Patent Application
20130197790
1. Field of Invention
The present invention relates to intelligent transportation systems (ITS) and road traffic control, and more particularly to a method and system detecting congestions, enhancing traffic performances, and real-time traffic surveillance.
2. Related Art
The size and complexity of transport problems continue to increase with the growth of cities, road networks, and the number of motor vehicles. The main issue in urban transportations is traffic congestion and poor traffic performances. Traffic congestion is due to several factors such as the infrastructure, the ratio of number of vehicles with respect to the capacity of the road network, and the traffic signaling to name a few. Moreover, road traffic depends heavily on the time of the day. Rush hours generally occur at the time people commute to and from work, 8 am and 4 pm, and around lunch time, 12 pm. This pattern makes road traffic non ergodic. Despite this problem, a decent amount of research effort was devoted to traffic flow modeling and simulation.
The demand in terms of road space continues to grow for the reasons mentioned above. If the status quo persists—no new roads are built or no structural nor organizational changes are made, congestions are unavoidable. Their impacts are important and multiple. They result in economic, social, and environmental costs. It is thus necessary to limit, or at least, to manage road congestions. This can be done either by limiting the request for traffic or by managing the flow of vehicles.
When dealing with road traffic analysis, both modeling and simulation are viable alternatives. However, depending on the nature of problems at hand, one alternative may overtake over the other.
Vehicular traffic problems are usually treated in the literature, and much research has focused on methodologies for the optimization and evaluation of transportation systems (S. Chen et al., A Multimodal Hierarchical-Based Model for Integrated Transportation Networks, Journal of Transportation Systems Engineering and Information Techonology, 9(6):130-135, 2009). Lozano et al. (An algorithm for the recognition of levels of congestion in road traffic problems”, Mathematics and Computers in Simulation, 79(6):1926-1934, 2009) present an algorithm for identifying levels of congestion in traffic problems. D'Ambrogio et al. (Simulation model building of traffic intersections”, Simulation Modeling: Practice and Theory, 17(4):625-640, 2009) propose a model for an urban road network made up of traffic intersections.
Other research presented an analytical queuing model that preserves finite capacity queues and uses parameters to investigate the correlation between the queues (C. Osorio et al., An analytic finite capacity queuing network model capturing the propagation of congestion and blocking; European Journal of Operational Research 196(3):996-1007, 2009). This model can be validated by mathematical methods and existing simulation results. Finally, some studies measure the size of the queues of road intersections in order to find points of congestion in urban networks (Liu et al., Real-time queue length estimation for congested signalized intersections, Transportations Research Part C, 17(4):412-427).
US Patent application No. 2011/0115648 A1 and WPO 2009/122107 A1 patent application, which are applied by Laurgeau et al., provide a method for computing actual travel times using vehicles in the road network. Their method is based on devices on board of vehicles and relays disposed across the road network. In the preferred embodiment of their invention, a set of relays are positioned in points known by their GPS coordinates; a vehicle containing a transmitting device and passing by a relay drops a message with the vehicle ID; the messages are aggregated and processed in a data processing center and actual run times are computed. While this method is good in computing actual vehicles run times, it has several drawbacks: it is very expensive to implement as it requires vehicles to carry transmitting devices; a set of relays has to be deployed to allow for message reception. Several vehicles traveling between two given points in the network may have variable actual run time according to the experience, state of mind, and mood of the driver. This leads to fluctuating measurements that may need to be smoothed or expressed as an average and standard deviation couple of data. While this method is better than other methods using magnetic loops, as it does neither need road works nor significant maintenance, it is still subject to the deployment of specific equipment, namely relays, and the acquisition of transmitting devices that should be mounted on all vehicles.
Therefore, there is a need for a method and apparatus which does not present the drawbacks of the mentioned conventional methods.
This invention is directed to a method and system to analyze traffic performance, to detect potential congestions and to analyze the propagation of congestions, to reconfigure the network in terms of road signs and markings, and to monitor the traffic in real-time.
In accordance with an aspect of this invention, traffic performance analysis requires representing the road network as a graph, where the vertices are road intersections and the edges are road sections connecting intersections. Satellite images can be used as a source data to build the graph. Image processing techniques such as road extraction may be applied to extract the roads. This intermediate data may be converted to a suitable format such as Open Street Map or any other format known to the one skilled in the art. The road map is then augmented with road signs and markings as well as information regarding the capacity of the road section. The mathematical representation of the road network can be extracted semi-automatically or automatically to build a graph—in the sense of graph theory. It is then easy to apply the maximum flow or the max flow min cut algorithms to detect congestions. Although the same procedure allows tracking down the congestion propagation, other methods are known to those skilled in the art. Indeed, other techniques based on queuing systems also allow for the road traffic analysis (Ouali et al., A Multiclass BCMP Queuing Modeling and Simulation-Based Road Traffic Flow Analysis, ACM Simultech, The Netherlands, 2011; Boris S. Kerner, Introduction to Modern Traffic Flow Theory and Control: the long road to three-phase traffic theory, Springer-Verlag Berlin Heidelberg, 2009).
In accordance with a further aspect of the invention, it is also desirable to process an existing urban road network that is subject to congestion. This is achieved through network reconfiguration. The operations described in this stage are semi-automatic, although they could be automatic. In this stage, a human traffic expert or an operator suggests to update locally the road signs and signaling to alleviate congestion. However, one needs to verify that this solution is viable.
This is done through discrete event simulation. The simulator is built in such a way that it takes into account the actual graph—with capacities, road signs, and signaling. The detection of congestions is confirmed when the actual flow reaches the capacity of the edge.
In accordance with another aspect of the invention, real-time traffic monitoring concerns the surveillance of the road traffic. A set of cameras is installed along the main boulevards, avenues, and expressways in the road network. The average speed of the traffic is measured by measuring the optical flow in the video. The traffic speed is compared to the locally defined speed limit: if the traffic speed is below a certain ratio of the speed limit, a human operator who is monitoring the traffic in control rooms or traffic information systems is requested to find the glitch downstream. The operator may change the timing of traffic lights or modify electronic traffic signs to solve the problem. The measured traffic speeds are aggregated to compute the estimated run times between two points of the road network according to the current traffic conditions. Estimated run times are displayed on variable-message signs. It should also be noted that it is possible to count the number of vehicles
The accompanying drawings show an embodiment having no limiting character. The invention will be described with reference to the accompanying drawings, wherein:
For purposes of explanation, specific embodiments are set forth to provide a thorough understanding of the present invention. However, it will be understood by one skilled in the art, from reading this disclosure, that the invention may be practiced without these specific details. Moreover, well-known elements, devices, process steps and the like are not set forth in detail in order to avoid obscuring the scope of the invention described.
The present invention provides a method and system for traffic performance analysis, congestion detection and network reconfiguration, and finally, real-time traffic monitoring and surveillance. The source data structure is in the form of a fully described graph. The graph is built for a particular road network or a city. To build the graph, satellite or aerial images of the city are processed in order to extract the road network. Other methods known to those skilled in the art are also available. Graph theory algorithms such as maximum flow or max-flow min-cut may be used to find critical points or hot spots in the graph. Hot spots are areas where the flow tends towards the capacity of an edge, which characterizes congestions. This analysis might be cumbersome for large graphs. Furthermore, the max flow algorithms do not allow considering the rate of generation and the rate of absorption in each edge. Discrete event simulation, however, is a convenient tool that may be used.
Once the congestions are found in the graph, a traffic expert suggests some changes to road signs and markings in order to overcome the congestions. For every suggested change, the simulator confirms whether or not the congestion has been definitely overcome or if it still persists.
It should be noted that the steps of locally updating the road signs and traffic light durations and verifying that these changes overcome congestions may also be performed automatically and algorithmically without departing from the scope of the invention. When all congestions are solved, the augmented graph data structure is saved. This data structure is used during traffic monitoring and surveillance so that when the traffic flow slows down and reaches a critical limit, the operator uses the graph data structure along with the simulator to determine the propagation of the congestion upstream. Consequently, the operator, who ideally is a traffic expert, changes road signs and/or traffic light durations temporarily to help improve the traffic conditions.
The graph data structure computation 10 is based on the usage of the digital augmented map 8, where roads intersections are vertices in the graph, and road sections are edges. Further information is appended to the graph data structure, such as the road signs and markings, traffic lights, edge capacity, rate of generation, and the rate of absorption. The graph data structure is stored in a memory 12. The detection and propagation of congestions 14 and 18 is based on the discrete event simulator embedded in a data processing system 16. The data processing system takes the graph data structure 10 as input and processes discrete events—vehicles traveling along road sections and negotiating intersections. Whenever a traffic flow gets closer to the corresponding edge capacity, there is a risk of congestion. The edge flow is saturated and the congestion is propagated upstream in the graph. At this stage we are only concerned by the analysis of the current state of the traffic conditions.
When the initial evaluation of the road network reveals the existence of potential or actual congestions, the data processing system 16 is used. A traffic expert examines locally the road signs, markings and traffic lights and makes local changes to overcome the congestion. The data processing system 16 re-processes the new changes within the graph data structure. This stage is depicted in
The last stage of the system is the traffic monitoring and surveillance. The flow chart in
The traffic speed is computed by processing videos acquired by surveillance cameras 42. The video processing takes place in the data processing center 44. If the traffic speed on road section 40 is found to fall below a threshold or a rate, the data processing center triggers an alarm 46 to a human operator. The data processing center 44 switches to display 48 the road section of interest to the operator for visual confirmation. The operator updates the road signs and the traffic light durations in the upstream road sections by sending new signaling to specific smart hardware 50 and 52 for example. Another embodiment of the invention is the detection of the speeding vehicle 38 and the extraction of plate number either automatically by processing the current frame or by a human operator.
While the invention has been described according to what is presently considered to be the most practical and preferred embodiments, it must be understood that the invention is not limited to the disclosed embodiments. Those ordinarily skilled in the art will understand that various modifications and equivalent structures and functions may be made without departing from the scope of the invention as defined in the claims. Therefore, the invention, as defined in the claims, must be accorded the broadest possible interpretation so as to encompass all such modifications and equivalent structures and functions.
This application claims the benefit of the filing date of provisional application No. 61/593,291, filed on Jan. 31, 2012. The contents of the provisional application are incorporated by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61593298 | Jan 2012 | US |
