Terahertz Sensor via Ultralow-Loss Dispersion-Flattened Polymer Optical Fiber: Design and Analysis

A novel cyclic olefin copolymer (COC)-based polymer optical fiber (POF) with a rectangular porous core is designed for terahertz (THz) sensing by the finite element method. The numerical simulations showed an ultrahigh relative sensitivity of 89.73% of the x-polarization mode at a frequency of 1.2 THz and under optimum design conditions. In addition to this, they showed an ultralow confinement loss of 2.18 × 10−12 cm−1, a high birefringence of 1.91 × 10−3, a numerical aperture of 0.33, and an effective mode area of 1.65 × 105 μm2 was obtained for optimum design conditions. Moreover, the range dispersion variation was within 0.7 ± 0.41 ps/THz/cm, with the frequency range of 1.0–1.4 THz. Compared with the traditional sensor, the late-model sensor will have application value in THz sensing and communication.

Powered flight in hatchling pterosaurs: evidence from wing form and bone strength

Competing views exist on the behaviour and lifestyle of pterosaurs during the earliest phases of life. A ‘flap-early’ model proposes that hatchlings were capable of independent life and flapping flight, a ‘fly-late’ model posits that juveniles were not flight capable until 50% of adult size, and a ‘glide-early’ model requires that young juveniles were flight-capable but only able to glide. We test these models by quantifying the flight abilities of very young juvenile pterosaurs via analysis of wing bone strength, wing loading, wingspan and wing aspect ratios, primarily using data from embryonic and hatchling specimens of Pterodaustro guinazui and Sinopterus dongi. We argue that a young Sinopterus specimen has been mischaracterised as a distinct taxon. The humeri of pterosaur juveniles are similar in bending strength to those of adults and able to withstand launch and flight; wing size and wing aspect ratios of young juveniles are also in keeping with powered flight. We therefore reject the ‘fly-late’ and ‘glide-early’ models. We further show that young juveniles were excellent gliders, albeit not reliant on specialist gliding. The wing forms of very young juveniles differ significantly from larger individuals, meaning that variation in speed, manoeuvrability, take-off angle and so on was present across a species as it matured. Juveniles appear to have been adapted for flight in cluttered environments, in contrast to larger, older individuals. We propose on the basis of these conclusions that pterosaur species occupied distinct niches across ontogeny.

Seamless postprocessing of multi-model NWP surface wind forecasts with deep learning

<p>Hourly wind forecasts from numerical weather prediction models suffer from a range of systematic and random errors that are to a great extent related to limitations in the model grid resolution. To correct for such biases, statistical postprocessing and downscaling procedures are commonly applied so to leverage the information provided by automatic wind measurements at the surface. More recently, such techniques have been reformulated in a machine learning framework so to profit from the increased availability of data and computational resources. The results reported in the literature are promising and call for a serious evaluation of their potential for operational forecasting.</p><p>However, there remain several scientific and more applied challenges that need to be addressed before such methods can transition to real-world applications. One such challenge relates to the availability of multiple ensemble forecasts for the same point in time and space, which raises the question of how the information can be efficiently and optimally handled during postprocessing, so to provide added value to the end-user without adding technical debt to the operational system.</p><p>We propose an approach where a single deep learning model is trained to postprocess a combination of three ensemble forecasting systems, namely the high-resolution regional COSMO model with two configurations, and the ECMWF IFS ENS global ensemble forecasting system. We will show how the training is set up to provide a robust postprocessing model that can account for real time scenarios that include missing data and late model runs, while the quality of the forecasts remains comparable to a single-model approach. We found that the flexibility of the deep learning architecture translates into a robust automatic postprocessing solution that limits the maintenance burden and improves the system’s reliability.</p>

A database development and analysis of selected side impact collisions in Toronto

This thesis is based on the initial phase of a project that developed an in-depth collision database and performed an analysis of police reported side-impact collisions for the City of Toronto intersections between 1998 and 2000. Currently, collision data exists through several different sources in Ontario. The development of a database involving the amalgamation of collision forms, the selection of data fields, and the collection of real collision data from selected, thoroughly investigated side impact collisions involving late model vehicles (1998 and newer), is described. For analysis, Statistical Analysis Software Release 8.02 was used to investigate causation and causal factors of side impact collisions. Statistically significant collision factors determined by fault propensity included apparent driver action, driver age, front seat passenger age, maximum posted speed, approximate vehicle speed, road character, and number of lanes. For intersection collision propensity, statistically significant findings included the system used, presence of flashing signals, intersection legs, roadway volume, and intersection leg road classifications. It is anticipated that the findings from this analysis can provide insight into significant factors in side-impact collisions that will be applied with greater focus to the in-depth collision database, once developed. Traffic accidents

A database development and analysis of selected side impact collisions in Toronto

This thesis is based on the initial phase of a project that developed an in-depth collision database and performed an analysis of police reported side-impact collisions for the City of Toronto intersections between 1998 and 2000. Currently, collision data exists through several different sources in Ontario. The development of a database involving the amalgamation of collision forms, the selection of data fields, and the collection of real collision data from selected, thoroughly investigated side impact collisions involving late model vehicles (1998 and newer), is described. For analysis, Statistical Analysis Software Release 8.02 was used to investigate causation and causal factors of side impact collisions. Statistically significant collision factors determined by fault propensity included apparent driver action, driver age, front seat passenger age, maximum posted speed, approximate vehicle speed, road character, and number of lanes. For intersection collision propensity, statistically significant findings included the system used, presence of flashing signals, intersection legs, roadway volume, and intersection leg road classifications. It is anticipated that the findings from this analysis can provide insight into significant factors in side-impact collisions that will be applied with greater focus to the in-depth collision database, once developed. Traffic accidents

Hermeneutic, Comparative, and Syncretic Philosophy: Or, On Ricoeurian, Confucian and Aztec Philosophy

Hermeneutic philosophy, and Paul Ricoeur’s formulation of hermeneutics in particular, faces a serious challenge, not from external sources, but from internal proponents of the program. In what might be called the Collapse Challenge, Ricoeur’s understanding of the hermeneutic circle is criticized for making use of structuralist methods that are no longer considered viable. Rather than look to replace Ricoeur’s work with an external model, the present essay draws on his late model of translation to suggest two viable paths forward beyond the Collapse Challenge. To develop these paths, the essay gives two concrete cases, one using Confucian philosophy, which is comparative, another using Aztec philosophy, which is syncretic.

Proposed Framework for Identifying and Predicting Operator Errors When using Advanced Vehicle Technologies

Advanced vehicle technologies include systems that are defined by the Society for Automotive Engineers as automated driving features or driver support features. The latter are increasingly available in late model vehicles in the form of advanced driver assistance systems (ADAS). ADAS features remove some responsibilities from drivers, but still depend on the drivers for safe operation. This can result in drivers committing errors while using ADAS, especially if their understanding of these systems, that is, their mental model, is incorrect. To understand how these systems could be used incorrectly it is necessary to have an insight into these mental models. One approach is to characterize users’ mental representations of systems based on the errors that they commit during system use. Such an approach necessitates a classification of potential errors that may be committed, and the underlying cognitive and behavioral reasons for such errors. To that end, a framework is proposed that can, among other goals, help predict user errors while using ADAS based on human factors and task analysis techniques. A methodology is detailed for mapping operator-system interactions using state diagrams, error identification techniques using task analysis are proposed, and a categorization scheme based on classic error taxonomies is described. This proposed framework can subsequently be expanded for error identification for a wider range and versions of ADAS, as well as for future automated driving systems (ADS). Moreover, the framework provides a systematic approach that can be used toward operationalizing mental models, forming the basis for structured user training, and for human-centered design of advanced vehicle technologies.