Search

Automation is a driving force in manufacturing, enabling quality and scalability during production and assembly. In contrast with the automotive industry, civil aerospace automation has traditionally lagged. However, the adoption rate of methodologies embracing automation for manufacturing, assembly, and testing is now accelerating, with new technologies being tested to enable reliable and safe assembly and inspection steps. This paper introduces a semi-automated system for quality control during the final production steps of single-aisle aircraft, namely after the automated assembly of hatrack and sidewall elements in the passengers' area, but before any seating elements are assembled in the environment. Quality control is performed using color and 3d cameras mounted on a custom holonomic mobile robot. The acquired data is processed for identifying geometrical or surface defects by using machine learning based models and 3D processing-based algorithms. The results are provided to an inspector officer using different on-site and off-site validation modalities. The obtained results enable us to affirm that the proposed solution looks very promising for semi-automatic quality control, and it can serve as a foundational framework for efficient manufacturing in the aerospace industry.

Carbon Fiber Reinforced Plastic (CFRP) is going to assume more and more importance in mechanical and aerospace engineering in the near future. This is due to its intrinsic high lightness and resistance, if compared to traditional metallic materials. CFRP is characterized by a proper production and repair technology. The tool showed in this paper is conceived to repair and verify the restoration of such kind of material. Its application is limited to the first part of the repair process: in jargon the scarfing operation, otherwise called “defect removal”. This operation is similar to a high-speed grinding.For this reason, the designed tool can be easily applied to other contexts, such as sanding, polishing or deburring.Basically, the ScarfingTool is composed by a high-speed electro-spindle and a 3D smart snapshot sensor. This tool is conceived to be as light as possible to be installed on a light collaborative robot. Moreover, its specific design assures great compliance that, as needed, can be regulated. A high efficiency suction inlet completes the tool and guarantees a clean environment during operation. The scarfing process has already been dealt with in [1] to which the interested reader is addressed.This work is licensed under CC BY-SA 4.0. (https://creativecommons.org/licenses/by-sa/4.0)

Anomaly detection is a topic of interest in several areas, ranging from Industry 4.0 to Energy Management, Smart Agriculture, Cybersecurity, and Bioinformatics. In a wide sense, detecting anomalies implies finding samples generated within a process that differs from its standard data generation mechanisms. Identifying these samples is extremely important for a variety of reasons, depending on the specific application and scenario, ranging from the minimization of production costs to maintaining the required safety standards. As such, the increasing availability of wide networks of sensors that yield large amounts of data characterizing the processes under observation allowed the large adoption of deep learning techniques, which proved worthy of attention due to their capability of identifying anomalies with large precision, accuracy and reproducibility. Consequently, there is an extensive need to consolidate research results to provide a common framework to understand the topic and ensure a common foundation to establish future research trends. To respond to this need, this work systematically reviews the state of the art of anomaly detection in Industry 4.0, evaluating gaps in the current knowledge and proposing future directions of interest. To pursue this objective, three main dimensions have been considered: the scenario where the anomaly detection methodologies were applied, the sensing equipment used to gather data characterizing the underlying process, and the algorithm employed to properly interpret the phenomena. The study was conducted following the PRISMA protocol, which allowed the identification of a relevant selection of papers by extracting a meaningful dataset of 78 papers of interest. The analysis highlighted the diffusion of autoencoders in several configurations and application scenarios, highlighting their effectiveness and flexibility for anomaly detection.

In this paper, we present a new generation of omnidirectional automated guided vehicles (omniagv) used for transporting materials within a manufacturing factory with the ability to navigate autonomously and intelligently by interacting with the environment, including people and other entities. This robot has to be integrated into the operating environment without significant changes to the current facilities or heavy redefinitions of the logistics processes already running. For this purpose, different vision-based systems and advanced methods in mobile and cognitive robotics are developed and integrated. In this context, vision and perception are key factors. Different developed modules are in charge of supporting the robot during its navigation in the environment. Specifically, the localization module provides information about the robot pose by using visual odometry and wheel odometry systems. The obstacle avoidance module can detect obstacles and recognize some object classes for adaptive navigation. Finally, the tag detection module aids the robot during the picking phase of carts and provides information for global localization. The smart integration of vision and perception is paramount for effectively using the robot in the industrial context. Extensive qualitative and quantitative results prove the capability and effectiveness of the proposed AGV to navigate in the considered industrial environment.

In the last decade, thanks to a widespread diffusion of powerful computing machines, artificial intelligence has been attracting the attention of the academic and industrial worlds. This review aims to understand how the scientific community is approaching the use of deep-learning techniques in a particular industrial sector, the railway. This work is an in-depth analysis related to the last years of the way this new technology can try to provide answers even in a field where the primary requirement is to improve the already very high levels of safety. A strategic and constantly evolving field such as the railway sector could not remain extraneous to the use of this new and promising technology. Deep learning algorithms devoted to the classification, segmentation, and detection of the faults that affect the railway area and the overhead contact system are discussed. The railway sector offers many aspects that can be investigated with these techniques. This work aims to expose the possible applications of deep learning in the railway sector established on the type of recovered information and the type of algorithms to be used accordingly.

Photo-identification is one of the best practices to estimate the abundance of cetaceans and, as such, it can help to obtain the biological information necessary to decision-making and actions to preserve the marine environment and its biodiversity. The Risso's dolphin is one of the least-known cetacean species on a global scale, and the distinctive scars on its dorsal fin proved to be extremely useful to photo-identify single individuals. The main novelty of this paper is the development of a new method based on deep learning, called Neural Network Pool (NNPool), and specifically devoted to the photo-identification of Risso's dolphins. This new method also includes the unique function of recognizing unknown vs known dolphins in large datasets with no interaction by the user. Moreover, the new version of DolFin catalogue, collecting Risso's dolphins data and photos acquired between 2013-2018 in the Northern Ionian Sea (Central-eastern Mediterranean Sea), is presented and used here to carry out the experiments. Results have been validated using a further data set, containing new images of Risso's dolphins from the Northern Ionian Sea and the Azores, acquired in 2019. The performance of the NNPool appears satisfying and increases proportionally to the number of images available, thus highlighting the importance of building large-scale data set for the application at hand.

Assessment is a fundamental part of teaching and learning. With the advent of online learning platforms, the concept of assessment has changed. In the classical teaching methodology, the assessment is performed by an assessor, while in an online learning environment, the assessment can also take place automatically. The main purpose of this paper is to carry out a study on Learning Analytics, focusing in particular on the study and development of methodologies useful for the evaluation of learners. The goal of this work is to define an effective learning model that uses Educational Data to predict the outcome of a learning process. Supervised statistical learning techniques were studied and developed for the analysis of the OULAD benchmark dataset. The evaluation of the learning process of learners was performed by making binary predictions about passing or failing a course and using features related to the learner’s intermediate performance as well as the interactions with the e-learning platform. The Random Forest classification algorithm and other ensemble strategies were used to perform the task. The performance of the models trained on the OULAD dataset was excellent, showing an accuracy of 95% in predicting the students’ learning assessment.

Knowing an accurate passengers attendance estimation on each metro car contributes to the safely coordination and sorting the crowd-passenger in each metro station. In this work we propose a multi-head Convolutional Neural Network (CNN) architecture trained to infer an estimation of passenger attendance in a metro car. The proposed network architecture consists of two main parts: a convolutional backbone, which extracts features over the whole input image, and a multi-head layers able to estimate a density map, needed to predict the number of people within the crowd image. The network performance is first evaluated on publicly available crowd counting datasets, including the ShanghaiTech part_A, ShanghaiTech part_B and UCF_CC_50, and then trained and tested on our dataset acquired in subway cars in Italy. In both cases a comparison is made against the most relevant and latest state of the art crowd counting architectures, showing that our proposed MH-MetroNet architecture outperforms in terms of Mean Absolute Error (MAE) and Mean Square Error (MSE) and passenger-crowd people number prediction.

In this paper we tackle the problem of indoor robot localization by using a vision-based approach. Specifically, we propose a visual odometer able to give back the relative pose of an omnidirectional automatic guided vehicle (AGV) that moves inside an indoor industrial environment. A monocular downward-looking camera having the optical axis nearly perpendicular to the ground floor, is used for collecting floor images. After a preliminary analysis of images aimed at detecting robust point features (keypoints) takes place, specific descriptors associated to the keypoints enable to match the detected points to their consecutive frames. A robust correspondence feature filter based on statistical and geometrical information is devised for rejecting those incorrect matchings, thus delivering better pose estimations. A camera pose compensation is further introduced for ensuring better positioning accuracy. The effectiveness of proposed methodology has been proven through several experiments, in laboratory as well as in an industrial setting. Both quantitative and qualitative evaluations have been made. Outcomes have shown that the method provides a final positioning percentage error of 0.21% on an average distance of 17.2 m. A longer run in an industrial context has provided comparable results (a percentage error of 0.94% after about 80 m). The average relative positioning error is about 3%, which is still in good agreement with current state of the art.

In this paper, an accurate range sensor for the three-dimensional reconstruction of environments is designed and developed. Following the principles of laser profilometry, the device exploits a set of optical transmitters able to project a laser line on the environment. A high-resolution and high-frame-rate camera assisted by a telecentric lens collects the laser light reflected by a parabolic mirror, whose shape is designed ad hoc to achieve a maximum measurement error of 10 mm when the target is placed 3 m away from the laser source. Measurements are derived by means of an analytical model, whose parameters are estimated during a preliminary calibration phase. Geometrical parameters, analytical modeling and image processing steps are validated through several experiments, which indicate the capability of the proposed device to recover the shape of a target with high accuracy. Experimental measurements show Gaussian statistics, having standard deviation of 1.74 mm within the measurable range. Results prove that the presented range sensor is a good candidate for environmental inspections and measurements.

In this article, we tackle the problem of developing a visual framework to allow the autonomous landing of an unmanned aerial vehicle onto a platform using a single camera. Specifically, we propose a vision-based helipad detection algorithm in order to estimate the attitude of a drone on which the camera is fastened with respect to target. Since the algorithm should be simple and quick, we implemented a method based on curvatures in order to detect the heliport marks, that is, the corners of character H. By knowing the size of H mark and the actual location of its corners, we are able to compute the homography matrix containing the relative pose information. The effectiveness of our methodology has been proven through controlled indoor and outdoor experiments. The outcomes have shown that the method provides high accuracies in estimating the distance and the orientation of camera with respect to visual target. Specifically, small errors lower than 1% and 4% have been achieved in the computing of measurements, respectively.

A high-resolution vision system for the inspection of drilling tools is presented. A triangulation-based laser scanner is used to extract a three-dimensional model of the target aimed to the fast detection and characterization of surface defects. The use of two orthogonal calibrated handlings allows the achievement of precisions of the order of few microns in the whole testing volume and the prevention of self-occlusions induced on the undercut surfaces of the tool. Point cloud registration is also derived analytically to increase to strength of the measurement scheme, whereas proper filters are used to delete samples whose quality is below a reference threshold. Experimental tests are performed on calibrated spheres and different-sized tools, proving the capability of the presented setup to entirely reconstruct complex targets with maximum absolute errors between the estimated distances and the corresponding nominal values below 12 μm.