Robots Get Smart About Approaching Objects Thanks to a New Flexible Skin-Type Sensor
Capable of sensing and distinguishing humans, metals, and plastics from nearly a foot away, the STPS sensor has already been commercialized.
A team of researchers from Sungkyunkwan University, working with Aidin Robotics for commercialization of their design, have created a new skin-like sensor for robotics projects that can sense the approach of objects and materials, including metals, plastics, and humans from nearly 12" away.
"This paper presents a sensor, called a skin-type dual proximity sensor (STPS), that can provide distance information in the form of impedance between a sensor and an approaching object before the collision," the team explains of its creation. "Meanwhile, a new method regarding the selection of the resonant frequency range for optimizing the measured distance is validated. The sensor adopts the combined sensing principle of inductive-capacitive proximity sensing. STPS can be fabricated with various dimensions and shapes, and easily attached to robot surfaces.
"Represented by the series of sensors studied with the dimensions of 100 x 100 x 2.75mm, the proposed sensor can detect an approaching human body up to 300mm away. In this study, sensor system modelling and simulation are conducted, followed by fabrication and signal processing methods. Finally, the performances of the sensor are experimentally validated, thoroughly analysed concerning various parameters, and demonstrated with the collaboration between human and robots."
While the sensor, constructed from coil-like electrodes, is flexible, it does have an impact on its performance: Its ability to measure electrical and magnetic field changes that indicate an object's approach weakens as the sensor's underlying surface curves — though still, Aiden Robotics claims, manages a 94 percent accuracy at a sensing range of up to 7.9" in its latest incarnation.
The team's work has been published in the journal IEEE Transactions on Industrial Electronics under closed-access terms; more information can be found in an interview with Aidin Robotics' Youn Haeng Lee on IEEE Spectrum.