EP1457944B1 - Process for supplying traffic information - Google Patents

Abstract

The map shows several roads (A1,A3,A4) with junctions. Straight bars are shown along or parallel to stretches of road between the junctions. Each bar may correspond to the number of reports (Mj 1-3) of vehicles passing a given point on each road. Route arrows with time measurements (tau12,23,31 j) are provided at the junctions. Time values are calculated according to given mathematical formulae.

Description

The invention relates to a method for providing traffic information for one or more specific route sections to a road user.

It is the traffic report known as dominant dosage form for current traffic information. This is a dataset that provides information on the location, extent, nature and possibly other characteristics of a traffic-related, usually the traffic flow obstructing facts and can be presented to the user in various ways: by reading aloud on the radio, by retrieval via a menu-driven speech dialogue system, by taking into account a navigation task, by displaying in text form, etc.

The DE 100 63 763 A1 describes a method and apparatus for generating output of route related Ronten information.

An important criterion for the quality of a traffic information in the form of a report is its relevance. The more time between the time when the message is valid and the time it is relevant to the user, the greater the likelihood that the message will no longer correspond in detail to the situation that the user actually subscribes to Finds place, if he goes in knowledge of the message still in the appropriate place.

1. 1. Es vergeht Zeit zwischen der Entstehung eines Ereignisses auf der Strasse und seiner erstmaligen Beobachtung und Meldung an eine zentrale Stelle.
2. 2. Das Abfassen einer geeigneten Verkehrsmeldung in der Zentrale benötigt Zeit.
3. 3. Die Verbreitung der Meldungen ist an Abläufe gebunden, die Zeit benötigen, z.B. weil sie periodisch sind.
4. 4. Nutzer tendieren häufig dazu anzunehmen, dass eine Verkehrsmeldung, die sie erreicht hat, unverändert gültig bleibt bis zu dem Zeitpunkt, zu dem sie den Ort des gemeldeten Ereignisses erreichen. Diese Annahme ist im Allgemeinen falsch.

There are many influencing factors which determine the said time, referred to below as the message delay. The 4 most important ones are mentioned here:

1. 1. Time passes between the occurrence of an event on the street and its first observation and reporting to a central location.
2. 2. The writing of a suitable traffic message in the control center takes time.
3. 3. The distribution of messages is tied to processes that take time, eg because they are periodic.
4. 4. Users often tend to assume that a traffic report they have received will remain valid until the time they reach the location of the reported event. This assumption is generally wrong.

In particular, by the point 4, the message delay gets a subjective and individual component. The existing distribution channels and terminals are aligned to traffic information in the form of messages.

The object of the invention is now to provide traffic reports in such a way that the road user receives the traffic information that is essential to him at the time of reaching a traffic-related event.

The invention solves the stated object of providing traffic reports according to the features of claim 1.
Advantageous embodiments are specified in the subclaims.

The essence of the invention is that not only a single message M , but a bundle of messages M _{i is} provided, wherein each message contained in the bundle describes the event at another time, namely the validity time T _i = t ₀ + iΔt, where t ₀ for the present (more precisely: the time of the provision of the message bundle) and Δ t for the time interval of the individual messages within the bundle. A typical value might be: Δ t = 5 min.

beigegeben werden, dessen Einträge die Reisezeiten angeben, welche zur Zeit T_i erforderlich sind, um vom Ende einer zu einem Bündel M_i ^(k) gehörenden Störung zum Anfang der zum Bündel M_i ^(l) gehörenden Störung zu gelangen.As supportive information, message bundles may be a bundle of travel time matrices ${\mathit{\tau }}_{\mathit{kl}}^{\left(l\right)}$

are added whose entries indicate the travel times which are required at the time T _{i in} order to arrive at the beginning of the disturbance belonging to the bundle M _i ⁽¹⁾ from the end of a disturbance belonging to a bundle M _i ^(k) .

welche den Index des zeitlich am nächsten liegenden Eintrags des Meldungsbündels zurückliefert.Since both the message bundle and the bundle of travel time matrices are temporally discrete objects, but their contents must be accessed at arbitrary times, the access function n ( t ) = i is required if T _i - ${\mathit{T}}_{\mathit{i}}\mathit{-}\frac{\mathit{1}}{\mathit{2}}\mathit{.delta.t}\mathit{\le }\mathit{t}\mathit{<}{\mathit{T}}_{\mathit{i}}\mathit{+}\frac{\mathit{1}}{\mathit{2}}\mathit{.delta.t}\mathit{.}$

which returns the index of the temporally closest entry of the message bundle.

For the application of the message bundle according to the invention, it is also advantageous if the individual messages each contain the attribute "transit time" τ ( M _i ), which indicates the time that is required at time T _i to pass through the fault belonging to M _i .

The most interesting entries in a message bundle are those whose validity dates are still in the future. To generate these entries, there are various methods of traffic forecast, which are not the subject of the present invention. For example, it is possible to treat an already known fault with a method for congestion course prognosis, ie, to determine the prospective position of the fault fronts at the validity points in question with knowledge of the inflows and outflows into the fault and from the fault. In this way, the growth or even the resolution of a disorder can be described over time and deposited in the message bundle. In addition, there is the possibility to extrapolate the traffic condition into the future due to the traffic volume or historical data even if no disturbance has been observed. Corresponding forecasts can be entered in the message bundle as "Probably congestion from ... to ... at ... o'clock" to explicitly mark the uncertainty of the forecast of a "traffic jam out of nowhere". Further quantification by specifying a congestion probability as "congestion of ... to ..., probability 75%" is conceivable.

Although the entry of historical values into the message bundle is not strictly required for dynamic navigation purposes, it may well be considered as a confidence-building measure for the user (or for other purposes such as traceability). If it is explicitly visible in the message bundle that an accident due to an accident has grown within the last 15 minutes from 5 to 7 to 9 km, the user can get a very clear picture of the situation.

The message bundle must be evaluated after it has been deployed. This can be done either by a service provider who knows enough about the user to be able to select the message relevant to him. But this can also happen only in the terminal or by the user.

The knowledge of the planned route is indeed conducive to the accuracy of the information reaching the user, but not absolutely necessary. Thus, the selection from the bundle, rather than distances and possible speeds in the road network, may also be done with airline distances and speeds assumed for certain road classes, should the user's route not be known at the time of his request.

The message bundle must be time-limited, ie the current index can not assume any value but is taken from a limited index set.
i ∈ [ i _min , ..., i _max ]. In this case, the lower limit i _{min can} be determined largely arbitrarily from the standpoint of the maximum amount of data to be transmitted or a reasonable time limit for historical information. The upper limit i _max, on the other hand, depends on the predictive capability of the involved data acquisition and processing system. Here, there is either the possibility, for example, based on service requirements, to define a fixed upper limit i _max and to mark the credibility of the corresponding forecast, for example by a value for the congestion probability (see above) or else to set minimum requirements for the quality of the forecast, resulting in a variable Forecast horizon and thus a variable upper limit ⁱ _max follows.

Finally, message bundles in a larger network need not be represented as such in terms of data technology. There is e.g. the possibility that two disturbances grow together into one or a fault separates into two. This does not make the presentation of the time course in the form of bundles impossible, but makes it more difficult. It is therefore conceivable also the bundling of descriptions of the entire network, which may contain different numbers of messages at each validity time. This facilitates deployment, but complicates the use of message bundles. It is therefore necessary to choose a data representation appropriate to the respective application.

Reference to the drawings, the invention will be explained in more detail below.

In the drawings, Figure 1 shows a road network with a ring structure and Figure 2 shows a section of this ring structure with a route of a user.

In Figure 1, the network is populated with the 3 message bursts M ⁽¹⁾ , M ⁽²⁾ and M ⁽³⁾ and with a bundle of travel time ^{matrices τ of} the form $${\mathit{\tau }}^{\left(i\right)}=\left(\begin{array}{ccc}-& {\mathit{\tau }}_{12}^{\left(i\right)}& {\mathit{\tau }}_{12}^{\left(i\right)}+\mathit{\tau }\left(M_{i\text{'}}^{\left(2\right)}\right)+{\mathit{\tau }}_{23}^{\left(i\text{'}\right)}\\ {\mathit{\tau }}_{23}^{\left(i\right)}+\mathit{\tau }\left(M_{i"}^{\left(3\right)}\right)+{\mathit{\tau }}_{31}^{\left(i\right)}& -& {\mathit{\tau }}_{23}^{\left(i\right)}\\ {\mathit{\tau }}_{31}^{\left(i\right)}& {\mathit{\tau }}_{31}^{\left(i\right)}+\mathit{\tau }\left(M_{\tilde{i}}^{\left(1\right)}\right)+{\mathit{\tau }}_{12}^{\tilde{i}}& -\end{array}\right)\mathrm{,}$$

FIG. 2 illustrates the case of a user who is at location x at time t ₀ and can calculate a route R that is occupied by the two message groups M ⁽¹⁾ and M ⁽²⁾ .

From the position of the user at the time of the request, the expected travel time can be determined until reaching the disturbance, T ₀ , described by the message burst M ⁽¹⁾ . With the aid of the access function n, it is therefore possible to select from the bundle that message which is likely to best describe the event at the time of the user's arrival, namely M ⁽¹⁾ _{n (t 0 + T 0 )} .
In order to extract the message which is suitable in the same sense from the next message bundle M ⁽²⁾ lying on the route, it is necessary to determine the time which is expected to pass until the arrival at M ⁽²⁾ . It can be seen that, until leaving M ^(1), the travel time T ' ₁ = T ₀ + τ ( M ⁽¹⁾ _{n ( t 0 + T 0 )} and consequently the travel time until reaching M ⁽²⁾ by T ₁ = T ' ₁ + τ ₁₂ ^{( n ( t 0 + T 1 )) is} given. Thus, from M ⁽²⁾ the message M ⁽²⁾ _{n ( t 0 + T 1 )} in order to come closest to the impression of the user on the spot. In this way any number of message bundles lying on the calculated route can be considered iteratively.

Claims (13)

1. Method for providing, to a road user, traffic information for one or more specific route sections,
   characterised in that
   there is generated for the road user, in the presence of a traffic-relevant event in an affected route section, a packet of traffic reports, each report contained in the packet describing the traffic-relevant event concerned at a different point in time within a defined period of time, and the message that is relevant for the road user being selected from this packet, which message is valid at a point in time that is relevant for the road user.
2. Method for providing traffic information as in claim 1,
   characterised in that
   the road user himself selects the relevant message.
3. Method for providing traffic information as in claim 1,
   characterised in that
   the terminal selects the message that is relevant for the road user.
4. Method for providing traffic information as in claim 1,
   characterised in that
   a service provider selects the message that is relevant for the road user.
5. Method for providing traffic information as in one of the preceding claims,
   characterised in that
   every traffic report contains a "transit time" attribute which gives the time required to pass through the traffic event described by the report.
6. Method for providing traffic information as in one of the preceding claims,
   characterised in that
   together with the packets of traffic reports there is sent a packet of transit time matrix information on the travel times, the entries for which give the transit times required to pass through each traffic event at the time given by the traffic report on routes from the starting point until the first notified event is reached, between two notified events, or from the final notified event until the destination is reached.
7. Method for providing traffic information as in one of the preceding claims,
   characterised in that
   the travel time information is likewise provided in the form of packets, the structure of which is the same as the packets of traffic reports, with the result that it is possible to calculate journey times on routes as the totals of travel times on clear sections and transit times through events.
8. Method for providing traffic information as in one of the preceding claims,
   characterised in that
   for selecting the relevant time for choosing the relevant traffic report from the packet of traffic reports, the access function n(t)=i is used, if $$T_i-{}^1{/}_2\mathrm{\Delta }\mathrm{t}\le t\le T_i+{}^1{/}_2\mathrm{\Delta }\mathrm{t}$$

   

   
   where T_i represents the validity time point and Δt represents the time intervals between individual reports.
9. Method for providing traffic information as in one of the preceding claims,
   characterised in that
   the packet of traffic reports also contains traffic reports whose validity time points are in the future.
10. Method for providing traffic information as in one of the preceding claims,
    characterised in that
    the traffic reports that refer to the future are provided with a probability value for the forecast.
11. Method for providing traffic information as in one of the preceding claims,
    characterised in that
    the packet of traffic reports also contains traffic reports whose validity time points are in the past.
12. Method for providing traffic information as in one of the preceding claims,
    characterised in that
    the selection of the relevant traffic report from the packet of traffic reports takes place by means of calculating straight line distances and assumed speeds for specific types of road.
13. Method for providing traffic information as in one of the preceding claims,
    characterised in that
    the traffic reports in the packets of traffic reports refer to an overall road network of a specific size.

Images