Unmanned Aerial Systems (UASs) have rapidly been integrated into the construction industry over the past few years, and their application is continually growing in this domain. The recent development in UAS regulations and technical capabilities have played a significant role in their popularity and wide deployment in various stages of the construction lifecycle. UASs could be used as a platform to enhance the construction practices in general; however, little is known about how construction professionals are adopting this technology in specific construction practices and the barriers they are facing for their successful implementation. The purpose of this study is to explore the current state-of-practice of UAS integration in construction from the industry professionals’ viewpoint. A comprehensive survey study was conducted in the United States to identify the practical construction UAS application areas, their adopted technologies, as well as the benefits and barriers encountered during their implementation. Responses (n=56) showed that most common UAS applications include progress monitoring, site planning, and site surveying/mapping. Rotary-wing vehicles and visual and thermal cameras were the most used platforms and onboard sensors, respectively. Saving time, improving accessibility to compromised spaces, and reducing cost while accomplishing construction tasks were highly regarded as UAS implementation benefits in construction. Participants also considered flying in various weather conditions, within confined or congested areas, as well as the advanced technical know-how requirements, and the increased liability and legal challenges as barriers to using UASs in construction-related tasks. By understanding UAS adoption in construction, this study provides a roadmap to better identify the industry needs and guide researchers and professionals in investigating application areas and barriers that might have maximum benefits for the construction industry in the United States.

/pdf/trends-benefits-and-barriers-of-unmanned-aerial-systems-in-34yv4d8idz.pdf

Trends, benefits, and barriers of unmanned aerial systems in the construction industry: a survey study in the United States

Recent advancements and developments in artificial intelligence (AI) based approaches have shifted the manufacturing practices towards the fourth industrial revolution, considered as Industry 4.0 practices. A positive impact of AI-based techniques on sustainability can be seen in manufacturing organisations’ at the system, product and process levels. Adopting AI-based strategies in manufacturing improves decision making, productivity and system performance. Despite sustainability and other benefits, the adoption of AI-based approaches in manufacturing organisations is still limited due to employees’ knowledge and digital skills. In the present time, due to the digitalisation of manufacturing activities, intelligent sensors, and supply chain activities, industries are facing challenges with the generation of high volume, different variety and velocity of data. This data can be helpful for manufacturing organisations to enhance their performance and sustainability. However, managing this big data is still a significant challenge due to a lack of knowledge and limited literature. Deep learning (DL) based models can be a suitable choice to provide advanced analytics tools for manufacturing data processing and analysis. However, literature on the DL is still limited in the manufacturing context with its relationship to sustainability. The present study discusses the evolution of DL approaches in manufacturing and different DL-based models. This study also highlights how DL-based approaches can enhance the sustainability performance of industries. In the study, primary research areas, i.e., fault diagnosis, quality management, and predictive maintenance, have been discussed. Finally, a conceptual DL-based framework is proposed for the manufacturing industries to enhance their sustainability performance in manufacturing activities. 

Deep learning for manufacturing sustainability: Models, applications in Industry 4.0 and implications

Machining quality control is a bottleneck operation as human inspectors and expensive equipment is needed in most operations. Automated quality assurance in the manufacturing industry has the potential to replace humans and lower the cost of the machined product. This paper presents the analysis of end-milled machined surfaces backed with experimental and deep learning model investigations. The effects of machining parameters like spindle speed, feed rate (table feed), depth of cut, cutting speed, and machining duration were investigated to find machined surface roughness using Taguchi orthogonal array. Following standard DOE, surface roughness and machined surface image were recorded for each machining experiment and categorized into four classes, viz. fine, smooth, rough, and coarse, based on the roughness value (Ra). The machined surface images were used to develop CNN models for surface roughness class prediction. Further, comparative studies among the five popular optimizers were performed. The results showed that the CNN model with the ‘Rectified Adam’ optimizer performed better amongst the optimizer pool, with the training and test accuracy of 96.30% and 92.91%, respectively. The proposed CNN model features a highly accurate and slim structure, potentially substituting human quality control procedures that employ expensive surface roughness measuring devices. 

/pdf/non-contact-surface-roughness-evaluation-of-milling-surface-2s6daxms.pdf

Non-contact surface roughness evaluation of milling surface using CNN-deep learning models

Comparative Study of Popular Deep Learning Models for Machining Roughness Classification Using Sound and Force Signals

Computer vision systems are successfully used for surface defect inspection. The use of computer vision systems to inspect surface defects in welding is an urgent task, as it allows inspection of defects in the weld and heat-affected zone after the welding cycle. The iRVision 3DL computer vision system can be used for this purpose. The angles at which the sensitivity of the robotic welding inspection system is at its best must be determined to evaluate the quality of the defect detection. When used in combination with industrial robots, vision systems have so far fulfilled only a minimal range of tasks in the process. They are reduced to the tasks of recognizing the object position relative to a given coordinate system of objects or coordinate system with the origin within the object itself. In this work, we investigate the possibility of using the iRVision 3DL vision system to solve the problem of analyzing the quality of welds obtained after welding for the presence of surface defects. The capability of the system to detect surface pores of butt welds on a specially prepared specimen is investigated. The possibility of robotic visual inspection using the iRVision 3DL computer vision system and surface pore detection is established. The most preferred angles of the FANUC robot position with the iRVision 3DL computer vision system mounted on it, at which the pores have pronounced bursts on the laser line, be preferable.

Sensitivity Analysis of Computer Vision System for Visual Control of Butt Joints Weld Defects

The safety of railways, the nation's main transportation network, is currently drawing attention. This is mainly because of recent terrorist attacks aimed at private multipurpose facilities in a nu...

Gijun Park

Papers

Development of a real-time security management system for restricted access areas using computer vision and deep learning

Development of a Surface Roughness Evaluation Method from Light and Shade Composition using Deep Learning