Pressure Mapping Using Nanocomposite-Enhanced Foam and Machine Learning

Abstract

Pressure mapping has garnered considerable interest in the healthcare and robotic industries. Low-cost and large-area compliant devices, as well as fast and effective computational algorithms, have been proposed in the last few years to facilitate distributed pressure sensing. One approach is to use electrical impedance tomography (EIT) to reconstruct the contact pressure distribution of piezoresistive materials. While tremendous success has been demonstrated, conventional algorithms may be unsuitable for real-time monitoring due to its computational demand and runtime. Moreover, the low resolution of reconstructed images is a well-known issue related to the regularization strategies typically employed for traditional EIT methods. Therefore, in this study, two different supervised machine learning (ML) approaches, namely, radial basis function networks and deep neural networks, were employed to efficiently solve the inverse EIT problem and improve the resolution of reconstructed pressure maps. The demonstration of high-resolution pressure mapping, specifically, for identifying pressure hotspots, was achieved using a carbon nanotube-based thin film integrated with foam.

Crack identification using smart paint and machine learning

“…Then, two different synthetic datasets were generated according the procedure explained in Section “Preparation of the training dataset.” The first dataset was built by considering single and multiple polygonal defects, with random shapes and positions in the region of interest, where conductivity was reduced by a random factor (0%–100%). More details about data generation for polygonal defects can be found in Quqa et al 55 The second dataset was generated considering randomly distributed line-shaped regions with zero conductivity in the sensing surface. Specifically, a crack was modeled as a random number of connected segments (between 1 and 10) with random length (max 2 cm) and inclination.…”

“…EIT has also shown to be effective in detecting strain states. 55 Embedded cracks near the free surface (e.g., delamination) or small cracks due to a tensile state may not involve disconnections in the sensing film. However, in general, they generate a variation in the surface strain, which changes the conductivity of the sensing film.…”

Crack identification using electrical impedance tomography and transfer learning

“…This strategy was proposed in previous studies (Quqa et al., 2022; Quqa et al, 2023) and will be used here to represent the reference performance of a DNN‐based method to solve the inverse ERT problem. This DNN will be referred to as “preliminary” DNN hereafter, as it is trained on synthetic data alone.…”

“…Previous studies showed that, in some cases, synthetic data alone could be enough to train neural networks for the identification of conductivity changes due to damage or strain variations when the sensing body is originally undamaged (Quqa et al., 2022). Nevertheless, while early damage might be identified with good accuracy (Quqa et al, 2023), the performance of neural networks typically degrades with damage severity (L. Chen, Gallet, et al., 2022; Hallaji et al., 2014; Seppänen et al., 2017).…”

Resolution enhancement of tongue tactile image based on deconvolution neural network

“…Besides, DL also has good migration ability and can be used to solve the problem of improving the resolution of complex tongue tactile imaging. Deep neural networks, such as deconvolution neural networks (DNNs), have shown great potential in improving the resolution of tactile perception imaging (Funabashi et al, 2022; Huang et al, 2020; Kim et al, 2021; Quqa et al, 2022). It can efficiently transform the distributed information of sparse original pressure contacts into high‐resolution tactile image close to tongue tactile perception.…”