EP1022703B1 - Method of organizing traffic flows to control optimally the traffic and to plan road networks according to the needs - Google Patents

Abstract

A road region being monitored is provided at all relevant entry and exit points with suitable detectors to detect individual vehicles. Identification codes are used to store all movement profiles as data sets in an analysis computer, or to recover data sets that have already been stored. The data sets are appended by the current movement profile of the vehicle concerned to enable a complex traffic flow analysis in the region being monitored.

Description

The invention relates to a method for the analysis of traffic flows for optimized traffic flow control and for the demand-oriented planning of road traffic networks, according to the preamble of patent claim 1.

The traffic flow analysis as a basis for traffic flow control and route planning is sometimes created by means of counting procedures. In the process, all vehicles of the traffic sections in question are usually counted manually during the period under consideration. This gives an indication of the time-dependent traffic density and accordingly any need for roadway construction. The method finds its limits in the detection of more complex dependencies, for example in the intersection area. The temporary placement of counting persons at all entrances and exits, which detect both vehicle density and traffic behavior of the vehicles, e.g. how many vehicles turn where or similar From this, it is possible to draw conclusions on, for example, traffic light control and the number of lanes by subsequent evaluation.

In extreme applications, this method is not usable if, for example, the distances are too large and unmanageable, more frequent lane changes of vehicles present and also high vehicle speeds and high traffic levels are present.
This is shown by way of example in FIG. A multi-lane motorway contains in a relatively short sequence three junctions a, b and c and three departures d, e and f. Although the traffic density could be recorded in all directions by manual or automatic counting of the passing vehicles, the actual cause of the congestion can not be determined, since this method makes no statement as to which traffic flows lead to which shares. Also a possible Extension of the motorway by another lane brings in the considered case no remedy if a high statistical traffic mix is present on all entrances and exits. The cause of congestion is then often in the mutual blocking of the vehicles at times mandatory or supposedly advantageous lane change, the theoretical throughput limit of the respective lanes and the arrivals and departures is far from achieved.

In the present example according to FIG. 1, the remedy of the traffic problem could exist in an additional connection ae (bypass) between the traffic routes a and e, as shown in FIG.
In practice, however, the complex analysis and traffic optimization in such cases for the aforementioned reasons is not possible, which can be seen from the daily situation on the highways.
This situation is to be remedied by the present invention.

US-A-5 801 943 discloses a method of traffic monitoring and simulation of a road network in which a plurality of vehicles moving in the road network are detected by intelligent sensors. The recorded vehicles are tracked individually and recorded their driving behavior. Based on this vehicle data is a traffic forecast, which is constantly checked by the actual movement of vehicles and corrected if necessary. The focus of this method is clearly on a simulation of traffic flow. To check the simulation results, it is sufficient to detect the driving behavior and the route of several vehicles by means of sensors. Not all vehicles are registered.

The present invention has for its object to propose a method on the basis of an automatic analysis of complex traffic flows for the purpose of optimized traffic flow control and needs-based planning of road networks is possible.

This problem is solved by the characterizing features of patent claim 1.

The inventive method is based on a comprehensive sensory detection of a traffic section or observation area to be considered, at least one detection at all arrival and access points of the traffic section, as well as on a vehicle-specific detection and traffic tracking and an automatic traffic analysis. Of primary interest here is the statistical distribution of traffic flows with regard to origin and destination of all individual vehicles, but also the statistical one Evaluation of lane change processes. On the basis of the analysis, a computer-aided traffic simulation with optimization and subsequent actual traffic route optimization can be carried out in the further development process.

The method requires the at least temporary installation of preferably optical sensors on all lanes, preferably at the boundaries of the viewing area and for tracking changes at accordingly interesting points within the viewing area, with a sufficiently reliable vehicle identification for statistical evaluation by reproducible criteria, such as the evaluation of the plate , or an alternative vehicle analysis according to size, color, number and characteristics of recognizable persons, noise behavior, or electronic detection mechanisms (eg responders) for the identification of vehicles or the like. For the purpose of acceleration and simplification of the EDP processes, the description / identification of the vehicles preferably takes place by means of parameter sets assigned to the optical criteria.

For a comfortable design of the evaluation procedure, in one embodiment a direct allocation and expansion of the local movement profile of the vehicles to already stored data sets of other measurement stations is provided, whereby a station independent superordinate evaluation process allows a permanent statistical evaluation of the complex traffic flow with all desired parameters and at least the analyzed values for be stored later in subsequent storage and use in resistant memory devices.

Another embodiment of the invention provides that statistical traffic flow data are saved in a long-term memory in such a way that they can be used for later computer-aided simulation of alternative traffic flow guidance models or modified traffic route guidance Optimization of the traffic flow in the traffic section / viewing area are usable.

In another technical realization it is envisaged that the process of vehicle identification by a method of permanent gapless vehicle tracking - by means of a software procedure based on pattern recognition methods - takes place in the entire traffic section / observation area, in which case the entire observation area is completely covered with corresponding sensors, e.g. optical sensors, and the collected data for evaluation and analysis to a common gapless record, e.g. an overall entry screen, for subsequent evaluation.

Figur 1:

beispielhaft einen betrachteten Verkehrsabschnitt mit mehreren Ein- und Ausfahrten;

Figur 2:

den Verkehrsabschnitt gemäss Figur 1 mit zusätzlicher Beipassverbindung;

Figur 3:

den Verkehrsabschnitt gemäss Figur 1 mit möglicher Anordnung von Erfassungseinrichtungen;

Figur 4:

Beispiel einer dezentralen Datenerfassung, und die Zusammenschaltung von optischen Erfassungseinrichtungen, Datenerfassungsrechnern (ER) und zentralem Auswerte- bzw. Analyserechner (AR).

The invention will be explained below using further drawing figures, which describe here only one possible embodiment in a schematic representation, with reference to the drawing figures further fields of application and advantages of the invention. Show it:

FIG. 1:

for example, a considered section of traffic with several entrances and exits;

FIG. 2:

the traffic section according to Figure 1 with additional Beipassverbindung;

FIG. 3:

the traffic section according to Figure 1 with a possible arrangement of detection devices;

FIG. 4:

Example of a decentralized data acquisition, and the interconnection of optical recording devices, data acquisition computers (ER) and central evaluation or analysis computer (AR).

FIGS. 1 and 2 have already been described in the introductory part, with a further example of a traffic section according to FIG. 1 being assumed.

Fig. 3 shows the preferred analysis positions in the considered traffic section, ie the preferred arrangement of the detection sensors here continuously with the numbers 1 - 12 marked. The detection sensors are arranged primarily at the entrances and exits of the traffic section and on the main route between the entrances and exits.
At each individual position 1-12, an analysis of the passing vehicles is carried out first, wherein the detection system of all access points 1, 7, 9 and 11, the task of identification as well as the identification and temporary storage of vehicles in a database including additional required parameters, such as Time, speed, lane, location, relative position to other vehicles etc.
At all remaining detection points 2-6, 8, 10 and 12, a search of the corresponding object in the applied data record of the central memory and the additional entry of all new motion-relevant data for each object is carried out after the respective recognition according to a defined method and assignment of the corresponding identification parameters.

In this way, a statistical evaluation of the traffic situation can take place permanently during operation (online), with a corresponding traffic profile over the time axis being able to be used as an evaluation criterion for, for example, a time-controlled traffic flow control for further analysis.

The arrangement of the preferably optical sensors 1-12, for example, on temporarily or permanently installed mounting bridges or poles in a corresponding height across the road, or for example with a corresponding speed limit between the lanes - but here with appropriate traffic degradation - done. If only one or two lanes are present, an arrangement of the sensors on the roadside is possible.

Additional criteria, such as the position between other stored vehicles can be used for detection.

Fig. 4 shows an example of decentralized data acquisition by means of optical sensors, e.g. In the example shown, a data acquisition computer (ER) acquires all possibly required track sensors (a, b, c) of a respective observation position and preprocesses the data in such a way that each detected vehicle is described individually with a defined parameter set that is identical for all data acquisition computers. which is then transmitted together with the required location and time-specific traffic flow parameters to the central evaluation or analysis computer (AR). The storage and allocation, as well as the statistical evaluation of all data then takes place in the central evaluation or analysis computer.

As an alternative to the described vehicle identification method, a method with object-related tracking of all vehicles is possible in the event that the entire observation area is monitored in closed form, so that there are no gaps in the vehicle tracking for an example based on optical detection analysis system.

Claims (7)

1. Process for the analysis of traffic flows for optimised traffic flow control and for demand-related planning of traffic networks, wherein a section of traffic to be examined or observation area is equipped at least at all relevant entry and exit points (a, b, c or d, e, f) with suitable recording devices (1-12), with the aid of which vehicles as such are recognised individually and reproducibly and stored by means of suitable identification code together with accompanying movement profile as a data record in an analysis computer (AR) or are recognised by the analysis computer in the case of data records already installed, wherein the data records are extended by the current movement profile of the relevant vehicles, as a result of which a complex traffic flow analysis is facilitated in the relevant section of traffic or observation area, characterised in that the recording devices (1-12) are arranged at the entry and exit points of the section of traffic and on the main route between the entry and exit points, in that first of all an analysis of the passing vehicles takes place at each individual recording device, wherein the recording devices of all access points (1, 7, 9, 11) have the task of identification and additionally of identification and temporary storage of the vehicles in a database together with additional required parameters, and a search of the corresponding object in the installed data record of the central memory and the additional entry of all new movement-relevant data for each object takes place at all remaining recording points (2-6, 8, 10, 12) after the particular recognition according to fixed process and assignment of the corresponding identification parameters.
2. Process according to claim 1, characterised in that for the purpose of traffic flow analysis within the section of traffic/observation area, recording devices are also provided at other analysis-relevant points.
3. Process according to claim 1 or 2, characterised in that optical sensors in the form of cameras and/or infrared detectors are used as recording devices.
4. Process according to one or more of the preceding claims, characterised in that the recognition and identification of the vehicles in the section of traffic/observation area takes place reproducibly in the course of the overall measurement with adequate security, wherein the recognition and identification is based on recognition of the number plate of the vehicle and/or other visual vehicle features and/or electronic recognition mechanisms, wherein the description/identification of the vehicles takes place by means of parameter blocks assigned to the recognised visual/electronic criteria.
5. Process according to one or more of the preceding claims, characterised in that direct assignment and extension of the local movement profile of the vehicles takes place to already stored data records of other recording devices, wherein a higher evaluation process which is independent of the recording devices facilitates a permanent statistical evaluation of the complex traffic course with all required parameters and at least the analysed values are stored for later further examination and use in resistant memory devices.
6. Process according to one or more of the preceding claims, characterised in that at least statistical traffic flow data are secured in a long-term memory such that they can be used for later computer-assisted simulation of alternative traffic flow control models or altered traffic route direction for the purpose of optimising the traffic flow in the section of traffic/observation area.
7. Process according to one or more of the preceding claims, characterised in that the process of vehicle identification takes place by a process of permanent gap-free vehicle tracking by means of a software procedure based on pattern recognition processes in the entire section of traffic/observation area examined, wherein the entire section of traffic or observation area is recorded without gaps using corresponding recording devices and the recorded data are collected for the purpose of evaluation and analysis to form a common gap-free data record for subsequent evaluation.

Images