복잡환경에서 다양한 형상을 가지는 식기의 안정적인 조작을 위한 파지정보 생성

Abstract

Recent advancements in robotics and artificial intelligence have enabled the integration of service robots into daily life, but the development of dish collection robots remains in its early stages. This study proposes a grasp information generation technique using a three-finger gripper for stable grasping of dishes with various shapes, categorized into circular, rectangular type-1, type-2, and type3. Mathematical formulations of corner constraints and the design of Search Space (SS), Graspable Space (GS), and gripper center movement ranges (rCSS, rCGS) were used to derive feasible grasping areas. Simulation experiments validated the proposed method, demonstrating its ability to stably grasp dishes in diverse environments. This study contributes to the advancement of dish collection robots for use in complex environments.