1,721,036 research outputs found
Validation of driving behaviour as a step towards the investigation of Connected and Automated Vehicles by means of driving simulators
Full text linkConnected and Automated Vehicles (CAVs) are likely to become an integral part of the traffic stream within the next few years. Their presence is expected to greatly modify mobility behaviours, travel demands and habits, traffic flow characteristics, traffic safety and related external impacts. Tools and methodologies are needed to evaluate the effects of CAVs on traffic streams, as well as the impact on traffic externalities. This is particularly relevant under mixed traffic conditions, where human-driven vehicles and CAVs will interact. Understanding technological aspects (e.g. communication protocols, control algorithms, etc.) is crucial for analysing the impact of CAVs, but the modification induced in human driving behaviours by the presence of CAVs is also of paramount importance. For this reason, the definition of appropriate CAV investigations methods and tools represents a key (and open) issue. One of the most promising approaches for assessing the impact of CAVs is operator in the loop simulators, since having a real driver involved in the simulation represents an advantageous approach. However, the behaviour of the driver in the simulator must be validated and this paper discusses the results of some experiments concerning car-following behaviour. These experiments have included both driving simulators and an instrumented vehicle, and have observed the behaviours of a large sample of drivers, in similar conditions, in different experimental environments. Similarities and differences in driver behaviour will be presented and discussed with respect to the observation of one important quantity of car-following, the maintained spacing
Lernen von Fahrermodellen zur Prognose urbaner Verkehrssituationen
Full text linkAn automated vehicle needs to be able to predict the future evolution of a perceived traffic situation to safely and comfortably interact with surrounding vehicles. This work focuses on generating predictions by executing a traffic simulation.
The advantage of this simulation-based prediction is that the predictions of all vehicles are constructed simultaneously and can interact with each other. Moreover, conditional predictions become possible, e.g., "How would the traffic situation evolve, if the automated vehicle merges in front of or behind another vehicle?"
The behavior model is crucial for the accuracy of the prediction. Therefore, this thesis investigates three approaches to learning a behavior model: Multi-Step Behavior Cloning, Reinforcement Learning, and Inverse Reinforcement Learning.
For Multi-Step Behavior Cloning, the behavior model is trained such that it selects an action sequence, and hence a trajectory, as similar as possible to human drivers, starting from the same initial situation. The training requires a differentiable simulation environment, which is introduced in this work.
In contrast, the training goal of Reinforcement Learning (RL) is to maximize a hand-defined reward function. With this, explicit goals can be formulated, such as avoiding collisions, remaining on the road, and maintaining safety distances. A modification of the method is proposed to represent different driving styles with one single behavior model, e.g., sporty or careful driving.
To model human driving with RL, the reward function must be adapted until the resulting trajectories are similar enough to human trajectories. This tedious procedure can be automatized with Inverse Reinforcement Learning (IRL). To this end, Adversarial Inverse Reinforcement Learning (AIRL) is employed. With the reconstructed reward function, the behavior model is trained in additional fictional critical situations to obtain a more robust model.
Finally, all trained models are compared under equal conditions in an untrained roundabout.
The IRL algorithms achieve the best results with collision rates below 1% and root mean squared prediction errors (RMSE) below 22m. RL and IRL reduce the collision rate compared to Behavior Cloning, because they directly penalize collisions beyond the goal of pure imitation.Ein automatisiertes Fahrzeug muss die Entwicklung einer wahrgenommenen Verkehrssituation vorhersagen können, damit es sicher und komfortabel mit anderen Fahrzeugen interagieren kann.
Diese Arbeit untersucht verschiedene Methoden, um Vorhersagen mit einer Simulation der Situation zu erzeugen.
Der simulationsbasierte Ansatz ist vorteilhaft, weil die Vorhersagen aller Fahrzeuge gleichzeitig aufgebaut werden und aufeinander reagieren können. Außerdem werden bedingte Vorhersagen möglich, z.B. "Wie entwickelt sich die Situation, wenn sich das automatisierte Fahrzeug vor oder hinter einem anderen Fahrzeug einfädelt?"
Das Verhaltensmodell der simulierten Fahrzeuge hat entscheidenden Einfluss auf die Genauigkeit der Vorhersage. Daher befasst sich diese Dissertation mit drei Ansätzen, um ein Verhaltensmodell zu lernen: Multi-Step Behavior Cloning, Reinforcement Learning und Inverse Reinforcement Learning.
Bei Multi-Step Behavior Cloning wird das Verhaltensmodell so trainiert, dass es ausgehend von derselben Ausgangssituation eine möglichst ähnliche Aktionsfolge und damit Trajektorie wie ein menschlicher Fahrer wählt. Für das Training wird eine differenzierbare Simulationsumgebung benötigt, die in dieser Arbeit vorgestellt wird.
Im Gegensatz dazu ist das Trainingsziel bei Reinforcement Learning (RL) die Maximierung einer händisch definierten Belohnungsfunktion. So können explizite Ziele vorgegeben werden, z.B., dass Fahrzeuge Kollisionen vermeiden, auf der Fahrbahn bleiben und Sicherheitsabstände einhalten. Die Methode wird erweitert, um mit einem Verhaltensmodell unterschiedliche Fahrverhalten zu repräsentieren, z.B. sportlichere oder vorsichtigere Fahrer.
Um menschliches Fahrverhalten mit RL nachzubilden, muss die Belohnungsfunktion so lange angepasst werden, bis die resultierenden Trajektorien ähnlich wie echte Trajektorien aussehen. Dieser aufwändige Prozess wird von Methoden des Inverse Reinforcement Learning (IRL) automatisiert. Hierfür wird unter anderem Adversarial Inverse Reinforcement Learning (AIRL) verwendet. Mit der rekonstruierten Belohnungsfunktion wird das Verhaltensmodell außerdem in fiktiven kritischen Situationen trainiert, um eine höhere Robustheit des Modells zu erreichen.
Abschließend werden alle trainierten Modelle unter gleichen Bedingungen in einem untrainierten Kreisverkehr verglichen. Hierbei schneiden die IRL-Algorithmen bei 10s-Vorhersagen mit Kollisionsraten unter 1% und Vorhersagefehlern (RMSE) unter 22m am besten ab.
RL und IRL verringern die Kollisionsrate im Vergleich zu Behavior Cloning, weil neben dem Ziel der Imitation des Verhaltens auch Kollisionen direkt bestraft werden
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Modeling driver control behavior in both routine and near-accident driving
Full text linkBuilding on ideas from contemporary neuroscience, a framework is proposed in which drivers’ steering and pedal behavior is modeled as a series of individual control adjustments, triggered after accumulation of sensory evidence for the need of an adjustment, or evidence that a previous or ongoing adjustment is not achieving the intended results. Example simulations are provided. Specifically, it is shown that evidence accumulation can account for previously unexplained variance in looming detection thresholds and brake onset timing. It is argued that the proposed framework resolves a discrepancy in the current driver modeling literature, by explaining not only the short-latency, well-tuned, closed-loop type of control of routine driving, but also the degradation into long-latency, ill-tuned open-loop control in more rare, unexpected, and urgent situations such as near-accidents
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
