Special Track Research Articles

Stability focused evaluation and tuning of special ground vehicle tracking algorithms

This paper deals with a special tracking problem when a ground vehicle should be tracked by a multicopter flying ahead of the vehicle. Pre-designed vehicle route is assumed and the UAV stops or slows down at every intersection to react to route changes. After introducing the problem, the methods applied in a real flight demonstration in the Smart City module of ZalaZONE proving ground are presented. Then new methods are introduced to possibly improve performance. The main focus of the article is the evaluation of the stability of the methods and the provision of tuning guidelines. All of the introduced methods is tuned based-on the guidelines considering real ground vehicle test data and the high fidelity simulation of the applied multicopter. The two best methods are compared in detail and guidelines of their applicability are provided.

Differences in mental health outcomes between adolescent elite athletes and peers not engaged in elite sport

Elite athletes are confronted with various challenges, pressures and stressors. Due to their status as student and athlete, adolescent elite athletes are exposed to a double burden of stress (pressure to be simultaneously successful in school and elite sport), which adds to the general risk for psychological problems in this developmental period. To facilitate the combination of elite sport and education, almost all cantons in Switzerland have established schools, which offer a special track for adolescent elite athletes. We examine (i) whether adolescent elite athletes differ from controls in a series of mental health outcomes, (ii) whether links between perceived stress and mental health symptoms are particularly strong among adolescent elite athletes, and (iii) whether perceived stress and training/competitive load interact to explain variance in mental health. Cross-sectional data from 156 athletes (134 from sport classes, 22 from regular classes) and 150 non-athletic controls (peers not engaged in elite sport) were analysed. All students were recruited from Swiss Olympic Partner schools, and attended grade 10-14. In total, 5–19% of the students reported moderate-to-severe mental health symptoms. With regard to depressive, burnout and insomnia symptoms, no significant differences were found between athletes and controls. A consistent link was observed between higher stress perceptions and more frequent mental health symptoms. However, engaging in elite sport did not seem to be an additional source of distress (or vulnerability factor) among adolescents with moderate-to-high stress levels. High training/competitive load did not amplify the association between perceived stress and mental health symptoms in elite athletes. Currently, no comprehensive framework or model of care exists to support and respond to mental health needs of young elite athletes. Our findings show that it is time to address these challenges to prevent mental health issues and ensure that young athletes can flourish while developing their full potential.

Learning to Design Fair and Private Voting Rules

Voting is used widely to identify a collective decision for a group of agents, based on their preferences. In this paper, we focus on evaluating and designing voting rules that support both the privacy of the voting agents and a notion of fairness over such agents. To do this, we introduce a novel notion of group fairness and adopt the existing notion of local differential privacy. We then evaluate the level of group fairness in several existing voting rules, as well as the trade-offs between fairness and privacy, showing that it is not possible to always obtain maximal economic efficiency with high fairness or high privacy levels. Then, we present both a machine learning and a constrained optimization approach to design new voting rules that are fair while maintaining a high level of economic efficiency. Finally, we empirically examine the effect of adding noise to create local differentially private voting rules and discuss the three-way trade-off between economic efficiency, fairness, and privacy. This paper appears in the special track on AI & Society.

Distribution of Physicians to Public Health Facilities and Factors Contributing to New Medical Graduates Serving in Public Facilities, 2016-2020, Thailand.

In Thailand, the shortage and maldistribution of physicians in the public sector is a critical public health problem. The aims of this study are to describe the retention situation of new physicians and to determine factors associated with new physicians' decision to serve in public facilities. Data of new physicians from the Human Resource Office of the Permanent Secretary System (HROPS) were analysed in order to describe the retention situation of new physicians. A retrospective cohort study was conducted to determine factors associated with new physicians' decision to remain in public service. The data were collected between Oct 2021 and Jan 2022. Cox regression was performed. According to the HROPS database, most new physicians remained in hospitals affiliated to the Office of the Permanent Secretary (OPS) of the Ministry of Public Health (MOPH) (83.5%). Among the physicians resigning from MOPH hospitals, 80.6% resigned due to career factors. From a retrospective cohort study, physicians in the special recruitment track had the lowest incidence rate of resigning. In terms of satisfaction, most physicians were least satisfied with salary levels and workload. The majority of physicians felt satisfied with being close to their families. The Cox regression revealed that physicians graduating from private medical schools or from abroad had a higher risk of leaving public facilities. To lessen the burden of physician shortage and maldistribution, the proportion of students recruited in the special track should be enhanced. High workload and inappropriate income should be addressed. Allocating new physicians to work in their hometown is recommended.

Calibration of Radar RCS Measurement Errors by Observing the Luneburg Lens Onboard the LEO Satellite.

Accurate radar RCS measurements are critical to the feature recognition of spatial targets. A calibration method for radar RCS measurement errors is proposed for the first time in the context of special target tracking by observing the Luneburg Lens onboard the LEO satellite. The Luneburg Lens has favorable RCS scattering properties for the radar microwave. Thus, the laboratory RCS measurements of the Luneburg Lens, with some fixed incident frequency and with different incident orientations for the radar microwave, will be implemented in order to build a database. The incident orientation for the radar microwave in the satellite body frame will be calculated by taking advantage of the precise orbit parameters, with errors only at the magnitude of several centimeters and within the actual satellite attitude parameters. According to the incident orientation, the referenced RCS measurements can be effectively obtained by the bilinear interpolation in the database. The errors of actual RCS measurements can thus be calibrated by comparing the referenced and the actual RCS measurements. In the RCS measurement experiment, which lasts less than 400 s, the actual RCS measurement errors of the Luneburg Lens are nearly less than 0 dBsm, which indicates that the RCS measurement errors of the spatial targets can be effectively calculated by the proposed calibration method. After the elaborated calibration, the RCS measurements of the spatial targets can be accurately obtained by radar tracking.

Highly Retentive, Anti-Interference, and Covert Individual Marking Taggant with Exceptional Skin Penetration.

The development of high‐performance individual marking taggants is of great significance. However, the interaction between taggant and skin is not fully understood, and a standard for marking taggants has yet to be realized. To achieve a highly retentive, anti‐interference, and covert individual marking fluorescent taggant, Mn2+‐doped NaYF4:Yb/Er upconversion nanoparticles (UCNPs), are surface‐functionalized with polyethyleneimine (PEI) to remarkably enhance the interaction between the amino groups and skin, and thus to facilitate the surface adhesion and chemical penetration of the taggant. Electrostatic interaction between PEI600‐UCNPs and skin as well as remarkable penetration inside the epidermis is responsible for excellent taggant retention capability, even while faced with robust washing, vigorous wiping, and rubbing for more than 100 cycles. Good anti‐interference capability and reliable marking performance in real cases are ensured by an intrinsic upconversion characteristic with a distinct red luminescent emission under 980 nm excitation. The present methodology is expected to shed light on the design of high‐performance individual marking taggants from the perspective of the underlying interaction between taggant and skin, and to help advance the use of fluorescent taggants for practical application, such as special character tracking.

Overview of the detection and tracking methods of the lab animals

This article presents an overview of several most common techniques and approaches for object detection and tracking. Today, the tracking task is a very common problem and it can appear in many aspects of our life. One particular case of using object tracking techniques can appear during a lab animal behavior study. Different experimental conditions and the need of certain data collection can require some special tracking techniques. Thus, a set of general approaches to object tracking techniques were considered, and their functionality and possibilities were tested in a real life experiment. In this paper, their basis and main aspects are presented. The experiment has demonstrated the advantages and disadvantages of the studied methods. Considering this, conclusions and recommendations to their usage cases were made.

Negotiating the “Maze”: SEN and the Transition From Lower Secondary Education in Austria

Austrian students with special educational needs (SEN) face many obstacles in the transition from lower to upper secondary education. Using administrative data from national statistics, we analyse the trajectories of these students focusing on two questions: First, what is the impact of the former setting on further pathways for students from special schools compared with mainstream schooling? Second, can low‐threshold training or apprenticeship projects (the “transition system”) compensate for educational disadvantages in former school careers and serve as a “second chance” or do they reinforce exclusionary practices by perpetuating “special tracks”? Regarding the first question, our research findings confirm those from several studies conducted in other German‐speaking countries that show advantages for graduates from mainstream education compared to those from special schools, as they face a lower risk of institutional exclusion. In respect of the second question, at first glance, our findings differ from prior research results. Participating in the transition system is associated with a slight increase in participation in upper secondary education, some increase in employment, and an important reduction concerning inactivity. As revealed by a regression analysis controlling for socio‐demographic characteristics, participation in this system has a distinct integrative influence. We conclude by hypothesising that this is due to the structure of the Austrian transition system offering pathways back to mainstream educational systems and formally recognised educational qualifications.

Measurement Feedback System for Intensive Neurorehabilitation after Severe Acquired Brain Injury

Outcome of acquired brain injury (ABI) and the potential for neurorehabilitation are subject to distinct heterogeneity between patients. Limited knowledge of the complex constellation of determinants at play interferes with the possibility to deploy precision medicine in neurorehabilitation. Measurement Feedback Systems (MFS) structure clinical data collection and deliver the measurement results as feedback to clinicians, thereby facilitating progress monitoring, promoting balanced patient-centered discussion and shared decision making. Accumulation of clinical data in the MFS also enables data-driven precision rehabilitation medicine. This article describes the development and implementation of a MFS for neurorehabilitation after ABI. The MFS consists of specialized measurement tracks which are developed together with representatives of each discipline in the multidisciplinary team. The MFS is built into a digital platform that automatically distributes measurements among clinicians, at predetermined time points during the inpatient treatment, outpatient treatment and follow-up. The results of all measurements are visualized in individual patient dashboards that are accessible for all clinicians involved in treatment. Since step-wise implementation, 124 patients have been registered on the MFS platform so far, providing an average of more than 200 new measurements per week. Currently, more than 15,000 clinical measurements are captured in the MFS. The current overall completion rate of measurements is 86,4%. This study shows that structured clinical assessment and feedback is feasible in the context of neurorehabilitation after severe ABI. The future directions are discussed for MFS data in our Health Intelligence Program, which aims at periodic care evaluation and the transition of neurorehabilitation care towards precision medicine.

Fingertip interactive tracking registration method for AR assembly system

Aiming at the problems of single input mode and lack of naturalness in the assembly process of existing AR systems, a tracking registration method of mobile AR assembly system is proposed based on multi-quantity and multi-degree of freedom natural fingertip interaction. Firstly, the real-time and stable tracking of hand area in complex environment is realized based on the hand region tracking; secondly, the fingertip detection and recognition based on K-COS and parallel vector is used to improve the precision and stability of fingertip recognition; thirdly, the special movement track of fingertip is recognized based on improved DTW algorithm, which has strong compatibility and feature gradient transformation for complex fingertip trajectory recognition; finally, through the real-time transformation of projection relationship between fingertip and virtual object, the interaction between fingertip and virtual object is made more natural and realistic. The experimental results show that in the complex environment of background, illumination, scale and rotation, the precision of fingertip detection and recognition is about 93%, and the precision of fingertip motion template matching is about 91%. The translation error of the registration method based on visual feature recognition is reduced by about 100pix compared with fingertip tracking registration method, and the efficiency of mobile AR-guided assembly method is improved by about 24.77% compared with the traditional manual assisted assembly method. These data verifies the strong interaction and practicability of the fingertips based on the user's multi-quantity and multi-degree of freedom features in the process of mobile AR guided assembly.

Viewpoint: Ethical By Designer - How to Grow Ethical Designers of Artificial Intelligence

Ethical concerns regarding Artificial Intelligence (AI) technology have fueled discussions around the ethics training received by AI designers. We claim that training designers for ethical behaviour, understood as habitual application of ethical principles in any situation, can make a significant difference in the practice of research, development, and application of AI systems. Building on interdisciplinary knowledge and practical experience from computer science, moral psychology and development, and pedagogy, we propose a functional way to provide this training.
 
 
 
 This article appears in the special track on AI & Society.
 
 

Integration, Creation, and Growth: A Path Forward for Sport Social Work Education

Sports fans and the public alike have seen athletes and sport organizations spark conversations focused on equity, mental health, and human rights. Social workers also are professionals that engage in advocacy to destigmatize mental health and to address complex social issues. To align advocacy efforts at the intersection of social work and sport, a small coalition of social workers and interdisciplinary partners convened to create the Alliance for Social Workers in Sports (ASWIS) in 2015. The ASWIS is leading the way in preparing social workers for practice in sport by offering the first Sport Social Work Certificate Program (SSW Certificate). Despite the development of the SSW Certificate, the demand, need, and interest in the sport social work specialization is growing. The purpose of this article is to outline a path forward for sport social work education by discussing strategies to (a) integrate sports into existing coursework; (b) create sport social work courses and field placements; and (c) develop specialization tracks that grow the sport social work workforce and platform for advancing social change through sport. Improving access to education and training can strengthen the specialization of sport social work, and will ultimately benefit athletes, teams, communities, and sport organizations.

Image Captioning as an Assistive Technology: Lessons Learned from VizWiz 2020 Challenge

Image captioning has recently demonstrated impressive progress largely owing to the introduction of neural network algorithms trained on curated dataset like MS-COCO. Often work in this field is motivated by the promise of deployment of captioning systems in practical applications. However, the scarcity of data and contexts in many competition datasets renders the utility of systems trained on these datasets limited as an assistive technology in real-world settings, such as helping visually impaired people navigate and accomplish everyday tasks. This gap motivated the introduction of the novel VizWiz dataset, which consists of images taken by the visually impaired and captions that have useful, task-oriented information. In an attempt to help the machine learning computer vision field realize its promise of producing technologies that have positive social impact, the curators of the VizWiz dataset host several competitions, including one for image captioning. This work details the theory and engineering from our winning submission to the 2020 captioning competition. Our work provides a step towards improved assistive image captioning systems.
 This article appears in the special track on AI & Society.

TO THE PROBLEM OF MODELING THE AERODYNAMICS OF SURFACE TRANSPORTATION VEHICLES

The article deals with the problems of modeling the aerodynamics of surface high-speed vehicles of winged layout moving at a short distance from the water surface. To support them in the air, the screen effect formed under the influence of proximity to the water surface should be used. The effect of ground proximity is observed in air transport in the takeoff and landing modes of aircraft. The proximity of the ground affects the formation of the aerodynamic characteristics of vehicles. Today, the study of the aerodynamics of advanced vehicles moving near the underlying surface is possible using the following approaches: full-scale tests; modeling on special tracks; wind tunnel modeling. Full-scale tests are usually used at the final stages of vehicle development. simulation on special tracks is associated with the complexity of obtaining aerodynamic characteristics and is not widely used. The most suitable is modeling using wind tunnels. The current level of equipment of experimental laboratories allows for correct aerodynamic studies of aircraft. However, when modeling aerodynamic processes around vehicles moving near the interface, complex problems arise with reproducing the conditions of motion dynamics. The paper describes in detail the methods of studying the influence of the proximity of the ground on the formation of aerodynamic characteristics of vehicles in wind tunnels. These include: a method using a fixed plate; a method of mirroring the model; a method of suction or blowing the boundary layer; a method using a moving belt. The possibility of their use for modeling the aerodynamics of high-speed surface vehicles of winged layout is evaluated. When modeling the flow of high-speed vehicles moving near the water surface, a number of difficulties arise due to the realization of real flow conditions, namely, taking into account the dynamics of the water surface. Ideally, relative motion is realized between the vehicle and the water surface. The review of methods used in modeling the aerodynamics of vehicles over a solid screen proved to be of little use for high-speed winged vehicles moving at a short distance from the water surface. Ensuring the similarity of relative motion is an extremely difficult task. It requires a reasonable analysis of physical processes and the search for ways to develop physical and mathematical models suitable for use in the design of high-speed surface vehicles with wing arrangement. These models should range from simple engineering techniques to techniques based on the full and averaged Navier-Stokes equations.

Influence of Nutritional Supplementation and Sports Training on the Physical Fitness of Track and Field Athletes

This study aims to understand the influence of nutritional supplementation and sports training on the physical fitness of track and field athletes. Twenty track and field athletes from Chongqing Normal University were supplemented with nutrition and trained for eight weeks. Water, sugar, and vitamins were supplemented. They were trained three times a week, two hours each time. One hour was for special track and field training, and one hour was for physical training. Before and after the experiment, the body composition, sports quality, and functional movement screening (FMS) of the athletes were tested. Compared with before the experiment, the muscle weight of the athletes increased, the fat-free weight increased (p < 0.05), the sports quality significantly improved, the endurance increased, the speed increased, the flexibility improved, the total score of FMS also increased significantly (p < 0.05). Nutritional supplementation and sports training can improve athletes’ body composition and sports quality and enhance limb flexibility and stability, which has a positive impact on physical fitness. They can be applied in practical training.

Perchlorate occurrence, sub-basin contribution and risk hotspots for drinking water sources in China based on industrial agglomeration method

Perchlorate is a persistent inorganic contaminant which has attracted wide attention because of its harmful effects on physical health. Despite the potential adverse effects to humans via drinking water, related research at the national scale in China are scarce. In this study, the occurrence of perchlorate in major river basins in China was investigated from 2009 to 2020. Generally, except for the Yangtze River Basin, perchlorate concentrations in the surface water of other river basins were low, ranging from < 0.01 to 8.53 μg/L. The results of a specialized field sampling and tracking program in the Yangtze River Basin in 2019 showed that the Xiangjiang River basin is the greatest contributor of perchlorate in the Yangtze River Basin, accounting for 58.63% of the total perchlorate input. Furthermore, based on correlation analysis between production enterprise information and measured concentrations in sampled sites, fireworks and explosives production industries were identified as the major sources of perchlorate contamination in surface water. The risk map showed that the central-southern part of China and the central part of Xinjiang province were risk hotspots for perchlorate contamination. The results gave insights into how to conduct more precise risk assessment and policy intervention towards prevention of perchlorate contamination.

Low Viscosity Epoxy Structural Adhesive Cured at Room Temperature for Crack Repair of Subway Tunnel

Epoxy structural adhesive has good mechanical properties, especially high adhesion, low shrinkage rate, and high stability, widely used in steel plate reinforcement, crack repair, bridge splicing, and concrete bonding, etc. However, due to the high cross-linking density and high internal stress after curing, its shortcomings, such as brittleness, the poor performance of fatigue resistance and heat resistance, limit its application in special track engineering. In this work, two kinds of epoxy structural adhesives (EPA-1, EPA-2) with low viscosity and room temperature curing were synthesized by selecting different bisphenol A epoxy resin and active diluent as component A and modified amine curing agent as component B. The results showed that EPA-1 and EPA-2 presented significant toughness compared with the classic construction adhesive (AralditeXH160). The yield strengths of EPA-1 and EPA-2 were 4.88 MPa and 13.13 MPa, and the shear strengths were 11.93 MPa and 13.08 MPa, respectively, showing good adhesive properties. In addition, the viscosities of EPA-1 and EPA-2 were 127 mPa·s and 308 mPa·s, respectively, and their decomposition temperatures were all above ∼280 °C, which indicated that the self-made epoxy structural adhesives could be used for the repair of cracks in the vibration subway tunnel.

Socially Responsible AI Algorithms: Issues, Purposes, and Challenges

In the current era, people and society have grown increasingly reliant on artificial intelligence (AI) technologies. AI has the potential to drive us towards a future in which all of humanity flourishes. It also comes with substantial risks for oppression and calamity. Discussions about whether we should (re)trust AI have repeatedly emerged in recent years and in many quarters, including industry, academia, healthcare, services, and so on. Technologists and AI researchers have a responsibility to develop trustworthy AI systems. They have responded with great effort to design more responsible AI algorithms. However, existing technical solutions are narrow in scope and have been primarily directed towards algorithms for scoring or classification tasks, with an emphasis on fairness and unwanted bias. To build long-lasting trust between AI and human beings, we argue that the key is to think beyond algorithmic fairness and connect major aspects of AI that potentially cause AI’s indifferent behavior. In this survey, we provide a systematic framework of Socially Responsible AI Algorithms that aims to examine the subjects of AI indifference and the need for socially responsible AI algorithms, define the objectives, and introduce the means by which we may achieve these objectives. We further discuss how to leverage this framework to improve societal well-being through protection, information, and prevention/mitigation.&#x0D; This article appears in the special track on AI &amp; Society.

Agent-Based Markov Modeling for Improved COVID-19 Mitigation Policies

The year 2020 saw the covid-19 virus lead to one of the worst global pandemics in history. As a result, governments around the world have been faced with the challenge of protecting public health while keeping the economy running to the greatest extent possible. Epidemiological models provide insight into the spread of these types of diseases and predict the effects of possible intervention policies. However, to date, even the most data-driven intervention policies rely on heuristics. In this paper, we study how reinforcement learning (RL) and Bayesian inference can be used to optimize mitigation policies that minimize economic impact without overwhelming hospital capacity. Our main contributions are (1) a novel agent-based pandemic simulator which, unlike traditional models, is able to model fine-grained interactions among people at specific locations in a community; (2) an RLbased methodology for optimizing fine-grained mitigation policies within this simulator; and (3) a Hidden Markov Model for predicting infected individuals based on partial observations regarding test results, presence of symptoms, and past physical contacts.&#x0D; This article is part of the special track on AI and COVID-19.

Ethics and Governance of Artificial Intelligence: Evidence from a Survey of Machine Learning Researchers

Machine learning (ML) and artificial intelligence (AI) researchers play an important role in the ethics and governance of AI, including through their work, advocacy, and choice of employment. Nevertheless, this influential group's attitudes are not well understood, undermining our ability to discern consensuses or disagreements between AI/ML researchers. To examine these researchers' views, we conducted a survey of those who published in two top AI/ML conferences (N = 524). We compare these results with those from a 2016 survey of AI/ML researchers (Grace et al., 2018) and a 2018 survey of the US public (Zhang &amp; Dafoe, 2020). We find that AI/ML researchers place high levels of trust in international organizations and scientific organizations to shape the development and use of AI in the public interest; moderate trust in most Western tech companies; and low trust in national militaries, Chinese tech companies, and Facebook. While the respondents were overwhelmingly opposed to AI/ML researchers working on lethal autonomous weapons, they are less opposed to researchers working on other military applications of AI, particularly logistics algorithms. A strong majority of respondents think that AI safety research should be prioritized and that ML institutions should conduct pre-publication review to assess potential harms. Being closer to the technology itself, AI/ML researchers are well placed to highlight new risks and develop technical solutions, so this novel attempt to measure their attitudes has broad relevance. The findings should help to improve how researchers, private sector executives, and policymakers think about regulations, governance frameworks, guiding principles, and national and international governance strategies for AI.&#x0D; This article appears in the special track on AI &amp; Society.