scispace - formally typeset
Search or ask a question
Author

Teng Wang

Bio: Teng Wang is an academic researcher from University of Kentucky. The author has contributed to research in topics: Level crossing & Transportation planning. The author has an hindex of 6, co-authored 19 publications receiving 141 citations. Previous affiliations of Teng Wang include Iowa State University & Texas A&M Transportation Institute.

Papers
More filters
Journal ArticleDOI
TL;DR: In this paper, the authors explore opinions regarding the perceived benefits and challenges of AVs among vulnerable road users, in particular pedestrians and bicyclists, and evaluate whether interaction experiences with AVs influence perceptions among vulnerable pedestrians and cyclists.

153 citations

Journal ArticleDOI
TL;DR: Vivid is presented, a mobile device-friendly indoor localization and navigation system that leverages visual cues as the cornerstone of localization and overcomes the difficulties brought by resource-intensive image processing tasks by leveraging the computation power at the extreme internet edges.
Abstract: Indoor localization and navigation have a great potential of application, especially in large indoor spaces where people tend to get lost. The indoor localization problem is the fundamental of an indoor navigation system. Existing research and commercial efforts have leveraged wireless-based approaches to locate users in indoor environments. However, the predominant wireless-based approaches, such as WiFi and Bluetooth, are still not satisfactory, either not supporting commodity devices, or being vulnerable to environmental changes. These issues make them hard to deploy and maintain. In this paper, we present Vivid, a mobile device-friendly indoor localization and navigation system that leverages visual cues as the cornerstone of localization. By leveraging the computation power at the extreme internet edges, Vivid to a large extent overcomes the difficulties brought by resource-intensive image processing tasks. We propose a grid-based algorithm that transforms the feature map into a grid, with which finding the path between two positions can be easily obtained. We also leverage deep learning techniques to assist in automatic map maintenance to adapt to the visual changes and make the system more robust. With edge computing, user privacy is preserved since the visual data is mainly processed locally and detected dynamic objects are removed immediately without saving to databases. The evaluation results show that: i) our system easily outperforms the existing solutions on COTS devices in localization accuracy, yielding decimeter-level error; ii) our choice of the system architecture is scalable and optimal among the available ones; iii) the automatic map maintenance mechanism effectively ameliorates the localization robustness of the system.

15 citations

Proceedings ArticleDOI
02 Apr 2014
TL;DR: In this paper, the authors report on the development of an accurate, low-cost and readily deployable sensor capable of rapid collection of 3D surface models of the crossing in its present state.
Abstract: Quality of surface is an important aspect affecting both the safety and the performance of at-grade rail-highway crossings. Roughness may increase the risk of crashes for both trains and automobiles. Varying grades in crossing profiles increase the likelihood of high-centered crossing collisions between train and truck [1]. The US DOT Railroad Highway Grade Crossing Handbook [2] suggests that rough surfaces could distract a driver’s attention from oncoming trains and that the unevenness of the crossing could result in a driver losing control of their vehicle resulting in a crash.No quantitative method currently exists to quickly and economically assess the condition of rail crossings in order to evaluate the long term performance of crossings and set a quantitative trigger for their rehabilitation. The conventional method to measure the surface of quality of crossings is based on expert judgment, whereby crossing surfaces are classified as poor, fair or good after an inspector visits and drives over the crossing. However, actual condition of the crossing could be different from the subjective rating. Poor condition rating crossings may not always present the most cost-effective locations for preventive maintenance to lower overall life-cycle costs. Conventional ratings may derive from driving a passenger car of pickup once over the crossing. Effects of various speed, on various vehicles (suspension), and at various places (laterally) cannot be determined or even estimated except at the smoothest of crossings. A quantifiable and extensible procedure is desired.With rapid advances in computer science, 3D sensing and imaging technologies, it seems logical that a cost-effective quantitative method could be developed to determine the need to rehabilitate rail crossings and assess long term performance. Fundamental to the quantification of crossing condition is the acquisition of an accurate 3D surface model of the crossing in its present state. This paper reports on the development of an accurate, low cost and readily deployable sensor capable of rapid collection of this 3D surface. The research is seen as a first step towards automating the crossing inspection process, ultimately leading to the quantification and estimation of future performance of rail crossing.© 2014 ASME

10 citations

Proceedings ArticleDOI
02 Jul 2019
TL;DR: An innovative machine learning approach to identify crash risk factors and find solutions to reduce the intersection-related crash frequency and severity caused by harmful lane-changing behaviors found no significant differences in the VAT values between the VAT model and the Lasso-LARS regression model.

8 citations

Journal ArticleDOI
TL;DR: In this article, the authors used 3D surface models and a customized vehicle dynamic model to predict accelerations experienced by highway vehicles using the crossing and validated with field-measured accelerations.
Abstract: Annually, more than 2,000 rail-highway crossing crashes in the United States result in nearly 300 fatalities. Crossing roughness is a concern for the motoring public from a comfort and vehicle maintenance perspective, and to highway authorities from a maintenance perspective. Roughness may even increase the risk of crossing crashes. However, with 216,000 rail-highway grade crossings in the United States, maintenance management is a large undertaking. Crossings deteriorate over time, sometimes rapidly, and life-cycle costs increase without preventive maintenance. However, though methods are available to quantify highway roughness, no method exists to quantitatively assess the condition of rail crossings. Because conventional inspection relies on qualitative judgment based on an inspector's perception of the crossing, effect on different vehicles and perception by other drivers are unknown. Further, roughness may be due to as-built geometry, deterioration, or a combination of both. A quantifiable and extensible procedure is thus desired. The article details the use of 3D surface models and a customized vehicle dynamic model to predict accelerations experienced by highway vehicles using the crossing. The model is validated with field-measured accelerations. RESULTS indicate good agreement between modeled and measured accelerations for a test vehicle at several speed ranges at two different locations. Language: en

7 citations


Cited by
More filters
01 Jan 2011
TL;DR: Results suggest that the likelihood of fatalities and severe injury is estimated to rise with the number of trailers, but fall with the truck length and gross vehicle weight rating (GVWR), and total crash costs of LCVs are lower than those of other trucks.
Abstract: Long-combination vehicles (LCVs) have significant potential to increase economic productivity for shippers and carriers by decreasing the number of truck trips, thus reducing costs. However, size and weight regulations, triggered by safety concerns and, in some cases, infrastructure investment concerns, have prevented large-scale adoption of such vehicles. Information on actual crash performance is needed. To this end, this work uses standard and heteroskedastic ordered probit models, along with the United States’ Large Truck Crash Causation Study, General Estimates System, and Vehicle Inventory and Use Survey data sets, to study the impact of vehicle, occupant, driver, and environmental characteristics on injury outcomes for those involved in crashes with heavy-duty trucks. Results suggest that the likelihood of fatalities and severe injury is estimated to rise with the number of trailers, but fall with the truck length and gross vehicle weight rating (GVWR). While findings suggest that fatality likelihood for two-trailer LCVs is higher than that of single-trailer non-LCVs and other trucks, controlling for exposure risk suggest that total crash costs of LCVs are lower (per vehicle-mile traveled) than those of other trucks.

163 citations

Journal ArticleDOI
TL;DR: In this article, a review of the existing studies on AV acceptance is presented, and the authors find that people in Europe and Asia have substantial differences in attitudes toward AVs and that safety is one of the most concerned factors of AVs.
Abstract: Excessive dependence on autonomous vehicles (AVs) may exacerbate traffic congestion and increase exhaust emissions in the future. The diffusion of AVs may be significantly affected by the public’s acceptance. A few factors that may affect people’s acceptance of AVs have been researched in the existing studies, one-third of which cited behavioral theories, while the rest did not. A total of seven factors with behavior theories are screened out that significantly affect the acceptance intention, including perceived ease of use, attitude, social norm, trust, perceived usefulness, perceived risk, and compatibility. Six factors without behavior theories are summed up that affect AV acceptance, namely safety, performance-to-price value, mobility, value of travel time, symbolic value, and environmentally friendly. We found that people in Europe and Asia have substantial differences in attitudes toward AVs and that safety is one of the most concerned factors of AVs by scholars and respondents. Public acceptance of the different types of AVs and consumers’ dynamic preferences for AVs are highlighted in the review too. The quality of literature is systematically assessed based on previously established instruments and tailored for the current review. The results of the assessment show potential opportunities for future research, such as the citation of behavior theories and access to longitudinal data. Additionally, the experimental methods and the utilization of mathematical and theoretical methods could be optimized.

124 citations

Journal ArticleDOI
TL;DR: In this article, the authors developed a theoretical model to identify the latent factors influencing public acceptance of autonomous vehicles and examined their interrelationships by applying three diverse research paradigms anchoring on innovation diffusion, customer utility and social psychology.

115 citations

Journal ArticleDOI
TL;DR: How social influence, system characteristics, and individual factors determine individual acceptance of autonomous driving is revealed, with implications for practitioners.

95 citations