In the not-to-distant future, a simple flick of the wrist will send your DJI Mavic Pro drone on its next aerial mission. A wave to the left, you’ll be able to command your drone to dart left. Clenching your hand into a fist – your Mavic drone will return back to its home base. This gesture-based, remote-less flight control is on the brink of becoming mainstream technology – all thanks to the research team at Empa Technology.


Empa’s Piezo-Resistive Fibers

According to the company’s most recent press release, Empa designed “a sensor made of piezo-resistive fibers” and when integrated with a wristband “measures wrist movements converting them into electrical signals.” Leading the team is head researcher Frank Clemens from the Laboratory of High-Performance Ceramics. The piezo-resistive fiber is a rather groundbreaking discovery and if used alongside other sensors, will revolutionize the way we interact with robots, drones, and other machines.

Modern sensors (cameras, accelerometers, gyroscopes) are primarily used to propel robots forward/backward, rotate them on command, and help them avoid obstacles. As we all know, they’re fairly limited with recognizing natural movements even though facial recognition and machine learning technology relies on them heavily. With that being said, motion sensors are confined to a particular range speed making them difficult to control with quick hand gestures or sudden body movements. With that being said, the combination of both motion and Empa sensor is a perfect team making remote-less control over devices possible.

“It takes a combination of different sensors to develop new concepts,” Clemens explains. “Only then can we spot and use movements that weren’t detectable with previous technologies.”

Gesture-Based Sensor Technology

Responding to the simplest of movements (like pointing of a finger), the Empa sensor is electroconductive – capable of recognizing the change in shapes for the sake of converting them into electrical signals. This is the magic surrounding Empa’s revolutionary fabric and when integrated into a smartwatch band, you now have a command center strapped to your wrist.

Still, in its development phase, Empa researchers attached its piezo-resistive fibers to a smartwatch. Unfortunately, there were some performance issues with their first prototype so team leaders Frank Clemens and Mark Meinykowycz came up with a different approach: “With the aid of additive manufacturing, we managed to integrate the sensor structure in non-textile materials,” Clemens said.

Additive Manufacturing

Turning to additive manufacturing and with the help from partners STBL Medical Research AG and Idezo, Empa’s team had a breakthrough. Using an algorithm that spoke to both motion and Empa sensors, researchers were able to program the technology to control drone flight with hand gestures. It’s basically a program that speaks to both the piezo-resistive fibers and the drone – making gesture-based operation a seamless experience.

With the algorithm still a work-in-progress and the goal centered around making this a fully functional technology, the team at Empa has passed their discovery onto the Bern University of Applied Sciences. Part of their bachelor’s program, students at the University will be using Empa’s fiber technology to build a series of gesture-based movement sequences. One ETH Zurich student has integrated the piezo-resistive sensor inside plaster – abandoning the wristwatch altogether.

Real World Application

With Empa’s fibers in the capable hands of students and researchers, it’s only a matter of time before we see the real-world application of their technology. “Together with our industry partner STBL Medical Research AG, we are currently discussing a potential industry implementation with partners from various sectors,” says Clemens. By simply strapping on a smartwatch and pointing a finger, we’ll soon be able to operate some of the world’s most innovative devices – whether that be a surveillance drone, a digital home assistant, or our computer displays.

Source: Empa Technology