stub Air-Powered Computer Memory Can Control Soft Robotic Movements - Unite.AI
Connect with us

Robotics

Air-Powered Computer Memory Can Control Soft Robotic Movements

Published

 on

Image: William Grover, UC Riverside

Engineers at the University of California – Riverside have developed an air-powered computer memory that can help control the movements of soft robots. The new system addresses one of the biggest challenges in soft robotics, which is the mismatch between pneumatics and electronics.

The research was published in PLOS One.

Pressurized air is used to move the soft, rubbery limbs and grippers of pneumatic soft robots, which are better at performing delicate tasks than traditional rigid robots. At the same time, they are also safer. 

The existing systems used to control pneumatic soft robots rely on electronic valves and computers to maintain the position of the robot’s moving parts. However, electronic parts increase the cost, size, and power demands of such robots.

The team was led by bioengineering doctoral student Shane Hoan, bioengineering professor William Grover, computer science professor Phillip Brisk, and mechanical engineer professor Konstantinos Karydis. The team took inspiration from the past to advance the field of soft robotics. 

“Pneumatic logic” was once used to control various different products like thermostats and climate control systems in the early 1990s. It relies on air instead of electricity, and the air flows through circuits or channels. Air pressure is also used to represent on/off or true/false, and in modern computers, this is represented by 1 and 0 in code, which then can control electrical charges.

Pneumatic soft robots are required to remember and maintain the positions of their moving parts, so the researchers set out to develop pneumatic logic “memory,” which could eliminate the need for the electronic memory currently used. 

Robot uses air-powered RAM to play piano

 

Creating the Pneumatic Random-Access Memory

The team created their pneumatic random-access memory, or RAM, chip by using microfluidic valves rather than electronic transistors. Originally designed to control the flow of liquids on microfluidic chips, the microfluidic valves are also able to control the air flow. The valves are sealed against a pressure differential, and this is true even when they are disconnected from an air supply line. This system created trapped pressure differentials, which act as memories and can maintain the states of a robot’s actuators. 

By relying on these dense arrays of valves, the robotic movements can perform advanced operations. At the same time, they reduce the need for expensive, bulky, and power-consuming electronic hardware. 

The team first modified the microfluidic valves so that they could handle larger air flow rates. They then produced an 8-bit pneumatic RAM chip that was capable of controlling larger and faster-moving soft robots before incorporating it into 3D-printed rubber hands. 

The pneumatic RAM relies on atmospheric-pressure air to represent a “0” or FALSE value, and vacuum is used to represent a “1” or TRUE value. When connected to atmospheric pressure, the soft robotic fingers extend, and when connected to vacuum, they contract.

The researchers were able to vary the combinations of atmospheric pressure and vacuum within the channels on the RAM chip to make the robot play notes, chords, and eventually an entire song on a piano. 

According to the team, this system could theoretically be used to operate other robots without the need of electronic hardware. It only requires a battery-powered pump to create a vacuum. 

There is also no risk of accidental overpressurization and robotic and control system failure thanks to the absence of positive pressure in the system. This means these robots are far safer around humans, so they could be used as wearable devices. 

 

Alex McFarland is a tech writer who covers the latest developments in artificial intelligence. He has worked with AI startups and publications across the globe.