On November 26, 2018, Stanford engineers developed an electronic glove containing a sensor, and it is expected to give the robotic arm the flexibility that humans consider to be ideal in the future.
In a paper published in Science Robotics, academician of the National Academy of Engineering and professor of chemical engineering at Stanford University, Bao Zhennan and his team demonstrated that the sensor can allow the robot arm to hold and not squirt delicate and fragile berries and ping pong ball. Bao Zhenan said that the sensor at the fingertips of the glove can simultaneously simulate two key properties of human flexibility: the strength and direction of the pressure.
She said: "This technology has helped us move the way to the robot to bring the sensory capabilities found in human skin one day in the future."
Researchers are perfecting technology to support their automatic control of sensors. But if it can be achieved, the sensor wearing the glove will be able to flexibly hold the egg with the thumb and forefinger without crushing or falling off.
This electronic hand is designed to simulate the human skin layer as a way of working together to give us extraordinary flexibility. The outer skin is filled with “sensors” (sensory nerves) that detect pressure, heat and other stimuli.
The “touch sensors” of the fingers and palms are especially rich. This "sensor" works with a skin sub-layer called the spinous layer.
The spinous layer in front of the basal layer of the skin is critical. When our fingers touch an item, the outer layer of the skin will be closer to the spine. The sensory nerve at the top of the spinous layer will be able to perceive a slight touch. More intense stress will force the outer layer of the skin closer to the spinous layer, triggering more sensory nerves, triggering a stronger touch.
But measuring pressure strength is only part of the function of the spinous layer. This uneven skin sublayer also helps to illustrate the direction of pressure or shear. For example, when a finger pushes north, the southern slope of the “hill” of the spinous layer will produce a strong signal.
For us humans to hold an egg gently and firmly with the thumb and forefinger, this shearing power has played a significant role. Clementine Boutry and Marc Negre led the development of electronic sensors that mimicked this human mechanism. Each sensor on the fingertip of the robot glove consists of three flexible layers. The top and bottom layers are electrically active, with rubber insulation in the middle, which work together.
The researchers laid a grid of wires on each of the opposing surfaces and rotated them perpendicular to each other to create a dense array of small sensing pixels. They also make the bottom layer as bumpy as the human skin.
The intermediate rubber insulator can simply keep the top and bottom electrodes spaced apart. But this separation is critical because electrodes that are close but not in contact can store electrical energy.
When the robot finger presses down and pushes the upper electrode toward the bottom layer, the stored electrical energy increases. The “hills” and “valleys” on the ground floor provide a way to map pressure intensity and direction to specific points in the vertical grid, just like human skin.
To test the technique, the researchers embed a three-layer sensor on the fingertips of a rubber glove and put it on the robot's hand. Their ultimate goal is to embed the sensor directly into the skin-like covering of the robotic arm.
In one experiment, the research team made the robotic hand wearing electronic gloves gently touch the berries (without pressure). They also use sensors to detect the appropriate shear forces to hold the table tennis (not falling off) and to move and lift the table tennis.
Bao Zhennan pointed out that through proper programming, the robotic hand wearing the touch glove can perform repetitive tasks, such as picking up eggs from the conveyor belt and putting them into the carton.
In addition, this technology can help robotic assisted surgery because precise touch control is critical to surgery. But Bao Zhennan's ultimate goal is to develop an advanced glove that automatically applies the proper force and safely handles the object without prior programming.
She pointed out: "We can program a robotic hand to make it touch the berries and blast. But we have a long way to go before we can touch and detect that it is a berry and let the robot pick it up."