In the shipping digitalisation process, the peak will be reached with the advent of a wholly autonomous and at the same time safe and reliable ship. Full autonomy could be obtained by two linked Artificial-Intelligence systems representing the ship navigator and the ship engineer that possess sensing and analysis skills, situational awareness, planning, and control capabilities. Many efforts have been made in developing onboard systems; however, the shore facilities are not ready yet to deal with these new technologies. The paper aims to present the innovative technologies and methodologies needed to develop a futuristic Vessel Traffic System. The proposed systems will aim at faultless data acquisition and processing, provide input to decision-making systems, and suggest evasive manoeuvre; to deal with hazards and systems failure without human intervention onboard. The system is composed of three different and interacting layers. The first is an artificially intelligent tool to detect and control autonomous ships, thanks to situation recognition and obstacle avoidance strategies. The second is an orchestration and management platform designed to coordinate the sensing/actuation infrastructure and the AI algorithms’ results made available by multiple ships, mustering edge, and distributed computing techniques to fulfil the specific harsh requirements of the sea environment. The final part is a holistic guidance-navigation-control framework to manage autonomous ships’ navigation in a crowded area. Eventually, a cyber-physical scenario, using both a ship digital-twin and a real model-scale ship, is suggested to test and validate the innovative system without the availability of a full-scale scenario.

The historical buildings of a nation are the tangible signs of its history and culture. Their preservation deserves considerable attention, being of primary importance from a historical, cultural, and economic point of view. Having a scalable and reliable monitoring system plays an important role in the Structural Health Monitoring (SHM): therefore, this paper proposes an Internet of Things (IoT) architecture for a remote monitoring system that is able to integrate, through the Virtual Reality (VR) paradigm, the environmental and mechanical data acquired by a wireless sensor network set on three ancient buildings with the images and context information acquired by an Unmanned Aerial Vehicle (UAV). Moreover, the information provided by the UAV allows to promptly inspect the critical structural damage, such as the patterns of cracks in the structural components of the building being monitored. Our approach opens new scenarios to support SHM activities, because an operator can interact with real-time data retrieved from a Wireless Sensor Network (WSN) by means of the VR environment.

Autonomous vehicles (AVs) generate a massive amount of multi-modal data that once collected and processed through Machine Learning algorithms, enable AI-based services at the Edge. In fact, only a subset of the collected data present informative attributes to be exploited at the Edge. Therefore, extracting such a subset is of utmost importance to limit computation and communication workloads. Doing that in a distributed manner imposes the AVs to cooperate in finding an agreement on which attributes should be sent to the Edge. In this work, we address such a problem by proposing a federated feature selection (FFS) algorithm where the AVs collaborate to filter out, iteratively, the less relevant attributes in a distributed manner, without any exchange of raw data, thought two different components: a Mutual-Information-based feature selection algorithm run by the AVs and a novel aggregation function based on the Bayes theorem executed on the Edge. The FFS algorithm has been tested on two reference datasets: MAV with images and inertial measurements of a monitored vehicle, WESAD with a collection of samples from biophysical sensors to monitor a relative passenger. The numerical results show that the AVs converge to a minimum achievable subset of features with both the datasets, i.e., 24 out of 2166 (99%) in MAV and 4 out of 8 (50%) in WESAD, respectively, preserving the informative content of data.

We are the witnesses of a technological ascent accomplished by the Internet of Things (IoT) in the new era of shared informatization. Smart cities make up one of the major areas of IoT applications. Despite several definitions for this new concept, the aim is to achieve a better use of public resources by increasing the quality of services and decreasing the costs of public services. Some initiatives of smart cities around the world have become reality, such as in [1]. Many smart components compose a smart city, such as transportation, energy, education, and buildings. A smart building includes services offered to the occupants, resources distributed to the city, and, in the case of historical buildings, information provided to municipalities about their state of health.

Emotion recognition has attracted a lot of interest in recent years in various application areas such as healthcare and autonomous driving. Existing approaches to emotion recognition are based on visual, speech, or psychophysiological signals. However, recent studies are looking at multimodal techniques that combine different modalities for emotion recognition. In this work, we address the problem of recognizing the user’s emotion as a driver from unlabeled videos using multimodal techniques. We propose a collaborative training method based on cross-modal distillation, i.e., “FedCMD” (Federated Cross-Modal Distillation). Federated Learning (FL) is an emerging collaborative decentralized learning technique that allows each participant to train their model locally to build a better generalized global model without sharing their data. The main advantage of FL is that only local data is used for training, thus maintaining privacy and providing a secure and efficient emotion recognition system. The local model in FL is trained for each vehicle device with unlabeled video data by using sensor data as a proxy. Specifically, for each local model, we show how driver emotional annotations can be transferred from the sensor domain to the visual domain by using cross-modal distillation. The key idea is based on the observation that a driver’s emotional state indicated by a sensor correlates with facial expressions shown in videos. The proposed “FedCMD” approach is tested on the multimodal dataset “BioVid Emo DB” and achieves state-of-the-art performance. Experimental results show that our approach is robust to non-identically distributed data, achieving 96.67% and 90.83% accuracy in classifying five different emotions with IID (independently and identically distributed) and non-IID data, respectively. Moreover, our model is much more robust to overfitting, resulting in better generalization than the other existing methods.

The upcoming sixth generation (6G) mobile networks require integration between terrestrial mobile networks and non-terrestrial networks (NTN) such as satellites and high altitude platforms (HAPs) to ensure wide and ubiquitous coverage, high connection density, reliable communications and high data rates. The main challenge in this integration is the requirement for line-of-sight (LOS) communication between the user equipment (UE) and the satellite. In this paper, we propose a framework based on actorcritic reinforcement learning and generative models for LOS estimation and traffic scheduling on multiple links connecting a user equipment to multiple satellites in 6G-NTN integrated networks. The agent learns to estimate the LOS probabilities of the available channels and schedules traffic on appropriate links to minimise end-to-end losses with minimal bandwidth. The learning process is modelled as a partially observable Markov decision process (POMDP), since the agent can only observe the state of the channels it has just accessed. As a result, the learning agent requires a longer convergence time compared to the satellite visibility period at a given satellite elevation angle. To counteract this slow convergence, we use generative models to transform a POMDP into a fully observable Markov decision process (FOMDP). We use generative adversarial networks (GANs) and variational autoencoders (VAEs) to generate synthetic channel states of the channels that are not selected by the agent during the learning process, allowing the agent to have complete knowledge of all channels, including those that are not accessed, thus speeding up the learning process. The simulation results show that our framework enables the agent to converge in a short time and transmit with an optimal policy for most of the satellite visibility period, which significantly reduces end-to-end losses and saves bandwidth. We also show that it is possible to train generative models in real time without requiring prior knowledge of the channel models and without slowing down the learning process or affecting the accuracy of the models.