Pub Date : 2021-10-29DOI: 10.1109/CVCI54083.2021.9661148
Xianjian Jin, Hang Yang, Zhiwei Li
Stable and efficient environmental perception in autonomous driving is an important prerequisite for path planning and behavior prediction. This paper proposes a vehicle identification framework based on LiDAR. After using the ground segmentation algorithm based on Gaussian process regression to complete high-quality ground segmentation, the adaptively density-based spatial clustering of applications with noise (A-DBSCAN) algorithm is used to cluster the remaining obstacle point clouds, and then the improved L-Shape algorithm is used for bounding box fitting. The experimental results based on the KITTI data set show that the framework has good stability under simple working conditions.
{"title":"Vehicle Detection Framework Based on LiDAR for Autonoumous Driving","authors":"Xianjian Jin, Hang Yang, Zhiwei Li","doi":"10.1109/CVCI54083.2021.9661148","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661148","url":null,"abstract":"Stable and efficient environmental perception in autonomous driving is an important prerequisite for path planning and behavior prediction. This paper proposes a vehicle identification framework based on LiDAR. After using the ground segmentation algorithm based on Gaussian process regression to complete high-quality ground segmentation, the adaptively density-based spatial clustering of applications with noise (A-DBSCAN) algorithm is used to cluster the remaining obstacle point clouds, and then the improved L-Shape algorithm is used for bounding box fitting. The experimental results based on the KITTI data set show that the framework has good stability under simple working conditions.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"123 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124398219","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 : 2021-10-29DOI: 10.1109/CVCI54083.2021.9661184
Fan Guo, K. Song, H. Xie
Smooth and accurate path tracking has a significant impact on safety and user-experience for self-driving buses. Path-tracking control, however, suffers from the uncertainties in the steering system, vehicle motion dynamics, and road-tire frictions, etc. In this paper, an extended state observer (ESO) based linear quadratic regulator (LQR) is proposed for the path-tracking of self-driving buses. In a departure from conventional LQR design, the bus motion dynamic is treated as a simple canonical model, the discrepancy of which from the real bus, caused by the nonlinearities and uncertainties, is lumped as a total disturbance state to be estimated by ESO online. With the total disturbance mitigated in the feedback loop in real-time, the bus motion dynamic is enforced to behave as the canonical model to be easily controlled by LQR. Experimental results show that relative to conventional LQR, the ESO-LQR design can enhance the transient response by 21%, and reduce the path-tracking error by up to 72 % in the presence of steering wheel angle offset.
{"title":"Extended State Observer Based Linear Quadratic Regulator for the Path-Tracking of Self-driving Buses","authors":"Fan Guo, K. Song, H. Xie","doi":"10.1109/CVCI54083.2021.9661184","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661184","url":null,"abstract":"Smooth and accurate path tracking has a significant impact on safety and user-experience for self-driving buses. Path-tracking control, however, suffers from the uncertainties in the steering system, vehicle motion dynamics, and road-tire frictions, etc. In this paper, an extended state observer (ESO) based linear quadratic regulator (LQR) is proposed for the path-tracking of self-driving buses. In a departure from conventional LQR design, the bus motion dynamic is treated as a simple canonical model, the discrepancy of which from the real bus, caused by the nonlinearities and uncertainties, is lumped as a total disturbance state to be estimated by ESO online. With the total disturbance mitigated in the feedback loop in real-time, the bus motion dynamic is enforced to behave as the canonical model to be easily controlled by LQR. Experimental results show that relative to conventional LQR, the ESO-LQR design can enhance the transient response by 21%, and reduce the path-tracking error by up to 72 % in the presence of steering wheel angle offset.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123052228","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 : 2021-10-29DOI: 10.1109/CVCI54083.2021.9661236
Ruitao Zhang, Yafei Wang, Shaoteng Wu
Loop closing is a crucial step to improve the accuracy of a SLAM system, which can reduce drift caused by frontend odometry and ensure global consistency of mapping process. In a traditional visual SLAM system, this is usually done by extracting and matching image features. The accuracy of the loop closure detection and the precision of the pose-constraint provided by loop closing is the bottleneck restricting the loop closing accuracy. Since image features is sensitive to large viewpoint changes, in this paper, we propose a novel neural network architecture VPD-Map, which takes visual pointcloud to provide a global 3D visual descriptor for fast loop closure detection and a top-view feature map for pose-constraint prior. Since the descriptor and feature map are based solely on visual pointcloud information, it is robust to viewpoint changes. VPD-Map can also serve as the backend of a visualSLAM system. Experiment on loop closure detection shows that this descriptor can perform viewpoint-free loop closure detection and outperforms traditional loop detection algorithms like bag-of-word model on the KITTI Odometry Dataset.
{"title":"VPD-Map: Visual Pointcloud Descriptor with Top-view Feature Map for Viewpoint-Free Loop Closure in Direct Visual Odometry","authors":"Ruitao Zhang, Yafei Wang, Shaoteng Wu","doi":"10.1109/CVCI54083.2021.9661236","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661236","url":null,"abstract":"Loop closing is a crucial step to improve the accuracy of a SLAM system, which can reduce drift caused by frontend odometry and ensure global consistency of mapping process. In a traditional visual SLAM system, this is usually done by extracting and matching image features. The accuracy of the loop closure detection and the precision of the pose-constraint provided by loop closing is the bottleneck restricting the loop closing accuracy. Since image features is sensitive to large viewpoint changes, in this paper, we propose a novel neural network architecture VPD-Map, which takes visual pointcloud to provide a global 3D visual descriptor for fast loop closure detection and a top-view feature map for pose-constraint prior. Since the descriptor and feature map are based solely on visual pointcloud information, it is robust to viewpoint changes. VPD-Map can also serve as the backend of a visualSLAM system. Experiment on loop closure detection shows that this descriptor can perform viewpoint-free loop closure detection and outperforms traditional loop detection algorithms like bag-of-word model on the KITTI Odometry Dataset.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"03 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127206069","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}
In the environment of intelligent and connected vehicles, the communication efficiency and information sharing level among vehicles and infrastructure get comprehensively improved. Aiming at the cooperative merging control at highway ramp, we present a vehicle cooperative control algorithm for distributed consensus considering communication delay among vehicles. First, we propose a group-based strategy and design a cost function to obtain the optimal vehicle passage order containing total travel time and delay. Then we model the merging vehicles formation as a multi-agent system and combine it with the multi-agent consensus theory to achieve the desired vehicle space and speed. Finally, based on the Webots simulation software and numerical analysis, we compare the traditional traffic signal control, Hamilton optimal control and the proposed method in terms of speed, position, arrival time and driving risk. The simulation experiment result shows that the efficiency of merging is increased by 47.3% and 11.7% on average respectively.
{"title":"Cooperation of Autonomous Vehicles at Merging Area Based on Multi-agent Consensus Control","authors":"Chen Zhao, Duanfeng Chu, Rukang Wang, Liping Lu, Kunyao Li, Chaozhong Wu","doi":"10.1109/CVCI54083.2021.9661126","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661126","url":null,"abstract":"In the environment of intelligent and connected vehicles, the communication efficiency and information sharing level among vehicles and infrastructure get comprehensively improved. Aiming at the cooperative merging control at highway ramp, we present a vehicle cooperative control algorithm for distributed consensus considering communication delay among vehicles. First, we propose a group-based strategy and design a cost function to obtain the optimal vehicle passage order containing total travel time and delay. Then we model the merging vehicles formation as a multi-agent system and combine it with the multi-agent consensus theory to achieve the desired vehicle space and speed. Finally, based on the Webots simulation software and numerical analysis, we compare the traditional traffic signal control, Hamilton optimal control and the proposed method in terms of speed, position, arrival time and driving risk. The simulation experiment result shows that the efficiency of merging is increased by 47.3% and 11.7% on average respectively.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"262 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114938689","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}
The combination of RTK and UWB is widely used for AGV positioning. In the case of large obstacle occlusion, GNSS-based localization methods may fail to obtain the accurate position of AGV. In this paper, we establish a new landmark dataset and put forward an auxiliary localization method based on landmark detection to improve the position accuracy, where a multi-task learning model is applied for both keypoint prediction and bounding box regression. We further propose the Keypoint Recovery Module (KRM) as a model-agnostic plug-in, to mitigate the challenge of missing rate. By this, the proposed approach is trained and validated on our proposed landmark dataset. Comparative experimental results show that the multi-task architecture in conjunction with KRM greatly enhances the accuracy of landmark detection, surpassing traditional methods.
{"title":"Visual marking-based AGV localization in parking lot","authors":"Dayi Tan, Wei Tian, Yuyao Huang, Qing Deng, Lu Xiong, Zhuoping Yu","doi":"10.1109/CVCI54083.2021.9661135","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661135","url":null,"abstract":"The combination of RTK and UWB is widely used for AGV positioning. In the case of large obstacle occlusion, GNSS-based localization methods may fail to obtain the accurate position of AGV. In this paper, we establish a new landmark dataset and put forward an auxiliary localization method based on landmark detection to improve the position accuracy, where a multi-task learning model is applied for both keypoint prediction and bounding box regression. We further propose the Keypoint Recovery Module (KRM) as a model-agnostic plug-in, to mitigate the challenge of missing rate. By this, the proposed approach is trained and validated on our proposed landmark dataset. Comparative experimental results show that the multi-task architecture in conjunction with KRM greatly enhances the accuracy of landmark detection, surpassing traditional methods.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"224 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122522325","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}
While the power battery playing a more and more important role in the society, its security is under an extremely strict requirement. But traditional fault diagnosis methods can’t ensure the safety of the batteries, since they are week in the detection of early minor battery fault. Therefore, a high-precision and low-cost battery diagnosis method is proposed. This method is based on the fusion of three individual sub-methods, that is modified correlation coefficient method, sample entropy and modified threshold method. The three sub-methods above are sensitive to different types of faults. By integrating their diagnosis result, the fusion method has a strong fault diagnosis ability for all types of faults. Meanwhile, it can largely shorten the early warning time of major faults. Result of experiments verifies that the method proposed does well in early minor fault diagnosis and can detect the fault at least 5 minutes in advance.
{"title":"An Early Battery Fault Diagnosis Method Based on Multi-Source Information Fusion Theory","authors":"Jinglun Li, Yunlong Shang, Xin Gu, Bin Duan, Yongzhe Kang, Chenghui Zhang","doi":"10.1109/CVCI54083.2021.9661260","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661260","url":null,"abstract":"While the power battery playing a more and more important role in the society, its security is under an extremely strict requirement. But traditional fault diagnosis methods can’t ensure the safety of the batteries, since they are week in the detection of early minor battery fault. Therefore, a high-precision and low-cost battery diagnosis method is proposed. This method is based on the fusion of three individual sub-methods, that is modified correlation coefficient method, sample entropy and modified threshold method. The three sub-methods above are sensitive to different types of faults. By integrating their diagnosis result, the fusion method has a strong fault diagnosis ability for all types of faults. Meanwhile, it can largely shorten the early warning time of major faults. Result of experiments verifies that the method proposed does well in early minor fault diagnosis and can detect the fault at least 5 minutes in advance.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"333 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122846617","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 : 2021-10-29DOI: 10.1109/CVCI54083.2021.9661159
Haorui Cheng, Hongjiu Yang, Z. Zuo, Hai Zhao
In this paper, an autonomous ground vehicle (AGV) motion system is developed considering full-state constraints by a Barrier Lyapunov functions (BLF) based finite-time motion control scheme. A BLF-based finite-time lateral controller is designed for the lateral subsystem in the AGV motion system with full-state constraints. A BLF-based finite-time longitudinal controller is proposed to limit the control deviation within a constraint for the longitudinal subsystem in the AGV motion system. Two nonlinear observers are designed to estimate the unmodeled dynamics in the lateral subsystem and the longitudinal subsystem of the AGV motion system. Effectiveness of the finite-time motion control scheme with full-state constraints is displayed by experimental results on an AGV experimental platform.
{"title":"Finite-Time Motion Control with Full-State Constraints for Autonomous Ground Vehicles on Curved Roads","authors":"Haorui Cheng, Hongjiu Yang, Z. Zuo, Hai Zhao","doi":"10.1109/CVCI54083.2021.9661159","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661159","url":null,"abstract":"In this paper, an autonomous ground vehicle (AGV) motion system is developed considering full-state constraints by a Barrier Lyapunov functions (BLF) based finite-time motion control scheme. A BLF-based finite-time lateral controller is designed for the lateral subsystem in the AGV motion system with full-state constraints. A BLF-based finite-time longitudinal controller is proposed to limit the control deviation within a constraint for the longitudinal subsystem in the AGV motion system. Two nonlinear observers are designed to estimate the unmodeled dynamics in the lateral subsystem and the longitudinal subsystem of the AGV motion system. Effectiveness of the finite-time motion control scheme with full-state constraints is displayed by experimental results on an AGV experimental platform.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"84 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114050015","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 : 2021-10-29DOI: 10.1109/CVCI54083.2021.9661183
Lisha Zhang, H. Bao, Feng Deng, Gangyan Li
Automatic pressure regulating valve is the core component of electro-pneumatic braking system of commercial vehicles, whose performance directly affects that of the brake system. In this paper, to realize intelligent braking, the design requirements of automatic pressure regulating valve controller are obtained through a comprehensive analysis of their application conditions and relevant standards. LabVIEW software is used to design a feedforward fuzzy PID (Proportional Integral Derivative) controller for new automatic pressure regulating valve, the test bench based on the response characteristics of which is constructed using a NI data acquisition card, sensors, and automatic pressure regulating valve, etc. The pressure response characteristics of automatic pressure regulating valve under the three control algorithms, namely PID control, feedforward PID control and feedforward fuzzy PID control， are verified by experiments. The experimental results under different algorithms are compared with relevant standards， and the results show that the feedforward fuzzy PID control algorithm provides the best pressure response characteristics of automatic pressure regulating valve with a shorter response time, a smaller overshoot and a better stability. The performance of automatic pressure regulating valve is improved in this paper so that the pressure of brake chamber can be better regulated dynamically, providing technical support for the precise control of commercial vehicle electro-pneumatic braking system.
自动调压阀是商用车电-气制动系统的核心部件，其性能直接影响到制动系统的性能。为了实现智能制动，本文通过对自动调压阀控制器的应用条件和相关标准的综合分析，得出了自动调压阀控制器的设计要求。利用LabVIEW软件设计了新型自动调压阀的前馈模糊PID (Proportional Integral Derivative，比例积分导数)控制器，并根据其响应特性，采用NI数据采集卡、传感器、自动调压阀等构成了试验台。通过实验验证了自动调压阀在PID控制、前馈PID控制和前馈模糊PID控制三种控制算法下的压力响应特性。将不同算法下的实验结果与相关标准进行对比，结果表明，前馈模糊PID控制算法具有响应时间短、超调量小、稳定性好的自动调压阀压力响应特性最佳。本文改进了自动调压阀的性能，使制动腔压力能够更好地动态调节，为商用车电气制动系统的精确控制提供技术支持。
{"title":"Controller of new automatic pressure regulating valve for electro-pneumatic braking system of commercial vehicles and its experimental study","authors":"Lisha Zhang, H. Bao, Feng Deng, Gangyan Li","doi":"10.1109/CVCI54083.2021.9661183","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661183","url":null,"abstract":"Automatic pressure regulating valve is the core component of electro-pneumatic braking system of commercial vehicles, whose performance directly affects that of the brake system. In this paper, to realize intelligent braking, the design requirements of automatic pressure regulating valve controller are obtained through a comprehensive analysis of their application conditions and relevant standards. LabVIEW software is used to design a feedforward fuzzy PID (Proportional Integral Derivative) controller for new automatic pressure regulating valve, the test bench based on the response characteristics of which is constructed using a NI data acquisition card, sensors, and automatic pressure regulating valve, etc. The pressure response characteristics of automatic pressure regulating valve under the three control algorithms, namely PID control, feedforward PID control and feedforward fuzzy PID control， are verified by experiments. The experimental results under different algorithms are compared with relevant standards， and the results show that the feedforward fuzzy PID control algorithm provides the best pressure response characteristics of automatic pressure regulating valve with a shorter response time, a smaller overshoot and a better stability. The performance of automatic pressure regulating valve is improved in this paper so that the pressure of brake chamber can be better regulated dynamically, providing technical support for the precise control of commercial vehicle electro-pneumatic braking system.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128559545","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}
With the concept of Intelligent Traffic(IT) proposed, intelligent connected road infrastructures have become a new trend. Intelligent Connected Signs(ICS) are one of products of IT, that can illustrate different colors and arrival time according to traffic flow on the ahead, as well as different colors included red, yellow, green which respectively represents congestion, slow traffic and smooth traffic. The author supposes that solutions to solve traffic congestion are traffic flow diverted in a range and drivers having enough time to change lanes smoothly. In order to verify the impact of ICS on traffic flow, the author utilizes PTV-VISSIM and Python as secondary traffic dynamic planning to compare total delay time and average travel time in a fixed route whether there is dynamic guidance. In the end, it is concluded that ICS greatly improve the congestion condition. Drivers’ lane-changing behavior is also influenced by the setting of ICS, so that the author utilizes CarSim, and Simulink software setting horizontal and longitudinal control strategies to build the driver’s lane-changing model. The output results of the speed after lane-changing and time of lane-changing are illustrated for the setting of ICS.
{"title":"Influence of Intelligent Connected Signs for Traffic Flow and Lane-Changing Behavior","authors":"Qishuai Xie, Yunshun Zhang, Minglei Gao, Jiaojiao Zhang","doi":"10.1109/CVCI54083.2021.9661142","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661142","url":null,"abstract":"With the concept of Intelligent Traffic(IT) proposed, intelligent connected road infrastructures have become a new trend. Intelligent Connected Signs(ICS) are one of products of IT, that can illustrate different colors and arrival time according to traffic flow on the ahead, as well as different colors included red, yellow, green which respectively represents congestion, slow traffic and smooth traffic. The author supposes that solutions to solve traffic congestion are traffic flow diverted in a range and drivers having enough time to change lanes smoothly. In order to verify the impact of ICS on traffic flow, the author utilizes PTV-VISSIM and Python as secondary traffic dynamic planning to compare total delay time and average travel time in a fixed route whether there is dynamic guidance. In the end, it is concluded that ICS greatly improve the congestion condition. Drivers’ lane-changing behavior is also influenced by the setting of ICS, so that the author utilizes CarSim, and Simulink software setting horizontal and longitudinal control strategies to build the driver’s lane-changing model. The output results of the speed after lane-changing and time of lane-changing are illustrated for the setting of ICS.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129086163","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 : 2021-10-29DOI: 10.1109/CVCI54083.2021.9661127
S. Liu, Shuai Zhao, Yongwang Shen, Yang Zhai, Xuliang Chen, Ziyi Liu
For autonomous vehicles, camera is a kind of commonly applied sensor. A vision-based scheme involving object detection, motion decision and tracking control is proposed in this paper for autonomous vehicles running on highways. A YOLO-DeepSORT network and a LaneNet network are adopted to realize object and lane detection. The decision task is carried out based on a designed finite state machine. To stabilize the tracking error, an enhanced pure-pursuit controller associated with lanes and an integral-separated PI controller are developed for the autonomous vehicle. At last, AD Chauffeur, a cloud simulation platform is utilized to verify the effectiveness, which is developed by Automotive Data of China (Tianjin) Co., Ltd. The experiment results not only verify that the performance of our vision-based strategy but also illustrate the functions of AD Chauffeur platform.
{"title":"Vision-Based Detection, Decision and Control for Autonomous Vehicles in AD Chauffeur","authors":"S. Liu, Shuai Zhao, Yongwang Shen, Yang Zhai, Xuliang Chen, Ziyi Liu","doi":"10.1109/CVCI54083.2021.9661127","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661127","url":null,"abstract":"For autonomous vehicles, camera is a kind of commonly applied sensor. A vision-based scheme involving object detection, motion decision and tracking control is proposed in this paper for autonomous vehicles running on highways. A YOLO-DeepSORT network and a LaneNet network are adopted to realize object and lane detection. The decision task is carried out based on a designed finite state machine. To stabilize the tracking error, an enhanced pure-pursuit controller associated with lanes and an integral-separated PI controller are developed for the autonomous vehicle. At last, AD Chauffeur, a cloud simulation platform is utilized to verify the effectiveness, which is developed by Automotive Data of China (Tianjin) Co., Ltd. The experiment results not only verify that the performance of our vision-based strategy but also illustrate the functions of AD Chauffeur platform.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123345728","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：