Search Results (7)

This paper offers a novel view for managing and controlling the movement of driverless, i.e., autonomous, vehicles by converting this movement to a simulated train movement moving on a rail track. It expands on the “virtual track” theory and creates a model for virtual track autonomous vehicle management and control based on the ideas and methods of railway train operation. The developed model and adopted algorithm allow for large-scale autonomous driving vehicle control on the highway while considering the temporal-spatial distribution of vehicles, temporal-spatial trajectory diagram optimization, and the management and control model and algorithm for autonomous vehicles, as design goals. The ultimate objective is to increase the safety of the road traffic environment when autonomous vehicles are operating in it together with human-driven vehicles and achieve more integrated and precise organization and scheduling of these vehicles in such mixed traffic conditions. The developed model adopted a “particle swarm” optimization algorithm that is tested in a hypothetical network pending a full-scale test on a real highway. The paper concludes that the proposed management and control model and algorithm based on the “virtual track” theory is promising and demonstrates feasibility and effectiveness for further development and future application. Full article

Obstacle intrusion is a serious threat to the safety of railway traffic. LiDAR point cloud 3D semantic segmentation (3DSS) provides a new method for unmanned rail-obstacle detection. However, the inevitable degradation of model performance occurs in complex weather and hinders its practical application. In this paper, a multi-modal contrastive learning (CL) strategy, named DHT-CL, is proposed to improve point cloud 3DSS in complex weather for rail-obstacle detection. DHT-CL is a camera and LiDAR sensor fusion strategy specifically designed for complex weather and obstacle detection tasks, without the need for image input during the inference stage. We first demonstrate how the sensor fusion method is more robust under rainy and snowy conditions, and then we design a Dual-Helix Transformer (DHT) to extract deeper cross-modal information through a neighborhood attention mechanism. Then, an obstacle anomaly-aware cross-modal discrimination loss is constructed for collaborative optimization that adapts to the anomaly identification task. Experimental results on a complex weather railway dataset show that with an mIoU of 87.38%, the proposed DHT-CL strategy achieves better performance compared to other high-performance models from the autonomous driving dataset, SemanticKITTI. The qualitative results show that DHT-CL achieves higher accuracy in clear weather and reduces false alarms in rainy and snowy weather. Full article

Non-signalized intersections have only ever been suitable for low traffic flow; however, with the development of autonomous driving technology and new control methods, the operation efficiency of this kind of intersection may be improved. In view of the shortcomings of existing non-signalized intersection control methods in multilane situations and inspired by railway trains, an interlocking-block intersection control model is proposed. In this study, vehicles between parallel lanes are combined into a few combos, and the combo shape can be determined according to a pairing model and the interlocking angle range, and the gaps between the front and rear vehicles are simulated as blocks in a railway system, which are added into the intersection control model as virtual blocked cars (VBCs) for optimization. In setting the optimization objectives, the connotation and realization of fairness are discussed. Experimental results show that compared with signalized intersections, roundabouts, and non-signalized intersections without control, the interlocking-block intersection control model greatly reduces vehicle delay. Compared with an existing model, the calculation speed in a multilane situation has been greatly improved, while the vehicle delay is similar. Full article

The transportation of dangerous goods by truck or railway multiplies the risk of harm to people and the environment when accidents occur. Many manufacturers are developing autonomous drones that can fly heavy cargo and safely integrate into the national air space. Those developments present an opportunity to not only diminish risk but also to decrease cost and ground traffic congestion by moving certain types of dangerous cargo by air. This work identified a minimal set of metropolitan areas where initial cargo drone deployments would be the most impactful in demonstrating the safety, efficiency, and environmental benefits of this technology. The contribution is a new hybrid data mining workflow that combines unsupervised machine learning (UML) and geospatial information system (GIS) techniques to inform managerial or investment decision making. The data mining and UML techniques transformed comprehensive origin–destination records of more than 40 commodity category movements to identify a minimal set of metropolitan statistical areas (MSAs) with the greatest demand for transporting dangerous goods. The GIS part of the workflow determined the geodesic distances between and within all pairwise combinations of MSAs in the continental United States. The case study of applying the workflow to a commodity category of dangerous goods revealed that cargo drone deployments in only nine MSAs in four U.S. states can transport 38% of those commodities within 400 miles. The analysis concludes that future cargo drone technology has the potential to replace the equivalent of 4.7 million North American semitrailer trucks that currently move dangerous cargo through populated communities. Full article

For smart mobility, autonomous vehicles, and advanced driver-assistance systems (ADASs), perception of the environment is an important task in scene analysis and understanding. Better perception of the environment allows for enhanced decision making, which, in turn, enables very high-precision actions. To this end, we introduce in this work a new real-time deep learning approach for 3D multi-object detection for smart mobility not only on roads, but also on railways. To obtain the 3D bounding boxes of the objects, we modified a proven real-time 2D detector, YOLOv3, to predict 3D object localization, object dimensions, and object orientation. Our method has been evaluated on KITTI’s road dataset as well as on our own hybrid virtual road/rail dataset acquired from the video game Grand Theft Auto (GTA) V. The evaluation of our method on these two datasets shows good accuracy, but more importantly that it can be used in real-time conditions, in road and rail traffic environments. Through our experimental results, we also show the importance of the accuracy of prediction of the regions of interest (RoIs) used in the estimation of 3D bounding box parameters. Full article

In recent years, the development of advanced systems and applications has propelled the adoption of autonomous railway traffic and train positioning, with several ongoing initiatives and experimental testbeds aimed at proving the suitability and reliability of the Global Navigation Satellite System signals and services, in this specific application domain. To satisfy the strict safety and accuracy requirements aimed at assuring the position solution’s integrity, availability, accuracy and reliability, recent proposals suggest the hybridization of the Global Navigation Satellite System with other technologies. The integration with localization techniques that are expected to be available with the upcoming fifth generation mobile communication networks is among the most promising approaches. In this work, different approaches to the design of hybrid positioning solutions for the railway sector are examined, under the perspective of the uncertainty evaluation of the attained results and performance. In fact, the way the uncertainty associated to the positioning measurements performed by different studies is reported is often not consistent with the Guide to the Expression of Uncertainty in Measurement, and this makes it very difficult to fairly compare the different approaches in order to identify the best emerging solution. Under this perspective, the review provided by this work highlights a number of open issues that should drive future research activities in this field. Full article

The development of advanced technologies has led to the emergence of autonomous vehicles. Herein, autonomous public transport (APT) systems equipped with prioritization measures are being designed to operate at ever faster speeds compared to conventional buses. Innovative APT systems are configured to accommodate prevailing passenger demand for peak as well as non-peak periods, by electronic coupling and decoupling of platooned units along travel corridors, such as the dynamic autonomous road transit (DART) system being researched in Singapore. However, there is always the trade-off between high vehicle speed versus passenger ride comfort, especially lateral ride comfort. This study analyses a new APT system within the urban context and evaluates its performance using microscopic traffic simulation. The platooning protocol of autonomous vehicles was first developed for simulating the coupling/decoupling process. Platooning performance was then simulated on VISSIM platform for various scenarios to compare the performance of DART platooning under several ride comfort levels: three bus comfort and two railway criteria. The study revealed that it is feasible to operate the DART system following the bus standing comfort criterion (a_y = 1.5 m/s²) without any significant impact on system travel time. For the DART system operating to maintain a ride comfort of the high-speed train (HST) and light rail transit (LRT), the delay can constitute up to ≈ 10% and ≈ 5% of travel time, respectively. This investigation is crucial for the system delay management towards precisely designed service frequency and improved passenger ride comfort. Full article