Deterioration Level Estimation Based on Convolutional Neural Network Using Confidence-Aware Attention Mechanism for Infrastructure Inspection

This paper presents deterioration level estimation based on convolutional neural networks using a confidence-aware attention mechanism for infrastructure inspection. Spatial attention mechanisms try to highlight the important regions in feature maps for estimation by using an attention map. The attention mechanism using an effective attention map can improve feature maps. However, the conventional attention mechanisms have a problem as they fail to highlight important regions for estimation when an ineffective attention map is mistakenly used. To solve the above problem, this paper introduces the confidence-aware attention mechanism that reduces the effect of ineffective attention maps by considering the confidence corresponding to the attention map. The confidence is calculated from the entropy of the estimated class probabilities when generating the attention map. Because the proposed method can effectively utilize the attention map by considering the confidence, it can focus more on the important regions in the final estimation. This is the most significant contribution of this paper. The experimental results using images from actual infrastructure inspections confirm the performance improvement of the proposed method in estimating the deterioration level.

Analysis on security-related concerns of unmanned aerial vehicle: attacks, limitations, and recommendations

<abstract> <p>Over time, the use of UAVs (unmanned aerial vehicles)/drones has increased across several civil and military application domains. Such domains include real-time monitoring, remote sensing, wireless coverage in disaster areas, search and rescue, product delivery, surveillance, security, agriculture, civil infrastructure inspection, and the like. This rapid growth is opening doors to numerous opportunities and conveniences in everyday life. On the other hand, security and privacy concerns for unmanned aerial vehicles/drones are progressively increasing. With limited standardization and regulation of unmanned aerial vehicles/drones, security and privacy concerns are growing. This paper presents a brief analysis of unmanned aerial vehicle's/drones security and privacy-related concerns. The paper also presents countermeasures and recommendations to address such concerns. While laying out a brief survey of unmanned aerial vehicles/drones, the paper also provides readers with up-to-date information on existing regulations, classification, architecture, and communication methods. It also discusses application areas, vulnerabilities, existing countermeasures against different attacks, and related limitations. In the end, the paper concludes with a discussion on open research areas and recommendations on how the security and privacy of unmanned aerial vehicles can be improved.</p> </abstract>

Integration of Micro-Structured Photovoltaic Cells into the Ultra-Light Wing Structure for Extended Range Unmanned Aerial Vehicles

There have been large developments in the unmanned aerial vehicles (UAV) industry over the last decade. Although UAV development was mainly for military related use in the beginning and despite there being fear surrounding the release of this technology to the open market for quite a long time, nowadays, there are a variety of applications where UAVs are used extensively, such as in agriculture, infrastructure inspection and monitoring, mobile retranslation relays for communications, etc. One of the weaknesses of electrically propelled UAVs is flight autonomy; there is often a difficult trade-of between the weight of the payload, batteries, and surface to be surveyed that is necessary to determine. There have been many attempts to use photovoltaic cells to increase the flight time for UAVs; however, a reliable solution has not yet been developed. The present paper presents improvements that have been conducted to extend the autonomy of electrically derived UAVs: instead of gluing photovoltaic cells on the wings, the new approach embeds the solar cells into the wing structure as well as develops a new wing that is significantly lighter to compensate for the weight added by the photovoltaic cells. It was demonstrated that by using this approach, a 33% increase in the flight time can be achieved with only one modified wing in a prototype vehicle.

Evaluating the effectiveness of drones in emergency situations: a hybrid multi-criteria approach

PurposeThe paper aims to build a customized hybrid multi-criteria model to identify the top three utilities of drones at both personal and community levels for two use cases: firefighting in high-rise buildings and logistic support.Design/methodology/approachA hybrid multi-criterion model that integrates fuzzy analytical hierarchy process (AHP), Best Worst, fuzzy analytical network process (ANP), fuzzy Decision-Making Trial and Evaluation Laboratory (DEMATEL) is used to compute the criteria weights. The weights are validated by a novel ensemble ranking technique further whetted by experts at the community and personal levels to two use cases.FindingsDrones' fire handling and disaster recovery utilities are the most important to fight fire in high-rise buildings at both personal and community levels. Similarly, drones' urban planning, municipal works and infrastructure inspection utilities are the most important for providing logistics support at personal and community levels.Originality/valueThe paper presents a novel multi-criteria approach, i.e. ensemble ranking, by combining the criteria ranking of individual methods – fuzzy AHP, Best-Worst, fuzzy ANP and fuzzy DEMATEL – in the ratio of optimal weights to each technique to generate the consolidated ranking. Domain experts also validate this ranking for robustness. This paper demonstrates a viable methodology to quantify the utilities of drones and their capabilities. The proposed model can be recalibrated for different use case scenarios of drones.

Localization System for Lightweight Unmanned Aerial Vehicles in Inspection Tasks

This paper presents a localization system for Unmanned Aerial Vehicles (UAVs) especially designed to be used in infrastructure inspection, where the UAVs have to fly in challenging conditions, such as relatively high altitude (e.g., 15 m), eventually with poor or absent GNSS (Global Navigation Satellite System) signal reception, or the need for a BVLOS (Beyond Visual Line of Sight) operation in some periods. In addition, these infrastructure inspection applications impose the following requirements for the localization system: defect traceability, accuracy, reliability, and fault tolerance. Our system proposes a lightweight solution combining multiple stereo cameras with a robotic total station to comply with these requirements, providing full-state estimation (i.e., position, orientation, and linear and angular velocities) in a fixed and time-persistent reference frame. Moreover, the system can align and fuse all sensor measurements in real-time at high frequency. We have integrated this localization system in our aerial platform, and we have tested its performance for inspection in a real-world viaduct scenario, where the UAV has to operate with poor or absent GNSS signal at high altitude.

Drone Deep Reinforcement Learning: A Review

Unmanned Aerial Vehicles (UAVs) are increasingly being used in many challenging and diversified applications. These applications belong to the civilian and the military fields. To name a few; infrastructure inspection, traffic patrolling, remote sensing, mapping, surveillance, rescuing humans and animals, environment monitoring, and Intelligence, Surveillance, Target Acquisition, and Reconnaissance (ISTAR) operations. However, the use of UAVs in these applications needs a substantial level of autonomy. In other words, UAVs should have the ability to accomplish planned missions in unexpected situations without requiring human intervention. To ensure this level of autonomy, many artificial intelligence algorithms were designed. These algorithms targeted the guidance, navigation, and control (GNC) of UAVs. In this paper, we described the state of the art of one subset of these algorithms: the deep reinforcement learning (DRL) techniques. We made a detailed description of them, and we deduced the current limitations in this area. We noted that most of these DRL methods were designed to ensure stable and smooth UAV navigation by training computer-simulated environments. We realized that further research efforts are needed to address the challenges that restrain their deployment in real-life scenarios.