Modeling and Calibration of Pressure-Sensing Insoles via a New Plenum-Based Chamber
Italo Belli,
Ines Sorrentino,
Simeone Dussoni,
Gianluca Milani,
Lorenzo Rapetti,
Yeshasvi Tirupachuri,
Enrico Valli,
Punith Reddy Vanteddu,
Marco Maggiali,
Daniele Pucci
Affiliations
Italo Belli
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Ines Sorrentino
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Simeone Dussoni
iCub Tech Facility, Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Gianluca Milani
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Lorenzo Rapetti
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Yeshasvi Tirupachuri
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Enrico Valli
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Punith Reddy Vanteddu
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Marco Maggiali
iCub Tech Facility, Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
Daniele Pucci
Artificial and Mechanical Intelligence Research Line, Istituto Italiano di Tecnologia (IIT), Center for Robotics and Intelligent Systems, 16163 Genova, Italy
This paper proposes a novel method to reliably calibrate a pair of sensorized insoles utilizing an array of capacitive tactile pixels (taxels). A new calibration setup is introduced that is scalable and suitable for multiple kinds of wearable sensors and a procedure for the simultaneous calibration of each of the sensors in the insoles is presented. The calibration relies on a two-step optimization algorithm that, firstly, enables determination of a relevant set of mathematical models based on the instantaneous measurement of the taxels alone, and, then, expands these models to include the relevant portion of the time history of the system. By comparing the resulting models with our previous work on the same hardware, we demonstrate the effectiveness of the novel method both in terms of increased ability to cope with the non-linear characteristics of the sensors and increased pressure ranges achieved during the experiments performed.