When Road Information Meets Data Mining: Precision Detection for Heading and Width of Roads

Abstract

Real-time road information plays a crucial role in enabling intelligent transportation systems (ITS) applications. With sufficient road information, the map of road topography can be built and updated more easily. Furthermore, many appealing ITS applications can be enabled accordingly. Aiming at improving the quality and update rate of road information, a hot topic today is how to mine information from global positioning systems (GPS) trajectories by the clustering-based methods. Such schemes, however, encounter two challenges: 1) GPS noise and 2) low sampling rate of GPS traces data. As a result, it is difficult to infer road information from these irregular clusters. To tackle the above issues, we directly mine useful road information, heading, and width of roads, for ITS applications from GPS point cloud, i.e., a set of GPS points. First, the distribution of GPS points is discussed and the least squares method (LSM) is demonstrated to be outstanding for mining the heading of the road under a huge number of GPS points. Second, the weighted approximation least squares method is proposed to improve the accuracy of the LSM. Furthermore, combining with relevant distribution features in GPS points, the data distribution variance-road width discrete model is proposed to mine road width from GPS point cloud. Finally, using real-world datasets, we demonstrate that these proposed methods can achieve satisfactory performance in practice. INDEX TERMSIntelligent transportation systems, road information, GPS data, data mining.The associate editor coordinating the review of this manuscript and approving it for publication was Zhong-Ke Gao. Chen et al.[10] apply a clustering-based method to mine road information from GPS data, aiming to reconstruct accurate digital maps for autonomous driving. Therefore, aiming to popularize intelligent transportation systems, it is necessary to improve data accuracy and timeliness of road information.Traditionally, this crucial road information is obtained through ground measurement [11]. It means moving devices equipped with sensors and visual systems are used to collect original information of road segments. This ground measurement is regarded as a time-consuming and expensive method. Recent years, with the research on ad hoc network, various individual mobile devices, such as unmanned aerial vehicles and robots [12], can collect road information simultaneously and share it with each other in this network. This way is effective to improve the collection rate for road information.

Real-Time Simulation Support for Real-Time Systems

“…Another promising direction to improve real-time simulation lies in the increasingly popular machine learning and deep learning models, which have been extensively in real-time systems to detect object location (Pan et al 2018), user activities (Lu et al 2019;Bhandari et al 2017), driver drowsiness (Zhang et al 2019), road conditions for vehicles (Zhou et al 2019;Xie et al 2018). As real-time systems are highly complex and essentially probabilistic, deep learning models can be used along with real-time simulators to improve the robustness and accuracy of simulation models.…”

Real-Time Simulation in Real-Time Systems: Current Status, Research Challenges and A Way Forward

“…Another promising direction to improve real-time simulation lies in the increasingly popular machine learning and deep learning models, which have been extensively in real-time systems to detect object location Pan et al (2018), user activites Lu et al (2019); Bhandari et al (2017), driver drowsiness Zhang et al (2019), road conditions for vehicles Zhou et al (2019); Xie et al (2018). As realtime systems are highly complex and essentially probalistic, deep learning models can be used along with real-time simulators to improve the robustness and accuracy of simulation models.…”

Wide-Attention and Deep-Composite Model for Traffic Flow Prediction in Transportation Cyber–Physical Systems

“…There are also diverse travelling manners from vehicular driving to bicycle-riding. Meanwhile, the increased traffic flows also result in the traffic congestion problem, which has become one of the main obstacles for urban development [1], [2]. The deployment of TCPS brings the opportunities to address this problem.…”

“…1(b), where the traffic flow is represented by the number of vehicles or bicycles during the t-th time interval from source A to destination B. Note that both vehicular traffic and cycling traffic are obtained from realistic datasets 1 . We can observe from Fig.…”