Parallel Jaw

Parallel jaw grippers, a common robotic end-effector, are the focus of ongoing research aimed at improving their grasping capabilities and applicability across diverse tasks. Current research emphasizes developing robust grasp detection methods, often employing neural networks and novel SO(3) representations to address inherent symmetries and improve consistency. This work also explores integrating advanced sensing, such as capacitive sensing, for enhanced precision manipulation of various materials, including liquids and granular media, and optimizing gripper design and grasp strategies for improved performance in cluttered environments and complex tasks like untangling. These advancements are crucial for expanding the capabilities of robotic systems in manufacturing, logistics, and other fields requiring dexterous manipulation.