Underactuated robot hands -- with fewer motors than joints -- have been around for decades; however, we've seen a surge of new designs in recent years. Personally, I attribute this trend to the availability of low-cost robot arms and associated open-source software (ie. ROS). In any case, underactuated hands offer numerous advantages in terms of cost, size, weight, and mechanical / electrical complexity while providing a large array of shape-adaptive grasps. In this article, I'd like to introduce you to two new underactuated robot gripers, the Lacquey "Fetch Hand" and the Willow Garage "Velo Gripper." Be sure to check out the photos and videos below.
Indeed, there has been a flurry of robot hand development in the last few years. Perhaps most exciting (to me) are the all the esoteric gripper technologies: Electroadhesive grippers being commercialized by GrabIt, jamming grippers created by iRobot (etal), and improved Bernoulli grippers.
But we've also seen a lot of new underactuated systems appear: Meka Robotics' H2 compliant hand, RobotIQ's 3-finger and 2-finger grippers, Aaron Dollar's SDM hands, and Kinova Robotics' Jaco robot hand
** Disclaimer: Meka Robotics & RobotIQ are Hizook sponsors.
Now we can add two more to that list: the Lacquey "FetchHand" and the Willow Garage "Velo Gripper."
This post was inspired by a recent conversation with Martijn Wisse, an associate professor at Delft University of Technology and also co-founder and CTO at Lacquey Robot Grasping Solutions. He had this to say about their new gripper:
What started a decade ago (for us) as an academic interest, quickly led to a long quest for robustness and reliability. Like many others, we were impressed by the many advantages of underactuation (fewer motors than joints): the automatic shape adaptation, the nice and equal contact force distribution, and the potentially low production cost. Little did we know that it would take years to develop our new (patented) actuation technology before we would actually have a (really) robust and affordable design (for sub-€1000) without the failure-prone cables often used for such grippers. But now, it’s here: the Lacquey FetchHand.
The FetchHand can grasp a wide range of objects with masses of up to 1 kilogram and diameters of up to 125 mm. Round, spherical, cubic or more complex shaped objects can be grasped without modifications. The FetchHand can fully enclose an object between all three fingers or grasp an object with its fingertips. It can grasp an object from the top or from the side. Even objects with a relatively slippery surface can be grasped, because FetchHand encloses an object with all fingers in a power grasp.
Just mount the gripper to your robot arm or test installation with a few bolts, connect the power cable to your control signal and you are good to go! The gripper has an on board force control that accepts any input between 12 and 30 VDC, while protecting the motor from electrical overload. Switching the sign of your signal makes FetchHand open and close, easy as that.
The FetchHand contains one motor and a patented mechanism that drives its three fingers with 2 phalanxes each. By accurately controlling the motor torque, the grasping force can be predicted for products within a certain size range. You don’t have to know exact finger positions or local contact forces (sensory feedback) to get a proper grip on the object. As long as you control the grasping-force, pretty much any object can be grasped without damage.
The FetchHand was developed by Lacquey Robot Grasping Solutions, a spinoff company to commercialize the underactuated gripping technology (with some assistance via SBIR grants). Lacquey focuses on grasping technology for vulnerable objects and objects that have a large variation in shape and size, such as fruits, vegetables, and household objects.
To quote from the Willow Garage Velo gripper website:
Parallel grippers are effective on a wide range of objects and tasks, executing fingertip grasps between two perfectly opposing pads. However, adding the ability to envelop objects can greatly increase the stability of the grasp in many situations. We explored the design space aiming to achieve both of these capabilities.The Velo has two fingers with two joints each, and is tendon-driven. Aiming to reduce the complexity and potential cost, we used a single actuator, resulting in an underactuated design. The gripper executes fingertip or enveloping grasps by passively adapting to the shape of the grasped object. Extension is passive, provided by springs.
The tendon-driven design allows for a modular implementation, with a tool-less quick-change interface between the kinematic module, composed of the palm and fingers, and the actuation module, containing the motor and the transmission. This interface allows easy experimentation with various types of kinematic components, and also enables easy replacement of the kinematic module should it become damaged.
Exciting new stuff in the world of gripper design.