|Project Title:||Intelligent and Integrated (I2) Gripper using Multisensors and actuators|
|Hosting Institution:||LSCM R&D Centre (LSCM)|
|Abstract:||Generally speaking, robotic arms and grippers are in-separable to carry out automatic tasks (e.g. pick and place). Robotic arms are electro-mechanical structures and programmable to perform repeatable movements/motions (e.g. moving forth and back between point A and B) within defined time. Robotic arms provide coarse location (X, Y and Z) in Cartesian Co-ordination system. Where, robotic grippers usually attached to the tips of robotic arms, namely “end-effectors”. The provide fine locations (x, y and z) and orientations (alpha, beta and gamma) to approach objects that are going to be grasped. Hence, robotic grippers enhance/enable high degree of freedoms (dof) to the manipulation of robotic arms. i.e. high dexterity.
There are many types of grippers - magnetic, pneumatic and motorized fingers to improve dexterous, manipulability and flexibility of robotic arms. Motorized grippers (or fingers) are most commonly seen. They are able to grasp objects with regular and irregular shapes regardless of object material characters/contents. The more the fingers, the more the dexterous it would be. But it is rather expensive. For example, a 5-fingers gripper may cost up to HK$500,000. On the other hand, pneumatic grippers are growing rapidly in recent years. It is able to achieve high dexterity, at a fraction of the cost, to its
counter-part (i.e. multi-finger grippers). However, pneumatic grippers require continuous supply of negativeair pressure (i.e. an air pump). It would make extra noise. In addition, it’s required that the object surface would be flat to reduce air leakage. Some grippers may target to pick specific object (e.g. electro-magnetic grippers) are likely to grasp objects with ferrite magnetic properties. However, a multi-actuation and sensor-rich gripper has yet been researched and developed.
This project aims to research and develop an intelligent and integrated gripper with multiple sensors and actuators for automatic sorting applications. Hence, proposed gripper is able to pick moving objects and detect its material characteristics (e.g. plastics, paper and so on) and irregular shapes on the conveyor
belt. Machine intelligence techniques will be researched and developed to identify object shapes and orientations (for grasping) to enhance efficiency and costeffectiveness.
|Project Coordinator:||Dr Ka Lun Fan|
|Approved Funding Amount:||HK$2.76M|
|Project Period:||01 Feb 2020 - 31 Jan 2021|