The maintenance and overhaul of rail transit vehicle frames are vital for ensuring the safe and reliable operation of equipment. AR-based maintenance has been proven to be cost-reducing and effective. Few studies explored the application of AR-based maintenance on subway overhaul. In this study, 74 participants were selected to conduct the underground maintenance AR recognition and obstacle avoidance experiment. And two-factor ANOVA, independent samples t-test, and non-parametric test were used for data statistics. The results show that a high-opacity AR interface has better recognition but affects obstacle avoidance. Facing different types of obstacles, AR interfaces with different layouts have significant differences in terms of recognizability and obstacle avoidance effects. Compared with the list layout, the matrix layout AR interface will make more effort for workers. The findings provide useful references for designing the visual information layout of AR interfaces and conducting usability testing of AR interfaces in complex scenarios such as underground maintenance.

Remote maintenance refers to the maintenance tasks performed by off-site personnel using communication devices. The quality of collaboration among off-site personnel has been a focus of research on remote maintenance. With the rapid development of virtual technology, remote collaboration has gradually transformed into a three-dimensional (3D) collaborative virtual environment (CVEs) based on mixed reality technology. 3D CVEs has improved the efficiency of the staff using real-time screen sharing and fusion of virtual and real information. The interaction modes provided by 3D CVEs also have deepened the sense of task participation of off-site personnel. Nevertheless, the addition of new technologies has had a complex impact on trust, communication, cohesion, and task performance among off-site personnel. In this study, we investigated the combined effects of intimacy between collaborators and the avatar interaction mode provided by a 3D CVEs on social presence, interpersonal trust, task performance, and task load during collaboration. The results showed that acquaintances were able to quickly establish good interpersonal trust among themselves and had better task performance even in a CVEs. For strangers, performing tasks through CVEs enhanced their social presence. In addition, the avatar had a more complex effect on the performance of members with different levels of intimacy in 3D CVEs.

Air traffic control (ATC) serves a critical role in the aviation industry and is responsible for the safety and efficiency of aircraft movement. High situation awareness (SA) involves continuously monitoring, understanding, and anticipating the state of the air traffic environment to ensure safe and efficient flight operations, which is critical for air traffic controllers (ATCOs) to prevent accidents. It is a challenge to accurately and promptly identify ATCOs’ situation awareness in a non-intrusive manner. This study aims to integrate learning-based methods with eye-tracking technology to assess the ATCOs’ amount of SA. Eye movement data from 26 participants were collected as they monitored aircraft on a simulated ATC radar screen and responded to freeze-probe queries targeting different levels of SA. Several conventional machine learning and deep learning models are trained using eye movement data and the models’ performance is extensively evaluated. Notably, a hybrid model of a convolutional neural network (CNN) and a long shortterm memory network (LSTM) has optimal performance. The CNN-LSTM model can learn useful features from the temporal sequential eye movement data and achieves an accuracy and F1 score of $\mathbf{9 2. 7 \%}$ and $\mathbf{9 0. 2 \%}$, respectively. The effectiveness and empirical implications of this learning-based method can contribute to future works of developing real-time SA monitoring systems to detect loss of SA in ATCOs, reducing human errors and improving aviation safety.

A common problem with segmentation of medical images using neural networks is the difficulty to obtain a significant number of pixel-level annotated data for training. To address this issue, we proposed a semi-supervised segmentation network based on contrastive learning. In contrast to the previous state-of-the-art, we introduce Min-Max Similarity (MMS), a contrastive learning form of dual-view training by employing classifiers and projectors to build all-negative, and positive and negative feature pairs, respectively, to formulate the learning as solving a MMS problem. The all-negative pairs are used to supervise the networks learning from different views and to capture general features, and the consistency of unlabeled predictions is measured by pixel-wise contrastive loss between positive and negative pairs. To quantitatively and qualitatively evaluate our proposed method, we test it on four public endoscopy surgical tool segmentation datasets and one cochlear implant surgery dataset, which we manually annotated. Results indicate that our proposed method consistently outperforms state-of-the-art semi-supervised and fully supervised segmentation algorithms. And our semi-supervised segmentation algorithm can successfully recognize unknown surgical tools and provide good predictions. Also, our MMS approach could achieve inference speeds of about 40 frames per second (fps) and is suitable to deal with the real-time video segmentation.