Researchers at Columbia University have developed a robotic skeleton equipped with 32 photodiodes and 30 adjacent LEDs and is covered by a squishy skin of reflective silicone, which keeps the device's own light in and outside light out.
When the robot finger touches an object, the soft exterior deforms, and the photodiodes detect changing light levels from the LEDs.
The system can determine where contact is being made with the finger, and the intensity of that contact.
The 32 photodiodes and the 30 LEDs produce 960 signals, a massive amount of data from a single poke.
The system relies on machine learning to analyze all of the information.
This type of tactile sensing can facilitate robot manipulation, and this new system is a significant improvement over previous robotic fingers that used electrodes overlaid with rubber to sense touch.
From Wired
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA
No entries found