An arc orienteering algorithm to find the most scenic path on a large-scale road network

Efficient fastest path computation on a large scale road network is an essential problem in logistics distribution, especially based on real-time traffic information. In this paper, we present an adaptive fastest path algorithm capable of efficiently accounting for the network with real-time traffic data. (1) Logistics vehicle routing between two cities usually choose the road network through the largest roads. According to this traveler character and the natural hierarchy present in road network, we classify the roads into largest roads and smaller ones. (2) Based on the largest roads and smaller roads, we can partition the road network into areas. (3) Finally, we propose an efficient A* algorithm that uses the road network partition to limit the search space and considers the real-time traffic information. Through an extensive experimental evaluation on real road network we show that our algorithm provides desirable routes, and that it is significantly fast and effective.

Ride sharing has been widely studied in academia and applied in mobility-on-demand (MoD) systems as a means of reducing the number of cars, congestion, and pollution by sharing empty seats. Solving this problem is challenging on large-scale road networks for the following two reasons: distance calculation on large-scale road networks is time consuming; and multi-request allocation and multi-point planning have been proved to be NP-hard problems. In this paper, we propose a clustering-based request matching and route planning algorithm Roo that considers spatial-temporal distances between ride requests on road networks. The Roo algorithm is evaluated with real-world taxi trajectory data and road networks from New York City and Beijing. The results show that Roo can save up to 50% of mileage by 1000 vehicles serving around 7000 trip requests in New York City between 7:40 am to 8:00 am with average waiting time of 4 minutes.

Co-evolutionary traffic signal control using reinforcement learning for road networks under stochastic capacity

Mobile users prefer to choose personalized travel routes using their mobile terminals. Keyword-aware Optimal Route Query (KORQ) is proposed to meet users' need because it not only considering the length of the route, but also considering the cost of the route and the points of interest covered by the route. As the number of points of interest in road networks increases sharply, the time and space complexity for preprocessing and route expansion rises dramatically, and the state of the art algorithms are not scalable. To address these issues, we propose an algorithm called KORAL, short for Keyword-aware Optimal Route Query Algorithm on Large-scale Road Networks. To reduce the overhead of preprocessing, KORAL partitions the road network into subgraphs, and stores only the information about the routes between subgraphs in preprocessing stages. In the rout expansion stage, KORAL uses a strategy called minimum budget pruning to prune infeasible vertices, which greatly speed up the route expansion. Experiments against 3 datasets of New York road network in a server with 16G RAM show that KORAL is scalable and breaks the limitation that the road network cannot exceed 100K vertices.

The area rate of the toll road network is designed to influence the regional road network traffic flow configuration and an important factor of goods vehicle path selection, transportation, and other related departments in formulating the related standard. The toll rate is usually combined with the local passenger and cargo loading conditions. Adjusting and optimizing the charge policy, optimization of processes and methods is still facing many difficulties. In this paper, the model parameters and toll status of various goods vehicles are considered, and an optimization algorithm of toll rates for different vehicles in a large-scale road network is proposed. By constructing a traffic demand prediction model based on multiple truck classification, this paper analyzes the distribution status of network trucks and the final charging under the influence of different rates. At the bottom of the algorithm is an analytical model composed of nonlinear equations whose dimensions scale linearly with the scale of the road network. The scale of the road network is independent of the path set and link attribute. The model is verified and the parameters are calibrated by a small network. Finally, the Road network is selected for empirical analysis in Xinjiang Autonomous Region. A case study shows that the analytical structure information provided by the analytic network model enables the algorithm to quickly identify high quality solutions, and the algorithm has better simulation accuracy and premium rate design advantages.

A dynamic spatial–temporal deep learning framework for traffic speed prediction on large-scale road networks

Much research has recently been devoted to the data management foundations of location-based mobile services. In one important scenario, the service users are constrained to a transportation network. As a result, query processing in spatial road networks is of interest. We propose a versatile approach to k nearest neighbor computation in spatial networks, termed the Islands approach. By offering flexible yet simple means of balancing re-computation and pre-computation, this approach is able to manage the trade-off between query and update performance. The result is a single, efficient, and versatile approach to k nearest neighbor computation that obviates the need for using several k nearest neighbor approaches for supporting a single service scenario. The experimental comparison with the existing techniques uses real-world road network data and considers both I/O and CPU performance, for both queries and updates.

Road networks are a type of spatial network, where edges may be associated with qualitative information such as road type and speed limit. Unfortunately, such information is often incomplete; for instance, OpenStreetMap only has speed limits for 13% of all Danish road segments. This is problematic for analysis tasks that rely on such information for machine learning. To enable machine learning in such circumstances, one may consider the application of network embedding methods to extract structural information from the network. However, these methods have so far mostly been used in the context of social networks, which differ significantly from road networks in terms of, e.g., node degree and level of homophily (which are key to the performance of many network embedding methods). We analyze the use of network embedding methods, specifically node2vec, for learning road segment embeddings in road networks. Due to the often limited availability of information on other relevant road characteristics, the analysis focuses on leveraging the spatial network structure. Our results suggest that network embedding methods can indeed be used for deriving relevant network features (that may, e.g, be used for predicting speed limits), but that the qualities of the embeddings differ from embeddings for social networks.

With the development of the automotive industry, urban transportation systems face increasingly complex challenges. To address the slow computation and low efficiency of the traditional Dijkstra's algorithm in large-scale road networks, this study proposes the following improvements: firstly, it employs binary, 4-ary, and pairing heap structures to accelerate shortest path calculations; secondly, it introduces a bidirectional strategy to reduce unnecessary node traversals; finally, real-world road network data for Guilin and Nanning obtained via OpenStreetMap (OSM) are used to compare computation times of unidirectional and bidirectional Dijkstra algorithms under different heap structures. The results demonstrate that the bidirectional search strategy significantly enhances efficiency in large-scale networks, while the binary heap structure provides optimal performance in scenarios involving frequent updates and traversals, effectively reducing computation time. This research provides efficient algorithmic support for urban traffic path planning and lays theoretical and data foundations for further study of multi-heap structures in shortest path problems.

With the advent of automated vehicles (AVs), new infrastructure planning concepts such as subnetworks of AV-ready roads have been proposed to improve the performance of transportation networks and to promote the adoption of AVs. However, these subnetworks should evolve over time in response to the growing AV demand, which necessitates a multi-stage modeling approach. In this study, we propose multi-stage deployment of AV-ready subnetworks and formulate it as a time-dependent network design problem, which is a bi-level mixed-integer programming problem. The lower level is a simultaneous travel mode and route choice equilibrium with continuous decision variables, and the upper level is a design problem including infrastructure investment decisions to determine which roads to upgrade and include in AV-ready subnetworks for mixed traffic. We use a case study of a real road network to demonstrate the concept. Since computational efficiency is a key factor for solving such large-scale problems, we develop two efficient and tailored evolutionary heuristics to solve the problem, and compare their performance to a computationally demanding Genetic-algorithm-based solution method. The results indicate that the proposed algorithms can efficiently solve this large-scale problem while satisfying constraints in all scenarios, and outperform Genetic algorithm, particularly in the scenario with larger number of stages. Moreover, in all scenarios, deployment of AV-ready subnetworks leads to improvements in network performance in terms of total travel time and cost. However, the improvements are always accompanied with increased total travel distance. The extent of changes depends on AV market penetration rate, AV-ready subnetwork density and timing of densification.

Road network deals with the development of a comprehensive plan for construction and operation of transportation facilities. In order to develop efficient and better transport facility, it is necessary to have a proper road network. In sustainable road network planning, planners put into consideration factors like gradients or slope, land-use and geology with community and governmental interest. These different considerations make the planning process complex and generate confusion in the decision making process. The use of geographic information system (GIS) and multi-criteria analysis (MCA) has helped planners to reduce complexity and to achieve desired and more accurate results. MCA prevents the imposition of criteria limit and gives opportunity to decision makers to enter their own judgments. This provides a better communication among the community for creating a more open choice for analysis and possible changes if necessary. In this study, road network has been analyzed with optimal least cost path algorithm of spatial analysis in GIS using different ancillary data layers and each layer weight-scoring has been computed with MCA in spatial decision support system (SDSS). The optimal least cost path would provide the best option with certainty and considers a gradient, connected neighbors, thematic cost and surface distance in three dimensional spaces. The path gradient can be adjusted as per the requirements, depending upon the terrain conditions and possible to design a more realistic route automatically with appropriate parameters.Nepalese Journal on Geoinformatics -13, 2014, Page: 34-40

Hybrid deep learning models for traffic prediction in large-scale road networks

The author describes technical and algorithmic solution of the task of building a spatial routing network for pedestrian panoramic survey points at monitoring urban infrastructure objects. Its urgency is caused by the necessity of making a panoramic measuring the yard, intra-block territories, as well as park areas and other objects which have no reliable data on the road network there. The matter of spatial routing network providing between panoramic images without reference to the road graph is considered and formulated. Based on the presented conditional variant of solving the task, as well as considering the problem in the view of graph theory, geometric conditions are compiled, the necessary sets of Geodata are determined, and the method of spatial routing network building is developed. Algorithm complexity analysis is performed. An intelligent GIS agent was developed providing filling of the DBMS table with spatial features, which are the object implementation of the spatial routing network in the GIS. Geoportal that helps the final user analyze and adjust the mentioned network with the introduction of changes to the DBMS on the server side in an interactive mode is presented.

Estimation of travel distances with the weighted ℓ p norm: Some empirical results

To facilitate the travel preparation process to a city, a lot of work has been done to recommend a POI or a sequence of POIs automatically to satisfy users' needs. However, most of the existing work ignores the issue of planning the detailed travel routes between POIs, leaving the task to online map services or commercial GPS navigators. Such a service or navigator in terms of suggesting the shortest travel distance or time, which cannot meet the diverse requirements of users. For instance, in the case of traveling by driving for leisure purpose, the scenic view along the travel routes would be of great importance to users, and a good planning service should put the sceneries of the route in higher priority rather than the distance or time taken. To this end, in this paper, we propose a novel framework called ScenicPlanner for route recommendation, leveraging a combination of geotagged image and check-in digital footprints from locationbased social networks (LBSNs). First, we enrich the road network and assign a proper scenic view score to each road segment to model the scenic road network, by extracting relevant information from geo-tagged images and check-ins. Then, we apply heuristic algorithms to iteratively add road segment and determine the travelling order of added road segments with the objective of maximizing the total scenic view score while satisfying the user-specified constraints (i.e., origin, destination and the total travel distance). Finally, to validate the efficiency and effectiveness of the proposed framework, we conduct extensive experiments on three real-world data sets from the Bay Area in the city of San Francisco, which contain a road network crawled from OpenStreetMap, more than 31 000 geo-tagged images generated by 1 571 Flickr users in one year, and 110 214 check-ins left by 15 680 Foursquare users in six months.