A Driving Scenario Representation for Scalable Real-Data Analytics with Neural Networks

Ries, Lennart; Langner, Jacob; Otten, Stefan; Bach, Johannes; Sax, Eric

doi:10.1109/ivs.2019.8813881

Cited by 13 publications

(7 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Examples of supervised prediction tasks used include modified Random Forests [17], Logistic Regressions [28], and Recurrent Neural Networks (RNNs) [40]. Furthermore, combinations of several approaches exist, such as the combination of supervised and unsupervised learning, to identify scenarios [18], [21], [60].…”

Section: A Process Of Scenario Generationmentioning

confidence: 99%

“…The categorization of representativity is performed between RA and RB, with RA depicting representativity within each scenario set or space, which is the case in 25 generation approaches [16], [23], [24], [29]- [32], [40]- [45], [47], [48], [53], [55], [60], [65]- [69], [72], [74]. To achieve this goal, filtering of the generated scenario set to reduce the number of scenarios is frequently performed.…”

Section: B Categorizationmentioning

confidence: 99%

“…Repeated algorithms in detail [34], [36], [37], [45], [50], [58]. Frequently applied algorithms for scenario detection include CNNs [17], [18], [25], [56], and RNNs [51], [60], [72]. RNNs use the results of the current step as the input for the next step, and may not require labelled data for training.…”

Section: B Categorizationmentioning

confidence: 99%

See 2 more Smart Citations

Categorizing Data-Driven Methods for Test Scenario Generation to Assess Automated Driving Systems

Bäumler,

Linke,

Prokop

2024

IEEE Access

View full text Add to dashboard Cite

This survey aims to provide an overview of various methods for generating data-driven test scenarios for assessing automated driving systems (ADS). The survey updates the overall process of scenario generation and categorizes the current methods using a systematic literature review of 64 studies identified between 2017 and 01/2023. Overall, we demonstrate that the data-driven scenario generation process should be updated by another process step, scenario fusion, leading to seven process steps: (1) scope definition, (2) primary data source selection, (3) primary data collection, (4) scenario identification, ( 5) scenario fusion, ( 6) scenario generation, and (7) scenario evaluation. "Scenario fusion" aims to fuse scenarios identified in different data sources for a better coverage of the ADSs' operational design domains (ODDs) and a more comprehensive scenario description. Moreover, we introduce an improved definition for the representativity of test scenario catalogs, which helps improve the collection of traffic data using sampling plans. Also, we show that real driving and police accident data are the most commonly used data input sources. Besides, we illustrate that the ODD is often not defined. Finally, we discuss that the standardization of test scenario generation is difficult because most methods do not address specific ADSs and test environments, and do not provide standardized interfaces. Overall, we recommend comparing existing approaches using the same input data and researching the mutual supplementation of the existing methods. Finally, pre-defined case studies, further standardized terminology, and standards for test execution and evaluation can help speed up the standardization process.INDEX TERMS Advanced driver assistance systems, autonomous driving, system validation, vehicle safety I. INTRODUCTION

show abstract

Section: A Process Of Scenario Generationmentioning

confidence: 99%

Section: B Categorizationmentioning

confidence: 99%

See 1 more Smart Citation

Categorizing Data-Driven Methods for Test Scenario Generation to Assess Automated Driving Systems

Bäumler,

Linke,

Prokop

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Многие исследования используют модели, основанные на методах управления поведением БТС (например, [13,15,21]). Такие подходы позволяют исследовать сценарии вождения (например, [19], [22]) и улучшать маневренность БТС [8,11]. С другой стороны, многие особенности, обусловленные психологией водителей ТС, влияющие на их поведение и маневрирование в интенсивном потоке, изучены недостаточно.…”

Section: Introductionunclassified

Improving the Maneuverability of Unmanned Vehicles at Various Road Network Configurations

Андраник

Левон²

2021

Artificial Societies

View full text Add to dashboard Cite

The article presents an approach to improving maneuverability of unmanned ground vehicles (UGVs) interacting with manned ground vehicles (MGVs) within the proposed multi-agent intelligent transport system (MITS) at various configurations of the digital road network. With the developed models, the abilities of improving maneuverability of agent-vehicles in MITS under various control scenarios for the most important characteristics of the system, such as, the arrival intensities of agents, speeds of UGVs and MGVs, clustering methods for the traffic density estimation and intelligent maneuvering are studied. The proposed MITS is implemented based on the reconfigurable digital road network (DRN). There is proposed an approach to improving the maneuverability of the UGVs, aimed to effectively maneuvering in changing lanes and choosing the best routes to bypass traffic congestions and other problem areas with providing the required level of output traffic and minimizing the number of potential traffic accidents in the MITS. The possibilities of improving maneuverability of UGVs at various configurations of the DRN (for example, the roundabouts, "Manhattan Lattice", etc.) are studied.

show abstract

“…При этом, для моделирования характеристик транспортных систем применяются различные методы, среди которых можно выделить использование паттернов (шаблонов) поведения различных участников дорожного движения (БТС, ОТС, пешеходов и т.д.) для эффективного прогнозирования динамики развития дорожной ситуации при различных сценарных условиях [10], модели многочастичного взаимодействия [9], методы машинного обучения и нейросетевого управления [8], модели, основанные на Гауссовских процессах движения [5] и др. К сожалению, общим недостатком подобных подходов является трудность оценки устойчивости транспортной системы по отношению к множественным риск-факторам, таким как, резкое снижение видимости на дорогах и аномальное поведение других участников дорожного движения.…”

unclassified

Scenario simulation of autonomous vehicles motion in Artificial Road Network using FLAME GPU

Сумбатович

Левонович

2021

Artificial Societies

View full text Add to dashboard Cite

Рис. 1. Общая схема организации искусственной дорожной сети.Важной характеристикой агента-ТС является его личное пространство, сжимаемое и расширяемое в зависимости от плотности дорожного потока относительно данного агента. Если значение плотности, то есть количество транспортных средств, расположенных вблизи агента-ТС превышает некоторый пороговый уровень (различный для ОТС и БТС), то радиус личного пространства агента резко увеличивается (из-за паники водителя) и стремления избежать аварии, вызывая эффекты «турбулентности» и «давки» (т.е. дорожные заторы) (рис. 1).На рисунке 2 представлена визуализация правил принятия решений агентом-БТС, предложенных в данной модели, при наличии дорожных заторов.

show abstract

A Driving Scenario Representation for Scalable Real-Data Analytics with Neural Networks

Cited by 13 publications

References 13 publications

Categorizing Data-Driven Methods for Test Scenario Generation to Assess Automated Driving Systems

Categorizing Data-Driven Methods for Test Scenario Generation to Assess Automated Driving Systems

Improving the Maneuverability of Unmanned Vehicles at Various Road Network Configurations

Scenario simulation of autonomous vehicles motion in Artificial Road Network using FLAME GPU

Contact Info

Product

Resources

About