New soft robotic gripper improves grasp force by 650%

March 26, 2021

Soft fabric helps improve robotic grip. (AP Photo/Aijaz Rahi)

Robots use conventional grippers made of metal and plastic to grasp and manipulate items and perform human-like tasks, though they are not optimized for holding fragile or irregular objects. In a new study, robotics engineers from Australia enhanced the quality of soft fabric grippers using a powerful adhesive inspired by gecko feet for more versatile, conformable and stronger grasping.

In a paper published March 6 in Sensors and Actuators, researchers explored the development of multifingered fabric-based soft grippers that are affordable, lightweight and can accommodate high gripping loads. The authors wanted to improve upon existing fabric grippers, which lack integrated variable stiffness and controlled adhesion. These grippers also mostly rely on molding elastic silicone rubber materials known as elastomers, which is not easily scalable.

"Robotic grippers are essential end effectors that are commonly used by humans for grasping and manipulating objects. For example, giant robotic arms that are used to lift heavy objects in factories or miniature grippers that are used in medicine and surgery," Thanh Nho Do, an assistant engineering professor at the University of New South Wales and lead author of the paper, said.

Do explained to The Academic Times that adding variable stiffness and controlled adhesion technologies can help improve the performance of fabric grippers in both load capacity and feedback ability. Controlled adhesion increases a grippers' load capacity by manipulating surface forces via electro-adhesion, gecko adhesion or the combination of both.

A variable stiffness actuator is made from a thermoplastic that softens at high temperatures and stiffens at cool temperatures. With a heating helical coil embedded inside the material, this allows the gripper to bend freely when soft but also withstand high loads when stiff. 

"Inspired by the way humans use fabric for clothing, which offer low cost[s] and highly conform to the human skin, we developed fabric-based soft grippers that are lightweight, robust and have high large-force production," he said.

Do said this work aims to contribute to soft robotics that are versatile, physically compliant and human-friendly. They have applications in the medical industry, including procedure settings where soft structures would help to safely interact with soft human tissue, and enable grasp and manipulate items for robotic surgery.

In experimental studies run at the Medical Robotics Lab at the University of New South Wales, Do and his co-researchers demonstrated that adding the variable stiffness filament and controlled adhesive improved the holding force of their soft fabric gripper up to 655% for gripping, which refers to vertical pulling, and 507% for the pull-out configuration, or horizontal pulling.

The physical design of the proposed robotic fabric gripper consists of four main parts: the variable stiffness, adhesive, main gripper and sensor. The team used a gecko-inspired adhesive made from silicone, which mimics the micro-structures on the toes of gecko lizards. This helps enhance the friction between the gripper surface and target objects to hold larger loads, Do said.

The main gripper is structured as a rubber tube inserted inside a fabric conduit, made of both stretchable and non-stretchable fabric, Do explained. When the rubber tube is inflated, the main gripper bends toward the non-stretchable side. 

The researchers also developed a new soft fabric sensor made of liquid metal and carbon particles to monitor the bending angles of the gripper when it bends. It is configured in the shape of a sheet, and can be integrated into the gripper easily, a feature not seen in other existing fabric grippers. 

"The materials employed by this gripper design are commercially available for a reasonable budget, enabling the gripper to be both cost-effective and have potential applications where both gentle grasping and high load capacity are required," the authors said in the paper.

Do emphasized that the fabrication process is also highly scalable, so the grippers could be fabricated with different sizes to fit a range of applications. Their gripper fingers are designed with a flat shape, which differs from other fabric grippers that have a fusiform shape, similar to a pointed oval or lemon that tapers off at both ends. 

"The flat shape provides the gripper fingers with stable surface contact instead of unstable point or edge contact as other fabric grippers," Do said. "[It] also adds more space and ease to incorporate other technologies for enhanced load capacity and feedback ability."

Experiments to test the gripper prototypes used objects such as cylindrical bars, ripened tomatoes and chocolate eggs. The researchers compared the performance of standard grippers without the variable stiffness or gecko adhesion against their design. 

The standard grippers had trouble with the slippery food objects, while the variable stiffness and gecko adhesion-enhanced gripper was able to grip objects that were slippery, larger than its gripping span, and weighed more than 140 times the gripper itself.

Now that the gripper design has proven to be effective, Do said the next step is to equip it with tactile sensors that can provide three-dimensional forces for monitoring contact force between the gripper and target objects. 

"This force data can then be combined with our recently developed soft haptics glove to provide the users and controllers with the feel of touch as if they were touching the objects themselves," Do said. Haptic technology is known as three-dimensional touch, and it creates the sensation of touch by applying forces, vibrations or motions to the user. 

"In addition to soft grippers, our lab is developing the soft skin stretch devices for haptics feedback and artificial muscle for wearable assistive devices and artificial organs," Do continued.

The study, "Soft robotic fabric gripper with gecko adhesion and variable stiffness," published March 6 in the Sensors and Actuators journal, was authored by Thanh Nho Do, Trung Thien Hoang, Jason Jia Sheng Quek, Mai Thanh Thai, Phuoc Thien Phan and Nigel Hamilton Lovell, the University of New South Wales.

We use cookies to improve your experience on our site and to show you relevant advertising.