Scanning the Issue

Azim Eskandarian
2020 IEEE transactions on intelligent transportation systems (Print)  
This article is devoted to surveying the latest progress on maritime traffic data mining technologies and maritime traffic forecasting technologies. Through the review, we highlight that maritime traffic patterns and knowledge base are useful for wide-spectrum domain applications such as traffic evaluation, visualization, forecasting, and anomaly detection. Maritime traffic forecasting is an essential component for better situation awareness, which contributes toward early collision alert,
more » ... ic hotspot detection, and serves as input for mitigation planning and actions. The development of maritime traffic research in pattern mining and traffic forecasting reviewed in this work affirms the importance of advanced maritime traffic studies and the great potential in sea transport to accommodate the implementation of the Internet of Things (IoT), Artificial Intelligence (AI) technologies, as well as knowledge engineering and big data computing solution. This article presents a review of motion planning techniques over the last decade for autonomous vehicles driving on highways. The state-of-the-art focuses on trajectory generation and decision making, and addresses lane changing, obstacle avoidance, car following, and merging situations. A novel classification, centered on the intrinsic attributes of the methods, is enhanced with a radar charts representation. Furthermore, a comparison table is proposed, following two purposes: 1) referring to an algorithm with an appropriate behavior regarding the problem description, as motion planning techniques are highly dependent on the constraints of perception, control, and environment and 2) giving insights on meta-algorithms for systemic treatment of motion planning. Finally, this article overviews the current and future challenges of implementing motion planning algorithms in a mixed environment with both manually driven and autonomous vehicles. This article provides a critical review of the state-of-theart pedestrian and evacuation dynamics. Types of typical data collection methods are classified and the connections and differences of three observation methods are explored. Pedestrians' complex behaviors characterized by the self-organization Digital Object Identifier 10.1109/TITS.2020.2989542 phenomena and movement data characterized by the fundamental diagram are then studied after the data collections, which can be used to calibrate and validate the pedestrian models. The mathematical models for pedestrian dynamics from both tactical level and operational level are also highlighted. The simulation data produced by the mathematical models could further reproduce pedestrian behaviors during the observations and contribute to decision makings for improving the evacuation efficiency. The applications of pedestrian models for behavior analysis, evacuation simulation, and layout design are also presented. Some challenges and future directions in the pedestrian and evacuation dynamics are also put forward. Estimating Travel Time Distributions by Bayesian Network Inference A. Prokhorchuk, J. Dauwels, and P. Jaillet Travel time estimation is a crucial task for intelligent transportation systems. To account for the uncertainty of travel times in an urban context, a path travel time distribution estimation model is presented. The proposed framework combines Gaussian copulas with Bayesian network inference to accurately estimate travel time distributions even from a sparse data set. Computational experiments are performed on GPS measurements from probe vehicles (taxis) in Singapore. The numerical results show that this approach produces distributions that are closer to empirical ones when compared to various baseline models. Relationships between estimation accuracy and factors such as path length, day of the week, and temporal data resolution are also investigated. This article recognizes the research gaps and difficulties in generating transition lines (the paths that pass through road intersection) in road intersections from mobile laser scanning (MLS) point clouds. The proposed method contains three modules: road surface detection, lane marking extraction, and transition line generation. The experimental results demonstrate that transition lines can be successfully generated for both T-and cross-intersections with promising accuracy. In the validation of lane marking extraction using the manually interpreted lane marking points, the method can achieve average precision, recall, and F1-score of 90.80%, 92.07%, and 91.43%, respectively. The success rate of the transition line generation is 96.5%. Furthermore, the bufferoverlay statistics (BOS) method validates that the proposed method can generate lane centerlines and transition lines within 20-cm-level localization accuracy from the MLS point clouds. 1524-9050
doi:10.1109/tits.2020.2989542 fatcat:hg6pjmk6zvfurjdqufxeslibpa