BEGIN:VCALENDAR VERSION:2.0 PRODID:-//linuxsoftware.nz//NONSGML Joyous v1.4//EN BEGIN:VEVENT SUMMARY:Skin-Inspired Magnetoresistive-based Tactile Sensor for Force Char acterization in Distributed Areas DTSTART:20241126T090000Z DTEND:20241126T110000Z DTSTAMP:20260811T171439Z UID:84fb7c8e-c818-46b6-9e09-9abd7ca6a3da SEQUENCE:1 CREATED:20241122T152000Z DESCRIPTION: Touch is a crucial sense for advanced organisms\, particularl y humans\, as it provides essential information about shape\, size\, tempe rature\, and texture. In robotics and automation\, the integration of tact ile sensors has become increasingly relevant\, enabling devices to properl y interact with their environment. This study aimed to develop a biomimeti c\, skin-inspired tactile sensor device capable of sensing applied force\, characterizing it in three dimensions\, and determining its point of appl ication.The device was designed as a 4x4 matrix of magnetoresistive sensor s\, wire-bonded to a PCB and encapsulated in epoxy. These sensors detect t he magnetic field from an overlayed magnetorheological elastomer\, compose d of Ecoflex and 5 μm neodymium-iron-boron ferromagnetic particles. Struc tural integrity tests showed that the device could withstand forces above 100 N\, with an epoxy distribution of 0.12 mL per sensor chip. A 3D moveme nt stage equipped with an indenting tip and force sensor was used to colle ct device data\, which was then used to train neural network models to pre dict force parameters. The magnitude-sensing model was trained on 31260 da ta points\, being able to accurately characterize force with a mean absolu te error between 0.07 and 0.17 N. The spatial sensitivity model was traine d on 171008 points and predicted the location of applied pressure with a m ean absolute error of 0.26 mm\, and 0.63 mm for points outside the testing range. LAST-MODIFIED:20241122T152000Z LOCATION:DF Seminar Room (2-8.3)\, 2nd floor of Physics Building URL:http://df.vps.tecnico.ulisboa.pt/en/events/skin-inspired-magnetoresist ive-based-tactile-sensor-for-force-characterization-in-distributed-areas/ X-ALT-DESC;FMTTYPE=text/html:

Touch is a crucial sense for advanced organisms\, particularly humans\, as it provides essen tial information about shape\, size\, temperature\, and texture. In roboti cs and automation\, the integration of tactile sensors has become increasi ngly relevant\, enabling devices to properly interact with their environme nt. This study aimed to develop a biomimetic\, skin-inspired tactile senso r device capable of sensing applied force\, characterizing it in three dim ensions\, and determining its point of application.

The device was designed as a 4x4 matrix of magnetoresistive se nsors\, wire-bonded to a PCB and encapsulated in epoxy. These sensors dete ct the magnetic field from an overlayed magnetorheological elastomer\, com posed of Ecoflex and 5 μm neodymium-iron-boron ferromagnetic particles. S tructural integrity tests showed that the device could withstand forces ab ove 100 N\, with an epoxy distribution of 0.12 mL per sensor chip.

A 3D movement stage equipped with an indenting tip and force sensor was used to collect device data\, which was then use d to train neural network models to predict force parameters. The magnitud e-sensing model was trained on 31260 data points\, being able to accuratel y characterize force with a mean absolute error between 0.07 and 0.17 N. T he spatial sensitivity model was trained on 171008 points and predicted th e location of applied pressure with a mean absolute error of 0.26 mm\, and 0.63 mm for points outside the testing range.

END:VEVENT END:VCALENDAR