Logo

Publikacije (46044)

Nazad
Zlatan Ajanović, Bakir Lacevic, Barys Shyrokau, M. Stolz, M. Horn

This paper presents a framework for fast and robust motion planning designed to facilitate automated driving. The framework allows for real-time computation even for horizons of several hundred meters and thus enabling automated driving in urban conditions. This is achieved through several features. Firstly, a convenient geometrical representation of both the search space and driving constraints enables the use of classical path planning approach. Thus, a wide variety of constraints can be tackled simultaneously (other vehicles, traffic lights, etc.). Secondly, an exact cost-to-go map, obtained by solving a relaxed problem, is then used by A*-based algorithm with model predictive flavour in order to compute the optimal motion trajectory. The algorithm takes into account both distance and time horizons. The approach is validated within a simulation study with realistic traffic scenarios. We demonstrate the capability of the algorithm to devise plans both in fast and slow driving conditions, even when full stop is required.

Weilian Song, Scott Workman, Armin Hadžić, Xu Zhang, Eric Green, Mei Chen, R. Souleyrette, Nathan Jacobs

This paper addresses the task of road safety assessment. An emerging approach for conducting such assessments in the United States is through the US Road Assessment Program (usRAP), which rates roads from highest risk (1 star) to lowest (5 stars). Obtaining these ratings requires manual, fine-grained labeling of roadway features in streetlevel panoramas, a slow and costly process. We propose to automate this process using a deep convolutional neural network that directly estimates the star rating from a street-level panorama, requiring milliseconds per image at test time. Our network also estimates many other roadlevel attributes, including curvature, roadside hazards, and the type of median. To support this, we incorporate taskspecific attention layers so the network can focus on the panorama regions that are most useful for a particular task. We evaluated our approach on a large dataset of real-world images from two US states. We found that incorporating additional tasks, and using a semi-supervised training approach, significantly reduced overfitting problems, allowed us to optimize more layers of the network, and resulted in higher accuracy.

Reproductive ability of the domestic cats, based on the experience, varies, which depends on the range of both internal and external factors. Like their wild counterparts, the domestic cats reproduce several times a year; hence, they are polyestric animals. Actuality of endocrinological testing, considering the extreme importance of the ovaries in the neuroendocrinology system and the importance of impact of different factors, climatic factors in particular from the aspect of the onset and detection of estrus, prompted us to determine the frequency of estrus during the period of one year, based on the level of estradiol concentration in blood serum.

A. Dedic, R. Braam

Every day physicians balance clinical information with medical test results when confronted with symptomatic patients. Sometimes they find themselves in an apparently contradicting situation in which a patient has persistent complaints while his or her medical tests are normal. In this issue of the Netherlands Heart Journal, Yokota et al. addressed this matter in the setting of stable angina pectoris [1]. The authors performed a retrospective analysis of all patients who had undergone nuclear myocardial perfusion imaging in their centre and selected those with a normal scan but with persistent or worsening complaints that compelled the treating physician to order an invasive angiogram. Out of more than 11,000 patients, 229 fulfilled the study criteria. The authors reported that in this highly selected group of patients a fairly high percentage (34%) had significant coronary artery disease despite a normal perfusion scan, which was defined as >50% stenosis in the left main coronary artery or >70% stenosis for other segments. In the majority of cases, it concerned single-vessel disease (60%), while only a minority (17%) had left main coronary artery disease or three-vessel disease. Coronary revascularisation was performed in 90% and most of them were free of symptoms after 7 years of follow-up. The authors found that older age, male sex, typical angina and previous PCI are independent predictors for the presence of severe stenosis on invasive angiography following a normal myocardial perfusion scan. As the study was conducted in a ‘pre-FFR era’ there was a low rate of invasive functional testing, which in part might explain the discordancy.

This paper presents a research on ash-related problems and emissions during co-firing low-rank Bosnian coals with different kinds of biomass; in this case woody sawdust and herbaceous energy crops Miscanthus. An entrained-flow drop tube furnace was used for the tests, varying fuel portions at a high co-firing ratio up to 30%wt woody sawdust and 10%wt Miscanthus in a fuel blend. The tests were supposed to optimize the process temperature, air distribution (including OFA) and fuel distributions (reburning) as function of SO2 and NOx emissions as well as efficiency of combustion process estimated through the ash deposits behaviors, CO emissions and unburnt. The results for 12 co-firing fuel combinations impose a reasonable expectation that the coal/biomass/Miscanthus blends could be successfully run under certain conditions not producing any serious ash-related problems. SO2 emissions were slightly higher when higher content of woody biomass was used. Oppositely, higher Miscanthus percentage in the fuel mix slightly decreases SO2 emissions. NOx emissions generally decrease with an increase of biomass co-firing rate. The study suggests that co-firing Bosnian coals with woody sawdust and Miscanthus shows promise at higher co-firing ratios for pulverized combustion, giving some directions for further works in co-firing similar multi-fuel combinations.

The theoretical legacy of Zygmunt Bauman is an inexhaustible source of inspiration for sociological analysis, particularly bearing in mind the scope of his work and the diverse range of modern day problems that this British-Polish author dealt with. The first part of this article examines the question of personal identity in liquid modernity, which is the starting point of Bauman’s work. Similar to some other authors, Bauman discusses the paradox of the individual who is not free in an individualized society. Bauman’s diagnosis carries pessimistic features which in some places correspond to insights developed in classical sociology. Bauman makes occasional and sporadic incursions into the pitfalls of conservative thought, particularly in relation to the dichotomies of individual versus community, individualism versus togetherness, and egoism versus solidarity. However, it seems that the author manages to skilfully avoid the inherent theoretical traps of sociology, turning towards cosmopolitan theory. The second part of this article presents the thesis that Bauman’s thought is in fact cosmopolitan, especially bearing in mind his final public appearances and writings. This argument is based in his description of global society that is simultaneously integrating and developing, and dramatically disintegrating and regressing. Bauman writes about violent killings and expulsions of people in the 21st century and their inability to find refuge in the Western and democratic world that promotes human rights. Recalling the crucial cosmopolitan principles of solidarity and hospitality, Bauman makes an appeal to progressive forces to consolidate and work on opening and reaffirming the “cosmopolitan condition” of contemporary society.

É. D. Souza, Jonatas Tavares Da Silva, R. Silva, Maria Francineila Pinheiro dos Santos

M. E. A. Pio, P. S. Mageste, C. Costa, Amanda Ferreira Pratti, Rogério de Oliveira Gonçalves

Nema pronađenih rezultata, molimo da izmjenite uslove pretrage i pokušate ponovo!

Pretplatite se na novosti o BH Akademskom Imeniku

Ova stranica koristi kolačiće da bi vam pružila najbolje iskustvo

Saznaj više