Interest In Autonomous Vehicles Research Articles

자율주행자동차 상용화에 대비한 자동차보험제도 연구

On May 7, 2016, in the state of Florida, USA, a fatal accident occurred involving an autonomous vehicle and a large truck, leading to the death of the driver. This incident heightened interest in autonomous vehicles and sparked discussions on their implications, including the challenges and liability issues in the event of accidents during autonomous operation. The application of existing legal frameworks, such as Article 750 of the Civil Code on tort liability and Article 3 of the Automobile Damage Compensation Guarantee Act on operator liability, to accidents involving autonomous vehicles raises questions about identifying responsible parties and compensating victims. This study examines the applicability of these legal principles to accidents occurring during autonomous driving and proposes the introduction of a No-Fault insurance system as a novel automotive insurance mechanism to address these issues. The analysis revealed that the No-Fault insurance system, a solution to the problems posed by the current fault-based liability system and already implemented in several advanced countries since the early 2000s, was considered but ultimately not adopted in South Korea due to challenges in amending existing legislation, limitations in calculating insurance rates, and concerns over moral hazard, among other issues. However, as autonomous vehicles are poised to integrate into our lives more rapidly than anticipated, and the social concerns regarding traffic accidents during autonomous operation deepen, it is imperative to form a task force comprising the Ministry of Land, Infrastructure, and Transport, the Financial Supervisory Service, the General Insurance Association, among others, before it's too late. This task force should engage in research and discussions, while building a consensus among the public, to prepare for the new automotive market driven by the autonomous vehicle industry.

Towards a Model of Snow Accretion for Autonomous Vehicles

Snow accumulation on surfaces exposed to adverse weather conditions has been studied over the years due to a variety of problems observed in different industry sectors, such as aeronautics and wind and civil engineering. With the growing interest in autonomous vehicles (AVs), this concern extends to advanced driver-assistance systems (ADAS). Weather stressors, such as snow and icing, negatively influence the sensor functionality of AVs, and their autonomy is not guaranteed by manufacturers during episodes of intense weather precipitation. As a basis for mitigating the negative effects caused by heavy snowfall, models need to be developed to predict snow accumulation over critical surfaces of AVs. The present work proposes a framework for the study of snow accumulation on road vehicles. Existing icing and snow accretion models are reviewed, and adaptations for automotive applications are discussed. Based on the new capabilities developed by the Weather on Wheels (WoW) program at Ontario Tech University, a model architecture is proposed in order to progress toward adequate snow accretion predictions for autonomous vehicle operating conditions, and preliminary results are presented.

The relevance of social factors in sharing a trip with strangers: Creating travel communities in the autonomous vehicles era

There has been a growing interest in autonomous vehicles (AVs) for the past few years due to technological innovation and their far-reaching expected impacts on urban mobility. The impact of AVs depends ultimately on how the technology will be adopted, e.g. privately or shared AVs (SAVs), and integrated into the socio-technical mobility system. In this study, we evaluate how social considerations and the AV technology may enhance new social arrangements for travel, especially for groups suffering from some disadvantages in accessing transport. Particularly, we explore a proposed new form of social arrangement, coined “Travel Community” (TC), to create shared AVs trips based on individual preferences. An online national representative sample in Israel was conducted (N = 1009). The data was analyzed using multivariate methods including multinomial logistic regression models for the willingness to adopt TC in two contexts: commuting (TCC) and travel leisure (TCL). The findings suggest that TC was well accepted by the participants under certain assumptions and could address latent demand for improving travel options for people with different constraints accessing transport, such as women, young, sporadic travelers, and low-income individuals. The personal social preferences option might help overcome psychological barriers regarding shared mobility, such as the fear of sharing the trip with strangers and a social value promoting a new kind of social interaction. Planners and stakeholders should consider social considerations' relevance (opportunities and risks) for promoting and creating new forms of mobility and improving urban sustainability and social inclusion.

On the Impact of Multiple Access Interference in LTE-V2X and NR-V2X Sidelink Communications

Developing radio access technologies that enable reliable and low-latency vehicular communications have become of the utmost importance with the rise of interest in autonomous vehicles. The Third Generation Partnership Project (3GPP) has developed Vehicle to Everything (V2X) specifications based on the 5G New Radio Air Interface (NR-V2X) to support connected and automated driving use cases, with strict requirements to fulfill the constantly evolving vehicular applications, communication, and service demands of connected vehicles, such as ultra-low latency and ultra-high reliability. This paper presents an analytical model for evaluating the performance of NR-V2X communications, with particular reference to the sensing-based semi-persistent scheduling operation defined in the NR-V2X Mode 2, in comparison with legacy sidelink V2X over LTE, specified as LTE-V2X Mode 4. We consider a vehicle platooning scenario and evaluate the impact of multiple access interference on the packet success probability, by varying the available resources, the number of interfering vehicles, and their relative positions. The average packet success probability is determined analytically for LTE-V2X and NR-V2X, taking into account the different physical layer specifications, and the Moment Matching Approximation (MMA) is used to approximate the statistics of the signal-to-interference-plus-noise ratio (SINR) under the assumption of a Nakagami-lognormal composite channel model. The analytical approximation is validated against extensive Matlab simulations that a show good accuracy. The results confirm a boost in performance with NR-V2X against LTE-V2X, particularly for high inter-vehicle distance and a large number of vehicles, providing a concise yet accurate modeling rationale for planning and adaptation of the configuration and parameter setup of vehicle platoons, without having to resort to extensive computer simulation or experimental measurements.

End-to-End Learning with Memory Models for Complex Autonomous Driving Tasks in Indoor Environments

The interest in autonomous vehicles has increased exponentially in recent years. While Lidar is a proven autonomous driving technology, end-to-end learning approaches have become popular as computer performance has improved. A fully end-to-end method—NVIDIA’s PilotNet has shown its ability to predict speed and steering angle with only camera images. This method achieved the Lidar-based methods’ performance in simple driving tasks. However, a significant drawback was no past spatiotemporal information, imposing an error-sensitive performance, especially in complex driving tasks. Spurred by this deficiency, this paper introduces two novel models: CNN + LSTM and CNN3D, aiming for complex autonomous driving tasks in indoor environments.

Early Detection of Gestures using Context-and-Gap-aware Predictive Pose Framework

With a growing interest in autonomous vehicles’ operation, there is an equally increasing need for efficient anticipatory gesture recognition systems for human-vehicle interaction. Existing gesture recognition algorithms have been primarily restricted to historical data. In this paper, we propose a novel Context-and-Gap-aware Pose Prediction Framework (CGAP2), which predicts the future pose data for anticipatory recognition of gestures in an online fashion. CGAP2 implements an encoder–decoder architecture paired with a pose prediction module to anticipate future frames followed by a shallow classifier. CGAP2’s pose prediction module uses 3D convolutional layers and is dependent on the number of pose frames supplied, the time difference between subsequent pose frames, and the number of predicted pose frames. The performance of CGAP2 is evaluated on the Human3.6M dataset with the MPJPE metric. For the pose prediction of 15 frames in advance, an error of 79 mm is achieved. The pose prediction module consists of only 26M parameters and can run at 50 FPS on the NVIDIA TITAN RTX. CGAP2 has a 1-s time advantage compared to other gesture recognition systems, which can be crucial for autonomous vehicles. Furthermore, this work formalizes the problem of anticipatory gesture recognition by introducing two hyper-parameters namely context and gap and a relationship is established between the two using carefully designed experiments in ablation testing. As per our knowledge, this work is the first to tackle all the challenges faced by anticipatory gesture recognition and also the first to post a performance benchmark on Human3.6M dataset.

Analysis of Korean Road Traffic Regulations to Establish Legal Layer for Evaluating Safety of Autonomous Driving

Currently, many governments, global automakers, and IT companies have been showing increasing interest in autonomous vehicles (AVs). Therefore, the research, development, and testing of AVs are underway worldwide and the development of scenarios for AV safety evaluation is emerging as an urgent issue in various nations and institutions. However, in Korea, research on an appropriate legal system for AVs is necessary because the existing one is insufficient for the development of a scenario for evaluating autonomous driving safety. In this study, the present legal system related to autonomous driving in Korea was selected to establish a five-layer format-based autonomous driving scenario for the PEGASUS project. The possibility of the expansion and improvement of the methodology was proposed by conducting a legal system analysis. Finally, an analysis of the related legal provisions for each layer was conducted to highlight the necessity of a legal system that is suitable for the era of autonomous driving in the future.

Deep learning-based robust positioning for all-weather autonomous driving

Interest in autonomous vehicles (AVs) is growing at a rapid pace due to increased convenience, safety benefits and potential environmental gains. Although several leading AV companies predicted that AVs would be on the road by 2020, they are still limited to relatively small-scale trials. The ability to know their precise location on the map is a challenging prerequisite for safe and reliable AVs due to sensor imperfections under adverse environmental and weather conditions, posing a formidable obstacle to their widespread use. Here we propose a deep learning-based self-supervised approach for ego-motion estimation that is a robust and complementary localization solution under inclement weather conditions. The proposed approach is a geometry-aware method that attentively fuses the rich representation capability of visual sensors and the weather-immune features provided by radars using an attention-based learning technique. Our method predicts reliability masks for the sensor measurements, eliminating the deficiencies in the multimodal data. In various experiments we demonstrate the robust all-weather performance and effective cross-domain generalizability under harsh weather conditions such as rain, fog and snow, as well as day and night conditions. Furthermore, we employ a game-theoretic approach to analyse the interpretability of the model predictions, illustrating the independent and uncorrelated failure modes of the multimodal system. We anticipate our work will bring AVs one step closer to safe and reliable all-weather autonomous driving.

Resistance to Cybersecurity Attacks in a Novel Network for Autonomous Vehicles

The increased interest in autonomous vehicles has led to the development of novel networking protocols in VANETs In such a widespread safety-critical application, security is paramount to the implementation of the networks. We view new autonomous vehicle edge networks as opportunistic networks that bridge the gap between fully distributed vehicular networks based on short-range vehicle-to-vehicle communication and cellular-based infrastructure for centralized solutions. Experiments are conducted using opportunistic networking protocols to provide data to autonomous trams and buses in a smart city. Attacking vehicles enter the city aiming to disrupt the network to cause harm to the general public. In the experiments the number of vehicles and the attack length is altered to investigate the impact on the network and vehicles. Considering different measures of success as well as computation expense, measurements are taken from all nodes in the network across different lengths of attack. The data gathered from each node allow exploration into how different attacks impact metrics including the delivery probability of a message, the time taken to deliver and the computation expense to each node. The novel multidimensional analysis including geospatial elements provides evidence that the state-of-the-art MaxProp algorithm outperforms the benchmark as well as other, more complex routing protocols in most of the categories. Upon the introduction of attacking nodes however, PRoPHET provides the most reliable delivery probability when under attack. Two different attack methods (black and grey holes) are used to disrupt the flow of messages throughout the network and the more basic protocols show that they are less consistent. In some metrics, the PRoPHET algorithm performs better when under attack due to the benefit of reduced network traffic.

ANALYSIS OF THE PERFORMANCE OF AN UWB-BASED COOPERATIVE POSITIONING FOR DIFFERENT CAR PLATOON CONFIGURATIONS

Abstract. The increasing interest in autonomous vehicles motivates the researches aiming at developing reliable positioning system also in conditions challenging for the Global Navigation Satellite Systems (GNSS), such as in urban canyons, tunnels, under quite dense vegetation. The uso of Ultra Wide-Band (UWB) systems is among the quite well known methods for providing reasonable positioning results without exploiting GNSS. UWB systems are typically used indoors, however their use can be of interest also outdoors, in particular when the need is to ensure good positioning results over a quite small area. This paper investigates the use of UWB systems for positioning in the case of terrestrial vehicles, and, more specifically, it focuses on checking the influence of car platoon configurations on the performance of an UWB cooperative positioning system. In the considered tests, where a high percentage of UWB communications was successful, the obtained results show that the car configuration can have a quite remarkable impact on the positioning performance, doubling the obtained median error.

Autonomous Vehicles and Machines 2022 Conference Overview and Papers Program

Advancements in sensing, computing, image processing, and computer vision technologies are enabling unprecedented growth and interest in autonomous vehicles and intelligent machines, from self-driving cars to unmanned drones, to personal service robots. These new capabilities have the potential to fundamentally change the way people live, work, commute, and connect with each other, and will undoubtedly provoke entirely new applications and commercial opportunities for generations to come. The main focus of AVM is perception. This begins with sensing. While imaging continues to be an essential emphasis in all EI conferences, AVM also embraces other sensing modalities important to autonomous navigation, including radar, LiDAR, and time of flight. Realization of autonomous systems also includes purpose-built processors, e.g., ISPs, vision processors, DNN accelerators, as well core image processing and computer vision algorithms, system design and architecture, simulation, and image/video quality. AVM topics are at the intersection of these multi-disciplinary areas. AVM is the Perception Conference that bridges the imaging and vision communities, connecting the dots for the entire software and hardware stack for perception, helping people design globally optimized algorithms, processors, and systems for intelligent “eyes” for vehicles and machines.

카메라와 라이다 센서 융합을 이용한 탐지 객체 제거기반 3D 지도 갱신 알고리즘

With the rapidly increasing research interest in autonomous vehicles, map update systems have become crucial. In the existing method, the original map is compared with the sensor data, and the newly changed map data is updated unconditionally. However, this is a simple iterative updating method that cannot distinguishing the landmarks (e.g., building, crosswalk, etc.). In this study, objects (i.e., people, cars, etc.) that are not related to the map are detected using the deep learning technique. The objects are excluded from the 3D data using a camera and LidarDAR sensor fusion. The remaining undetected 3D data is compared with the map data, and the map is updated by adding new landmarks and simultaneously removing the missing landmarks. The location accuracy is increased by localization based on the updated map. Compared to the original map, this proposed deep-learning-based method can reduces the error by up to 1.5 m. Thus, this proposed method is expected to aid in the advancement of the map update system.

How Imitation Learning and Human Factors Can Be Combined in a Model Predictive Control Algorithm for Adaptive Motion Planning and Control.

Interest in autonomous vehicles (AVs) has significantly increased in recent years, but despite the huge research efforts carried out in the field of intelligent transportation systems (ITSs), several technological challenges must still be addressed before AVs can be extensively deployed in any environment. In this context, one of the key technological enablers is represented by the motion-planning and control system, with the aim of guaranteeing the occupants comfort and safety. In this paper, a trajectory-planning and control algorithm is developed based on a Model Predictive Control (MPC) approach that is able to work in different road scenarios (such as urban areas and motorways). This MPC is designed considering imitation-learning from a specific dataset (from real-world overtaking maneuver data), with the aim of getting human-like behavior. The algorithm is used to generate optimal trajectories and control the vehicle dynamics. Simulations and Hardware-In-the-Loop tests are carried out to demonstrate the effectiveness and computation efficiency of the proposed approach.

Cost analysis of autonomous battery electric field tractors in agriculture

Interest in the electrification of agricultural vehicles is increasing along with growing interest in autonomous vehicles. Individual technologies have been well-explored, but not their combined use and the effects on agricultural fieldwork. In this study, cost analysis was conducted based on a simulated vehicle system with 50 kW self-driving battery-electric drive (BED) tractors. The analysis included battery degradation due to cycling and the cost of inadequate machine capacity, as these factors are suspected to be problems for electric tractors. A dynamic discrete-event vehicle system model, a linear timeliness model and a one-dimensional battery cell ageing model were assumed. Costs obtained were compared with those of contemporary manned diesel-based systems. BED systems had equal or lower annual costs compared to conventional manned diesel-based systems; this was due to lower costs for fuel and maintenance, while providing adequate capacity and lower energy usage. Sensitivity analysis showed that operating costs were of greater significance than investment costs. The generally more expensive investment costs of BED systems were outweighed by the reduced operating costs for several different BED system systems. Battery degradation costs and timeliness were influential, but not sufficient to make the system uncompetitive. The synergistic effect of vehicular autonomy and BED outweighed several of the drawbacks of BED systems, such as frequent recharging, increased transport and reduced consecutive work time. • Cost analysis was performed for a simulated agricultural, self-driving BEV system. • Timeliness and battery degradation costs were not detrimental to BED. • BEV systems had lower or equal annual costs compared to equivalent diesel systems. • Operating costs had a larger impact on annual costs than investment costs. • Autonomy mitigated BEV drawbacks (frequent recharging, reduced work time).

All work and no play? Autonomous vehicles and non-commuting journeys

ABSTRACT People travel by car for a wide variety of reasons. A large proportion of household travel is for non-commuting purposes, including social and recreational journeys. The emergence and (potential) diffusion of highly automated vehicles, also known as autonomous vehicles (AVs), could transform the way (some) people work and travel. Should they become mainstream, AVs could reshape patterns of leisure travel. To date, however, the impacts and implications of AVs beyond commuting trips have received minimal attention from transport scholarship. This paper presents a state-of-the-art review of literatures on AVs. It follows PRISMA guidelines and synthesises 63 papers on AV travel focusing on non-commuting journeys, including travel for purposes of leisure, tourism, shopping and visiting friends and relatives. Given the economic importance of the tourism sector and its inherent focus on non-commuting journeys, this analysis is supplemented with a review of the extent to which national tourism strategies of countries leading AV deployment include reference to AVs. The paper reveals an overwhelming focus on commuting journeys in existing AV studies as less than one-fifth of the reviewed academic sources include non-commuting as part of their wider analysis. The review's further key findings are that the interest of publics in AVs for leisure journeys appears to exceed that for commuting, sharing vehicles will be less likely when AVs are used for leisure and there is an absence of recognition in the literature that certain non-commuting journeys will require a lower SAE level of automation. Surprisingly, analysis of the national tourism strategies of countries most prepared to meet the challenges of AVs shows that just three countries make specific reference to AVs within their national tourism strategies. The paper contributes to setting future AV policy agendas by concluding that two gaps must be narrowed: one, the distance between how academic studies predominantly conceive of AV use (commuting) and articulated public interest in AVs for non-commuting journeys; and two, the lack of readiness in certain national tourism strategies to accommodate AVs. As non-commuting journeys are likely to represent some of the earliest trip purposes for which AVs could be adopted, the paper points to the potential barriers to AV uptake by remaining focused on a limited set of trip purposes.

Should the Government Be Allowed to Take Control over Your Car as Part of a Disaster Management Plan?

Introduction: With the Safety Ensuring Lives Future Deployment and Research in Vehicle Evolution (SELF DRIVE) Act in the United States, there is a growing interest in autonomous vehicles (AVs). One avenue of innovation would be to use them to mobilize and coordinate response efforts during natural disasters. This study uses an earthquake response in an urban, developed setting as a hypothetical example case study. In this hypothetical scenario, private AVs would be mobilized to help rescue victims from collapsed structures. Methods: A Markov model compared an intervention arm with AVs to a status quo arm using a hypothetical cohort of American earthquake victims. The three possible health states were trapped but alive, rescued and alive, and dead. The cycle length of the Markov model was 6 h. Results: The cost of deploying AVs was $90,139 relative to $87,869 in status quo arm. Using AVs produced an incremental cost of $2269 (95% credible interval (CI) = $−12,985–$8959). Victims have 7.33 quality-adjusted life years (QALYs) in the intervention arm compared to 7.20 QALYs in the status quo arm, resulting in an incremental gain of 0.13 (95% CI = −0.73–2.19) QALYs. The incremental cost-effectiveness ratio (ICER) was $16,960/QALY gained (95% CI = cost-saving–$69,065/QALY). Discussion: The mobilization of private AVs in the setting of an earthquake has the potential to save money and reduce the loss of life. AVs may advance emergency management competencies.

Single-shot Coded Diffraction System for 3D Object Shape Estimation

The three-dimensional (3D) shape reconstruction problem of an object is a task of high interest in autonomous vehicles, detection of moving objects, and precision agriculture. A common methodology to recover the 3D shape of an object is using its optical phase. However, this approach involves solving a non-convex computationally demanding inverse problem known as phase retrieval (PR) in a setup that records coded diffraction patterns (CDP). Usually, the acquisition of several snapshots from the scene is required to solve the PR problem. This work proposes a single-shot 3D shape estimation technique using the optical phase of the object from CDP. The presented approach consists on accurately estimating the optical phase of the object by low-passfiltering the leading eigenvector of a carefully constructed matrix. Then, the estimated phase is used to infer the 3D object shape. It is important to mention that the estimation procedure does not involve a full time demanding reconstruction of the objects. Numerical results on synthetic data demonstrate that the proposed methodology closely estimates the 3D surface of an object with a normalized Mean-Square-Error of up to 0.27, under both noiseless and noisy scenarios. Additionally, the proposed method requires up to 60% less measurements to accurately estimate the 3D surface compared to a state-of-the-art methodology.

Capturing the behavioural determinants behind the adoption of autonomous vehicles: Conceptual frameworks and measurement models to predict public transport, sharing and ownership trends of self-driving cars

Autonomous Vehicles (AVs) have the potential to make motorized transport safer and more sustainable, by integrating clean technologies and supporting flexible shared-mobility services. Leveraging this new form of transport to transform mobility in cities will depend fundamentally on public acceptance of AVs, and the ways in which individuals choose to use them, to meet their daily travel needs. Empirical studies exploring public attitudes towards automated driving technologies and interest in AVs have emerged in the last few years. However, within this strand of research there is a paucity of theory-driven and behaviourally consistent methodologies to unpack the determinants of user adoption decisions with respect to AVs. In this paper, we seek to fill this gap, by advancing and testing four conceptual frameworks which could be deployed to capture the range of possible behavioural influences on individuals’ AV adoption decisions. The frameworks integrate socio-demographic variables and relevant latent behavioural factors, including perceived benefits and perceived ease of use of AVs, public fears and anxieties regarding AVs, subjective norm, perceived behavioural control, and attitudinal factors covering the environment, technology, collaborative consumption, public transit and car ownership. We demonstrate the utility and validity of the frameworks, by translating the latent variables into indicator items in a structured questionnaire, and administering it online to a random sample of adult individuals (n = 507). Using the survey data in confirmatory factor analyses, we specify and demonstrate scale reliability of indicator items, and convergent and discriminant validity of relationships among latent variables. Ultimately, we advance four measurement models. These theory-grounded measurement models are intended for application in research aimed at understanding and predicting (a) AV interest and adoption intentions, and (b) user adoption decisions regarding three different AV modes: ownership, sharing and public transport.

IEEE 802.11bd & 5G NR V2X: Evolution of Radio Access Technologies for V2X Communications

With the rising interest in autonomous vehicles, developing radio access technologies (RATs) that enable reliable and low-latency vehicular communications has become of paramount importance. Dedicated short-range communications (DSRCs) and cellular V2X (C-V2X) are two present-day technologies that are capable of supporting day-1 vehicular applications. However, these RATs fall short of supporting communication requirements of many advanced vehicular applications, which are believed to be critical in enabling fully autonomous vehicles. Both the DSRC and C-V2X are undergoing extensive enhancements in order to support advanced vehicular applications that are characterized by high reliability, low latency, and high throughput requirements. These RAT evolutions-the IEEE 802.11bd for the DSRC and NR V2X for C-V2X-can supplement today's vehicular sensors in enabling autonomous driving. In this paper, we survey the latest developments in the standardization of 802.11bd and NR V2X. We begin with a brief description of the two present-day vehicular RATs. In doing so, we highlight their inability to guarantee the quality of service requirements of many advanced vehicular applications. We then look at the two RAT evolutions, i.e., the IEEE 802.11bd and NR V2X, outline their objectives, describe their salient features, and provide an in-depth description of key mechanisms that enable these features. While both, the IEEE 802.11bd and NR V2X, are in their initial stages of development, we shed light on their preliminary performance projections and compare and contrast the two evolutionary RATs with their respective predecessors.

Shared versus private mobility: Modeling public interest in autonomous vehicles accounting for latent attitudes

Emerging autonomous vehicles (AVs) and shared mobility systems per se will transform urban passenger transportation. Coupled together, shared AVs (SAVs) can facilitate widespread use of shared mobility services by providing flexible public travel modes comparable to private AV. Hence, it may be conjectured that future urban mobility is likely an on-demand service and AV private ownership is unappealing. Nonetheless, it is still unclear what observable and latent factors will drive public interest in (S)AVs, the answer to which will have important implications on transportation system performance. This paper aims to jointly model public interest in private AVs and multiple SAV configurations (carsharing, ridesourcing, ridesharing, and access/egress mode) in daily and commute travels with explicit treatment of the correlations across the (S)AV types. To this end, multivariate ordered outcome models with latent variables are employed, whereby latent attitudes and preferences describing traveler safety concern about AV, green travel pattern, and mobility-on-demand savviness are accounted for using structural and measurement equations. Drawing from a stated preference survey in the State of Washington, important insights are gained into the potential user groups based on the socio-economic, built environment, and daily/commute travel behavior attributes. Key policies are also offered to promote public interest in (S)AVs by scrutinizing the marginal effects of the latent variables.