AU755096B2

AU755096B2 - Method of updating a traffic route network map and map- supported method for generating vehicle guidance information

Info

Publication number: AU755096B2
Application number: AU27641/00A
Authority: AU
Inventors: Fredy Eckstein; Helmut Gimmler; Klaus-Peter Kuhn; Klaus Mezger; Thomas Renger
Original assignee: DaimlerChrysler AG
Current assignee: Mercedes Benz Group AG
Priority date: 1999-04-15
Filing date: 2000-04-10
Publication date: 2002-12-05
Anticipated expiration: 2020-04-10
Also published as: AU2764100A; EP1045224A3; DE19916967C1; EP1045224A2; JP2000321081A

Description

-1- METHOD OF UPDATING A TRAFFIC ROUTE NETWORK MAP AND MAP-SUPPORTED METHOD FOR GENERATING VEHICLE GUIDANCE INFORMATION Field of the Invention The invention refers to a method of updating a data-processing-supported traffic route network map according to the characterizing clause of Claim 1, and a method of generating vehicle guidance information according to the characterizing clause of io Claim 3, in which such a traffic route network map can be used.

.oo..i Background Art A "data-processing-supported traffic route network map" is understood to mean a map, S" which can be managed independently using data processing, and which is of a traffic is route network which can be driven by vehicles, e.g. so-called digital maps. The maps include route or local data, meaning the data which defines the course of the sections of the route network which are available to be driven, and assigned attribute data, i.e. data about additional properties of the various route sections. These attributes traditionally include, for instance, information which is displayed along the actual traffic routes by traffic signs, such as route type, possibility of driving in a direction, speed restrictions etc., but also information about the traffic situation to be expected, e.g. so-called load curve information. In all cases, the attribute data can be stored in the map with reference to time, i.e. the value of an attribute can be different for different seasons or days or even times of day. On the basis of the data from such a traffic route network map, vehicle guidance information can then be generated, e.g. in the form of route selection information about the choice of a favourable route to a destination and/or in the form of vehicle control information to support the driver's vehicle guidance or [R:\LIBCC]52950.doc:wxb 28 944/DE/1 2 for automatic vehicle guidance.

By using one or more test vehicles, i.e. vehicles which are equipped to record the required up-to-date local and attribute data, it is possible to implement a running update of the traffic route network map, or at least one which takes place at reasonably short intervals, e.g.

hourly or daily. In particular, the vehicles which exist and are used in any case can be used as test vehicles, so that in general no special drives just for the purpose of map updating are required.

A generic method of updating a digital road map is described in Patent Document DE 195 25 291 Cl. In the method which is described there, the attributes consist practically exclusively of information about the S permissibility of driving a road section in at least one direction. The attribute values which are stored in the digital map for this purpose are updated on the basis of "20 current attribute values, which are recorded by the test vehicle while taking account of its continuously determined current geographical position. This happens, for instance, by recognizing corresponding traffic signs via a video camera, or implicitly by a traffic computer 25 which processes the data recognizing that a test vehicle has driven a road section in the relevant direction. To exclude wrong guidance to destinations and traffic steering, all road sections, for which no current attribute data about the permissibility of driving has been recorded within a specified interval, can be set to the "Driving not permitted" attribute value. Whereas the position of the test vehicle in certain cases must be determined exactly, e.g. by a GPS determination with additional DGPS Correction, optionally followed by socalled map matching, a correspondingly exact local assignment of the attributes is not addressed there, and 28 944/DE/1 3 also not required for the attributes being considered.

In a method which is described in Patent Application

DE

196 50 844 Al for determining driving route data, attribute assignment for a digital route network map by static and dynamic parameters is provided. The static parameters include, in particular, structural features of the traffic routes, such as number of lanes, presence of an ascending or descending section and road type. The dynamic parameters include information about the traffic position or traffic flow to be expected, e.g. the number and average speed of vehicles in a given route section and period. Whereas the dynamic parameters are updated on the basis of measurement data which is recorded by vehicles, the values of the static parameters are not adjusted in this way.

From Patent Application DE 37 00 552 Al, storing safetyrelevant information as attributes for a digital road map is known in the art. These attributes include, among others, the presence of ascending and descending sections and curves, and are determined empirically in advance, when the map is produced, by special test vehicles.

S.Later, when a route section is driven by a vehicle which .25 is equipped with the map, the stored safety-relevant information for the current route section is displayed to the driver.

From Patent Document DE 43 44 369 C2, a method of consumption-oriented limitation of the driving performance of a vehicle propulsion mechanism during a drive to a desired destination is known in the art. In this method, a traffic computer, during the drive, repeatedly determines a permitted target route consumption to travel the route to the destination, on the basis of a measured current energy stock, and from this derives a target value for limitation of the driving performance, in such a way that the smaller the determined target route consumption, the more the driving performance is limited. For this purpose, characteristic curves for energy consumption, which are specified differently for each route section, are used, depending on, among other things, s whether the relevant route section has a hilly, mountainous, uphill, or slightly mountainous, downhill topography, for which purpose appropriate route information is recorded in advance, e.g. during an appropriate test drive.

In Patent Document DE 195 06 364 C2, a vehicle control system is described, with which it is possible to guide a road vehicle autonomously using a digital road map, in which S•fictitious road nodes are stored, preferably including junctions, crossings and curves of the road. The vehicle control system derives the curvature of the road section in front of the vehicle from this node information of the digital map, and controls a corresponding steering angle and a suitable driving speed.

There exists a need for a method of updating a traffic route network map and a mapsupported method of generating vehicle guidance information of the type which was mentioned at the start, through which a traffic route network map which remains up to date can be provided, which is particularly suitable for prior estimation of the driving 20 energy consumption and/or as the basis for determining predictive vehicle control actions, and through which vehicle guidance information, which can include, in particular, vehicle .control information which supports the driver or makes autonomous vehicle guidance possible, energy-consumption-optimizing vehicle control information, or energyconsumption-dependent route selection information, can be generated.

Summary of the Invention In accordance with one aspect of the present invention, there is disclosed a method of updating a data-processing-supported traffic route network map containing route data about a traffic route network which is driven on by at least one test vehicle, said method comprising the steps of: recording current route data and current attribute data by a said test vehicle, wherein said attribute data includes topographical data comprising slopes and curves of g AL te sections locally exact with the position and strength of slopes and curves of said trac route network; and [R:\LIBCC]52950.doc:wxb updating the traffic route network map using recorded route data and recorded attribute data, wherein said one or more test vehicles are linked to mutually exchange data with various road vehicles and information obtained is combined into said traffic route network map.

In accordance with another aspect of the present invention, there is disclosed a method of generating vehicle guidance information or actions, said method comprising the steps of: determining a vehicle's current position; reading data from a data-processing-supported updated traffic route network map comprising route data about a traffic route network that can be driven by a vehicle; assigning attribute data to the traffic route network map, said attribute data including at least locally exact topographical data of the slopes or curves of route sections of the traffic route network; and S 15 generating, simultaneously with step advance vehicle guidance information determined by the topographical data for a specifiable forecast route network area, wherein said the traffic route network map is updated by recording current route data and attribute data of the route section of the traffic route network by at least one test vehicle, that is linked to mutually exchange data with various said road vehicles.

S• •"In the case of the map updating method, it is characteristically provided that locally exact :topographical data, also called topological data, about the slopes and/or curves of at least a part of the route sections of the stored traffic route network is used as attribute data, and additionally further, traditional attribute data can be taken into account depending on the case. The locally exact topographical and possibly other attribute data is stored in the map and updated on the basis of current measurement data, which refers to it, from a test vehicle. "Locally exact" means that the topographical data does not just give topological information about the mere existence of a sloping section, which is here understood to include a descending section, or a curve, somewhere on a route section, also called a route edge, between two route nodes, but goes further, to give information about its exact geographical position on the route edge and its strength, e.g. in the form of the slope or curvature value. The traditional attribute data which may also be provided, such as typical traffic sign information, is preferably stored and updated in this locally exact way in the route network map. It is understood that the attributes can also, as usual, be stored [R:\LIBCC]52950.doc:wxb -6time-related. They are also normally direction-specific, i.e. they can have different values at the same route network position for different driving directions.

With this procedure, when the route guidance changes, it is possible to bring the traffic route network map to the appropriately updated state for roadworks or a new route section very quickly and independently. Because of the locally exact attribute data which it contains, a traffic route network map which is always kept up to date in this way can be used as the basis for generating vehicle routing information which requires such information, e.g. for energy-consumption-optimizing vehicle guidance actions. It is lo understood here that updating includes the initial production of a traffic route network S•map and the initial taking into account of a newly added road section.

In a development of the map updating method, as data of a given attribute, as well as the current value and position, previously recorded values or positions or magnitudes derived from them, such as an average value, a minimum value and a maximum value, or other statistical quantities, and the number of previously recorded values taken into account, S: should be stored. This can be used for a statistical attribute update, in which the attribute .value and place which are held in the map are not replaced directly by a newly recorded attribute value, but by a relativized attribute value and place, into which the new attribute o. 20 value goes together with the statistical data which is additionally held in the map. In this way gross updating errors caused by previous wrong measurements can be prevented.

The method of generating vehicle guidance information characteristically makes use of attribute data of a data-processing-supported traffic route network map, which includes locally exact data of the network geometry, as explained above, and may include additional, traditional attribute data, depending on the case. On the basis of such a map, information which belongs to a specifiable forecast route network area, i.e. an area of the traffic route network which extends from the current vehicle position to a specifiable forecast limit in the intended driving direction, is read and used to generate vehicle guidance information which depends on it. By taking account of the attribute data, which refers exactly to places, highly exact and up-to-date advance information can be provided for vehicle guidance, e.g. acceleration in advance shortly before reaching an uphill >section, deceleration in advance shortly before reaching a downhill section or curve, [R:\LIBCC]52950.doc:wxb -7reduction of energy consumption, finding a favourable driving route to a destination for power consumption, etc.

The generated vehicle guidance information may include parameters about the expected energy consumption for a drive to a specifiable destination on the relevant route network sections, using, in particular, the stored information about the extent and strength of ascending and descending sections and other consumption-relevant attributes. With this procedure, it is possible to estimate in advance whether a desired destination can still be reached with the present energy stock, or it is necessary to fill the tank for energy. It is understood that the destination does not have to be specified permanently, but can be kept •variable.

*In an embodiment of the method, an area in front of the current vehicle position is used as the forecast area, i.e. an area directly in front of the vehicle, with a length of typically a s few tens to a few hundreds of metres, and for this the associated route-related attribute data is read, e.g. to generate energy-consumption-optimizing information to control the .•vehicle, i.e. to affect its lengthways and sideways dynamics. This may consist of pure display information for the driver or information for an autonomous vehicle guidance system, which then makes the corresponding vehicle control interventions independently.

S.Brief Description of the Drawings Advantageous embodiments of the invention are shown in the drawings and are described below. In the drawing: Fig. 1 shows a schematic representation of a communication structure of a car, with the ability to update a digital, attribute-labelled road map, and to generate vehicle guidance information, Fig. 2 shows a schematic function block representation of storing and reading attribute values of the digital road map according to Fig. 1, Fig. 3 shows a schematic representation of a process to store a recorded attribute ,value for the digital road map according to Fig. 1, [R:\LIBCC]52950.doc:wxb Fig. 4 shows a schematic function block representation of a consumption assistant unit of the vehicle of Fig. 1, and Fig. 5 shows a schematic representation of a road section in front of the vehicle of Fig. 1, to represent how the forecasting function of the consumption assistant unit of Fig. 4 works, on the basis of the learned attributes.

Detailed Description Fig. 1 shows, in schematic representation, a car 1, which is equipped to carry out variants, which are explained in ease*: o o• [R:\LLBCC]52950.doc:wxb 28 944/DE/1 9 more detail below, of the method according to the invention, and the communication structure which is implemented for it. The car 1 contains various data processing components, in particular an on-board computer with associated peripherals, which are of traditional construction unless otherwise stated below, and therefore shown in Fig. 1 only symbolically by a notebook computer 2. The on-board data processing electronics 2 includes, in particular, as well as the actual on-board or vehicle computer, a destination guidance system with associated location devices, and on-board sensors to record measured values of attributes, i.e. route-related additional information.

15 To determine the vehicle position, the location devices obtain (D)GPS signals from appropriate satellites 3 in the usual way via an associated unidirectional satellite communication channel 4. Via a bidirectional ~communication channel 5, which may be, for instance, a 20 satellite or GSM communication channel, the vehicle 1 is :i linked for data exchange, directly or via satellite 6, to other road vehicles and/or a traffic centre. The abovementioned on-board sensors capture momentary values of specified terrain or traffic attributes, as symbolized by 25 an arrow E, which can then be used to update a dataprocessing-supported traffic route network map, in this example a digital road map. The digital road map may be implemented in the on-board electronics 2 of the vehicle 1 itself, or in a centre which is linked to it, or in another vehicle. In other words, the vehicle 1 is a test vehicle, with which, during the drive, measurement data about the current vehicle position and associated terrain and traffic attributes is gained continuously and used to update the digital road map accordingly. The digital road map which is kept up to date continuously in this way is then used by the on-board computer 2 to provide 28 944/DE/I 10 forecasting information about the terrain and traffic characteristics of the route section in front of the vehicle 1 as information which is relevant to vehicle guidance, as symbolized by an arrow A.

Since the procedure according to the invention can be implemented by suitable use of hardware components which are common in modern vehicles, practically any such vehicle can be used as the test vehicle. Therefore, the collection and use of the route network information and additional attributes can be done in each vehicle autonomously, i.e. in one vehicle only, without exchanging information with other vehicles or a centre.

A high number of test vehicles makes the digital map more -15 up to date. On the other hand, the information which is ~gained in various vehicles can also be combined by mutual exchange and/or transmission to the centre. By common telematics techniques using the abovementioned communication channels, the current data of the route osoo network map can be made available to all system participants, i.e. all participating vehicles and associated centres if any. For instance, it is possible to provide collection of locally exact attribute data G..'which is currently recorded in different vehicles in one 25 centre, which accordingly updates a data-processingsupported route network map which it keeps centrally and makes the current map data available to the connected vehicles.

The essence of this type of attribute assignment, i.e.

assignment of attribute values to route sections of the digital road map, is that it is not merely an assignment of measured attribute values to the associated route section between two route nodes, but is locally exact, also called "metre-exact" below. This means, as shown schematically in Fig. 2, that an attribute value which is 28 944/DE/1 11 recorded at a particular point P along an edge 7 of the road network, i.e. a direction-specific route section between two network nodes 8, 9, is not assigned to the edge 7 as a whole, but exactly to the capture position p on it.

In particular, attributes about the geographical properties of the road network are provided, with reference to the geographical co-ordinates including the height, i.e. the slope progression. Not only the occurrence of ascending and descending sections and curves is registered in the digital map, but also their strength, e.g. in the form of the captured value for the slope or radius of curvature. In this way, in particular, an almost continuous assignment of carriageway slope and curvature attributes to the road network can be provided.

Additionally, depending on the application, other attributes, which themselves are known in the art, but 20 here are stored in locally exact form, can be taken into S. account, such as speed attributes about a typical average vehicle speed or a speed restriction giving a maximum or minimum speed, typical road instruction sign attributes *e*such as about entering a town, the start of a motorway, road reserved for motor vehicles or major road, typical prohibition sign attributes such as about prohibited overtaking, prohibited through roads or prohibition of certain vehicles, e.g. motor bicycles or lorries, socalled POI (point of interest) attributes, such as breakdown help, first aid, a police station, a hotel, a service area or petrol station, restriction attributes, e.g. a maximum through speed, a maximum vehicle weight or a restriction of use by time, driving direction attributes about restrictions of the permitted driving directions, and in particular traffic situation attributes about the traffic situation or state to be 28 944/DE/1 12 expected on average, such as free traffic, lively traffic, strong traffic, rush hour, stop and go traffic, etc.

It is understood that these attributes can be captured and stored in the digital map in relation to time, i.e.

for the same road location at different times different values for the attribute can be held in the map. This is particularly important for the traffic situation attributes, the values of which can be subject to strong periodic variations. The attributes are also preferably direction-specific, i.e. for the same road location different values of the same attribute are possible for different driving directions. The attribute values which 15 are recorded at a particular place P can then be stored e°"in the digital map together with the place data for this place P, i.e. with the data which identifies it in the route network, in the form of its geographical coordinates, supplemented by a unique identification if necessary. If no value is yet present for the attribute, it can be filled for the first time. This also applies in the case that a road section which is not yet included in the stored road network is driven, and can then be listed •in the digital map for the first time, together with the attribute values which are recorded there. In other words, this method of updating the digital map includes its initial generation or updating on the basis of a network structure basic graph.

If a current attribute value is measured and captured, but a different attribute value is already stored in the map, the map updating can consist of a simple replacement of the previous value by the new value. However, to exclude wrong attribute assignment because of measurement errors to a large extent, the update preferably happens in a statistically relativized form. For this purpose, an 28 944/DE/1 13 updating algorithm which is provided forms the attribute value which is to be newly stored in the map from several previously recorded place-specific, time-specific and direction-specific values using a statistical method, e.g. in the form of a sliding average value. For this purpose, as well as the currently applicable attribute value, which is determined in a statistically relativized way, other, previously determined values of the attribute are held in the map, such as previously measured values or statistical quantities, which summarize all the values with relevant parameters, e.g. a value, a maximum value and an average value, and the number of such held values.

Similarly, the map learns the place data associated with 15 the attributes in a way which is place-specific, but not specific to a fixed place, to avoid an incorrect attribute position being assigned to a captured attribute because of a position measurement error. For this purpose, it is again provided that a valid, relativized 20 position value should be determined from all the known positions for a given attribute within a tolerance to be specified, using a suitable, traditional statistical method, and the attribute should be assigned to the S• relevant position. If an attribute of a type which is known to the system is captured outside the specified tolerance, e.g. more than 100m from the previously known attribute, the two attributes are not combined, but the newly captured attribute is stored for the first time in the database of the map, which can be updated, and thus can and does learn.

The map updating preferably also includes the fact that at least a part of the attributes in the map are not held in the map and assigned to a specified position with no time limit. If a previously learned attribute is not detected several times by the on-board sensors of the 28 944/DE/1 14 test vehicles driving past the relevant position, the learned attribute is deliberately removed from the database of the map according to a suitable, specifiable algorithm. The map which is updated in this way takes account of, for instance, moving roadworks, for which at a particular time there is a route restriction, which is recognized and learned as an attribute, but is cancelled or moved along the route at a later time. As a further improvement, it is possible to provide that local and attribute data, irrespective of the statistical learning method, is provided for an application only if it has a specifiable degree of trust. The degree of trust here *indicates the reliability of attribute detection, i.e.

.how strongly detection of the attribute may be prevented by errors in the sensors, calculation or evaluation errors., road-side information sources such as traffic signs being covered, or environmental effects. It is also possible to take account of how often, how exactly etc.

it was possible to determine an attribute. The 20 statistical quantities can again be used for evaluation.

In other words, beginning with a completely or mostly empty digital map, with the method according to the invention the whole basic graph, i.e. the traffic route network which can be driven, can be learned through the drives of the test vehicles only, and the desired attributes can be assigned to it, metre-exact, first time. The attributes and basic graph can be corrected by secure differences. In this way, the learning map can learn new crossing layouts, diversions, etc. and change, delete or gradually "forget" existing basic graph elements. The map which is continuously updated in this way can be made available to all vehicles which participate in the system, directly or via a centre, or used locally. The information which it contains can be used in various ways. For this purpose, as illustrated in 28 944/DE/1 15 the right-hand block of Fig. 2, after entry of the vehicle position as determined by the location devices, e.g. in the form of geographical co-ordinates, the stored values of corresponding attributes can be read, and preferably not just the attributes which are stored for this place itself, but also the attributes which are included in a specifiable forecast area in front of the vehicle in the direction of driving. The last-mentioned action makes it possible to provide information for predictive vehicle guidance.

The typical progress of an attribute assignment event is illustrated in Fig. 3. The right-hand half of the figure -"shows a section of the road network which can be driven, o 15 the left-hand half contains, as an example, data of a specific attribute assignment, in a function block.

Driving a circular route 10, marked with a dashed line, is taken as an example. The vehicle position is continuously determined, using a GPS measured value 20 followed by map matching. By using map matching, the GPS measured values along the circular route 10, which because of measurement errors may differ slightly from the true vehicle position, and are given with a dotted *o*line 11 in Fig. 3 as an example, are mapped onto the route network in the digital map. For this purpose, the measured values of the location co-ordinates, together with the vehicle kilometre reading, are passed to the subsystem which updates the digital road map, and the subsystem uses them to form a capture circle 12, which contains the possible road edges for the measured vehicle position. The map-learning subsystem then determines, by map matching or a suitable traditional coupling algorithm, the road edge within the capture circle 12 which ought to correspond to the road section which is actually being driven, and thus excludes irrelevant road edges as far as possible. Then, on the basis of several 28 944/DE/1 16 successive position determinations, the driving direction on the determined road edge is established. For instance, in Fig. 3 the association of the current vehicle position with an edge with identity number 4711 and a first driving direction (direction 1) is assumed. If the attribute to be stored is time-specific, the current time is also taken into account.

In parallel with the determination of the vehicle position, for every attribute to be stored in the map the current attribute value for the momentary vehicle position is recorded. For instance, Fig. 3 shows the determination of a slope value of 6.3% for the "slope progression" attribute. From the captured attribute value, a new attribute value, which corresponds to it or is statistically derived from it and earlier attribute values which were recorded at this place, is determined, assigned to the defined position and stored in the digital map.

oo As an example, in Figs. 4 and 5 a possible application of the learning digital road map as described above is illustrated in the context of a consumption assistant function. "Consumption assistant" here means a subsystem of the vehicle on-board electronics, which generates predictive energy-consumption-optimizing vehicle guidance information, with which the driver or an autonomous vehicle guidance system can adjust the driving style in good time to the course of the route in front of the vehicle, in such a way that the energy consumption for the drive is kept as low as possible. In particular, before curves and uphill and downhill sections, but also other interferences with driving, consumption-reducing vehicle control actions can be initiated in good time.

As well as other, traditional components, as the 28 944/DE/1 17 specialist will understand immediately from the following description of the functioning of the consumption assistant 13, the consumption assistant subsystem 13 contains, in particular, as mentioned above, a map 14 of the type described above, which can learn and be updated.

The consumption assistant 13 obtains the necessary vehicle position information from the position measurement device or a navigation system 15. The consumption assistant 13 also obtains current attribute values 16 for the attributes of the map 14 in the form of values which are recorded by the appropriate sensors and :delivered via data bus lines. This includes the possibility of optical recognition using an image recording and analysis system, also called "computer seeing". With the recorded data, the "learning map" core function, which updates the map 14 as required from the consumption assistant 13 or another subsystem, is supplied.

20 The learning map 14 has a predictive function, part of which is that for a specified area in front of the vehicle in the direction of driving, in this example about 100m, the attribute values which are stored in this oo area for the driven road edge can be provided, in particular the course of the route. The consumption assistant 13 then uses the information which is thus available about the terrain characteristics in front of the vehicle to generate predictive, energy-consumptionreducing vehicle control actions, such as braking, accelerating and/or steering actions.

Fig. 5 shows a specific example of an application of such a prediction process by the consumption assistant 13. By reading the appropriate locally exact attribute values from the learning road map 14, the consumption assistant 13 obtains, for instance, the information that measured 28 944/DE/1 18 1 from the current vehicle position, after 10m a downhill section begins, this ends after another 20m and has an average downhill slope of a minimum downhill slope of 4.3% and a maximum downhill slope of that 50m in front of the vehicle there is a stop sign, and 75m in front of the vehicle an uphill section which is 17m long and has a 7% uphill slope begins. 85m in front of the vehicle there is a town entrance sign, which in a known way also shows a speed limit. With this information, the consumption assistant 13 is able to generate predictive vehicle control actions, using which this route section can be driven with the lowest possible energy consumption. In particular, for this purpose the oooe consumption assistant 13, shortly before reaching the 15 downhill section or stop sign, can request or initiate deactivation of the accelerator, and before reaching the uphill section can request or initiate gentle acceleration in good time, taking account of the speed restriction following the town entrance. In the same way, 20 when the consumption assistant 13 detects a curve, it can request or initiate appropriate deactivation of the accelerator or braking in preparation, depending on the radius of curvature.

The vehicle control information which is output by the consumption assistant 13 can be used as driving instructions for the driver, and in this case they are displayed to the driver in suitable form. In vehicles which are equipped with an autonomous vehicle guidance system, the vehicle guidance actions which are generated by the consumption assistant 13 can be used for consumption-optimized autonomous guidance of the vehicle.

It is understood that the extent of the forecast area can have any desired value or can be chosen variably depending on the situation, e.g. depending on the speed, larger for higher speeds than for lower speeds.

28 944/DE/1 19 In a further embodiment, the consumption assistant can be designed to determine energy-consumption-optimized driving routes to a destination. In this case, it reads from the digital map 14 the values of attributes which are relevant to consumption for the relevant route network area between the vehicle starting point and the destination, and on the basis of this attribute data determines a driving route to the destination with the lowest possible energy consumption, using a traditional route searching method.

oo Alternatively to the shown example, instead of the road map 14 which can learn and be updated, a non-learning road map can be used for the consumption assistant 13.

15 This road map contains the required attribute data which is relevant to energy consumption as information which is .stored in advance, on the basis of test drives, and is not continuously updated.

:i 20 It is understood that each system component which is •g*mentioned for the consumption assistant function of Fig.

4 can be implemented in hardware or software as desired.

In a further application of a route network map which is updated according to the invention, vehicle-type-specific attribute data is stored in it, and used to generate corresponding vehicle-type-specific guidance information, for instance to prevent use of the traffic route by particular vehicles such as utility vehicles, particularly dangerous goods transporters, at specified times or on specified weekdays. In a further application, the route network map which can be updated can be used to implement an intelligent speed control system, which continuously sets the highest permitted speed on the route which is being driven, depending on the vehicle type and if applicable the time. As well as the attribute 28 944/DE/1 20 data mentioned above, in particular information about town entrance signs and traffic light systems, the switching states of which can be detected by "computer seeing", can be taken into account.

It is understood that this invention is not just suitable for road traffic technology as described, but also for rail, air and shipping traffic technology or combinations of them.

ae...

e ae co*e *r ft

Claims

1. A method of updating a data-processing-supported traffic route network map containing route data about a traffic route network which is driven on by at least one test vehicle, said method comprising the steps of: recording current route data and current attribute data by a said test vehicle, wherein said attribute data includes topographical data comprising slopes and curves of route sections locally exact with the position and strength of slopes and curves of said traffic route network; and updating the traffic route network map using recorded route data and recorded •attribute data, fee: wherein said one or more test vehicles are linked to mutually exchange data with various road vehicles and information obtained is combined into said traffic route 0 network map. 0

2. The method according to claim 1, wherein, for each attribute, multiple measured 00 0.values previously recorded for each attribute, or statistical quantities derived from said multiple measured values, are held and used together with the latest measured value to 0•°0 00:6determine a current attribute value to be stored.

3. The method according to either one of claim 1 or claim 2, wherein said at least *0 .i one test vehicle is linked to a traffic centre via a bi-directional communication channel for data exchange.

4. A method of generating vehicle guidance information or actions, said method comprising the steps of: determining a vehicle's current position; reading data from a data-processing-supported updated traffic route network map comprising route data about a traffic route network that can be driven by a vehicle; assigning attribute data to the traffic route network map, said attribute data including at least locally exact topographical data of the slopes or curves of route sections of the traffic route network; and generating, simultaneously with step advance vehicle guidance information n ',determined by the topographical data for a specifiable forecast route network area, [R:\LIBCC]52950.doc:wxb I\ -22- wherein said the traffic route network map is updated by recording current route data and attribute data of the route section of the traffic route network by at least one test vehicle, that is linked to mutually exchange data with various said road vehicles.

5. The method according to claim 4, comprising the further steps of: estimating the energy consumption expected based on the vehicle's current position, a specifiable destination, and the topographical data available for the relevant area; and generating route choice information as a result of the energy consumption of the relevant route network sections, as advance vehicle guidance information.

6. The method according to either one of claim 4 or claim 5, comprising the further :step of: providing advance vehicle guidance information by generating or triggering 15 energy-consumption-optimising vehicle control information or actions, determined by the topographical data read for the forecast route network area ahead of the vehicle's current position.

7. A method of updating a data-processing-supported traffic route network map, said method being substantially as herein described with reference to Figs. 1 to 5 of the accompanying drawings.

8. A method of generating vehicle guidance information or actions, said method being substantially as herein described with reference to Figs. 1 to 5 of the accompanying drawings. DATED this twenty-eighth Day of August, 2002 DaimlerChrysler AG Patent Attorneys for the Applicant SPRUSON FERGUSON [R:\LIBCC]52950.doc:wxb