Wearable devices are nowadays a common occurrence in the consumer industry. Typical applications range from listening to music (e.g. Bluetooth headphones, iPod), through health and fitness monitoring (e.g. Fitbit, Apple watch), to devices for virtual reality (e.g. Oculus Rift, HTC Vive, Google cardboard). A common pattern that can be seen from this list is that while there is a good number of devices that target the sense of vision and hearing, there are not many devices that target the sense of touch. While haptic devices are still not common in the consumer industry, there is a rich research community that focuses on their design and application. Many of these devices are still grounded, but some of them are meant as wearable technology: the shift towards wearable haptics has been particularly strong these last few years, in Europe in particular with the WEARHAP project focusing specifically on this technology. This thesis aims to provide a contribution in this context, both for haptic sensing and tactile stimulus delivery. Wearable sensing is addressed with the introduction of ThimbleSense, a system of shells assembled around six-axes force and torque sensors designed to be worn on fingertips during manipulation of objects, providing a complete force measurement for each finger as well as the position of contacts through intrinsic tactile sensing. Furthermore, in order to enable the collection of measurements in a completely wearable manner, I designed a passive hand exoskeleton, named ExoSense, to be used to measure the position of fingertips while ThimbleSense is being worn. A study where ThimbleSense was worn during tasks reproducing daily living activities is also presented, to investigate the effect of tactile impairment induced by the presence of rigid shells. Finally, applications of wearable sensing to robotics are shown, namely a study where ThimbleSense was used to implement a reflex grasp control to prevent slippage and a reactive grasp approach where a soft manipulator grasps an object during a handover task relying on tactile information coming from an Inertial Measurement Units glove. For what concerns the design of wearable haptic interfaces, my contributions are twofold. The first research topic focuses on the design of wearable haptic interfaces for softness rendering. Initially modeling and characterization of the human fingertip during tactile interaction is discussed, with a focus on a tactile flow model for softness discrimination. A finite element model of the fingertip is presented, tested for soundness against a newly developed bench test to evaluate mechanical properties of the fingertip, and compared with existing work on the subject. This model is then used to quantitatively evaluate the validity of the tactile flow hypothesis. Then, a wearable softness rendering device (named W-FYD), designed from the tactile flow model, is described; with this device the user is able to interact with the softness display in an active or passive way, a characteristic that makes it unique with respect to other finger tip wearable haptic displays. The mechanical design of the W-FYD is described and its effectiveness is evaluated in perceptual experiments. Finally, studies done with two wearable haptic feedback devices for assistive robotics are presented, focusing on conveying information on the artificial hand opening in prosthetics. The first device, named HapPro, was developed as part of the work done in this thesis and is based on a rolling contact used to convey proprioceptive feedback on the opening of an artificial hand. The second device, called the Rice Haptic Rocker, is a pre-existing skin stretch device that was developed at Rice University, and was used in a joint work. For both devices effectiveness of their use to convey proprioception with a compliant robotic hand (Pisa/IIT SoftHand) was investigated, both from a quantitative point of view through perceptual experiments and from a qualitative point of view through a Likert-style survey. Preliminary tests with amputees were also performed.

Touch on the go: wearable haptics for sensing and augmented perception

2018

Abstract

Wearable devices are nowadays a common occurrence in the consumer industry. Typical applications range from listening to music (e.g. Bluetooth headphones, iPod), through health and fitness monitoring (e.g. Fitbit, Apple watch), to devices for virtual reality (e.g. Oculus Rift, HTC Vive, Google cardboard). A common pattern that can be seen from this list is that while there is a good number of devices that target the sense of vision and hearing, there are not many devices that target the sense of touch. While haptic devices are still not common in the consumer industry, there is a rich research community that focuses on their design and application. Many of these devices are still grounded, but some of them are meant as wearable technology: the shift towards wearable haptics has been particularly strong these last few years, in Europe in particular with the WEARHAP project focusing specifically on this technology. This thesis aims to provide a contribution in this context, both for haptic sensing and tactile stimulus delivery. Wearable sensing is addressed with the introduction of ThimbleSense, a system of shells assembled around six-axes force and torque sensors designed to be worn on fingertips during manipulation of objects, providing a complete force measurement for each finger as well as the position of contacts through intrinsic tactile sensing. Furthermore, in order to enable the collection of measurements in a completely wearable manner, I designed a passive hand exoskeleton, named ExoSense, to be used to measure the position of fingertips while ThimbleSense is being worn. A study where ThimbleSense was worn during tasks reproducing daily living activities is also presented, to investigate the effect of tactile impairment induced by the presence of rigid shells. Finally, applications of wearable sensing to robotics are shown, namely a study where ThimbleSense was used to implement a reflex grasp control to prevent slippage and a reactive grasp approach where a soft manipulator grasps an object during a handover task relying on tactile information coming from an Inertial Measurement Units glove. For what concerns the design of wearable haptic interfaces, my contributions are twofold. The first research topic focuses on the design of wearable haptic interfaces for softness rendering. Initially modeling and characterization of the human fingertip during tactile interaction is discussed, with a focus on a tactile flow model for softness discrimination. A finite element model of the fingertip is presented, tested for soundness against a newly developed bench test to evaluate mechanical properties of the fingertip, and compared with existing work on the subject. This model is then used to quantitatively evaluate the validity of the tactile flow hypothesis. Then, a wearable softness rendering device (named W-FYD), designed from the tactile flow model, is described; with this device the user is able to interact with the softness display in an active or passive way, a characteristic that makes it unique with respect to other finger tip wearable haptic displays. The mechanical design of the W-FYD is described and its effectiveness is evaluated in perceptual experiments. Finally, studies done with two wearable haptic feedback devices for assistive robotics are presented, focusing on conveying information on the artificial hand opening in prosthetics. The first device, named HapPro, was developed as part of the work done in this thesis and is based on a rolling contact used to convey proprioceptive feedback on the opening of an artificial hand. The second device, called the Rice Haptic Rocker, is a pre-existing skin stretch device that was developed at Rice University, and was used in a joint work. For both devices effectiveness of their use to convey proprioception with a compliant robotic hand (Pisa/IIT SoftHand) was investigated, both from a quantitative point of view through perceptual experiments and from a qualitative point of view through a Likert-style survey. Preliminary tests with amputees were also performed.
5-giu-2018
Italiano
Bicchi, Antonio
Scilingo, Enzo Pasquale
Bianchi, Matteo
Università degli Studi di Pisa
File in questo prodotto:
File Dimensione Formato  
EdoardoBattaglia_PhD_RelazioneAttivit.pdf

Open Access dal 16/06/2021

Tipologia: Altro materiale allegato
Dimensione 3.35 MB
Formato Adobe PDF
3.35 MB Adobe PDF Visualizza/Apri
TouchOnTheGo_EdoardoBattaglia_PhDThesis_Final.pdf

Open Access dal 16/06/2021

Tipologia: Altro materiale allegato
Dimensione 81.93 MB
Formato Adobe PDF
81.93 MB Adobe PDF Visualizza/Apri

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/133395
Il codice NBN di questa tesi è URN:NBN:IT:UNIPI-133395