Random outbreaks of infectious diseases in the past have left a persistent impact on societies. Currently, COVID-19 is spreading worldwide and consequently risking human lives. In this regard, maintaining physical distance has turned into an essential precautionary measure to curb the spread of the virus. In this paper, we propose an autonomous monitoring system that is able to enforce physical distancing rules in large areas round the clock without human intervention. We present a novel system to automatically detect groups of individuals who do not comply with physical distancing constraints, i.e., maintaining a distance of 1 m, by tracking them within large areas to re-identify them in case of repetitive non-compliance and enforcing physical distancing. We used a distributed network of multiple CCTV cameras mounted to the walls of buildings for the detection, tracking and re-identification of non-compliant groups. Furthermore, we used multiple self-docking autonomous robots with collision-free navigation to enforce physical distancing constraints by sending alert messages to those persons who are not adhering to physical distancing constraints. We conducted 28 experiments that included 15 participants in different scenarios to evaluate and highlight the performance and significance of the present system. The presented system is capable of re-identifying repetitive violations of physical distancing constraints by a non-compliant group, with high accuracy in terms of detection, tracking and localization through a set of coordinated CCTV cameras. Autonomous robots in the present system are capable of attending to non-compliant groups in multiple regions of a large area and encouraging them to comply with the constraints.

Industry 4.0 has revolutionized the use of physical and digital systems while playing a vital role in the digitalization of maintenance plans for physical assets in an optimal way. Road network conditions and timely maintenance plans are essential in the predictive maintenance (PdM) of a road. We developed a PdM-based approach that uses pre-trained deep learning models to recognize and detect the road crack types effectively and efficiently. We, in this work, explore the use of deep neural networks to classify roads based on the amount of deterioration. This is done by training the network to identify various types of cracks, corrugation, upheaval, potholes, and other types of road damage. Based on the amount and severity of the damage, we can determine the degradation percentage and have a PdM framework where we can identify the intensity of damage occurrence and, thus, prioritize the maintenance decisions. The inspection authorities and stakeholders can make maintenance decisions for certain types of damages using our deep learning-based road predictive maintenance framework. We evaluated our approach using precision, recall, F1-score, intersection-over-union, structural similarity index, and mean average precision measures, and found that our proposed framework achieved significant performance.

No abstract