Lei Deng och Hong Tao, Elektroteknik

Abstract:
Nowadays, home service robots have become a research hot-spot in the field of robotics, and delivering small objects is the most important research task of home service robots. Grasping is an essential component of robotic manipulation and focuses on obtaining complete control over the motion of an object. The grasping task is divided into object detection, grasp planning, and grasp control.
 
This thesis aims to analyze the manipulation strategies of different objects applied to the TIAGo robot. Firstly, we implement object detection using different algorithms and strategies, including object color detection with OpenCV, point cloud library clustering method, and unseen object clustering method for object segmentation. Secondly, we realize and analyze robotic grasp planning. Thirdly, we implement robot motion control using different strategies. Consequently, we integrate the whole manipulation strategies implementation and realize the complete grasping task of different objects on the TIAGo robot.
 
Welcome to the presentation!

Lei Deng and Hong Tao