Introduction to the Special Issue on Artificial Intelligence and Cyber-Physical Systems - Part 2

This is Part 2 of the Special Issue on Artificial Intelligence and Cyber-Physical Systems (CPSs). Following the first part of our Special Issue [1], the second part features nine papers at the intersection of AI and CPS.

Smart Home is an important application of CPSs. A home automation system monitors and controls home attributes such as lighting, climate, entertainment systems, and appliances to provide better comfort, care, and safety while reducing costs. Three articles in this issue push the boundary of research in this domain.

Due to the growing prevalence of noncommunicable diseases in the senior population, a key application in the domain of smart home is the detection of cognitive issues based on sensor data. In “FreeSia: A Cyber-Physical System for Cognitive Assessment through Frequency-Domain Indoor Locomotion Analysis ”, Riboni et al. proposed a novel cyber-physical system for cognitive assessment in smart-homes. They introduced novel feature extraction techniques for frequency-based locomotion features, and adopted state-of-the-art machine learning algorithms for short- and long-term cognitive evaluation. In “Learning from Non-Experts: An Interactive and Adaptive Learning Approach for Appliance Recognition in Smart Homes ”, Codispoti et al. developed a Stream-based Active Learning (SAL) algorithm, called K-Active-Neighbors (KAN), for the problem of household appliance recognition. Differently from previous approaches, KAN jointly learns the user behavior and the appliance signatures. KAN dynamically adjusts the querying strategy to increase accuracy by considering the user availability as well as the quality of the collected signatures. In “ASHRAY: Enhancing Water-usage Comfort in Developing Regions Using Data-driven IoT Retrofits ”, Alizai et al. proposed Ashray, an IoT-inspired, intelligent system to minimize the exposure of water to the elements, thereby maintaining its temperature close to that of the groundwater. Ashray learns the water demand patterns of a household and pumps water into the overhead tank only when necessary. The predictive, machine learning based approach of Ashray improves water thermal comfort and has the potential to reduce energy costs for millions of households in developing countries.

Formal Verification is a powerful tool in proving the correctness of complex CPSs, which traditionally are often developed in a model-based development (MBD) paradigm. When introducing machine learning into CPSs, new challenges arise for formal verification. Three articles in this issue address these new challenges.

In “A Framework for Identification and Validation of Affine Hybrid Automata from Input-Output Traces ”, Johnson et al. proposed a framework for inferring and validating models of deterministic hybrid systems with linear ordinary differential equations (ODEs) from input/output execution traces. The framework contains algorithms for the approximation of continuous dynamics in discrete modes, estimation of transition conditions, and inference of automata mode merging. In “On Modularity in Reactive Control Architectures, with an Application to Formal Verification ”, Biggar et al. proposed a graph-structured control architecture and showed how it generalises some reactive control architectures that are popular in artificial intelligence and robotics, specifically Teleo-Reactive programs (TRs), Decision Trees (DTs), Behavior Trees (BTs) and Generalised Behavior Trees (k-BTs). In “Collaborative Rover-Copter Path Planning and Exploration with Temporal Logic Specifications Based on Bayesian Update Under Uncertain Environments ”, Hashimoto et al. investigated collaborative rover-copter path planning and exploration with temporal logic specifications under uncertain environments. The objective of the rover is to complete a mission expressed by a syntactically co-safe linear temporal logic (scLTL) formula, while the objective of the copter is to actively explore the environment and reduce its uncertainties, aiming at assisting the rover and enhancing the efficiency of the mission completion.

Reliability is a fundamental requirement for complex CPSs. Three articles in this issue aim to improve reliability through low-latency monitoring, out-of-distribution detection, and preventive maintenance.

In “Fog-supported Low Latency Monitoring of System Disruptions in Industry 4.0: A Federated Learning Approach ”, Sahnoun et al. designed a new monitoring tool for system disruption related to the localization of mobile resources. In “Efficient Out-of-Distribution Detection Using Latent Space of β-VAE for Cyber-Physical Systems ”, Ramakrishna et al. tackled the problem that the sampled observations used for training the model may never cover the entire state space of the physical environment, and as a result, the system will likely operate in conditions that do not belong to the training distribution. These conditions that do not belong to training distribution are referred to as Out-of-Distribution (OOD). Detecting OOD conditions at runtime is critical for the safety of CPS. The authors proposed an approach to design and train a single β-Variational Autoencoder detector with a partially disentangled latent space sensitive to variations in image features to detect OOD images and identify the most likely feature(s) responsible for the OOD. In “A Hybrid Deep Learning Framework for Intelligent Predictive Maintenance of Cyber-Physical Systems ”, Sai et al. proposed a practical and effective hybrid deep learning multi-task framework, which integrates the advantages of convolutional neural network (CNN) and long short-term memory (LSTM) neural network, to reflect the relatedness of remaining useful life prediction with health status detection process in the CPS environment. The proposed framework can provide strong support for the health management and maintenance strategy development of complex multi-object systems.

There are many emerging challenges and promising solutions at the intersection of AI and CPS. We hope that this special issue serves as a first step in formalizing these challenges and developing mathematically well-founded solutions. Since this is a very actively researched area, we received a large number of quality submissions. Thus, this special issue consists of two parts. This part 2 concludes our special issue.

Again, we would like to thank the people who contributed to the success of this special issue. We thank all the authors for their valuable contribution, and thank all the reviewers for their constructive feedback to the authors. In particular, we would like to thank the editor-in-chief, Prof. Tei-Wei Kuo and Prof. Chenyang Lu for their tremendous support, and Ms. Gita Delsing for her continuous support in the review and publication process.