Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8916976
T. M. Hale, R. Doorley, Michael O'Byrne, Giacinto Rittgers, V. Pakrashi, Bidisha Ghosh
This paper analyses the influence of passing events on the risk perception and behavior (e.g. heart rate, cycling speed etc.) of cyclists through a quasi-natural experiment in urban multimodal signalized transport network. An instrumented bicycle was used by participants in real traffic conditions, recording the passing distance of other road users, the speed and location of the bicycle, the heart rate of the participant and also the self-reported risk level. Additionally, a stereographic camera system was developed and evaluated to test its viability as a method of measuring passing speed and relative distance of vehicles while overtaking cyclists. The factors studied showed that passing events relate to an increased heart rate and reduced cycling speed, but not an increased perceived level of risk. The present study also demonstrated the stereographic camera system is capable of collecting data regarding speed and distance. However, the method used for this experiment requires considerable time and effort to perform the necessary data processing.
{"title":"How Do Passing Events Influence the Perception of Risk Among Cyclistsƒ","authors":"T. M. Hale, R. Doorley, Michael O'Byrne, Giacinto Rittgers, V. Pakrashi, Bidisha Ghosh","doi":"10.1109/ITSC.2019.8916976","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8916976","url":null,"abstract":"This paper analyses the influence of passing events on the risk perception and behavior (e.g. heart rate, cycling speed etc.) of cyclists through a quasi-natural experiment in urban multimodal signalized transport network. An instrumented bicycle was used by participants in real traffic conditions, recording the passing distance of other road users, the speed and location of the bicycle, the heart rate of the participant and also the self-reported risk level. Additionally, a stereographic camera system was developed and evaluated to test its viability as a method of measuring passing speed and relative distance of vehicles while overtaking cyclists. The factors studied showed that passing events relate to an increased heart rate and reduced cycling speed, but not an increased perceived level of risk. The present study also demonstrated the stereographic camera system is capable of collecting data regarding speed and distance. However, the method used for this experiment requires considerable time and effort to perform the necessary data processing.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"93 1","pages":"2355-2360"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76227062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8916896
Deepika Ravipati, Kenny Chour, Abhishek Nayak, T. Marr, Sheelabhadra Dey, Alvika Gautam, S. Rathinam, Swaminathan Gopalswamy
Infrastructure Enabled Autonomy (IEA) is a new paradigm in autonomous vehicles research that aims at distributed intelligence architecture by transferring the core functionalities of sensing and localization to infrastructure. This paradigm is also promising in designing scalable systems that enable autonomous car platooning on highways. This paper gives a detailed description about the experimental realization of IEA and techniques devised to localize a vehicle in such a setup. A reliable camera calibration technique for such an experimental setup is discussed, followed by a technique to transform 2D image coordinates to 3D world coordinates. In this research, localization information is received from on-board vehicle sensors like GPS/IMU, and (2) localized vehicle position data derived from deep learning, and 2D to 3D coordinate transformations on the real-time camera feeds and (3) lane detection data from infrastructure cameras. This data is fused together utilizing an Extended Kalman Filter (EKF) to obtain reliable estimates of the position of the vehicle at 50 Hz. This position information is then used to control the vehicle with an objective of following a prescribed path. Extensive simulation and experimental results are also presented to corroborate the performance of the proposed approach.
{"title":"Vision Based Localization for Infrastructure Enabled Autonomy","authors":"Deepika Ravipati, Kenny Chour, Abhishek Nayak, T. Marr, Sheelabhadra Dey, Alvika Gautam, S. Rathinam, Swaminathan Gopalswamy","doi":"10.1109/ITSC.2019.8916896","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8916896","url":null,"abstract":"Infrastructure Enabled Autonomy (IEA) is a new paradigm in autonomous vehicles research that aims at distributed intelligence architecture by transferring the core functionalities of sensing and localization to infrastructure. This paradigm is also promising in designing scalable systems that enable autonomous car platooning on highways. This paper gives a detailed description about the experimental realization of IEA and techniques devised to localize a vehicle in such a setup. A reliable camera calibration technique for such an experimental setup is discussed, followed by a technique to transform 2D image coordinates to 3D world coordinates. In this research, localization information is received from on-board vehicle sensors like GPS/IMU, and (2) localized vehicle position data derived from deep learning, and 2D to 3D coordinate transformations on the real-time camera feeds and (3) lane detection data from infrastructure cameras. This data is fused together utilizing an Extended Kalman Filter (EKF) to obtain reliable estimates of the position of the vehicle at 50 Hz. This position information is then used to control the vehicle with an objective of following a prescribed path. Extensive simulation and experimental results are also presented to corroborate the performance of the proposed approach.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"49 1","pages":"1638-1643"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74809112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8917309
S. Ware, Antonis F. Lentzakis, R. Su
In this paper, we present a simulation modeling framework that can accommodate multiple classes of travelers and integrates several distinct features, which in turn can be associated with each of the traveler classes, thus providing flexibility and a so-called bird’s-eye view to any potential user. Concretely, we integrate into the multi-class region-based dynamic traffic model, called multi-class Network Transmission Model (McNTM), several features, including a public transit diversion component, as well as routing methods associated with different traveler classes. Three distinct traveler classes are defined, the 1st class of travelers equipped with autonomous vehicles, the 2nd traveler class comprising of RGIS-equipped, conventional vehicles and the 3rd traveler class comprising of unequipped, conventional vehicles. Certain assumptions are made for each traveler class. The gain in overall performance for the case where 1st and 2nd class travelers are present in the system, ranges from 0.78% - 23.43%. Region-based routing methods employed by the 1st and 2nd class respectively, not only benefit overall network performance, but with their respective market penetration rates exceeding certain thresholds, can prove beneficial to the individual performance of other traveler classes.
{"title":"A Simulation Modeling Framework with Autonomous Vehicle Region-based Routing and Public Transit Diversion Integration","authors":"S. Ware, Antonis F. Lentzakis, R. Su","doi":"10.1109/ITSC.2019.8917309","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8917309","url":null,"abstract":"In this paper, we present a simulation modeling framework that can accommodate multiple classes of travelers and integrates several distinct features, which in turn can be associated with each of the traveler classes, thus providing flexibility and a so-called bird’s-eye view to any potential user. Concretely, we integrate into the multi-class region-based dynamic traffic model, called multi-class Network Transmission Model (McNTM), several features, including a public transit diversion component, as well as routing methods associated with different traveler classes. Three distinct traveler classes are defined, the 1st class of travelers equipped with autonomous vehicles, the 2nd traveler class comprising of RGIS-equipped, conventional vehicles and the 3rd traveler class comprising of unequipped, conventional vehicles. Certain assumptions are made for each traveler class. The gain in overall performance for the case where 1st and 2nd class travelers are present in the system, ranges from 0.78% - 23.43%. Region-based routing methods employed by the 1st and 2nd class respectively, not only benefit overall network performance, but with their respective market penetration rates exceeding certain thresholds, can prove beneficial to the individual performance of other traveler classes.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"55 1","pages":"2626-2632"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72856622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8917346
Johannes Potzy, Sophie Feinauer, Karl-Heinz Siedersberger, K. Bengler
The vision to integrate automated vehicles into manual traffic motivates to investigate automated merging in dense traffic. To gain an easy, distinct and interpretable behavior for interacting traffic this study investigates release conditions of lane changes into small gaps in a within-subject design on a test track with 39 participants. To generate standardized situations all merging maneuvers are performed automatically. The study is divided into two parts. In the first part participants validate different times headway between the participant’s vehicle and automated vehicle under different situational parameters (deceleration to target gap, velocity and existence of road work). In the second part, participants release the lane change of the automated vehicle themselves, when they expected it to merge. Here, in addition to part one, the automated vehicle adjusted velocity to the target gap with weak and strong deceleration. The results show that participants prefer an efficient lane change of the automatic vehicle, where interacting traffic has to react as little as possible. Compliance with safety distances is not decisive. The required times headway between automated and interacting vehicles decreases with higher velocity and in lane narrowing situations. The study contributes to the design of vehicle behaviour that can enhance the acceptance of automated vehicles in mixed-traffic.
{"title":"Manual Drivers’ Evaluation of Automated Merging Behavior in Dense Traffic: Efficiency Matters","authors":"Johannes Potzy, Sophie Feinauer, Karl-Heinz Siedersberger, K. Bengler","doi":"10.1109/ITSC.2019.8917346","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8917346","url":null,"abstract":"The vision to integrate automated vehicles into manual traffic motivates to investigate automated merging in dense traffic. To gain an easy, distinct and interpretable behavior for interacting traffic this study investigates release conditions of lane changes into small gaps in a within-subject design on a test track with 39 participants. To generate standardized situations all merging maneuvers are performed automatically. The study is divided into two parts. In the first part participants validate different times headway between the participant’s vehicle and automated vehicle under different situational parameters (deceleration to target gap, velocity and existence of road work). In the second part, participants release the lane change of the automated vehicle themselves, when they expected it to merge. Here, in addition to part one, the automated vehicle adjusted velocity to the target gap with weak and strong deceleration. The results show that participants prefer an efficient lane change of the automatic vehicle, where interacting traffic has to react as little as possible. Compliance with safety distances is not decisive. The required times headway between automated and interacting vehicles decreases with higher velocity and in lane narrowing situations. The study contributes to the design of vehicle behaviour that can enhance the acceptance of automated vehicles in mixed-traffic.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"86 1","pages":"3454-3460"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73051185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8917002
Patrick Hart, Leonard Rychly, A. Knoll
Many current behavior generation methods struggle to handle real-world traffic situations as they do not scale well with complexity. However, behaviors can be learned off-line using data-driven approaches. Especially, reinforcement learning is promising as it implicitly learns how to behave utilizing collected experiences. In this work, we combine policy-based reinforcement learning with local optimization to foster and synthesize the best of the two methodologies. The policy-based reinforcement learning algorithm provides an initial solution and guiding reference for the post-optimization. Therefore, the optimizer only has to compute a single homotopy class, e.g. drive behind or in front of the other vehicle. By storing the state-history during reinforcement learning, it can be used for constraint checking and the optimizer can account for interactions. The post-optimization additionally acts as a safety-layer and the novel method, thus, can be applied in safety-critical applications. We evaluate the proposed method using lane-change scenarios with a varying number of vehicles.
{"title":"Lane-Merging Using Policy-based Reinforcement Learning and Post-Optimization","authors":"Patrick Hart, Leonard Rychly, A. Knoll","doi":"10.1109/ITSC.2019.8917002","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8917002","url":null,"abstract":"Many current behavior generation methods struggle to handle real-world traffic situations as they do not scale well with complexity. However, behaviors can be learned off-line using data-driven approaches. Especially, reinforcement learning is promising as it implicitly learns how to behave utilizing collected experiences. In this work, we combine policy-based reinforcement learning with local optimization to foster and synthesize the best of the two methodologies. The policy-based reinforcement learning algorithm provides an initial solution and guiding reference for the post-optimization. Therefore, the optimizer only has to compute a single homotopy class, e.g. drive behind or in front of the other vehicle. By storing the state-history during reinforcement learning, it can be used for constraint checking and the optimizer can account for interactions. The post-optimization additionally acts as a safety-layer and the novel method, thus, can be applied in safety-critical applications. We evaluate the proposed method using lane-change scenarios with a varying number of vehicles.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"1 1","pages":"3176-3181"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73105555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8916928
K. Rezaee, P. Yadmellat, M. Nosrati, E. Abolfathi, Mohammed Elmahgiubi, Jun Luo
Competent multi-lane cruising requires using lane changes and within-lane maneuvers to achieve good speed and maintain safety. This paper proposes a design for autonomous multi-lane cruising by combining a hierarchical reinforcement learning framework with a novel state-action space abstraction. While the proposed solution follows the classical hierarchy of behavior decision, motion planning and control, it introduces a key intermediate abstraction within the motion planner to discretize the state-action space according to high level behavioral decisions. We argue that this design allows principled modular extension of motion planning, in contrast to using either monolithic behavior cloning or a large set of handwritten rules. Moreover, we demonstrate that our state-action space abstraction allows transferring of the trained models without retraining from a simulated environment with virtually no dynamics to one with significantly more realistic dynamics. Together, these results suggest that our proposed hierarchical architecture is a promising way to allow reinforcement learning to be applied to complex multi-lane cruising in the real world.
{"title":"Multi-lane Cruising Using Hierarchical Planning and Reinforcement Learning","authors":"K. Rezaee, P. Yadmellat, M. Nosrati, E. Abolfathi, Mohammed Elmahgiubi, Jun Luo","doi":"10.1109/ITSC.2019.8916928","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8916928","url":null,"abstract":"Competent multi-lane cruising requires using lane changes and within-lane maneuvers to achieve good speed and maintain safety. This paper proposes a design for autonomous multi-lane cruising by combining a hierarchical reinforcement learning framework with a novel state-action space abstraction. While the proposed solution follows the classical hierarchy of behavior decision, motion planning and control, it introduces a key intermediate abstraction within the motion planner to discretize the state-action space according to high level behavioral decisions. We argue that this design allows principled modular extension of motion planning, in contrast to using either monolithic behavior cloning or a large set of handwritten rules. Moreover, we demonstrate that our state-action space abstraction allows transferring of the trained models without retraining from a simulated environment with virtually no dynamics to one with significantly more realistic dynamics. Together, these results suggest that our proposed hierarchical architecture is a promising way to allow reinforcement learning to be applied to complex multi-lane cruising in the real world.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"32 1","pages":"1800-1806"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73180753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8916969
Teng Liu, Bin Tian, Yunfeng Ai, Long Chen, Fei Liu, Dongpu Cao, Ning Bian, Feiyue Wang
As a combination of various kinds of technologies, autonomous vehicles could complete a series of driving tasks by itself, such as perception, decision-making, planning and control. Since there is no human driver to handle the emergency situation, future transportation information is significant for automated vehicles. This paper proposes different methods to forecast the time series for autonomous vehicles, which are the nearest neighborhood (NN), fuzzy coding (FC) and long short term memory (LSTM). First, the formulation and operational process for these three approaches are introduced. Then, the vehicle velocity is regarded as a case study and the real-world dataset is utilized to predict future information via these techniques. Finally, the performance, merits and drawbacks of the presented methods are analyzed and discussed.
{"title":"Dynamic States Prediction in Autonomous Vehicles: Comparison of Three Different Methods","authors":"Teng Liu, Bin Tian, Yunfeng Ai, Long Chen, Fei Liu, Dongpu Cao, Ning Bian, Feiyue Wang","doi":"10.1109/ITSC.2019.8916969","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8916969","url":null,"abstract":"As a combination of various kinds of technologies, autonomous vehicles could complete a series of driving tasks by itself, such as perception, decision-making, planning and control. Since there is no human driver to handle the emergency situation, future transportation information is significant for automated vehicles. This paper proposes different methods to forecast the time series for autonomous vehicles, which are the nearest neighborhood (NN), fuzzy coding (FC) and long short term memory (LSTM). First, the formulation and operational process for these three approaches are introduced. Then, the vehicle velocity is regarded as a case study and the real-world dataset is utilized to predict future information via these techniques. Finally, the performance, merits and drawbacks of the presented methods are analyzed and discussed.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"67 1","pages":"3750-3755"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74613806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8917151
Meifang Yang, Xin Nie, R. W. Liu
Captured images in maritime video surveillance under non-uniform illumination conditions easily suffer from low contrast and details loss. The low-quality images may significantly result in negative effects in practical applications, e.g., target detection, recognition, classification and tracking, etc. Increasing attention has recently been paid to improve the quality of low-light images via computer vision techniques. In this paper, we propose to establish a two-step luminance estimation framework to enhance low-light images. In particular, the coarse luminance is firstly estimated using traditional Max-RGB which extracts the highest pixel values in each color channel. The refined luminance is obtained by introducing a weighted variational model which has the capacities of structure-preserving and texture-smoothing. Based on the estimated well-constructed luminance, the enhanced low-light images are obtained by combining Retinex model with its extended version. The image quality is further improved through a BM3D-based denoising approach. Experimental results on both synthetic and realistic low-light images have demonstrated the satisfactory imaging performance in terms of quantitative and qualitative evaluations.
{"title":"Coarse-to-Fine Luminance Estimation for Low-Light Image Enhancement in Maritime Video Surveillance","authors":"Meifang Yang, Xin Nie, R. W. Liu","doi":"10.1109/ITSC.2019.8917151","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8917151","url":null,"abstract":"Captured images in maritime video surveillance under non-uniform illumination conditions easily suffer from low contrast and details loss. The low-quality images may significantly result in negative effects in practical applications, e.g., target detection, recognition, classification and tracking, etc. Increasing attention has recently been paid to improve the quality of low-light images via computer vision techniques. In this paper, we propose to establish a two-step luminance estimation framework to enhance low-light images. In particular, the coarse luminance is firstly estimated using traditional Max-RGB which extracts the highest pixel values in each color channel. The refined luminance is obtained by introducing a weighted variational model which has the capacities of structure-preserving and texture-smoothing. Based on the estimated well-constructed luminance, the enhanced low-light images are obtained by combining Retinex model with its extended version. The image quality is further improved through a BM3D-based denoising approach. Experimental results on both synthetic and realistic low-light images have demonstrated the satisfactory imaging performance in terms of quantitative and qualitative evaluations.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"42 4","pages":"299-304"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72596325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8917275
K. Abe, H. Fujii, S. Yoshimura
Traffic simulation is utilized to solve traffic-related problems. Microscopic simulations can describe individual vehicles and thus reproduce detailed vehicle behavior. To use a simulator, traffic demand should be estimated in the form of an origin-destination (OD) matrix. The simulator and OD estimation models must be consistent. In addition, microscopic models are sensitive to congestion, and can thus easily produce unexpected congestion. Here, we propose a simulator-embedded OD estimation method that uses congestion sensing. We minimize the residual between the observed and simulated link traffic volumes with some constraints regarding congestion. If a link is judged to be congested, we use resistance in a constraint in the optimization problem, which is determined by the number of the stuck vehicles at each link. Use of the resistance prevents excessively large traffic demand for that link. This congestion sensing mitigates unrealistic congestion in the estimated traffic demand.
{"title":"Estimation of Traffic Demand Corresponding to Observed Link Traffic Volume in Microscopic Simulation","authors":"K. Abe, H. Fujii, S. Yoshimura","doi":"10.1109/ITSC.2019.8917275","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8917275","url":null,"abstract":"Traffic simulation is utilized to solve traffic-related problems. Microscopic simulations can describe individual vehicles and thus reproduce detailed vehicle behavior. To use a simulator, traffic demand should be estimated in the form of an origin-destination (OD) matrix. The simulator and OD estimation models must be consistent. In addition, microscopic models are sensitive to congestion, and can thus easily produce unexpected congestion. Here, we propose a simulator-embedded OD estimation method that uses congestion sensing. We minimize the residual between the observed and simulated link traffic volumes with some constraints regarding congestion. If a link is judged to be congested, we use resistance in a constraint in the optimization problem, which is determined by the number of the stuck vehicles at each link. Use of the resistance prevents excessively large traffic demand for that link. This congestion sensing mitigates unrealistic congestion in the estimated traffic demand.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"96 1","pages":"2220-2225"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73916727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ITSC.2019.8917176
Jorge Beltrán, Irene Cortés, Alejandro Barrera, Jesús Urdiales, Carlos Guindel, F. García, A. D. L. Escalera
LiDAR devices have become a key sensor for autonomous vehicles perception due to their ability to capture reliable geometry information. Indeed, approaches processing LiDAR data have shown an impressive accuracy for 3D object detection tasks, outperforming methods solely based on image inputs. However, the wide diversity of on-board sensor configurations makes the deployment of published algorithms into real platforms a hard task, due to the scarcity of annotated datasets containing laser scans. We present a method to generate new point clouds datasets as captured by a real LiDAR device. The proposed pipeline makes use of multiple frames to perform an accurate 3D reconstruction of the scene in the spherical coordinates system that enables the simulation of the sweeps of a virtual LiDAR sensor, configurable both in location and inner specifications. The similarity between real data and the generated synthetic clouds is assessed through a set of experiments performed using KITTI Depth and Object Benchmarks.
{"title":"A Method for Synthetic LiDAR Generation to Create Annotated Datasets for Autonomous Vehicles Perception","authors":"Jorge Beltrán, Irene Cortés, Alejandro Barrera, Jesús Urdiales, Carlos Guindel, F. García, A. D. L. Escalera","doi":"10.1109/ITSC.2019.8917176","DOIUrl":"https://doi.org/10.1109/ITSC.2019.8917176","url":null,"abstract":"LiDAR devices have become a key sensor for autonomous vehicles perception due to their ability to capture reliable geometry information. Indeed, approaches processing LiDAR data have shown an impressive accuracy for 3D object detection tasks, outperforming methods solely based on image inputs. However, the wide diversity of on-board sensor configurations makes the deployment of published algorithms into real platforms a hard task, due to the scarcity of annotated datasets containing laser scans. We present a method to generate new point clouds datasets as captured by a real LiDAR device. The proposed pipeline makes use of multiple frames to perform an accurate 3D reconstruction of the scene in the spherical coordinates system that enables the simulation of the sweeps of a virtual LiDAR sensor, configurable both in location and inner specifications. The similarity between real data and the generated synthetic clouds is assessed through a set of experiments performed using KITTI Depth and Object Benchmarks.","PeriodicalId":6717,"journal":{"name":"2019 IEEE Intelligent Transportation Systems Conference (ITSC)","volume":"239 1","pages":"1091-1096"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73963497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：