«application of geodetic and information technologies in the physical planning of territories»

Дата канвертавання17.01.2017
Памер28.56 Kb.



Sofia, 09 – 10 November, 2000

Everyday Mapping of Traffic Conditions – An Urban Planning Tool

P. Savvaidis; I.M. Ifadis; K. Lakakis

Thessaloniki, Greece


A basic part of a completed urban plan is a traffic study and research. A major problem in these researches is the availability of traffic data and the cost in money and time of their updating. In this paper, the concept of using a number of public vehicles as sensors for mapping and monitoring the traffic conditions in an urban area is described. The fleet management system used for this task is working by using GPS technology and DGPS methodology. In addition, a GIS software platform is used for providing network analysis capabilities. The major scope is to collect primary traffic data, like transit times, by using GPS as an accurate timer with spatial reference. Also, an accurate digital map of the street network of the area can be generated. By using collected traffic data and employing proper transformation algorithms, standard traffic characteristics as traffic volume, traffic density, traffic capacity, etc. can be obtained.

1. Introduction

A major problem in designing a traffic plan for an urban area is the availability of traffic data and the cost in time and money of their updating. The traffic conditions are characterized or can be simulated by a number of parameters, such as mean vehicle speed, traffic volume, traffic density, traffic capacity etc. [1]. Until recently, the traditional ways of obtaining traffic data were basically observations carried out at certain points or the movement of a vehicle along traffic routes and the use of recording instrumentation. All the above mentioned techniques, although simple in practice, are time consuming, need a large number of observers to cover an entire road network and provide data only for the period of the measurements. Updating the observed data in a future occasion due to changes in traffic conditions (something very common in developing urban areas) requires the organization of a new survey, while the results will again soon be out of date. In order to overcome these problems, a land navigation system developed in the Laboratory of Geodesy, School of Civil Engineering under the name “VECON” System (VEhicle COntrol and Navigation System) [2] was used in a pilot project as the mean for the determination of the basic traffic parameters in the downtown part of the city of Thessaloniki. VECON System is basicly a Fleet Management System, but it has been designed so that it can be used both for building its own digital map and for continuous monitoring traffic conditions in the sense of travel time and travel speed computations.

2. Structure of the VECON System
The VECON System consists of two parts, the control station and the vehicle sensors (fig.1). These two main parts consist of other smaller units. Thus, we have:
a) The control station (fig. 2, 3), which involves:

  1. A base GPS station. In this case the base GPS station is the continuous reference GPS station that has been operational by the Laboratory of Geodesy since the end of 1999 [3].

  2. A Geographical Information System and other software for processing of data and computations.

  1. A communication unit (UHF radio network).

  2. Operator(s).

b) The vehicle sensors (fig. 4), each of which involves:

  1. A GPS receiver (RTCM ready).

  2. A communication unit.

  3. An operator (driver).

The operation of the VECON system, which is a real time system, is based on the typical DGPS procedure [4]. The GPS receiver of the continuous reference station is mounted on a point with known coordinates and produces differential corrections comparing the known coordinates with those given continuously from GPS observations. These differential corrections are transmitted to the vehicle through the communication systems in RTCM format.

Another GPS receiver mounted on each vehicle continuously estimates the position of the vehicle with an accuracy of about 15-30m. Using the differential corrections these estimations can be improved up to 1-3m. This final precision is satisfactory for determining the position of the vehicle along its route [5], [6]. The corrected coordinates of the vehicle in the NMEA format are transmitted back to the control station and feed the Geographical Information System, where several operations, such as animation of the movement of the vehicle, shortest path computations etc. can be done.

When incorporating the system for the first time, the GIS operations are based on available existing digital maps of the area of interest with details about the road network. The return signals with the precise vehicle positions give the opportunity to the system to compute and display a new digital map of the area under investigation with the accuracy of the DGPS. Depending on the number of vehicle sensors and their travelling along the urban road network, the system gradually builds its own map in a continuous and self-feeded way. The produced digital database is the most proper for navigation because the navigation data experience the same error sources as the digital database itself.

Another possibility that arises from the GPS module of the system is the continuous recording of time data. This comes from the fact that GPS is a very accurate timekeeper with spatial reference. The obtained vehicle positions always refer to specific times easily found in the resulting files. As a result, by using the appropriate software, the time needed for each vehicle sensor to travel between two nodes of the road network can be computed.

Figure 1. General framework of the Control Operation Center in an urban area

Figure 2. The antennas of the Thessaloniki Continuous Reference GPS Station

and the UHF radio network

Figure 3. The continuous reference GPS receiver, the GIS station and part of the telecommunication system.

Finally, during navigation, the system will continuously receive new data and thus continuously update its dynamic models of road layout and travel times. The basic mathematical models used in the presented system (VECON) for the processing of the incoming DGPS data have been described elsewhere [7], [8]. By employing these models, a spatial digital map of the road network of the urban area can be developed due to the positioning data from all vehicles as well as a time digital map for all road segments travelled due to the respective time data.

3. Employing VECON for the determination of traffic data
The ability of VECON System to record positional and time data was used for the determination of traffic data in a pilot project covering the downtown part of the city of Thessaloniki. Three vehicle sensors were used to investigate the efficiency of the system. The vehicles travelled along the roads of the area under consideration at different time zones during the day, according to the expected traffic conditions, for a total of several days (fig. 5). The road network was divided into segments starting and ending at the intersections of the roads (or nodes of the network). For each segment, the travel time between the starting and the ending node was computed from the GPS time data. Knowing the distance between the nodes and the respective travel time, the speed of the vehicle along the particular road segment was computed. Due to the rather small distances between the nodes, we can accept that the above derived speed values represent the mean vehicle speed for that part of the road.

Figure 4. The vehicle unit including a GPS receiver, a UHF transceiver

and a radio modem

Traffic volume Q (vehicles/hour) can be related to mean vehicle speed V both theoretically and empirically. For example, this computation can be done by using the equations proposed by Greenshields (eq. 1) or by Greenberg (eq. 2) [1]:

Q = 122 V – 1.65 V2 (1)

Q = 227 V e –V/17.2 (2)
Traffic density K (vehicles/km or mile) can be also computed from empirical equations giving the value of K as a function of the mean vehicle speed V (the Greenberg relation (eq. 3)) or connecting K to the traffic volume Q which can be derived from the speed data (eq. 4) [1]:
K = 227 e –0.0581 V (3)
Q = 74 K – 0.61 K2 (4)
Following the computation of traffic volume or/and traffic density and the GPS time data, more traffic parameters can be determined, such as traffic capacity for a road segment (vehicles/specific period of time).

The key factor in the above discussion is the ability of the vehicle sensors of the VECON System to record travel times. The VECON vehicles are driven along the road network collecting data. In our case, the collected data are time and corrected in real time coordinates of the trajectories of the vehicles. In a more simple way, similar computations can be performed with recording of GPS raw data and post – processing software. In both cases, the data base of observations will contain the necessary information for traffic data calculations.

Figure 5. An example of the employment of a vehicle sensor for observations

along a series of road network segments

The application of the VECON System for the measurement of time data can be done:

  1. By employing a limited number of vehicle sensors travelling along the road network of an urban area according to a pre-determined plan.

  2. By employing an already operational vehicle fleet with units which travel in the roads.

The use of urban transportation units such as buses will give information about the traffic conditions along main roads, while the use of taxi vehicles will cover the main and secondary roads in a random way. Taxi vehicles can very soon gather a lot of data all during the day. Furthermore, the continuous travelling of taxi vehicles will continually enrich the time database of the system, thus updating the available traffic data on an everyday basis.

4. Conclusions

The above described system can be a very efficient solution for vehicle fleet management and navigation along with mapping the everyday traffic conditions in an urban area. The VECON System has the following very important characteristics that define it as a valuable tool for recording traffic parameters:

  1. It functions as a self-governed closed system because it can produce the needed digital geographical database from its functions/operation/incoming information.

  2. It records travel times with the help of the GPS system for every road segment driven through by each vehicle sensor, thus providing a digital time database from which traffic parameters can be also computed.

  3. It can make use of existing vehicle fleets in an urban area, rapidly collecting data and covering the whole road network in the simplest way.

  4. It can be defined as a dynamic system because, during the everyday operation, it continuously updates and corrects its databases, optimizing in this way their effectiveness.

  5. It can be used for real time monitoring of traffic conditions under disaster or emergency situations where all the previously available data become invalid or unreliable.

  6. The initial cost of the system (control station hardware and software) might look slightly expensive but its expansion is rather economical and the cost of operation minimal.


  1. Frantzeskakis I. and Giannopoulos G.: Transportation Planning and Traffic Engineering, Paratiritis Publ. Co., Thessaloniki, 1986.

  2. Savvaidis P., I. Ifadis, and K. Lakakis: Use of a Fleet Management System for Monitoring Traffic Conditions After a Major Earthquake in an Urban Area, Proceedings of 22nd Urban and Regional Data Management Symposium, Delft, 2000.

  3. Savvaidis P., I. Ifadis, and K. Lakakis: Thessaloniki Continuous Reference GPS Station: Initial Estimation of Position, presented at the EGS XXV General Assembly, Nice, France, 2000.

  4. Zhao Y.: Vehicle Location and Navigation Systems, Artech House Inc., 1997.

  5. Savvaidis P.: Automatic Vehicle Location Systems and Their Perspectives of Applications, Proceeding of the First Hellenic Highway Conference, pp. 629-638, Larisa, Greece, 1995.

  6. Savvaidis P. and Katsambalos K.: Determination of a Vehicle’s Route Using GPS Techniques, Journal of the Northern Greece Association of Rural and Surveying Engineers, vols. 38 (pp 26-27) and 39 (pp 21-30), Thessaloniki, Greece, 1995.

  7. Savvaidis P. and Lakakis K.: Use of a Fleet Management System for Minimizing Response Time in Emergency Cases, Proceedings of Conference on Emergency Response after Seismic Disasters, Thessaloniki, 1999.

  8. Lakakis K.: Land Vehicle Navigation in an Urban Area by Using GPS and GIS Technologies, Phd Thesis, Aristotle University of Thessaloniki, Department of Civil Engineering, Thessaloniki, Greece, June, 2000.

Dr. Eng. Paraskevas D. Savvaidis is Professor and Director of the Laboratory of Geodesy, Department of Civil Engineering, Aristotle University of Thessaloniki (AUTH), Univ. Box 465, 54006 Thessaloniki, Greece. Tel. +30 31 995724, Fax +30 31 996159 E-mail: psav@civil.auth.gr.

Dr. Eng. Ioannis M. Ifadis is Assistant Professor in the Laboratory of Geodesy, Department of Civil Engineering, AUTH, Univ. Box 465, 54006 Thessaloniki, Greece. Tel. +30 31 995745, Fax +30 31 996159 E-mail: ifadis@civil.auth.gr.

Dr. Eng. Konstantinos Lakakis is Research Assistant in the Laboratory of Geodesy, Department of Civil Engineering, AUTH, Univ. Box 465, 54006 Thessaloniki, Greece. Tel. +30 31 995812, Fax +30 31 996159 E-mail: lakakis@civil.auth.gr.

База данных защищена авторским правом ©vuchoba.org 2019
звярнуцца да адміністрацыі

    Галоўная старонка