Publications News

The VR program is oriented around talk therapy and peer support, two of psychiatry’s traditional frontline defenses against PTSD. Developed by an international team of psychologists in partnership with the VR platform 8agora, each participant—including a therapist, who guides each session—appears as a customizable avatar. In a pilot session which was recorded and posted to YouTube, Balashova and two other Ukrainian refugees currently living in Bucharest discussed some of the more mundane challenges they’ve encountered since relocating to a new country.  Another woman lamented the difficulties of adapting to the transit system in Bucharest. The subject matter of the conversation was easygoing by design. “People with PTSD sometimes don't react well when other people mention a traumatic event that they themselves went through,” says Dr. Cezar Giosan, Associate Professor of Psychology, University of Bucharest, who helped design the virtual peer support group.

Lisa Orlandi, CEO of 8agora, said in a message to VentureBeat that we’ll see an early push of metaverse processing and applications into the cloud.
“If you look at metaverse companies today, the heavy compute requirements and rendering are downloaded onto the user device, but this model does not scale well to billions of people across the globe,” Orlandi said. “This will need to be pushed to a multicloud infrastructure (similar to what Amazon or Netflix are doing to stream in the cloud). This also means that data centers will need to ramp up their compute capacity and continue to build out their infrastructure to support the high-speed, high-compute requirements.” But she noted it will be a challenge to do this kind of processing in the cloud in a sustainable way as power consumption for these heavy compute-intensive environments and bandwidth requirements to the user will increase significantly. “Even when you look at Nvidia’s streaming, where they render in the cloud, the user still needs to download an app (it’s not web-based),” Orlandi said.

This paper describes authors’ research in developing collaborative virtual reality classrooms for laboratory and distance learning. The proposed platform was developed to facilitate the implementation of virtual reality labs that can be used by students, locally or by remote access from any device. The platform was developed to facilitate the implementation of virtual reality labs that can be used by students, locally or by remote access from any device (computer, tablet, smart TV or game console). The architecture of the application for MUVE is client server, multiple servers can be used for content, and from the client application installed on the laboratory computers it will be possible to access these servers preferentially depending on the desired applications.

The paper presents a set of deep learning algorithms for detecting vibration anomalies in bearings using multivariate time series on datasets provided by Case Western Reserve University. The study considers a problem of multiclassification of the condition of the bearings depending on the type of defect, but also on the degree of defect, considering only punctual defects in an incipient phase. Once the data sets are correctly labeled and the algorithms are trained on this data, they can accurately predict the type and the size of defect. The model with the best results in the set is RNN - CNN (Recurrent Neural Network with Convolutions) giving an accuracy greater than 97% in all (load) cases.

Rapid prototyping, especially through 3D Printing, for small and medium-sized products, managed to improve the design phase by the quality of the prototype obtained and by the short times of its realization. Recently, virtual immersion tends to replace 3D Printing technology. Virtual Reality Aided Design (VRAD) significantly reduces production times, costs and especially allows the rapid modification of the prototype, which is impossible to achieve in 3D Printing. Our article presents the study of a Compressed Air Vehicle (CAV) prototype participating in the international academic competition PNEUMOBILE; it is a study carried out by the UPBAIR team of UPB through virtual immersion.

Identifying the “health state” of the equipment is the domain of condition monitoring. The paper proposes a study of two models: DNN (Deep Neural Network) and CNN(Convolutional Neural Network) over an existent dataset provided by Case Western Reserve University for analyzing vibrations in fault diagnosis. After the model is trained on the windowed dataset using an optimal learning rate, minimizing the cost function, and is tested by computing the loss, accuracy and precision across the results, the weights are saved, and the models can be tested on other real data. The trained model recognizes raw time series data collected by micro electro-mechanical accelerometer sensors and detects anomalies based on former times series entries.

This paper describes some results of authors' research for developing a Virtual Reality collaborative platform for remote teams able to work in a fully equipped environment that resemble as close as possible a real office. The platform allows real-time interaction between user’s avatars, using high quality surround audio messaging, gestures and physics and includes a large number of tools for collaborative working such as: remote desktop, videoconference, white-board, screen sharing, slide presenter, synchronized video player, web applications, individual and group teleporting, animated 3D model viewer, advanced user management, IoT telemetry data visualization, smart factory digital twin integration. The platform was implemented for few business cases for virtual offices, conferences, eLearning, trade shows and as a virtual reality interface for a digital tween.

This paper describes authors’ research in developing collaborative virtual reality applications as an interface for monitoring big data by creating a digital twin of the factory and sync the movement of virtual machines with the real ones. The platform allows an interactive reading of the sensor telemetry and processes data, maintenance information and access to a large technical library. For data acquisition and reports, a novel image data method was used. The data values that are encoded as pixel colors of images, using different encoding methods for each data type (text, integer, float, Boolean) are also encrypted using an image as a symmetric encryption key and are stored in the cloud in a time base folder structure, assuring a better data compression, security and speed, compared with the existing solutions based on JSON and NoSQL databases. The platform allows the remote access from the VR environment to the machines consoles and allows parametrization and remote commands

Moving to Industry 4.0 involves the collection of massive amounts of data and the development of big data applications that can ensure a quick data flow between different systems, including massive amounts of data and information collection from smart sensors, and sending them to cloud applications that allow real-time data monitoring and processing. Securing and protecting the transmitted data represents a big issue to be discussed and resolved. In the paper, we propose a new method of data encoding and encryption for cloud applications using PNG format images. The proposed method is described in comparison with one of the classical methods of data encoding and transmission used currently. The paper includes a case study in which the proposed method was used to collect and transmit data from an automated waste collection system.

Anomaly detection is a crucial analysis topic in the field of Industry 4.0 data mining as well as knowing what is the probability that a specific machine to go down due to a failure of a component in the next time interval. In this article, we used time series data collected from machines, from both classes - time series data which leads up to the failures of machines as well as data from healthy operational periods of the machine. We used telemetry data, error logs from still operational components, maintenance records comprising historical breakdowns and replacement component to build and compare several different models. The validation of the proposed methods was made by comparing the actual failures in the test data with the predicted component failures over the test data.

The present paper presents a new method for sending data to the cloud from a Client OPC UA Application, using PNG images as a medium. The data values encoded as pixel colors of images, using different encoding methods for each data type are also encrypted using an image as a symmetric encryption key and are stored in the cloud in a time base folder structure. This method of sending and storing the data using an encrypted image format assure a better data compression, security and speed, compared with the actual JSON format which is text based and open. Based on the stored image files can be generated some index image files containing the data values of each property (sensor value) for a chunk of time, in this way reducing the total number of images and increase the compression of data, because of the small allocation space on disk and similarity of data that are better compressed by the LZW compression algorithm of PNG image format.

IOTIA (Internet of Things and Industrial Automation) Platform is a computerized maintenance management system which manages all type of maintenance within a company: breakdown maintenance, preventive maintenance, predictive maintenance and prognostics and health management, aimed at improving maintenance operations, coordinotion of teams and communication between production and maintenance departments. This software application, unlike existing solutions on the market, is intended to facilitate monitoring, prediction and maintenance interventions at the component / subassembly level and not at the machine level, by creating a virtual configurator based on which production lines can be subsequently defined by simply instantiating all the existing machines in the factory, which have been previously defined and configured in this configurator.

The present paper presents a new working instruction system placed on the assembly line, composed by few synchronized video displays for providing interactive procedural instructions and a QR (Quick Response) code web-based application for gathering additional information and publishing comments, requests, and evaluation. This system helps operators memorize and perform operations correctly, respecting the recommended time at each workstation, ultimately increasing productivity and reducing scrapping on the assembly line. Two questionnaire based surveys were conducted in order to evaluate the proposed video working instruction system and to evaluate the work carried out within the workstations. Results indicate that the proposed system is more efficient, increases productivity, produces a lower level of waste and is accessible to users.

Companies, professionals and industry experts are using trade shows to share information and to build long-lasting business relationships. A trade show could have some disadvantages such as: higher price, difficulty to attend for potential partners due to distance, cost and time. An alternative could be the use of virtual reality exhibitions. In this paper, research results for developing a virtual reality exhibition platform are presented. The platform provides collaborative and advertisement tools enabling the possibility to interact and run complex activities remotely in a virtually immersive environment. The authors have developed some tools that extends the High Fidelity platform capabilities by designing, modelling and texturing the virtual environment, consisting in a futuristic city, a large exhibition pavilion and the exhibition booths using Maxon Cinema 4D and Substance Painter for modelling

In the paper, research results for developing a virtual reality broadcasting studio are presented. the proposed platform provides users with a fully automated television studio, including tools like automated camera system, multi video switchers, props and text prompters. The studio allows the productions of video shows exclusively in VR using avatars in a complete virtual immersive space and live broadcasting of live or prerecorded materials to the vr domains or outside to the real world. The platform includes a very good spatial audio mixer in which each user's live voice and ambient sounds included in the platform are amplified or attenuated and directed into the surround channels, depending on the spatial position of the user's avatar in relation to them. The platform also allows the creation of spatial audio areas with different sound attenuations.

This paper describes some results of authors' research for developing a social virtual reality collaborative platform. The platform provides users with a wealth of collaborative tools that make it possible to interact and run complex activities remotely in a virtually immersive environment. From the facilities developed by our team on the virtual platform, we mention 3D animated product layouts and animated 3D CAD model visualization. This tool allows importing and visualization of complex animated CAD models, exported as baked FBX from the 3D designing application. This tool allows an easy way to import and study the models and trigger various animations based on some developed scripts. The platform includes a realistic physics simulation system for avatar movement and 3D tracking of user movement, based on HTC Vive and Oculus Rift trackers or Leap-Motion controller.

Increasing the level of waste electrical and electronic equipment (weee) separate collection represents a priority in the context of high requirements for material resources. Nowadays, weee collecting systems are not fully automated, human operators are still used for weee identification, weighing, sorting, transporting and storing. in order to obtain a full automated weee collecting system, the waste identification part is the hardest to achieve. In this paper, some preliminary research results regarding a software development and implementation for the weee identification, weighing and sorting are presented. The main components of the weee identification system are described, such as: a vision system, a metal components detection system, an integrated electronic weighing scale, a pos printer and a reading interface.

This paper describes some results of authors' research in machine reading at scale as a support for self-learning, which combines the challenges of document retrieval (finding the relevant articles) with that of machine comprehension of text (identifying the answer spans from those articles). Our approach combines a search component based on bigram hashing and TF-IDF (term frequency–inverse document frequency) matching with a multi-layer recurrent neural network model trained to detect answers in Wikipedia paragraphs.

One of the pylons of the Industry 4.0 is augmented and virtual reality. It can improve the perception on the industrial processes in real time and more. The purpose of these research is to obtain a starting point regarding the actual abilities to enlarge the use of VR HMDs assembling into a unique application, Virtual and Augmented Reality with Gesture Interaction. The project relies on the Leap Motion controller. This controller have an integrated infrared depth camera wich can be used as an input video feed for stereoscopic view on Oculus Rift, in the same time, projecting the raw images of your hands into a 3D mesh that can interact with other objects in real-world space (in our case a 3D model of a complex product). The pass through from the twin Leap Motion cameras covers the part of the mesh such that each camera assures a distinct view of the hands, letting you see the actual depth.

This paper describes some results of authors' research in machine vision, as a key technology, essential in connecting the manufacturing processes with the digital twin of the manufacturing architecture. The focus is on designing a system that uses machine vision in combination with deep learning algorithms. The result is a human-like vision system for real-time analysis and interpretation of all video streams of cameras to continuously detect and count the various products on the production line. All detected features and counting information are available in real-time by augmenting on a video screen and streamed to the cloud where they can be easily processed and stored.

This paper introduces an immersive platform for distance learning. It integrates a lot of useful instruments for e-Learning like: browsing web, screen sharing, video players, virtual library, live streaming, webcam, videoconferences, VoIP, chat, whiteboard, automatic checking systems, and interactive testing evaluation systems. Our platform, TIMV Plus, allows the import of 3D objects in FBX and OBJ format. This type of objects can be obtained by using any 3D software or by being imported from Google Warehouse or Google Blocks, so that more 3D components can continuously emerge on the platform, up to each user’s will. Given the fact that this platform uses a server-client architecture, any new modification is saved on server, and the other users can visualize in real time the new components.

Augmented Reality (AR) is a technology widely used in smart manufacturing, which makes data visible in real time, while the user is interacting with the real world. This paper describes a new web based AR application developed by the authors for smart manufacturing architectures, providing portability to any device, free from any existing AR platforms based on reading and interpreting markers placed on different components of industrial machines. The application provides useful real-time information to the operators: technical manuals, operating diagrams, maintenance history, components availability in the warehouse or in supplier’s stock with the possibility of direct order, training videos for maintenance and work safety. Also, being connected to the smart manufacturing software and performing easier data analysis, the application is especially useful for time crucial predictive maintenance.