Advance shows promise for ‘prescribing robots’ designed to deliver drugs or help with rescue missions Artificial intelligence and robotics news

A team of UCLA engineers and their colleagues have developed a new design strategy and 3D printing technology to build robots in one step.

A study describing the progress, as well as building and displaying a variety of small robots that walk, maneuver and jump, has been published in Science.

This breakthrough made it possible to manufacture all the mechanical and electronic systems needed to power the robot in one go through a new type of 3D printing process for technically active, multifunctional materials (also called metamaterials). Once 3D-printed, a “meta-bot” will be able to propel, move, sense and make a decision.

Printed metamaterials consist of an internal network of sensory, mobile, and structural elements and can move on their own by following programmed commands. With the internal motion and sensor grid already in place, the only external component needed is a small battery to power the robot.

said the study’s lead author, Xiaoyu (Rayne) Zheng, assistant professor of civil and environmental and mechanical and aerospace engineering at UCSD’s Samueli School of Engineering. “With complex movements, multiple sensing patterns and programmable decision-making capabilities, all tightly integrated, it is like a biological system in which nerves, bones, and tendons work in tandem to carry out controlled movements.”

The team demonstrated integration with an onboard battery and controller for fully autonomous operation of the 3D-printed robots – each the size of a fingernail. According to Zheng, who is also a fellow at UCLA’s California NanoSystems Institute, the methodology could lead to new designs for biomedical robots, such as self-guided endoscopes or small swimming robots, which can emit ultrasound and move close to blood vessels to deliver doses. pharmaceutical. to specific target sites within the body.

These “descriptive robots” can also explore dangerous environments. In a collapsed building, for example, a swarm of these tiny robots armed with built-in sensor parts can quickly get into tight spaces, assess threat levels and aid in rescue efforts by finding people trapped under rubble.

Most robots, regardless of size, are typically built in a series of complex manufacturing steps that integrate limbs, electronics, and active components. The process results in heavier weights, larger sizes and lower output power compared to the robots that can be built using this new method.

The key to an all-in-one method led by UCLA is the design and printing of piezoelectric metamaterials — a class of complex materials that can change shape and move in response to an electric field. where The creation of an electric charge as a result of physical forces.

The use of active materials capable of translating electricity into motion is nothing new. However, these materials generally have limitations in their range of motion and travel distance. They must also be connected to gearbox-type transmission systems in order to achieve the desired movements.

By contrast, the robotic materials developed by the University of California — each about the size of a penny — consist of complex piezoelectric and structural elements designed to bend, bend, warp, rotate, expand or contract very quickly.

The team also introduced a methodology for designing these robotic materials so that users can create their own models and print materials directly into the robot.

“This allows actuation elements to be precisely arranged throughout the robot for fast, complex, extended movements over different types of terrain,” said study lead author Huachen Cui, a UCLA postdoctoral researcher at Zheng Additive Manufacturing and Metamaterials Lab. “Thanks to the bi-directional piezoelectric effect, the robotic materials can also sense the torsions themselves, detect obstructions via echo and ultrasonic emissions, as well as respond to external stimuli via a feedback control loop that determines how the robots move, how fast they move, and what target they are moving to. »

Using this technology, the team built and demonstrated three “descriptive robots” with different capabilities. One robot can move around S-shaped corners and randomly placed obstacles, another can escape in response to the impact of contact, while the third can walk over rough terrain and even make small jumps.

Other authors of the UCSD study are graduate students Desheng Yao, Ryan Hensley, Zhenping Xu, and Houtian Lu. Postdoctoral Researcher Ariel Calderon; Development engineering assistant Zhen Wang. The other authors are Shida Davaria, a research associate at Virginia Tech. Patrick Mercier, Associate Professor of Electrical and Computer Engineering at the University of California, San Diego; and Pablo Tarazzaga, professor of mechanical engineering at Texas A&M University.

The research was supported by a Young Faculty Award and a Director’s Fellowship Award from the US Defense Advanced Research Projects Agency (DARPA), with additional funding from the US Naval Research Office, the Air Force Office of Scientific Research and the National Science Foundation.

Advances include 3D printing technologies previously developed by Zheng and Hensleigh when they were both researchers at Virginia Tech, which holds the patent. The researchers plan to file an additional patent through the UCLA Technology Development Group for the new methodology being developed at UCLA.

Leave a Comment