A Deep-Learned E-Skin Decodes Complex Human Motion

The technology behind artificial intelligence is regularly enhancing and is able to be implemented into a number of industries, including the healthcare sector, auto-trading using Forex robots and even voice search through Siri and Alexa. As there is becoming an increase in demand for a deep-learned e-skin, many technology firms and academic institutions are conducting research surrounding a deep-learned e-skin to see how well it is able to decode complex human motions.

By having a deep-learned e-skin that is able to decode complex human motions, many are hoping that this will be able to enhance the medical/healthcare sector, saving more lives. With AI already benefitting industries, such as forex robots in auto-trading enabling users to save both time and effort, researchers are looking to enhance the ability of artificial intelligence in the future.

How Does It Work?

A deep learning powered single-strained electronic skin sensor is able to capture human motion from a distance. The sensor is placed on a wrist and is able to decode complex five-finger motions in real time, using a virtual 3D hand mirroring original motions. The deep neural network is enhanced by RSL (rapid situation learning), which ensures a stable operation. This is regardless of the position on the sensor on the surface of the skin.

The Research Behind A Deep-Learned E-Skin

The research behind the deep-learned e-skin was led by Professor Sungho Jo, who collaborated with Professor SeunghwanKo to design the new measuring system. This system takes signals that are corresponding to different finger motions that generate cracks in metal nanoparticle films through laser technology. The sensor patch used was attached to the wrist of a user to detect the movement of the fingers.

What Is The Concept Of The Research?

Where did the concept of the research begin? The concept originated from the idea that pinpointing a single area would be efficient for identifying movements through affixing sensors to each joint and muscle. In order for the targeting strategy to work, the research needed to accurately capture signals from a variety of areas where they all converge.

The rapid situation learning (RSL) system is able to collect data from arbitrary parts on the wrist and can automatically train the model in a real-time demonstration with a virtual 3D hand mirroring the original motions. Researchers used laser-induced nanoscale cracking to help to enhance the sensitivity of the sensor used. The sensory system can also work by tracking the motion of the entire body by using a small sensory network and using indirect remote measurement of human motions.

The team behind the research said that they focused primarily on two tasks whilst developing the sensor. They firstly analysed the sensor signal patterns and then mapped the latent vectors to finger motions metric spaces.

The research team have announced that the system is expandable to be used on other body parts, including extracting gait motions from the pelvis. It looks as those this technology is expected to now provide a tiring point in motion tracking, soft robotics and health-monitoring.

Why Is An E-Skin Beneficial?

So, why was this research important and why do we need to improve e-skin technology? An e-skin, also known as an electronic skin, actually has a variety of benefits, including helping those with disabilities. An e-skin has many properties that make it beneficial for a number of sectors, including health. From flexibility to stretching characteristics, antennas, light-emitting diodes and compact power sources, an e-skin also has sensors that are able to measure strain, temperature and stress when applied to different points of a patients’ body. The e-skin is currently in a nascent stage and researchers are intending on developing a synthetic skin that can heal itself, just like natural skin does. Currently, there are four potential health care applications for an e-skin; prosthetics that are able to mimic the sense of touch, monitors for life signs (this includes temperature and respiration rate), drug delivery and wound care. Many major technology firms and academic institutions are currently pursuing each of these applications.

With there being a current rise in diseases such as diabetes, cancer and heart disease, this continues to drive the need for researchers to improve health monitoring. Through the Internet of Things making home health care a possibility, it’s hoping that e-skin will be able to help remotely located clinicians to manage chronic diseases and enable them to give patients more independence to monitor their own wellness and health.

As you can see, the e-skin has got many places in society already and will soon become very sought after technology, especially for the health sector. As research is continued to be conducted by technology firms and academic institutions, it’s exciting to see the future of the e-skin, in particular after recent research identifying that it can decode complex human motion.

Share this post