In recent years,advancements in autonomous vehicle technology have accelerated,promising safer and more efficient transportation systems.However,achieving fully autonomous driving in challenging weather conditions,par...In recent years,advancements in autonomous vehicle technology have accelerated,promising safer and more efficient transportation systems.However,achieving fully autonomous driving in challenging weather conditions,particularly in snowy environments,remains a challenge.Snow-covered roads introduce unpredictable surface conditions,occlusions,and reduced visibility,that require robust and adaptive path detection algorithms.This paper presents an enhanced road detection framework for snowy environments,leveraging Simple Framework forContrastive Learning of Visual Representations(SimCLR)for Self-Supervised pretraining,hyperparameter optimization,and uncertainty-aware object detection to improve the performance of YouOnly Look Once version 8(YOLOv8).Themodel is trained and evaluated on a custom-built dataset collected from snowy roads in Tromsø,Norway,which covers a range of snow textures,illumination conditions,and road geometries.The proposed framework achieves scores in terms of mAP@50 equal to 99%and mAP@50–95 equal to 97%,demonstrating the effectiveness of YOLOv8 for real-time road detection in extreme winter conditions.The findings contribute to the safe and reliable deployment of autonomous vehicles in Arctic environments,enabling robust decision-making in hazardous weather conditions.This research lays the groundwork for more resilient perceptionmodels in self-driving systems,paving the way for the future development of intelligent and adaptive transportation networks.展开更多
With the advantage of fast calculation and map resources on cloud control system(CCS), cloud-based predictive cruise control(CPCC) for heavy trucks has great potential to improve energy efficiency, which is significan...With the advantage of fast calculation and map resources on cloud control system(CCS), cloud-based predictive cruise control(CPCC) for heavy trucks has great potential to improve energy efficiency, which is significant to achieve the goal of national carbon neutrality. However, most investigations focus on the on-board predictive cruise control(PCC) system,lack of research on CPCC architecture under CCS. Besides, the current PCC algorithms have the problems of a single control target and high computational complexity, which hinders the improvement of the control effect. In this paper, a layered architecture based on CCS is proposed to effectively address the realtime computing of CPCC system and the deployment of its algorithm on vehicle-cloud. In addition, based on the dynamic programming principle and the proposed road point segmentation method(RPSM), a PCC algorithm is designed to optimize the speed and gear of heavy trucks with slope information. Simulation results show that the CPCC system can adaptively control vehicle driving through the slope prediction, with fuel-saving rate of 6.17% in comparison with the constant cruise control. Also,compared with other similar algorithms, the PCC algorithm can make the engine operate more in the efficient zone by cooperatively optimizing the gear and speed. Moreover, the RPSM algorithm can reconfigure the road in advance, with a 91% roadpoint reduction rate, significantly reducing algorithm complexity.Therefore, this study has essential research significance for the economic driving of heavy trucks and the promotion of the CPCC system.展开更多
Safety performance functions（SPFs） are crucial to science-based road safety management.Success in developing and applying SPFs, apart data quality and availability, depends fundamentally on two key factors： the val...Safety performance functions（SPFs） are crucial to science-based road safety management.Success in developing and applying SPFs, apart data quality and availability, depends fundamentally on two key factors： the validity of the statistical inferences for the available data and on how well the data can be organized into distinct homogeneous entities. The latter aspect plays a key role in the identification and treatment of road sections or corridors with problems related to safety. Indeed, the segmentation of a road network could be especially critical in the development of SPFs that could be used in safety management for roadway types, such as motorways（freeways in North America）, which have a large number of variables that could result in very short segments if these are desired to be homogeneous. This consequence, from an analytical point of view, can be a problem when the location of crashes is not precise and when there is an overabundance of segments with zero crashes. Lengthening the segments for developing and applying SPFs can mitigate this problem, but at a sacrifice of homogeneity. This paper seeks to address this dilemma by investigating four approaches for segmentation for motorways, using sample data from Italy. The best results were obtained for the segmentation based on two curves and two tangents within a segment and with fixed length segments. The segmentation characterized by a constant value of all original variables inside each segment was the poorest approach by all measures.展开更多
The study of impacts of down-up hill road segment on the density threshold of traffic shock formation in ring road vehicular flow is helpful to the deep understanding of sags’bottleneck effect.Sags are freeway segmen...The study of impacts of down-up hill road segment on the density threshold of traffic shock formation in ring road vehicular flow is helpful to the deep understanding of sags’bottleneck effect.Sags are freeway segments along which the gradient increases gradually in the traffic direction.The main aim of this paper is to seek the density threshold of shock formation of vehicular flow in ring road with down-up hill segment,because down-up hill roadway segment is a source to cause capacity reduction that is an attractive topic in vehicular traffic science.To seek the density threshold numerically,a viscoelastic continuum model[1]is extended and used.To solve the model equations,a fifth-order weighted essentially non-oscillatory scheme for spatial discretization,and a 3rd order Runge-Kutta scheme for time partial derivative term are used.Validation by existing observation data and the Navier-Stokes like model[2]extended as EZM is done before conducting extensive numerical simulations.For ring road vehicular flow with three separated down-up hill segments,it is found that the density threshold of shock formation decreases monotonically with the relative difference of free flow speed,this variation can be simply fitted by a third order polynomial.展开更多
Vehicle operating speed is a crucial factor for road safety,as it strictly affects occurrence and severity of crashes.Usually,85th percentile of the operating speed distributions(i.e.,V85)in free-flow traffic conditio...Vehicle operating speed is a crucial factor for road safety,as it strictly affects occurrence and severity of crashes.Usually,85th percentile of the operating speed distributions(i.e.,V85)in free-flow traffic condition is widely accepted as a base value of consistency evaluation for homogenous portion of existing roads.Although the computation of V85 is simple,many road authorities cannot collect speed data for each road.Therefore,providing prediction models could be a useful tool to investigate the relationship between V85 and road characteristics.The literature proposed several models to account it.However,to the best of our knowledge,the effects of some road geometric characteristics,road markings and signs,traffic data,type of terrain and the simultaneous consideration of different road categories on the V85 prediction were not completely analyzed.This paper fills this gap by isolating key variables that mostly affect V85.In doing so,60000+car spot speed data were collected along the county road network of the province of Brescia(Italy),and then processed by multiple regression models.The main findings show that V85 increases owing to the presence of a wider or paved shoulder,visible road median markings,a higher number of lanes and a higher percentage of cars with respect to the total traffic flow.Conversely,V85 decreases as the road axis curvature,the number of accesses and rate of forbidden overtaking increase.In addition,the presence of visible road external markings and the surrounding mountainous terrain contribute to decreasing V85.The overall findings may support road authorities to verify roads’operating conditions and,possibly,adjust the speed limits,especially for existing roads.展开更多
文摘In recent years,advancements in autonomous vehicle technology have accelerated,promising safer and more efficient transportation systems.However,achieving fully autonomous driving in challenging weather conditions,particularly in snowy environments,remains a challenge.Snow-covered roads introduce unpredictable surface conditions,occlusions,and reduced visibility,that require robust and adaptive path detection algorithms.This paper presents an enhanced road detection framework for snowy environments,leveraging Simple Framework forContrastive Learning of Visual Representations(SimCLR)for Self-Supervised pretraining,hyperparameter optimization,and uncertainty-aware object detection to improve the performance of YouOnly Look Once version 8(YOLOv8).Themodel is trained and evaluated on a custom-built dataset collected from snowy roads in Tromsø,Norway,which covers a range of snow textures,illumination conditions,and road geometries.The proposed framework achieves scores in terms of mAP@50 equal to 99%and mAP@50–95 equal to 97%,demonstrating the effectiveness of YOLOv8 for real-time road detection in extreme winter conditions.The findings contribute to the safe and reliable deployment of autonomous vehicles in Arctic environments,enabling robust decision-making in hazardous weather conditions.This research lays the groundwork for more resilient perceptionmodels in self-driving systems,paving the way for the future development of intelligent and adaptive transportation networks.
基金supported by the National Key Research and Development Program (2021YFB2501003)the Key Research and Development Program of Guangdong Province (2019B090912001)the China Postdoctoral Science Foundation (2020M680531)。
文摘With the advantage of fast calculation and map resources on cloud control system(CCS), cloud-based predictive cruise control(CPCC) for heavy trucks has great potential to improve energy efficiency, which is significant to achieve the goal of national carbon neutrality. However, most investigations focus on the on-board predictive cruise control(PCC) system,lack of research on CPCC architecture under CCS. Besides, the current PCC algorithms have the problems of a single control target and high computational complexity, which hinders the improvement of the control effect. In this paper, a layered architecture based on CCS is proposed to effectively address the realtime computing of CPCC system and the deployment of its algorithm on vehicle-cloud. In addition, based on the dynamic programming principle and the proposed road point segmentation method(RPSM), a PCC algorithm is designed to optimize the speed and gear of heavy trucks with slope information. Simulation results show that the CPCC system can adaptively control vehicle driving through the slope prediction, with fuel-saving rate of 6.17% in comparison with the constant cruise control. Also,compared with other similar algorithms, the PCC algorithm can make the engine operate more in the efficient zone by cooperatively optimizing the gear and speed. Moreover, the RPSM algorithm can reconfigure the road in advance, with a 91% roadpoint reduction rate, significantly reducing algorithm complexity.Therefore, this study has essential research significance for the economic driving of heavy trucks and the promotion of the CPCC system.
基金made possible by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC)
文摘Safety performance functions（SPFs） are crucial to science-based road safety management.Success in developing and applying SPFs, apart data quality and availability, depends fundamentally on two key factors： the validity of the statistical inferences for the available data and on how well the data can be organized into distinct homogeneous entities. The latter aspect plays a key role in the identification and treatment of road sections or corridors with problems related to safety. Indeed, the segmentation of a road network could be especially critical in the development of SPFs that could be used in safety management for roadway types, such as motorways（freeways in North America）, which have a large number of variables that could result in very short segments if these are desired to be homogeneous. This consequence, from an analytical point of view, can be a problem when the location of crashes is not precise and when there is an overabundance of segments with zero crashes. Lengthening the segments for developing and applying SPFs can mitigate this problem, but at a sacrifice of homogeneity. This paper seeks to address this dilemma by investigating four approaches for segmentation for motorways, using sample data from Italy. The best results were obtained for the segmentation based on two curves and two tangents within a segment and with fixed length segments. The segmentation characterized by a constant value of all original variables inside each segment was the poorest approach by all measures.
基金supported by National Natural Science Foundation of China(NSFC No.11972341)fundamental research project of Lomonosov Moscow State University’s Mathematical models for multi-phase media and wave processes in natural,technical and social systems.
文摘The study of impacts of down-up hill road segment on the density threshold of traffic shock formation in ring road vehicular flow is helpful to the deep understanding of sags’bottleneck effect.Sags are freeway segments along which the gradient increases gradually in the traffic direction.The main aim of this paper is to seek the density threshold of shock formation of vehicular flow in ring road with down-up hill segment,because down-up hill roadway segment is a source to cause capacity reduction that is an attractive topic in vehicular traffic science.To seek the density threshold numerically,a viscoelastic continuum model[1]is extended and used.To solve the model equations,a fifth-order weighted essentially non-oscillatory scheme for spatial discretization,and a 3rd order Runge-Kutta scheme for time partial derivative term are used.Validation by existing observation data and the Navier-Stokes like model[2]extended as EZM is done before conducting extensive numerical simulations.For ring road vehicular flow with three separated down-up hill segments,it is found that the density threshold of shock formation decreases monotonically with the relative difference of free flow speed,this variation can be simply fitted by a third order polynomial.
基金the research agreement research agreement signed in 2019 by the University of Brescia and the Province of Brescia(Prot.668/2019).
文摘Vehicle operating speed is a crucial factor for road safety,as it strictly affects occurrence and severity of crashes.Usually,85th percentile of the operating speed distributions(i.e.,V85)in free-flow traffic condition is widely accepted as a base value of consistency evaluation for homogenous portion of existing roads.Although the computation of V85 is simple,many road authorities cannot collect speed data for each road.Therefore,providing prediction models could be a useful tool to investigate the relationship between V85 and road characteristics.The literature proposed several models to account it.However,to the best of our knowledge,the effects of some road geometric characteristics,road markings and signs,traffic data,type of terrain and the simultaneous consideration of different road categories on the V85 prediction were not completely analyzed.This paper fills this gap by isolating key variables that mostly affect V85.In doing so,60000+car spot speed data were collected along the county road network of the province of Brescia(Italy),and then processed by multiple regression models.The main findings show that V85 increases owing to the presence of a wider or paved shoulder,visible road median markings,a higher number of lanes and a higher percentage of cars with respect to the total traffic flow.Conversely,V85 decreases as the road axis curvature,the number of accesses and rate of forbidden overtaking increase.In addition,the presence of visible road external markings and the surrounding mountainous terrain contribute to decreasing V85.The overall findings may support road authorities to verify roads’operating conditions and,possibly,adjust the speed limits,especially for existing roads.
