Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569545
Anna Feldhütter, Dominik Kroll, K. Bengler
Although fatigue's negative impact on driving performance is well known from manual driving, its effect on the take-over performance during the transition from conditionally automated driving to manual driving is still uncertain. The effect of fatigue on the take-over performance was examined in a driving simulator study with 47 participants assigned to two conditions: fatigued or alert. In the corresponding condition (fatigued or alert), the desired driver state was promoted by specific measures (e.g, daytime, caffeinated beverages, physical exercise). In the fatigued condition, the take-over situation was triggered once participants reached a certain high level of fatigue. Two trained, independent observer assessed fatigue with the support of a technical fatigue assessment system based on objective eyelid-closure metrics (e.g, PERCLOS). In the alert condition, participants drove conditionally automated for a fixed 5-minute period. Results showed no significant difference between participants' take-over times in the two conditions. However, fatigued participants were significantly more burdened and stressed during the take-over situation than were alert participants and manifested less confident behavior when coping with the situation. This behavior may negatively affect the transition from conditionally automated driving to manual driving in more complex situations and merits further examination.
{"title":"Wake Up and Take Over! The Effect of Fatigue on the Take-over Performance in Conditionally Automated Driving","authors":"Anna Feldhütter, Dominik Kroll, K. Bengler","doi":"10.1109/ITSC.2018.8569545","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569545","url":null,"abstract":"Although fatigue's negative impact on driving performance is well known from manual driving, its effect on the take-over performance during the transition from conditionally automated driving to manual driving is still uncertain. The effect of fatigue on the take-over performance was examined in a driving simulator study with 47 participants assigned to two conditions: fatigued or alert. In the corresponding condition (fatigued or alert), the desired driver state was promoted by specific measures (e.g, daytime, caffeinated beverages, physical exercise). In the fatigued condition, the take-over situation was triggered once participants reached a certain high level of fatigue. Two trained, independent observer assessed fatigue with the support of a technical fatigue assessment system based on objective eyelid-closure metrics (e.g, PERCLOS). In the alert condition, participants drove conditionally automated for a fixed 5-minute period. Results showed no significant difference between participants' take-over times in the two conditions. However, fatigued participants were significantly more burdened and stressed during the take-over situation than were alert participants and manifested less confident behavior when coping with the situation. This behavior may negatively affect the transition from conditionally automated driving to manual driving in more complex situations and merits further examination.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124001725","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569536
Jashojit Mukhtarjee, K. Praveen, V. Madumbu
The visual quality of an image captured by vision systems can degrade significantly under adverse weather conditions. In this paper we propose a deep learning based solution to improve the visual quality of images captured under rainy and foggy circumstances, which are among the prominent and common weather conditions that attribute to bad image quality. Our convolutional neural network(CNN), NVDeHazenet learns to predict both the original signal as well as the atmospheric light to finally restore image quality. It outperforms the existing state of the art methods by evaluation on both synthetic data as well as real world hazy images. The deraining CNN, NVDeRainNet shows similar performance on existing rain datasets as the state of the art. On natural rain images NVDeRainNet shows better than state of the art performance. We show the use of perceptual loss to improve the visual quality of results. These networks require considerable amount of data under adverse weather conditions and their respective ground truth for training. For this purpose we use a weather simulation framework to simulate synthetic rainy and foggy environments. This data is augmented with existing rain datasets to train the networks.
{"title":"Visual Quality Enhancement Of Images Under Adverse Weather Conditions","authors":"Jashojit Mukhtarjee, K. Praveen, V. Madumbu","doi":"10.1109/ITSC.2018.8569536","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569536","url":null,"abstract":"The visual quality of an image captured by vision systems can degrade significantly under adverse weather conditions. In this paper we propose a deep learning based solution to improve the visual quality of images captured under rainy and foggy circumstances, which are among the prominent and common weather conditions that attribute to bad image quality. Our convolutional neural network(CNN), NVDeHazenet learns to predict both the original signal as well as the atmospheric light to finally restore image quality. It outperforms the existing state of the art methods by evaluation on both synthetic data as well as real world hazy images. The deraining CNN, NVDeRainNet shows similar performance on existing rain datasets as the state of the art. On natural rain images NVDeRainNet shows better than state of the art performance. We show the use of perceptual loss to improve the visual quality of results. These networks require considerable amount of data under adverse weather conditions and their respective ground truth for training. For this purpose we use a weather simulation framework to simulate synthetic rainy and foggy environments. This data is augmented with existing rain datasets to train the networks.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125888126","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569373
Maximilian Graf, O. Speidel, Julius Ziegler, K. Dietmayer
Behavioral-specific trajectory planning for automated vehicles is an intensively explored research topic. Many situations in daily traffic, e.g. following a leading vehicle or stopping behind it, require knowledge about how the scene may evolve. In recent years, much effort has been put into developing driver models to predict traffic scenes as realistic as possible according to human behavior. In this paper, we present a method for behavioral-specific trajectory planning using dedicated driver models. The main idea is to first calculate a reference trajectory using a suitable model to achieve the desired behavior and then to incorporate this reference trajectory into an optimal control problem to obtain an acceleration- and jerk-optimal trajectory. A major strength of this method is in the small computation time, since the problem is formalized as a quadratic optimization problem and can thus be efficiently solved in real time, even for a huge number of optimization variables.
{"title":"Trajectory Planning for Automated Vehicles using Driver Models","authors":"Maximilian Graf, O. Speidel, Julius Ziegler, K. Dietmayer","doi":"10.1109/ITSC.2018.8569373","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569373","url":null,"abstract":"Behavioral-specific trajectory planning for automated vehicles is an intensively explored research topic. Many situations in daily traffic, e.g. following a leading vehicle or stopping behind it, require knowledge about how the scene may evolve. In recent years, much effort has been put into developing driver models to predict traffic scenes as realistic as possible according to human behavior. In this paper, we present a method for behavioral-specific trajectory planning using dedicated driver models. The main idea is to first calculate a reference trajectory using a suitable model to achieve the desired behavior and then to incorporate this reference trajectory into an optimal control problem to obtain an acceleration- and jerk-optimal trajectory. A major strength of this method is in the small computation time, since the problem is formalized as a quadratic optimization problem and can thus be efficiently solved in real time, even for a huge number of optimization variables.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124725872","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569643
Peng Liu, B. Paden, Ü. Özgüner
Motion planning for autonomous vehicles requires spatio-temporal motion plans (i.e. state trajectories) to account for dynamic obstacles. This requires a trajectory tracking control process which faithfully tracks planned trajectories. In this paper, a control scheme is presented which first optimizes a planned trajectory and then tracks the optimized trajectory using a feedback-feedforward controller. The feedforward element is calculated in a model predictive manner with a cost function focusing on driving performance. Stability of the error dynamic is then guaranteed by the design of the feedback-feedforward controller. The tracking performance of the control system is tested in a realistic simulated scenario where the control system must track an evasive lateral maneuver. The proposed controller performs well in simulation and can be easily adapted to different dynamic vehicle models. The uniqueness of the solution to the control synthesis eliminates any nondeterminism that could arise with switching between numerical solvers for the underlying mathematical program.
{"title":"Model Predictive Trajectory Optimization and Tracking for on-Road Autonomous Vehicles","authors":"Peng Liu, B. Paden, Ü. Özgüner","doi":"10.1109/ITSC.2018.8569643","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569643","url":null,"abstract":"Motion planning for autonomous vehicles requires spatio-temporal motion plans (i.e. state trajectories) to account for dynamic obstacles. This requires a trajectory tracking control process which faithfully tracks planned trajectories. In this paper, a control scheme is presented which first optimizes a planned trajectory and then tracks the optimized trajectory using a feedback-feedforward controller. The feedforward element is calculated in a model predictive manner with a cost function focusing on driving performance. Stability of the error dynamic is then guaranteed by the design of the feedback-feedforward controller. The tracking performance of the control system is tested in a realistic simulated scenario where the control system must track an evasive lateral maneuver. The proposed controller performs well in simulation and can be easily adapted to different dynamic vehicle models. The uniqueness of the solution to the control synthesis eliminates any nondeterminism that could arise with switching between numerical solvers for the underlying mathematical program.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128315007","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569341
Ming-Chou Shen, Hanyang Hu, Bohua Sun, W. Deng
On ramp merging scenario is critical to mobility and safety of highway traffic. The existing planning algorithms mainly focus on longitudinal motion coordination for collision avoidance. However, it has been shown that with lateral lane change in main lanes, the road capacity can be greatly enhanced. In this article, we propose an integrated framework for lane change decision making and longitudinal motion planning. Vehicles estimate the time they arrive merge region to guarantee collision avoidance, and optimal control based heuristics are calculated to make lane change decisions. To achieve better performance, we propose a cooperative decision making and planning algorithm. Numerical simulations show the efficiency of the proposed planning algorithm over simple sequential planning policy and scenarios without lane change.
{"title":"Heuristics Based Cooperative Planning for Highway On-Ramp Merge","authors":"Ming-Chou Shen, Hanyang Hu, Bohua Sun, W. Deng","doi":"10.1109/ITSC.2018.8569341","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569341","url":null,"abstract":"On ramp merging scenario is critical to mobility and safety of highway traffic. The existing planning algorithms mainly focus on longitudinal motion coordination for collision avoidance. However, it has been shown that with lateral lane change in main lanes, the road capacity can be greatly enhanced. In this article, we propose an integrated framework for lane change decision making and longitudinal motion planning. Vehicles estimate the time they arrive merge region to guarantee collision avoidance, and optimal control based heuristics are calculated to make lane change decisions. To achieve better performance, we propose a cooperative decision making and planning algorithm. Numerical simulations show the efficiency of the proposed planning algorithm over simple sequential planning policy and scenarios without lane change.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129668099","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569418
John Martinsson, N. Mohammadiha, Alexander Schliep
The safety of autonomous vehicles needs to be verified and validated by rigorous testing. It is expensive to test autonomous vehicles in the field, and therefore virtual testing methods are needed. Generative models of maneuvers such as cut-ins, overtakes, and lane-keeping are needed to thoroughly test the autonomous vehicle in a virtual environment. To train such models we need ground truth maneuver labels and obtaining such labels can be time-consuming and costly. In this work, we use a mixture of hidden Markov models to find clusters in maneuver trajectories, which can be used to speed up the labeling process. The maneuver trajectories are noisy, asynchronous and of uneven length, which make hidden Markov models a good fit for the data. The method is evaluated on labeled data from a test track consisting of cut-ins and overtakes with favorable results. Further, it is applied to natural data where many of the clusters found can be interpreted as driver maneuvers under reasonable assumptions. We show that mixtures of hidden Markov models can be used to find motion patterns in driver maneuver data from highways and country roads.
{"title":"Clustering Vehicle Maneuver Trajectories Using Mixtures of Hidden Markov Models","authors":"John Martinsson, N. Mohammadiha, Alexander Schliep","doi":"10.1109/ITSC.2018.8569418","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569418","url":null,"abstract":"The safety of autonomous vehicles needs to be verified and validated by rigorous testing. It is expensive to test autonomous vehicles in the field, and therefore virtual testing methods are needed. Generative models of maneuvers such as cut-ins, overtakes, and lane-keeping are needed to thoroughly test the autonomous vehicle in a virtual environment. To train such models we need ground truth maneuver labels and obtaining such labels can be time-consuming and costly. In this work, we use a mixture of hidden Markov models to find clusters in maneuver trajectories, which can be used to speed up the labeling process. The maneuver trajectories are noisy, asynchronous and of uneven length, which make hidden Markov models a good fit for the data. The method is evaluated on labeled data from a test track consisting of cut-ins and overtakes with favorable results. Further, it is applied to natural data where many of the clusters found can be interpreted as driver maneuvers under reasonable assumptions. We show that mixtures of hidden Markov models can be used to find motion patterns in driver maneuver data from highways and country roads.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"2014 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127471427","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569638
M. Nishigaki, R. Mose, Osamu Takahata, Hideki Imafuku, Hironori Aoygai
Advanced driver assistance systems (ADAS) for cars have been in market for a few decades and gaining popularity. The automation level for these systems are getting higher over years and automated driving is expected to be launched in near future. Advantage of those systems is not only for safety, but also for reducing the workload in driving. Especially the system which allows drivers to leave their hands off the steering wheel is considered to provide additional benefits to drivers compared to the system requiring hands on the steering wheel or to manual driving. The one of the additional benefits is the mental workload reduction, in other words stress level reduction, by free of their hands in driving. In this paper, we propose the method to measure mental workload which allows quantitative comparison in stress level between hands off and hands on the steering wheel system including manual driving, taking individual initial stress level and temporal change of stress level in a day into account. The proposed method is relatively easier on the measurement procedure and not requires complex measurement tools. In this sense, it fits to evaluate ADAS systems with actual driving. We report with experiments that our proposed method is effective for the purpose of evaluating the mental workload reduction for highly advanced driver assistance systems.
{"title":"Quantitative evaluation on mental worklosd reduction for hands free driving","authors":"M. Nishigaki, R. Mose, Osamu Takahata, Hideki Imafuku, Hironori Aoygai","doi":"10.1109/ITSC.2018.8569638","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569638","url":null,"abstract":"Advanced driver assistance systems (ADAS) for cars have been in market for a few decades and gaining popularity. The automation level for these systems are getting higher over years and automated driving is expected to be launched in near future. Advantage of those systems is not only for safety, but also for reducing the workload in driving. Especially the system which allows drivers to leave their hands off the steering wheel is considered to provide additional benefits to drivers compared to the system requiring hands on the steering wheel or to manual driving. The one of the additional benefits is the mental workload reduction, in other words stress level reduction, by free of their hands in driving. In this paper, we propose the method to measure mental workload which allows quantitative comparison in stress level between hands off and hands on the steering wheel system including manual driving, taking individual initial stress level and temporal change of stress level in a day into account. The proposed method is relatively easier on the measurement procedure and not requires complex measurement tools. In this sense, it fits to evaluate ADAS systems with actual driving. We report with experiments that our proposed method is effective for the purpose of evaluating the mental workload reduction for highly advanced driver assistance systems.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"335 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132441406","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569961
Claudio Ruch, S. Hörl, Emilio Frazzoli
In an autonomous mobility-on-demand (AMoD) system, customers are transported by autonomously driving vehicles in an on-demand fashion. Although these AMoD systems will soon be introduced to cities, their quantitative analysis from a fleet operational and city planning viewpoint remains challenging due to the lack of dedicated analysis tools. In this paper, we introduce AMoDeus, an open-source software package for the accurate and quantitative analysis of autonomous mobility-on-demand systems. AMoDeus uses an agent-based transportation simulation framework to simulate arbitrarily configured AMoD systems with static or dynamic demand. It includes standard benchmark algorithms, fleet efficiency and service level analysis methods and a dedicated graphical viewer that allows in-depth insights into the system. Together with AMoDeus, we publish a typical simulation scenario based on taxi traces recorded in San Francisco. It can be used to test novel fleet control algorithms or as a basis to model more complex transportation research scenarios.
{"title":"AMoDeus, a Simulation-Based Testbed for Autonomous Mobility-on-Demand Systems","authors":"Claudio Ruch, S. Hörl, Emilio Frazzoli","doi":"10.1109/ITSC.2018.8569961","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569961","url":null,"abstract":"In an autonomous mobility-on-demand (AMoD) system, customers are transported by autonomously driving vehicles in an on-demand fashion. Although these AMoD systems will soon be introduced to cities, their quantitative analysis from a fleet operational and city planning viewpoint remains challenging due to the lack of dedicated analysis tools. In this paper, we introduce AMoDeus, an open-source software package for the accurate and quantitative analysis of autonomous mobility-on-demand systems. AMoDeus uses an agent-based transportation simulation framework to simulate arbitrarily configured AMoD systems with static or dynamic demand. It includes standard benchmark algorithms, fleet efficiency and service level analysis methods and a dedicated graphical viewer that allows in-depth insights into the system. Together with AMoDeus, we publish a typical simulation scenario based on taxi traces recorded in San Francisco. It can be used to test novel fleet control algorithms or as a basis to model more complex transportation research scenarios.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131954834","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569910
Nigel Williams, Guoyuan Wu, K. Boriboonsomsin, M. Barth, Samer A. Rajab, Sue Bai
Conventional lane change warning and automated lane changing systems detect other vehicles using on-board sensors such as camera, radar, and ultrasonic sensors. With the advent of Connected Vehicle (CV) technology, wireless communication (e.g, Dedicated Short Range Communications, or DSRC) becomes another option for “sensing” surrounding vehicles. In particular, DSRC does not have the line-of-sight limitation of ranging sensors and thus can “see” traffic farther ahead, which lends itself well to anticipating the movements of nearby vehicles. We have developed an algorithm that uses such data to predict whether a desired lane change will result in an unsafe situation, and prevents the lane change if that is the case. The effectiveness was evaluated in the microscopic traffic simulator VISSIM using a freeway network that has been well calibrated with rush hour traffic data. System performance in terms of safety was estimated using the Surrogate Safety Assessment Model (SSAM) under a variety of traffic scenarios (different congestion levels, penetration rates of connected vehicles and application-equipped vehicles). Preliminary tests showed that the proposed algorithm can reduce the number of potential traffic conflicts by up to 30%, with higher reductions at higher traffic volumes and higher percentages of application-equipped vehicles.
{"title":"Anticipatory Lane Change Warning using Vehicle-to-Vehicle Communications","authors":"Nigel Williams, Guoyuan Wu, K. Boriboonsomsin, M. Barth, Samer A. Rajab, Sue Bai","doi":"10.1109/ITSC.2018.8569910","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569910","url":null,"abstract":"Conventional lane change warning and automated lane changing systems detect other vehicles using on-board sensors such as camera, radar, and ultrasonic sensors. With the advent of Connected Vehicle (CV) technology, wireless communication (e.g, Dedicated Short Range Communications, or DSRC) becomes another option for “sensing” surrounding vehicles. In particular, DSRC does not have the line-of-sight limitation of ranging sensors and thus can “see” traffic farther ahead, which lends itself well to anticipating the movements of nearby vehicles. We have developed an algorithm that uses such data to predict whether a desired lane change will result in an unsafe situation, and prevents the lane change if that is the case. The effectiveness was evaluated in the microscopic traffic simulator VISSIM using a freeway network that has been well calibrated with rush hour traffic data. System performance in terms of safety was estimated using the Surrogate Safety Assessment Model (SSAM) under a variety of traffic scenarios (different congestion levels, penetration rates of connected vehicles and application-equipped vehicles). Preliminary tests showed that the proposed algorithm can reduce the number of potential traffic conflicts by up to 30%, with higher reductions at higher traffic volumes and higher percentages of application-equipped vehicles.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132319374","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 : 2018-11-01DOI: 10.1109/ITSC.2018.8569426
Pranamesh Chakraborty, Anuj Sharma, C. Hegde
Early detection of incidents is a key step to reduce incident related congestion. State Department of Transportation (DoTs) usually install a large number of Close Circuit Television (CCTV) cameras in freeways for traffic surveillance. In this study, we used semi-supervised techniques to detect traffic incident trajectories from the cameras. Vehicle trajectories are identified from the cameras using state-of-the-art deep learning based You Look Only Once (YOLOv3) classifier and Simple Online Realtime Tracking (SORT) is used for vehicle tracking. Our proposed approach for trajectory classification is based on semi-supervised parameter estimation using maximum-likelihood (ML) estimation. The ML based Contrastive Pessimistic Likelihood Estimation (CPLE) attempts to identify incident trajectories from the normal trajectories. We compared the performance of CPLE algorithm to traditional semi-supervised techniques Self Learning and Label Spreading, and also to the classification based on the corresponding supervised algorithm. Results show that approximately 14% improvement in trajectory classification can be achieved using the proposed approach.
早期发现事件是减少事件相关拥塞的关键步骤。国家交通部门通常在高速公路上安装大量闭路电视(CCTV)摄像机进行交通监控。在这项研究中，我们使用半监督技术来检测来自摄像头的交通事故轨迹。车辆轨迹从摄像头中识别，使用最先进的基于You Look Only Once (YOLOv3)的深度学习分类器，并使用Simple Online Realtime Tracking (SORT)进行车辆跟踪。我们提出的弹道分类方法是基于半监督参数估计的最大似然估计。基于机器学习的对比悲观似然估计(CPLE)试图从正常轨迹中识别事件轨迹。我们将CPLE算法的性能与传统的半监督学习和标签扩展技术进行了比较，并与基于相应监督算法的分类进行了比较。结果表明，该方法可使轨迹分类精度提高约14%。
{"title":"Freeway Traffic Incident Detection from Cameras: A Semi-Supervised Learning Approach","authors":"Pranamesh Chakraborty, Anuj Sharma, C. Hegde","doi":"10.1109/ITSC.2018.8569426","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569426","url":null,"abstract":"Early detection of incidents is a key step to reduce incident related congestion. State Department of Transportation (DoTs) usually install a large number of Close Circuit Television (CCTV) cameras in freeways for traffic surveillance. In this study, we used semi-supervised techniques to detect traffic incident trajectories from the cameras. Vehicle trajectories are identified from the cameras using state-of-the-art deep learning based You Look Only Once (YOLOv3) classifier and Simple Online Realtime Tracking (SORT) is used for vehicle tracking. Our proposed approach for trajectory classification is based on semi-supervised parameter estimation using maximum-likelihood (ML) estimation. The ML based Contrastive Pessimistic Likelihood Estimation (CPLE) attempts to identify incident trajectories from the normal trajectories. We compared the performance of CPLE algorithm to traditional semi-supervised techniques Self Learning and Label Spreading, and also to the classification based on the corresponding supervised algorithm. Results show that approximately 14% improvement in trajectory classification can be achieved using the proposed approach.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130843857","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：