Aspects related to the maintenance scheduling have become a crucial problem especially in those sectors where the fault of a component can compromise the operation of the entire system, or the life of a human being. Current systems have the ability to warn only when the failure has occurred causing, in the worst case, an offline period that can cost a lot in terms of money, time, and security. Recently, new ways to address the problem have been proposed thanks to the support of machine learning techniques, with the aim to predict the Remaining Useful Life (RUL) of a system by correlating the data coming from a set of sensors attached to several components. In this paper, we present a machine learning approach by using LSTM networks in order to demonstrate that they can be considered a feasible technique to analyze the "history" of a system in order to predict the RUL. Moreover, we propose a technique for the tuning of LSTM networks hyperparameters. In order to train the models, we used a dataset provided by NASA containing a set of sensors measurements of jet engines. Finally, we show the results and make comparisons with other machine learning techniques and models we found in the literature.

In recent years, statistics have confirmed that the number of elderly people is increasing. Aging always has a strong impact on the health of a human being; from a biological of point view, this process usually leads to several types of diseases mainly due to the impairment of the organism. In such a context, healthcare plays an important role in the healing process, trying to address these problems. One of the consequences of aging is the formation of pressure ulcers (PUs), which have a negative impact on the life quality of patients in the hospital, not only from a healthiness perspective but also psychologically. In this sense, e-health proposes several approaches to deal with this problem, however, these are not always very accurate and capable to prevent issues of this kind efficiently. Moreover, the proposed solutions are usually expensive and invasive. In this paper we were able to collect data coming from inertial sensors with the aim, in line with the Human-centric Computing (HC) paradigm, to design and implement a non-invasive system of wearable sensors for the prevention of PUs through deep learning techniques. In particular, using inertial sensors we are able to estimate the positions of the patients, and send an alert signal when he/she remains in the same position for too long a period of time. To train our system we built a dataset by monitoring the positions of a set of patients during their period of hospitalization, and we show here the results, demonstrating the feasibility of this technique and the level of accuracy we were able to reach, comparing our model with other popular machine learning approaches.

Diagnosis of plant health conditions is gaining significant attention in smart agriculture. Timely recognition of early symptoms of a disease can help avoid the spread of epidemics on the plantations. In this regard, most of the existing solutions use some AI techniques on smart edge devices (IoTs or intelligent Cyber Physical Systems), typically equipped with a hardware like sensors and actuators. However, the resource constraints on such devices like energy (power), memory and computation capability, make the execution of complex operations and AI algorithms (neural network models) for disease detection quite challenging. To this end, compression and quantization techniques offer viable solutions to reduce the memory footprint of neural networks while maximizing performance on the constrained devices. In this paper, we realized a real intelligent CPS on top of which we implemented an AI application, called Deep Leaf running on a microcontroller of the STM32 family, to detect coffee plant diseases with the help of a Quantized Convolutional Neural Network (Q-CNN) model. We present a quantitative analysis of Deep Leaf by comparing five different deep learning models: a 32-bit floating point model, a compressed model, and three different types of quantized models exhibiting differences in terms of accuracy, memory utilization, average inference time, and energy consumption. Experimental results show that the proposed Deep Leaf detector is able to correctly classify the plant health condition with an accuracy of 96%, thus demonstrating the feasibility of our approach on a Smart Edge platform.

The industrial Internet of Things technology had a very strong impact on the realization of smart frameworks for detecting anomalous behaviors that could be potentially dangerous to a system. In this regard, most of the existing solutions involve the use of artificial intelligence models running on edge devices, such as intelligent cyber physical systems typically equipped with sensing and actuating capabilities. However, the hardware restrictions of these devices make the implementation of an effective anomaly detection algorithm quite challenging. Considering an industrial scenario, where signals in the form of multivariate time-series should be analyzed to perform a diagnosis, echo state networks (ESNs) are a valid solution to bring the power of neural networks into low complexity models meeting the resource constraints. On the other hand, the use of such a technique has some limitations when applied in unsupervised contexts. In this article, we propose a novel model that combines ESNs and autoencoders (ESN-AE) for the detection of anomalies in industrial systems. Unlike the ESN-AE models presented in the literature, our approach decouples the encoding and decoding steps and allows the optimization of both the processes while performing the dimensionality reduction. Experiments demonstrate that our solution outperforms other machine learning approaches and techniques we found in the literature resulting also in the best tradeoff in terms of memory footprint and inference time.

Recently, even the road industry has been involved in an evolutionary process, inspired by the novel ‘Industry 4.0’. This transition moves the attention towards big-data and Internet of Things concepts, determining novel data flows and relevant management efforts. In smart roads, modern instrumented vehicles and networks of sensors will provide frequent helpful measures (traffic, weather, condition, accidents, mechanical performance, etc.), with reduced efforts and costs. However, this process is currently at the preliminary phases and several issues raise. Since the management and elaboration of huge amounts of data represent a relevant novel issue, in this study, an original web platform for collection and analysis of road performance data is proposed. This platform can acquire and process several data classes and support maintenance activity planning. This paper focuses on pavement maintenance, to provide a reliable decision support tool for road agencies, alternatively feedable by modern survey equipment and, in future, widespread sensors (when effective smart road sensors are installed on the main highways). The platform has been tested, in a preliminary form, on an existing motorway, considering high-performance survey systems data, with interesting and positive results.