Traffic congestion assessment based on street level data for on-edge deployment

With the increasing technological advancements and fields such as the industrial internet, autonomous driving, smart homes, intelligent transportation, and smart cities, all may benefit greatly from edge applications. These fields demand greater speed and efficiency for computing, network transmission, and user interaction in order to keep up with the growth. Stemming from the unique speed-up advantages that the edge computation model gives over traditional cloud computation, it is a suitable technology for deployment on the roads and streets, making edge a prime candidate to process traffic data. As previously mentioned, intelligent transportation, autonomous driving and smart cities all require intense computation to ensure they function correctly, and deploying edge on roads serves to address this issue. We explore the KITTI data set from the perspective of analyzing traffic congestion on complex roads. Our analysis shows: 1) based on object bounding boxes expressed within the data, each instantaneous situation can be separated into non-congestion and congestion; 2) the mid-point distances between bounding boxes effectively determines the congestion level. To the best of our knowledge, this is the first proposed use of a non-continuous timestamped data set for traffic congestion detection.

[1]  Jianxiong Xiao,et al.  DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[2]  Ji Wan,et al.  Multi-view 3D Object Detection Network for Autonomous Driving , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Weisong Shi,et al.  Edge Computing: Vision and Challenges , 2016, IEEE Internet of Things Journal.

[4]  Qing Yang,et al.  Cooper: Cooperative Perception for Connected Autonomous Vehicles Based on 3D Point Clouds , 2019, 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS).

[5]  Andreas Geiger,et al.  Vision meets robotics: The KITTI dataset , 2013, Int. J. Robotics Res..

[6]  Qi Wang,et al.  Locality constraint distance metric learning for traffic congestion detection , 2018, Pattern Recognit..

[7]  Sanja Fidler,et al.  Monocular 3D Object Detection for Autonomous Driving , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Yu-Wing Tai,et al.  Accurate Single Stage Detector Using Recurrent Rolling Convolution , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Chan-Tong Lam,et al.  A real-time traffic congestion detection system using on-line images , 2017, 2017 IEEE 17th International Conference on Communication Technology (ICCT).

[10]  Lakshminarayanan Subramanian,et al.  Road traffic congestion in the developing world , 2012, ACM DEV '12.

[11]  Qi Chen,et al.  Low-Latency High-Level Data Sharing for Connected and Autonomous Vehicular Networks , 2019, 2019 IEEE International Conference on Industrial Internet (ICII).