Over the past few years, roboticists have been working tirelessly to create robotic grippers that can effectively pick up and manipulate various objects. While many grippers are inspired by the versatility of the human hand, they often come with complex mechanisms and advanced programming tools that make them impractical for wide-scale deployment. However, researchers from Purdue University and MIT have recently developed a simpler yet equally effective robotic gripper that challenges this notion.
The robotic gripper developed by Yu She and his team features only 5 degrees of freedom (DOF), making it much simpler than traditional high DOF dexterous robotic hands. Despite its reduced complexity, this gripper can still perform intricate in-hand manipulations, thanks to an integrated vision-based tactile sensor that provides crucial sensory data. By bridging the gap between high DOF robotic hands and one DOF grippers, this new design opens up possibilities for more cost-effective and energy-efficient robots.
The key feature of this new gripper is its ability to combine the simplicity of control with a high level of dexterity during complex manipulation tasks. The gripper consists of two fingers attached to a base, with each finger equipped with a linear actuator and a rotational servo motor. The addition of a GelSight mini vision-based tactile sensor on one of the fingertips provides valuable information about the grasped object’s geometry, orientation, and grip force.
In initial real-world experiments, the researchers found that this simplified gripper outperformed many traditional grippers with only a few DOF. It excelled in tasks such as singulation and scooping, showcasing its potential for handling complex object manipulation tasks effectively. The team plans to further develop this gripper and explore its capabilities in even more challenging tasks that are currently beyond the reach of other robotic grippers.
The introduction of this simplified robotic gripper opens up new possibilities for the future of robotic systems. Its streamlined design, combined with a focus on enhancing dexterity and control, could lead to the development of more efficient and cost-effective solutions for object manipulation tasks. This research not only highlights the potential of simplified robotic grippers but also paves the way for the creation of similar systems that prioritize simplicity and effectiveness in robotic design.
The work done by Yu She and his team demonstrates the importance of striking a balance between complexity and simplicity in robotic design. By creating a gripper that simplifies control while maintaining high dexterity, they have paved the way for more efficient and cost-effective robotic solutions in the future. This innovative approach to robotic gripper design could inspire new advancements in the field of robotics and lead to the development of more versatile and capable robots for a wide range of applications.