Hasil untuk "Special situations and conditions"

Tasmeen Zaman Ornee, Md Kamran Chowdhury Shisher, Clement Kam et al.

In this study, we consider a problem of remote safety monitoring, where a monitor pulls status updates from multiple sensors monitoring several safety-critical situations. Based on the received updates, multiple estimators determine the current safety-critical situations. Due to transmission errors and limited channel resources, the received status updates may not be fresh, resulting in the possibility of misunderstanding the current safety situation. In particular, if a dangerous situation is misinterpreted as safe, the safety risk is high. We study the joint design of transmission scheduling and estimation for multi-sensor, multi-channel remote safety monitoring, aiming to minimize the loss due to the unawareness of potential danger. We show that the joint design of transmission scheduling and estimation can be reduced to a sequential optimization of estimation and scheduling. The scheduling problem can be formulated as a Restless Multi-armed Bandit (RMAB) , for which it is difficult to establish indexability. We propose a low-complexity Maximum Gain First (MGF) policy and prove it is asymptotically optimal as the numbers of sources and channels scale up proportionally, without requiring the indexability condition. We also provide an information-theoretic interpretation of the transmission scheduling problem. Numerical results show that our estimation and scheduling policies achieves higher performance gain over periodic updating, randomized policy, and Maximum Age First (MAF) policy.

Shuhao Qi, Qiling Aori, Luyao Zhang et al.

To enable autonomous driving in interactive traffic scenarios, various model predictive control (MPC) formulations have been proposed, each employing different interaction models. While higher-fidelity models enable more intelligent behavior, they incur increased computational cost. Since strong interactions are relatively infrequent in traffic, a practical strategy for balancing performance and computational overhead is to invoke an appropriate controller based on situational demands. To achieve this approach, we first conduct a comparative study to assess and hierarchize the interactive capabilities of different MPC formulations. Furthermore, we develop a neural network-based classifier to enable situation-aware switching among controllers with different levels of interactive capability. We demonstrate that this situation-aware switching can both substantially improve overall performance by activating the most advanced interactive MPC in rare but critical situations, and significantly reduce computational load by using a basic MPC in the majority of scenarios.

Jan Zimmermann, Jörg Mönnich, Michael Scherl et al.

By using an automated braking system, such as the Automatic Emergency Brake (AEB), crashes can be avoided in situations where the driver is unaware of an imminent collision. However, conventional AEB systems detect potential collision adversaries with onboard sensor systems, such as radars and cameras, that may fail in non-line-of-sight situations. By leveraging vehicle-to-everything (V2X) communication, information regarding an approaching vehicle can be received by the ego vehicle at an early point in time, even if the opponent vehicle is occluded by a view obstruction. In this work, we consider a 2-stage braking cascade, consisting of a partial brake, triggered based on V2X information, and a sensor-triggered AEB. We evaluate its crash avoidance performance in real-world crash situations extracted from the German In-Depth Accident Study (GIDAS) database using an accident simulation framework. The results are compared against a sensor-triggered AEB system and a purely V2X-triggered partial brake. To further analyze the results, we identify the crash cause for each situation in which the brake function under test could not prevent the crash. The simulation results show a high added benefit of the V2X-enhanced braking systems compared to the exclusive use of visual-based sensor systems for automated collision prevention.

Anna BYZOVA

The article analyzes the legislation in the field of regulation of the legal status of military justice in Ukraine. The concepts of "legal status", "special period", "crisis situations that threaten national security" are investigated. The conclusion was made about the necessity of amending the legislation of Ukraine in order to consolidate the powers of the military justice authorities during a special period and in crisis situations.

Dong-Hoon Yang, Eun Ran Kim

Raquel B Robinson, Anya Osborne, Chen Ji et al.

Most currently accepted approaches to evaluating Research through Design (RtD) presume that design prototypes are finalized and ready for robust testing in laboratory or in-the-wild settings. However, it is also valuable to assess designs at intermediate phases with mid-fidelity prototypes, not just to inform an ongoing design process, but also to glean knowledge of broader use to the research community. We propose 'formative situations' as a frame for examining mid-fidelity prototypes-in-process in this way. We articulate a set of criteria to help the community better assess the rigor of formative situations, in the service of opening conversation about establishing formative situations as a valuable contribution type within the RtD community.

Kaiwen Zhou, Chengzhi Liu, Xuandong Zhao et al.

Multimodal Large Language Models (MLLMs) are rapidly evolving, demonstrating impressive capabilities as multimodal assistants that interact with both humans and their environments. However, this increased sophistication introduces significant safety concerns. In this paper, we present the first evaluation and analysis of a novel safety challenge termed Multimodal Situational Safety, which explores how safety considerations vary based on the specific situation in which the user or agent is engaged. We argue that for an MLLM to respond safely, whether through language or action, it often needs to assess the safety implications of a language query within its corresponding visual context. To evaluate this capability, we develop the Multimodal Situational Safety benchmark (MSSBench) to assess the situational safety performance of current MLLMs. The dataset comprises 1,820 language query-image pairs, half of which the image context is safe, and the other half is unsafe. We also develop an evaluation framework that analyzes key safety aspects, including explicit safety reasoning, visual understanding, and, crucially, situational safety reasoning. Our findings reveal that current MLLMs struggle with this nuanced safety problem in the instruction-following setting and struggle to tackle these situational safety challenges all at once, highlighting a key area for future research. Furthermore, we develop multi-agent pipelines to coordinately solve safety challenges, which shows consistent improvement in safety over the original MLLM response. Code and data: mssbench.github.io.

So Yeon Min, Xavi Puig, Devendra Singh Chaplot et al.

Language is never spoken in a vacuum. It is expressed, comprehended, and contextualized within the holistic backdrop of the speaker's history, actions, and environment. Since humans are used to communicating efficiently with situated language, the practicality of robotic assistants hinge on their ability to understand and act upon implicit and situated instructions. In traditional instruction following paradigms, the agent acts alone in an empty house, leading to language use that is both simplified and artificially "complete." In contrast, we propose situated instruction following, which embraces the inherent underspecification and ambiguity of real-world communication with the physical presence of a human speaker. The meaning of situated instructions naturally unfold through the past actions and the expected future behaviors of the human involved. Specifically, within our settings we have instructions that (1) are ambiguously specified, (2) have temporally evolving intent, (3) can be interpreted more precisely with the agent's dynamic actions. Our experiments indicate that state-of-the-art Embodied Instruction Following (EIF) models lack holistic understanding of situated human intention.

Chinthani Sugandhika, Chen Li, Deepu Rajan et al.

Graph based representation has been widely used in modelling spatio-temporal relationships in video understanding. Although effective, existing graph-based approaches focus on capturing the human-object relationships while ignoring fine-grained semantic properties of the action components. These semantic properties are crucial for understanding the current situation, such as where does the action takes place, what tools are used and functional properties of the objects. In this work, we propose a graph-based representation called Situational Scene Graph (SSG) to encode both human-object relationships and the corresponding semantic properties. The semantic details are represented as predefined roles and values inspired by situation frame, which is originally designed to represent a single action. Based on our proposed representation, we introduce the task of situational scene graph generation and propose a multi-stage pipeline Interactive and Complementary Network (InComNet) to address the task. Given that the existing datasets are not applicable to the task, we further introduce a SSG dataset whose annotations consist of semantic role-value frames for human, objects and verb predicates of human-object relations. Finally, we demonstrate the effectiveness of our proposed SSG representation by testing on different downstream tasks. Experimental results show that the unified representation can not only benefit predicate classification and semantic role-value classification, but also benefit reasoning tasks on human-centric situation understanding. We will release the code and the dataset soon.

Binh Thang Tran, Minh Tu Nguyen, Minh Tam Nguyen et al.

Objectives We assessed the prevalence of stress, anxiety, and depression among adolescents living in families with separated or divorced parents in Hue City, Vietnam and identified factors associated with these conditions. Methods This cross-sectional study enrolled 309 adolescents, aged 12 to 17 years, living in families with separated or divorced parents in Hue City, Vietnam. The depression anxiety stress scale-21 (DASS-21) was used to measure stress, anxiety, and depression. Predictors of overall and individual mental health problems were identified using ordered and binary logistic regression, respectively. Results The DASS-21 scale revealed a 49.2% prevalence of stress, while anxiety and depression had s prevalence rates of 61.5%. Among participants, 42.4% experienced all 3 mental health issues. Several factors were identified as significant predictors of mental health problems, including poor to average economic status (adjusted odds ratio [aOR], 2.00; 95% confidence interval [CI], 1.21–3.31; p=0.007); being in high school (aOR, 5.02; 95% CI, 2.93–8.60; p<0.001); maternal occupation of teacher, healthcare professional, or official (aOR, 2.39; 95% CI, 1.13–5.03; p=0.022); longer duration of family separation or divorce (aOR, 1.24; 95% CI, 1.05–1.45; p=0.009); living with one’s mother (aOR, 1.69; 95% CI, 1.03–2.76; p=0.04); alcohol consumption (aOR, 1.70; 95% CI, 0.99–2.92; p=0.050); and being bullied (aOR, 5.33; 95% CI, 1.10–25.69; p=0.037). Most of these factors were associated with stress, anxiety, and depression. Additionally, smoking was associated with stress. Conclusion Adolescents with separated or divorced parents were at increased risk of stress, anxiety, and depression. The findings of this study provide important implications for prevention programs.

Anna-Clara Ivarsson, Elin Fransén, Ioanna Broumou et al.

Abstract Background Light microscopy and rapid diagnostic tests (RDT) have long been the recommended diagnostic methods for malaria. However, in recent years, loop-mediated isothermal amplification (LAMP) techniques have been shown to offer superior performance, in particular concerning low-grade parasitaemia, by delivering higher sensitivity and specificity with low laboratory capacity requirements in little more than an hour. In this study, the diagnostic performance of two LAMP kits were assessed head-to-head, compared to highly sensitive quantitative real time PCR (qPCR), in a non-endemic setting. Methods In this retrospective validation study two LAMP kits; Alethia® Illumigene Malaria kit and HumaTurb Loopamp™ Malaria Pan Detection (PDT) kit, were evaluated head-to-head for detection of Plasmodium-DNA in 133 biobanked blood samples from suspected malaria cases at the Clinical Microbiology Laboratory of Region Skåne, Sweden to determine their diagnostic performance compared to qPCR. Results Of the 133 samples tested, qPCR detected Plasmodium DNA in 41 samples (defined as true positives), and the two LAMP methods detected 41 and 37 of those, respectively. The results from the HumaTurb Loopamp™ Malaria PDT kit were in complete congruence with the qPCR, with a sensitivity of 100% (95% CI 91.40–100%) and specificity of 100% (95% CI 96.07–100%). The Alethia® Illumigene Malaria kit had a sensitivity of 90.24% (95% CI 76.87–97.28) and a specificity of 95.65% (95% CI 89.24–98.80) as compared to qPCR. Conclusions This head-to-head comparison showed higher performance indicators of the HumaTurb Loopamp™ Malaria PDT kit compared to the Alethia® illumigene Malaria kit for detection of malaria.

Beatriz L. Gómez

Satoru Kobayashi, Younghee Hahn, Brett Silverstein et al.

Diabetes is a major risk factor for a variety of cardiovascular complications, while diabetic cardiomyopathy, a disease specific to the myocardium independent of vascular lesions, is an important causative factor for increased risk of heart failure and mortality in diabetic populations. Lysosomes have long been recognized as intracellular trash bags and recycling facilities. However, recent studies have revealed that lysosomes are sophisticated signaling hubs that play remarkably diverse roles in adapting cell metabolism to an ever-changing environment. Despite advances in our understanding of the physiological roles of lysosomes, the events leading to lysosomal dysfunction and how they relate to the overall pathophysiology of the diabetic heart remain unclear and are under intense investigation. In this review, we summarize recent advances regarding lysosomal injury and its roles in diabetic cardiomyopathy.

Mattia Bianchi, Liam Anderson, Thomas Brownlee et al.

This study aimed to investigate the effect of a combined jump and sprint training program, two sessions a week for 6 weeks, on sprinting, change of directions (COD) and jumping performance in semiprofessional soccer players. Twenty soccer players were enrolled in this randomized controlled trial (age 20±2 years, body mass 74.3±5.9 kg). Players were randomized into two groups such as training group (TG, n = 10 players) or control group (CG, n = 10 players). Physical tests were performed before and after 6 weeks of training such as sprint 10 m, sprint 30 m, 505-COD test and standing long jump (LJ). The two groups performed the same training except for the combined jump and sprint training performed twice a week by TG. After 6 weeks of training, between-group analysis reported statistical difference in favor of the TG in sprint 10 m (p = 0.015, η 2 = 0.295, large), sprint 30 m (p < 0.001, η 2 = 0.599, large), in 505-COD (p = 0.026, η 2 = 0.154, large), and LJ (p = 0.025, η 2 = 0.027, small). These data indicate that combined sprint and jump training, when performed twice a week, for the duration of 6 weeks, in addition to the regular team training, can improve specific physical performance in male soccer players. This study has shown that a volume increment of 10% after 3 weeks of training can be a suitable training dose progression and that a combination of 64–70 jumps and 675–738 m of sprinting training per session can yield benefits in sprint, COD and jump performance.

Vega García López

Tras una primera “Introducción conceptual” bajo un paradigma de trabajo-salud que integra todos los elementos que explican su interconexión (condiciones de empleo, servicios sanitarios, prevención, daños a la salud, causalidad, responsabilidad…) los autores y autoras nos conducen al complejo mundo de la salud laboral desde la visión clásica de los riesgos laborales y los daños hasta una visión holística que aborda los distintos dispositivos del Sistema de Salud y los condicionantes sociales del empleo. Todo abordado con un ENFOQUE DE SALUD PÚBLICA que busca la salud y bienestar de la población trabajadora. Aunque la perspectiva de la salud laboral en el Sistema Público de Salud ya se describía en nuestro país en la LGS’86 (Ley General de Salud 14/1986) y LGSP’11 (Ley General de Salud Pública 33/2011), todavía está insuficientemente desarrollada, y tal como se enfoca en el libro, es necesario considerar los riesgos laborales como determinantes de salud e imprescindible la coordinación con los Servicios de Salud Laboral.  Recorriendo la publicación(1), se aborda la PREVENCION de RIESGOS, desde los más evidentes, de seguridad que causan la patología traumática aguda, hasta otros más silentes como los químicos, biológicos o físicos, de los que cuesta tomar conciencia por sus consecuencias a más largo plazo (ej. Cáncer laboral, hipoacusia…) e incluye los de naturaleza psicosocial que son los que producen mayor merma en la percepción global de la salud. Advierte de la existencia de trabajadores ESPECIALMENTE SENSIBLES a los riesgos que normativamente establecen unos límites permisibles no válidos para ellos (estado biológico, embarazo, edad límite…). Asimismo, recuerda la necesaria PARTICIPACION del personal trabajador, legalmente protegida y fundamental en la implicación en la prevención de riesgos laborales.  También, reflexiona sobre la VIGILANCIA DE LA SALUD, creyendo necesario conceptualizar los Criterios de Aptitud y reconocimientos iniciales, sobre todo. Considera que es necesario tomar conciencia de los daños, más allá de los legalmente reconocidos (lesiones por accidentes de trabajo y enfermedades profesionales) aquellos relacionados con el trabajo y que, con frecuencia, se atienden en el Sistema Público de Salud (ej. Sucesos centinela) y advierte de la necesidad de revisar a la luz de la evidencia científica los PROTOCOLOS de vigilancia de la salud y los aspectos éticos que aseguren el respeto a la confidencialidad , dignidad y voluntariedad del trabajador. Incluye, además, el Sistema Público de Salud como complemento a los Servicios que tienen encomendada la Vigilancia de la Salud, para la detección precoz de la patología laboral, su consecuente notificación y protección a través de los sistemas de aseguramiento de las lesiones por accidentes de trabajo y enfermedades profesionales, y las ENCUESTAS de condiciones de trabajo y de salud que reflejan la percepción que los trabajadores.  Reserva espacio, además, para la prevención de la INCAPACIDAD laboral que, aunque cuenta con un sistema garantista de subsidio, lamentablemente retira a la población trabajadora del mundo laboral precozmente.  Destacar, el reservado para la Salud MENTAL en ambas vertientes, agravamiento de problemas personales en ambientes hostiles y los problemas generados por el trabajo (desde estrés postraumático hasta ideaciones suicidas), sin olvidar los problemas de reconocimiento y la estigmatización social y laboral que provoca. Así considera necesaria la concienciación y abordaje en coordinación con el sistema público de salud.  Además, enumera los trabajadores VULNERABLES: por el trabajo INFORMAL, no protegido y muy frecuentes en trabajadoras domésticas (de cuidados y mujeres)  Inmigrantes: mayor dificultad de acceso al mercado laboral, peores condiciones, y mayor accidentabilidad. Colectivos ESPECIALES: trabajadoras domésticas y del Sector Agrario. Dedica un extenso capítulo al CANCER laboral (en el que existe evidencia de asociación exposición y neoplasia) y PROFESIONAL, el legalmente reconocido y varía según los países. Desafía a estimar la fracción atribuible real del origen laboral, ya que si para los 14 principales cancerígenos laborales es del 3,9%, queda mucho recorrido para la investigación y adopción de medidas que eviten su infradeclaración (registros de expuestos, sistemas de reconocimiento e indemnización..). Reflexiona de como la pandemia COVID ha visibilizado a los trabajadores esenciales, fundamentalmente sanitarios, el entorno laboral como lugar de especial riesgo y el papel de los SPRL en el control de la salud de los trabajadores. Concluye con la consideración de la PROMOCION de la Salud como otra extensión de la salud laboral que contribuirá al bienestar y salud en el futuro del trabajo Concepto NIOSH (Total Workers Health) y con el cambio en la naturaleza del trabajo y del empleo (teletrabajo, automatización de tareas, coworking, empleo flexible…), el cambio de la población laboral (envejecimiento, diversidad…) que exige, además, un ENFOQUE ampliado de la salud laboral como “Salud INTEGRAL “ del trabajador.  En resumen, se trata de una lectura imprescindible para quienes trabajan en salud laboral en estos tiempos de “crisis” que exigen cambio de paradigma… guiado por expertos.

Foad Alzoughool, Suhad Abumweis, Lo’ai Alanagreh et al.

Objectives The aim of this study was to evaluate the association of pre-existing cardiovascular comorbidities, including hypertension and coronary heart disease, with coronavirus disease 2019 (COVID-19) severity and mortality. Methods PubMed, ScienceDirect, and Scopus were searched between January 1, 2020, and July 18, 2020, to identify eligible studies. Random-effect models were used to estimate the pooled event rates of pre-existing cardiovascular disease comorbidities and odds ratio (OR) with 95% confidence intervals (95% CIs) of disease severity and mortality associated with the exposures of interest. Results A total of 34 studies involving 19,156 patients with COVID-19 infection met the inclusion criteria. The prevalence of pre-existing cardiovascular disease in the included studies was 14.0%. Pre-existing cardiovascular disease in COVID-19 patients was associated with severe outcomes (OR, 4.1; 95% CI, 2.9 to 5.7) and mortality (OR, 6.1; 95% CI, 2.9 to 12.7). Hypertension and coronary heart disease increased the risk of severe outcomes by 3 times (OR, 3.2; 95% CI, 2.0 to 3.6) and 2.5 times (OR, 2.5; 95% CI, 1.7 to 3.8), respectively. No significant publication bias was indicated. Conclusion COVID-19 patients with pre-existing cardiovascular comorbidities have a higher risk of severe outcomes and mortality. Awareness of pre-existing cardiovascular comorbidity is important for the early management of COVID-19.

Yan Ding, Xiaohan Zhang, Saeid Amiri et al.

Automated task planning algorithms have been developed to help robots complete complex tasks that require multiple actions. Most of those algorithms have been developed for "closed worlds" assuming complete world knowledge is provided. However, the real world is generally open, and the robots frequently encounter unforeseen situations that can potentially break the planner's completeness. This paper introduces a novel algorithm (COWP) for open-world task planning and situation handling that dynamically augments the robot's action knowledge with task-oriented common sense. In particular, common sense is extracted from Large Language Models based on the current task at hand and robot skills. For systematic evaluations, we collected a dataset that includes 561 execution-time situations in a dining domain, where each situation corresponds to a state instance of a robot being potentially unable to complete a task using a solution that normally works. Experimental results show that our approach significantly outperforms competitive baselines from the literature in the success rate of service tasks. Additionally, we have demonstrated COWP using a mobile manipulator. The project website is available at: https://cowplanning.github.io/, where a more detailed version can also be found. This version has been accepted for publication in Autonomous Robots.

Getayeneh Antehunegn Tesema, Achamyeleh Birhanu Teshale

Abstract Background Despite the remarkable decrease in infant mortality rate in most countries, the rate of decline is slow and it remains unacceptably high in Sub-Saharan Africa. The progress in infant mortality in Ethiopia is far below the rate needed to achieve the Sustainable Development Goal. Understanding the residential inequality and spatiotemporal clusters of infant mortality is essential to prioritize areas and guide public health interventions. Therefore, this study aimed to investigate the residential inequality and spatial patterns of infant mortality in Ethiopia. Methods A secondary data analysis was done based on the Ethiopian demographic and health surveys conducted in 2000, 2005, 2011, and 2016. A total weighted sample of 46,317 live births was included for the final analysis. The residential inequality was assessed by calculating the risk difference in infant mortality rates between urban and rural live births and presented using a forest plot. For the spatial patterns of infant mortality, the SaTScan version 9.6 and ArcGIS version 10.6 statistical software were used to identify the spatial patterns of infant mortality. Results The study revealed that the infant mortality rate significantly declined from 96.9 per 1000 live births [95% CI 93.6, 104.2] in 2000 to 48.0 per 1000 live births [95% CI 44.2, 52.2] in 2016 with an annual rate of reduction of 3.2%. The infant mortality rate has substantial residential inequality over time, which is concentrated in the rural area. The spatial distribution of infant mortality was significantly clustered at the national level in survey periods (global Moran’s I, 0.04–0.081, p value < 0.05). In 2000, the most likely clusters were found in east Afar and at the border areas of south Amhara and north Oromia regions (LLR = 7.61, p value < 0.05); in 2005, at the border areas of Southern Nations Nationalities and People and in the entire Amhara region (LLR = 10.78, p value< 0.05); in 2011, at Southern Nations Nationalities and People and Gambella regions (LLR = 6.63, p value< 0.05); and in 2016, at east Oromia and northeast Somali regions (LLR = 8.38, p value < 0.05). Conclusion In this study, though infant mortality has shown remarkable reduction, infant mortality remains a major health care concern and had significant spatial variation across regions. Besides, the study found that infant mortality was highly concentrated in rural areas. Identifying the hotspot areas of infant mortality would help in designing effective interventions to reduce the incidence of infant mortality in these areas. Therefore, the findings highlighted that public health interventions should target rural areas and identified hotspot areas to reduce the incidence of infant mortality.

In Hye Park, Yustika Sya’bandari, Yang Liu

Abstract Background Although some previous studies have reported the impact of cultural factors on individuals’ cognition and decision making, a shortage of research has led to this comparison study for Chinese and Korean elderly, a growing population with depression. This study aimed to explore depression levels in Chinese and South Korean elderly individuals by focusing on testing the generalizability of the Geriatric Depression Scale (GDS). Methods The data of 493 community-dwelling Chinese and Korean elderly individuals over the age of 60 years were used to examine GDS. To test the dimensionality, item quality, and reliability of the GDS, the item response theory, Rasch analysis was performed. The detection of differential item functioning (DIF) of the GDS between the two countries was determined by performing a hybrid ordinal logistic regression. Results The four-dimensional framework of the GDS, categorized into agitation, cognitive concerns, dysphoria, and vigor/withdrawal was fit for measuring depression levels in Chinese and Korean elderly individuals. In addition, good item quality and reliability of the GDS indicate that almost all items in this scale contribute to measuring the intended trait. Meanwhile, 18 out of 28 items of the GDS were detected as country-related DIF with five items having a large effect size. Conclusions Although China and Korea are close geographically and culturally, the item bias shown by severe country-related DIF implies that different cultural backgrounds impact how the elderly interpret GDS items. The cultural issues related to the specific DIF items, the implication to accuracy of individual scores estimation, and the optimal decision to treat individuals were discussed.

Thierry Deruyttere, Victor Milewski, Marie-Francine Moens

Current technology for autonomous cars primarily focuses on getting the passenger from point A to B. Nevertheless, it has been shown that passengers are afraid of taking a ride in self-driving cars. One way to alleviate this problem is by allowing the passenger to give natural language commands to the car. However, the car can misunderstand the issued command or the visual surroundings which could lead to uncertain situations. It is desirable that the self-driving car detects these situations and interacts with the passenger to solve them. This paper proposes a model that detects uncertain situations when a command is given and finds the visual objects causing it. Optionally, a question generated by the system describing the uncertain objects is included. We argue that if the car could explain the objects in a human-like way, passengers could gain more confidence in the car's abilities. Thus, we investigate how to (1) detect uncertain situations and their underlying causes, and (2) how to generate clarifying questions for the passenger. When evaluating on the Talk2Car dataset, we show that the proposed model, \acrfull{pipeline}, improves \gls{m:ambiguous-absolute-increase} in terms of $IoU_{.5}$ compared to not using \gls{pipeline}. Furthermore, we designed a referring expression generator (REG) \acrfull{reg_model} tailored to a self-driving car setting which yields a relative improvement of \gls{m:meteor-relative} METEOR and \gls{m:rouge-relative} ROUGE-l compared with state-of-the-art REG models, and is three times faster.