Given the recent advancements in autonomous driving functions, one of the main challenges is safe and efficient operation in complex traffic situations such as road junctions. There is a need for comprehensive testing, either in virtual simulation environments or on real-world test tracks. This paper presents a novel data analysis method including the preparation, analysis and visualization of car crash data, to identify the critical pre-crash scenarios at T- and four-legged junctions as a basis for testing the safety of automated driving systems. The presented method employs k-medoids to cluster historical junction crash data into distinct partitions and then applies the association rules algorithm to each cluster to specify the driving scenarios in more detail. The dataset used consists of 1056 junction crashes in the UK, which were exported from the in-depth "On-the-Spot" database. The study resulted in thirteen crash clusters for T-junctions, and six crash clusters for crossroads. Association rules revealed common crash characteristics, which were the basis for the scenario descriptions. The results support existing findings on road junction accidents and provide benchmark situations for safety performance tests in order to reduce the possible number parameter combinations.
This paper presents a new validation method for automated driving systems at road junctions. The method comprises the clustering of critical traffic scenarios at junctions as well as a simulation and evaluation framework to validate those scenarios. The safety performance indicators selected and implemented in the framework can be seen as a new reference for conducting virtual tests at junctions. The applicability of the framework is demonstrated by an experiment based on a selected car-to-car collision scenario. Considering the current progression of automated transport, this work is highly relevant for virtual testing procedures and is an important step towards approval and certification of automated vehicles.
This paper investigates the use of smartphones in vehicle fleets for identifying high-risk locations in a road network, before a crash may have happened. A novel method is proposed on how to use smartphone GPS and motion sensor data to automatically recognize critical car driving situations and near-misses such as emergency braking, evasion manoeuvres or sudden driving speed changes. In the area of Vienna, Austria, approximately 200 hours of driving data were collected with a dedicated smartphone app, from about 100 drivers covering more than 8,000 km. Additionally, various near-miss manoeuvres were measured on a closed test track under controlled conditions. In post-processing, this data was analysed in terms of driver-specific thresholds for critical driving situations. Results show that by using this modelling approach, critical situations can be accurately identified and geographically located with smartphones. An interface to traffic management would allow near-miss information to be used along accident data in the improvement of safety and efficiency of a traffic system. A combination of the proposed method with digital maps enables future applications for traffic and fleet managers, such as a "road safety hazard map".
Previous numerical simulations have suggested that the area adjacent to Itaipu Lake in Southern Brazil is significantly affecting the local thermal regime through development of a lake breeze. This has led to concerns that soybean growth and development, and consequently yield, has been affected by the creation of the artificial lake in this important agricultural region, but a systematic climatological study of the thermal effects of Itaipu Lake has not been conducted. The objectives of this study were to assess the spatial pattern of minimum and maximum air temperatures in a 10-km-wide area adjacent to Itaipu Lake as affected by distance from the water. Measurements were conducted over 3 years in seven transects along the shore of Itaipu Lake, with five weather stations placed in each transect. Phenological observations in soybean fields surrounding the weather stations were also conducted. Generalized additive models for location, scale, and shape (GAMLSS) analysis indicated no difference in the temperature time series as distance from water increased. Semivariograms showed that the random components in the air temperature were predominant and that there was no spatial structure to the signal. Wind direction measured over the three growing seasons demonstrated that, on average, the development of a lake breeze is limited to a few locations and a few hours of the day, supporting the temporal and spatial analysis. Phenological observations did not show differences in the timing of critical soybean stages. We suggest that the concerns that soybean development is potentially affected by the presence of Itaipu Lake are not supported by the thermal environment observed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.