Pub Date : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8964902
Wenbo Yu, Yong Sun, Ruilin Zhou, Xingjian Liu
Deep learning models in Autonomous Driving perception tasks commonly use supervised learning methods and thus highly depend on labeled data. Training with more labeled data tends to bring better results, which highlights the meaning of data augmentation. Currently there are two difficulties when doing data augmentation. Firstly, it is time consuming to manually label the collected raw data. The second issue is that the diversity of a dataset is limited by the collection environment and time. In this paper, we proposed to use the current state of the art Multimodal Unsupervised Image-to-Image Translation (MUNIT) to generate synthesized images from labeled data. One of the benefits is that the generated data are automated labeled since they share the same ground truth with the raw data. Then we used the augmentation dataset to do different tasks including drivable area detection and object detection to prove that the data could be used to improve the performance of convolution neural networks (CNNs). We also designed an auto labelling tool that people could do labelling with the help of the improved CNN. The whole process is like a close loop that finishes labelling tasks while making progresses by itself. Generally speaking, our approach introduces an auto labelling pipeline based on unsupervised image-to-image translation to increase the amount and diversity of labeled data.
{"title":"GAN Based Method for Labeled Image Augmentation in Autonomous Driving","authors":"Wenbo Yu, Yong Sun, Ruilin Zhou, Xingjian Liu","doi":"10.1109/ICCVE45908.2019.8964902","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8964902","url":null,"abstract":"Deep learning models in Autonomous Driving perception tasks commonly use supervised learning methods and thus highly depend on labeled data. Training with more labeled data tends to bring better results, which highlights the meaning of data augmentation. Currently there are two difficulties when doing data augmentation. Firstly, it is time consuming to manually label the collected raw data. The second issue is that the diversity of a dataset is limited by the collection environment and time. In this paper, we proposed to use the current state of the art Multimodal Unsupervised Image-to-Image Translation (MUNIT) to generate synthesized images from labeled data. One of the benefits is that the generated data are automated labeled since they share the same ground truth with the raw data. Then we used the augmentation dataset to do different tasks including drivable area detection and object detection to prove that the data could be used to improve the performance of convolution neural networks (CNNs). We also designed an auto labelling tool that people could do labelling with the help of the improved CNN. The whole process is like a close loop that finishes labelling tasks while making progresses by itself. Generally speaking, our approach introduces an auto labelling pipeline based on unsupervised image-to-image translation to increase the amount and diversity of labeled data.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"238 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121871151","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965116
Christoph Schmittner, Sebastian Chlup, Arndt Bonitz, Martin Latzenhofer, M. Hofer, Carina Kloibhofer, Thomas Raab, Edvin Spahovic, Thomas Doms
Cyber security and privacy are major challenges for interconnected automatic driving cars which rely on confidentiality, integrity, availability, authenticity, accountability as well as privacy of a cooperative intelligent transport system (C-ITS). The cars require detailed data about the environment on different levels to generate a comprehensive overview of the current traffic situation in real time to ensure their safe and secure movement. The availability, integrity, authenticity and accountability of data and its processing are a prerequisite for automated and interconnected driving. Additionally, confidentiality and privacy are main requirements for using C-ITS services. Cyber security is not only necessary for an efficient traffic management. Co-operative functions and coordinative autonomy is mandatory, since successful intentional attacks on a C-ITS in fact threaten human lives. There is a fundamental need for a comprehensive harmonization of safety and security aspects from the infrastructure provider's perspective to ensure awareness and trust - and not solely from an original equipment manufacturer's (OEM) view. Especially cyber security is essential to make these cooperative traffic structures and autonomous interconnected driving technologies confidently available to society. In this paper, the approach for component identification, scoping, use case development, and the applied risk management method as preliminary work for the development of a comprehensive C-ITS cybersecurity reference framework is discussed. These steps are part of a currently ongoing research project.
{"title":"Preliminary Considerations for a Cooperative Intelligent Transport System Cybersecurity Reference Architecture","authors":"Christoph Schmittner, Sebastian Chlup, Arndt Bonitz, Martin Latzenhofer, M. Hofer, Carina Kloibhofer, Thomas Raab, Edvin Spahovic, Thomas Doms","doi":"10.1109/ICCVE45908.2019.8965116","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965116","url":null,"abstract":"Cyber security and privacy are major challenges for interconnected automatic driving cars which rely on confidentiality, integrity, availability, authenticity, accountability as well as privacy of a cooperative intelligent transport system (C-ITS). The cars require detailed data about the environment on different levels to generate a comprehensive overview of the current traffic situation in real time to ensure their safe and secure movement. The availability, integrity, authenticity and accountability of data and its processing are a prerequisite for automated and interconnected driving. Additionally, confidentiality and privacy are main requirements for using C-ITS services. Cyber security is not only necessary for an efficient traffic management. Co-operative functions and coordinative autonomy is mandatory, since successful intentional attacks on a C-ITS in fact threaten human lives. There is a fundamental need for a comprehensive harmonization of safety and security aspects from the infrastructure provider's perspective to ensure awareness and trust - and not solely from an original equipment manufacturer's (OEM) view. Especially cyber security is essential to make these cooperative traffic structures and autonomous interconnected driving technologies confidently available to society. In this paper, the approach for component identification, scoping, use case development, and the applied risk management method as preliminary work for the development of a comprehensive C-ITS cybersecurity reference framework is discussed. These steps are part of a currently ongoing research project.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"242 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133492004","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965152
Michele Menarini, Pasquale Marrancone, Giammarco Cecchini, A. Bazzi, B. Masini, A. Zanella
Vehicles will be equipped with short-range wireless technologies with the aim to improve safety and traffic efficiency. Novel applications are thus being implemented for future cars and trucks, and one of the main issues is how to conduct tests and optimizations in an effective way, limiting the need to perform costly and time consuming experiments on the road. To cope with this issue, we have implemented a simulator with hardware-in-the-loop (HIL), called TRUDI, where the hardware and the implemented applications are tested in the laboratory by injecting virtual positions of the vehicles with the support of a traffic simulator. TRUDI acts as a man-in-the-middle between the communication module and the application itself, making it possible to perform tests with the real devices and providing as an output a system ready for the road. Using TRUDI, it is possible to check the application with a few vehicles and real wireless devices or many vehicles using simulated communication components before moving to experiments on the road. As an example use case, an application for the intersection management is presented, where the driver is warned of the presence and speed of other vehicles approaching the same junction.
{"title":"TRUDI: Testing Environment for Vehicular Applications Running with Devices in the Loop","authors":"Michele Menarini, Pasquale Marrancone, Giammarco Cecchini, A. Bazzi, B. Masini, A. Zanella","doi":"10.1109/ICCVE45908.2019.8965152","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965152","url":null,"abstract":"Vehicles will be equipped with short-range wireless technologies with the aim to improve safety and traffic efficiency. Novel applications are thus being implemented for future cars and trucks, and one of the main issues is how to conduct tests and optimizations in an effective way, limiting the need to perform costly and time consuming experiments on the road. To cope with this issue, we have implemented a simulator with hardware-in-the-loop (HIL), called TRUDI, where the hardware and the implemented applications are tested in the laboratory by injecting virtual positions of the vehicles with the support of a traffic simulator. TRUDI acts as a man-in-the-middle between the communication module and the application itself, making it possible to perform tests with the real devices and providing as an output a system ready for the road. Using TRUDI, it is possible to check the application with a few vehicles and real wireless devices or many vehicles using simulated communication components before moving to experiments on the road. As an example use case, an application for the intersection management is presented, where the driver is warned of the presence and speed of other vehicles approaching the same junction.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132091446","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965218
A. Wischnewski, Johannes Betz, B. Lohmann
One of the key aspects in racing is the ability of the driver to find the handling limits of the vehicle to minimize the resulting lap time. Many approaches for raceline optimization assume the tire-road friction coefficient to be known. However, this neglects the fact that the ability of the system to realize such a race trajectory depends on complex interdependencies between the online trajectory planner, the control systems and the non-modelled uncertainties. In general, a high quality control system can approach the physical limit more reliable, as it applies less corrective actions. We present a model-free learning method to find the minimum achievable lap-time for a given controller using online adaption of a scale factor for the maximum longitudinal and lateral accelerations in the online trajectory planner. In contrast to existing concepts, our approach can be applied as an extension to already available planning and control algorithms instead of replacing them. We demonstrate reliable and safe operation for different vehicle setups in simulation and demonstrate that the algorithm works successfully on a full-size racecar.
{"title":"A Model-Free Algorithm to Safely Approach the Handling Limit of an Autonomous Racecar","authors":"A. Wischnewski, Johannes Betz, B. Lohmann","doi":"10.1109/ICCVE45908.2019.8965218","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965218","url":null,"abstract":"One of the key aspects in racing is the ability of the driver to find the handling limits of the vehicle to minimize the resulting lap time. Many approaches for raceline optimization assume the tire-road friction coefficient to be known. However, this neglects the fact that the ability of the system to realize such a race trajectory depends on complex interdependencies between the online trajectory planner, the control systems and the non-modelled uncertainties. In general, a high quality control system can approach the physical limit more reliable, as it applies less corrective actions. We present a model-free learning method to find the minimum achievable lap-time for a given controller using online adaption of a scale factor for the maximum longitudinal and lateral accelerations in the online trajectory planner. In contrast to existing concepts, our approach can be applied as an extension to already available planning and control algorithms instead of replacing them. We demonstrate reliable and safe operation for different vehicle setups in simulation and demonstrate that the algorithm works successfully on a full-size racecar.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132370580","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965092
Manuel Koschuch, Walter Sebron, Z. Szalay, Á. Török, Hans Tschürtz, I. Wahl
With the advent of cypher-physical (systems of) systems, new challenges for safety and security arise. Especially in the context of autonomous driving we are currently facing a complex environment, where security problems can easily result in safety-relevant issues, and vice versa. There have been multiple approaches in the past to combine the approaches from safety and security best practices into a combined view, all with their individual challenges. We propose a fully integrated approach, combining safety with security and modelling their complex interactions. In this work we start by giving a thorough definition of the basic terms and concepts used in safety and security, in order to identify similarities and differences. We then propose and outline a combined view on the safety and security causal chains and define their interdependencies.
{"title":"Safety & Security in the Context of Autonomous Driving","authors":"Manuel Koschuch, Walter Sebron, Z. Szalay, Á. Török, Hans Tschürtz, I. Wahl","doi":"10.1109/ICCVE45908.2019.8965092","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965092","url":null,"abstract":"With the advent of cypher-physical (systems of) systems, new challenges for safety and security arise. Especially in the context of autonomous driving we are currently facing a complex environment, where security problems can easily result in safety-relevant issues, and vice versa. There have been multiple approaches in the past to combine the approaches from safety and security best practices into a combined view, all with their individual challenges. We propose a fully integrated approach, combining safety with security and modelling their complex interactions. In this work we start by giving a thorough definition of the basic terms and concepts used in safety and security, in order to identify similarities and differences. We then propose and outline a combined view on the safety and security causal chains and define their interdependencies.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129710484","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965037
S. Neumeier, Michael Höpp, Christian Facchi
Numerous issues on the way to autonomous vehicles have already been solved. Nevertheless, there are further problems preventing the introduction of autonomous driving features of higher SAE levels. Remote control of vehicles by human operators located in dedicated operation centers, Teleoperated Driving, can help to overcome the problems of autonomous driving. To enable functional Teleoperated Driving, existent network technology has to be utilized. These cellular networks suffer from variable performance. However, testing Teleoperated Driving and its algorithms in real-world scenarios is costly and potentially dangerous. Virtual testing is an approach to mitigate these obstacles. This paper introduces OpenROUTS3D, an open-source driving simulator initially developed for Teleoperated Driving, but expandable to various use cases.
{"title":"Yet Another Driving Simulator OpenROUTS3D: The Driving Simulator for Teleoperated Driving","authors":"S. Neumeier, Michael Höpp, Christian Facchi","doi":"10.1109/ICCVE45908.2019.8965037","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965037","url":null,"abstract":"Numerous issues on the way to autonomous vehicles have already been solved. Nevertheless, there are further problems preventing the introduction of autonomous driving features of higher SAE levels. Remote control of vehicles by human operators located in dedicated operation centers, Teleoperated Driving, can help to overcome the problems of autonomous driving. To enable functional Teleoperated Driving, existent network technology has to be utilized. These cellular networks suffer from variable performance. However, testing Teleoperated Driving and its algorithms in real-world scenarios is costly and potentially dangerous. Virtual testing is an approach to mitigate these obstacles. This paper introduces OpenROUTS3D, an open-source driving simulator initially developed for Teleoperated Driving, but expandable to various use cases.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126260033","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965213
G. Balázs, W. Stechele
This paper investigates neural network architectures that fuse feature-level data of radar and vision sensors in order to improve automotive environment perception for advanced driver assistance systems. Fusion is performed with occupancy grids, which incorporate sensor-specific information mapped from their individual detection lists. The fusion step is evaluated on three types of neural networks: (1) fully convolutional, (2) auto-encoder and (3) auto-encoder with skipped connections. These networks are trained to fuse radar and camera occupancy grids with the ground truth obtained from lidar scans. A detailed analysis of network architectures and parameters is performed. Results are compared to classical Bayesian occupancy fusion on typical evaluation metrics for pixel-wise classification tasks, like intersection over union and pixel accuracy. This paper shows that it is possible to perform grid fusion of feature-level sensor data with the proposed system architecture. Especially the auto-encoder architectures show significant improvements in evaluation metrics compared to classical Bayesian fusion method.
{"title":"Deep Grid Fusion of Feature-Level Sensor Data with Convolutional Neural Networks","authors":"G. Balázs, W. Stechele","doi":"10.1109/ICCVE45908.2019.8965213","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965213","url":null,"abstract":"This paper investigates neural network architectures that fuse feature-level data of radar and vision sensors in order to improve automotive environment perception for advanced driver assistance systems. Fusion is performed with occupancy grids, which incorporate sensor-specific information mapped from their individual detection lists. The fusion step is evaluated on three types of neural networks: (1) fully convolutional, (2) auto-encoder and (3) auto-encoder with skipped connections. These networks are trained to fuse radar and camera occupancy grids with the ground truth obtained from lidar scans. A detailed analysis of network architectures and parameters is performed. Results are compared to classical Bayesian occupancy fusion on typical evaluation metrics for pixel-wise classification tasks, like intersection over union and pixel accuracy. This paper shows that it is possible to perform grid fusion of feature-level sensor data with the proposed system architecture. Especially the auto-encoder architectures show significant improvements in evaluation metrics compared to classical Bayesian fusion method.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124805992","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965028
G. Dupont, J. D. Hartog, S. Etalle, A. Lekidis
Modern vehicles are complex safety critical cyber physical systems, that are connected to the outside world, with all security implications it brings. Different network intrusion detection systems (NIDSs) proposed for the CAN bus, the predominant type of in-vehicle network, to improve security are hard to compare due to disparate evaluation methods adopted. In this paper we provide the means to compare CAN NIDSs on equal footing and evaluate the ones detailed in the literature. Based on this we observe some limitation of existing approaches and why in the CAN setting it is intrinsically difficult to distinguish benign from malicious payload. We argue that “meaning-aware” detection (a concept we define) which is challenging (but perhaps not impossible) to create for this setting.
{"title":"Evaluation Framework for Network Intrusion Detection Systems for In-Vehicle CAN","authors":"G. Dupont, J. D. Hartog, S. Etalle, A. Lekidis","doi":"10.1109/ICCVE45908.2019.8965028","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965028","url":null,"abstract":"Modern vehicles are complex safety critical cyber physical systems, that are connected to the outside world, with all security implications it brings. Different network intrusion detection systems (NIDSs) proposed for the CAN bus, the predominant type of in-vehicle network, to improve security are hard to compare due to disparate evaluation methods adopted. In this paper we provide the means to compare CAN NIDSs on equal footing and evaluate the ones detailed in the literature. Based on this we observe some limitation of existing approaches and why in the CAN setting it is intrinsically difficult to distinguish benign from malicious payload. We argue that “meaning-aware” detection (a concept we define) which is challenging (but perhaps not impossible) to create for this setting.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129404085","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 : 2019-11-01DOI: 10.1109/iccve45908.2019.8965015
{"title":"[Front matter]","authors":"","doi":"10.1109/iccve45908.2019.8965015","DOIUrl":"https://doi.org/10.1109/iccve45908.2019.8965015","url":null,"abstract":"","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130756371","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 : 2019-11-01DOI: 10.1109/ICCVE45908.2019.8965220
S. Hasan, Muhammed Abdullah Al Ahad, Irfan Šljivo, A. Balador, Svetlana Girs, Elena Lisova
Recent development in wireless technology enabling communication between vehicles led to introduction of the concept of Cooperative Adaptive Cruise Control (CACC), which uses wireless vehicle-to-vehicle communication and aims at string stable behavior in a platoon of vehicles. Degradation cascades have been proposed as a way to maintain a certain level of the system functionality in presence of failures. Such degradation behaviour is usually controlled by a runtime/state manager that performs fault detection and transitions the system into states where it will remain acceptably safe. In this paper, we propose a dynamic controller manager that focuses on both safety and performance of the system. In particular, it monitors the channel quality within the platoon and reacts by degrading platoon performance in presence of communication failures, or upgrading the performance when the communication quality is high enough. The reaction can include, e.g., adjusting the inter-vehicle distance or switching to another suitable platoon controller to prevent collisions. We focus on the functional and operational safety and evaluate the performance of the dynamic controller manager under different scenarios and settings in simulation experiments to demonstrate that it can avoid rear-end collisions in a platoon, continue platooning operation even in dense traffic scenarios where the state-of-the-art controllers fail to do so.
{"title":"A Fault-Tolerant Controller Manager for Platooning Simulation","authors":"S. Hasan, Muhammed Abdullah Al Ahad, Irfan Šljivo, A. Balador, Svetlana Girs, Elena Lisova","doi":"10.1109/ICCVE45908.2019.8965220","DOIUrl":"https://doi.org/10.1109/ICCVE45908.2019.8965220","url":null,"abstract":"Recent development in wireless technology enabling communication between vehicles led to introduction of the concept of Cooperative Adaptive Cruise Control (CACC), which uses wireless vehicle-to-vehicle communication and aims at string stable behavior in a platoon of vehicles. Degradation cascades have been proposed as a way to maintain a certain level of the system functionality in presence of failures. Such degradation behaviour is usually controlled by a runtime/state manager that performs fault detection and transitions the system into states where it will remain acceptably safe. In this paper, we propose a dynamic controller manager that focuses on both safety and performance of the system. In particular, it monitors the channel quality within the platoon and reacts by degrading platoon performance in presence of communication failures, or upgrading the performance when the communication quality is high enough. The reaction can include, e.g., adjusting the inter-vehicle distance or switching to another suitable platoon controller to prevent collisions. We focus on the functional and operational safety and evaluate the performance of the dynamic controller manager under different scenarios and settings in simulation experiments to demonstrate that it can avoid rear-end collisions in a platoon, continue platooning operation even in dense traffic scenarios where the state-of-the-art controllers fail to do so.","PeriodicalId":384049,"journal":{"name":"2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130903665","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：