Robust Underwater Grasping of Sloped Objects with a Waterproof Passive Adaptive Gripper

Imagine a robot arm reaching into a kitchen sink to pick up a wet, slippery mug. In the dry air of a workshop, a standard mechanical hand might grip the handle firmly. But underwater, the rules change. Water acts as a lubricant, stripping away the friction that usually keeps objects from sliding. When a robot tries to lift a tilted or uneven item, like a cone-shaped cup or a bowl with a curved rim, the lack of grip causes the object to roll or slide out of the hand. This is not just a minor inconvenience; it is a fundamental barrier preventing service robots from performing everyday chores like loading dishwashers or clearing wet counters. For a machine to succeed in these wet, domestic environments, it needs a way to hold on that does not rely on the high friction we take for granted on land.

Researchers at the University of Illinois have developed a solution to this specific problem: a waterproof robotic hand designed to grasp sloped, slippery objects with surprising stability. The device, which they call a passive adaptive gripper, does not use complex sensors or powerful motors to constantly adjust its grip. Instead, it relies on clever mechanical design that allows the hand to physically reshape itself to fit the object it is holding. The core of the invention is a combination of two simple but effective mechanisms. First, the fingers are mounted on a joint that allows them to tilt and rotate freely. Second, the tips of the fingers are covered in soft pads filled with tiny particles, similar to ground coffee, that can change from soft to hard depending on how much they are squeezed.

The need for such a device became clear when the researchers looked at the items we use every day. They measured the angles of common kitchenware, such as measuring cups, mugs, and sauce bottles, and found that many have sloped sides. When a standard, rigid robotic finger touches a sloped surface, it often makes contact at just a single point or a thin line. In a wet environment, this small contact area is a recipe for failure. As soon as the robot tries to lift the object, the water reduces the friction, and the object begins to roll or slide sideways, causing the robot to drop it. The researchers found that simply making the fingers softer was not enough, because a soft finger that cannot change its overall angle still presents a fixed orientation to a tilted object, leaving the torque that causes the slip unaddressed.

To solve this, the team built a gripper where the fingers can pivot. When the hand closes on a tilted cup, the joint allows the finger to rotate until its surface is flat against the side of the cup. This simple movement transforms a precarious, unstable point of contact into a broad, stable surface contact. By aligning the finger with the slope, the hand eliminates the sideways force that would otherwise push the object out of its grasp. This global self-alignment happens automatically, without any computer telling the joint where to move. It is a purely mechanical response to the shape of the object.

Once the finger has aligned itself, the second mechanism takes over. The soft pads on the fingertips are filled with small particles. When the gripper first touches the object, these pads are soft and flexible, allowing them to mold perfectly to the shape of the cup or bowl, increasing the area of contact. As the robot squeezes tighter, the particles inside the pad are forced together until they lock into place, a process known as jamming. This transition turns the soft pad into a rigid, load-bearing structure that resists sliding. The researchers also added a special texture to the surface of these pads: a series of grooves. These channels help squeeze out the thin layer of water that often gets trapped between the hand and the object, allowing the pad to make direct contact with the surface and regain friction.

The team tested their invention by pulling various objects out of a water tank to see how much force was required to make them slip. They compared their new design against standard rigid hands and hands with fixed joints. The results were clear. When gripping a tilted object underwater, the new gripper held on with significantly more strength than the rigid alternatives. In one set of tests, the adaptive hand held objects with an average force of 9.75 newtons, while a fixed-joint version managed only 7.71 newtons. The improvement was even more dramatic when comparing the particle-filled pads to a solid, non-jamming pad. The jamming pads held objects with nearly double the force of the rigid ones, proving that the ability to lock the grip in place was crucial for underwater stability.

The experiments also revealed how different shapes affected the grip. For objects with steep slopes, the ability of the finger to rotate was the most important factor, as it allowed the hand to find a stable angle. For straight, cylindrical objects, the grooved texture on the pads provided the best grip, likely because the grooves helped interlock with the surface. The researchers tested a wide variety of items, from plastic cups and metal cans to plates and bowls. In a demonstration of the gripper's capabilities, the robot successfully picked up and moved these items underwater, even when they were floating or buoyant. While some lightweight objects shifted slightly due to the water currents and their own tendency to float, none of them were dropped. The gripper maintained a secure hold on every item, from a small plastic cup to a large frying pan.

The success of this design lies in its simplicity and its ability to work without external power sources for the adaptation itself. The hand does not need a vacuum pump or a complex network of tubes to change its stiffness; the squeezing motion of the motor is enough to trigger the jamming effect. The entire system is sealed to be waterproof, including the camera that helps the robot see what it is holding. This integration of a self-aligning joint and a variable-stiffness pad creates a robust solution for a problem that has long hindered underwater robotics. By addressing both the angle of the object and the slippery nature of the surface, the gripper offers a practical path forward for robots that need to operate in the wet, messy reality of human kitchens. The work suggests that for robots to truly live alongside us, they must be able to adapt to the world as it is, not just the world as it is in a dry laboratory.