Abstract
Robots frequently need to work in human environments and handle many different types of objects. There are two problems that make this challenging for robots: human environments are typically cluttered, and the multi-finger robot hand needs to grasp and to lift objects without knowing their mass and damping properties. Therefore, this study combined vision and robot hand real-time grasp control action to achieve reliable and accurate object grasping in a cluttered scene. An efficient online algorithm for collision-free grasping pose generation according to a bounding box is proposed, and the grasp pose will be further checked for grasp quality. Finally, by fusing all available sensor data appropriately, an intelligent real-time grasp system was achieved that is reliable enough to handle various objects with unknown weights, friction, and stiffness. The robots used in this paper are the NTU 21-DOF five-finger robot hand and the NTU 6-DOF robot arm, which are both constructed by our Lab.
Highlights
Rapid technology development is enabling intelligent robots to be used in many fields, such as medicine, the military, agriculture, and industry
Recognition and grasping of unknown objects in a cluttered scene have been very challenging to robots
This section describes how the NTU 6-DOF robot arm [5,6,35,36,37] and the NTU five-finger robot hand were equipped with additional hardware and software to enable the resultant grasp of unknown objects
Summary
Rapid technology development is enabling intelligent robots to be used in many fields, such as medicine, the military, agriculture, and industry. A robot’s ability is a key function to grasp and manipulate an object that helps people with complicated tasks. In order to provide daily support by using humanoid hands and arms [1,2], robots must have the ability to grasp a variety of unseen objects in human environments [3]. A common gripper has the limitation of not being able to grasp a great variety of objects. Studies need to be focused on using a multi-fingered robot hand to grasp objects with different shapes. This study attempted to develop a grasping system that is fast, robust and does not need a model of the object beforehand in order to reduce reliance on preprogrammed behaviors.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have