construction sites are reviewed.