GB2470357A

GB2470357A - Disambiguation of location inputs to a navigation system

Info

Publication number: GB2470357A
Application number: GB0908503A
Authority: GB
Inventors: Kurt Dusterhoff
Original assignee: Denso Corp
Current assignee: Denso Corp
Priority date: 2009-05-18
Filing date: 2009-05-18
Publication date: 2010-11-24
Also published as: FR2945623A1; GB0908503D0; FR2945623B1; DE102010029083A1

Abstract

A navigation apparatus performs a location defining process wherein ambiguities in inputted location information are resolved via later inputted location information. The apparatus receives a settlement name S4-3, and searches its map records for settlements matching the received name to create a candidate list location S4-5. If a settlement name is not unique, the map records point to a disambiguation record which defines a tree structure to possible locations arising from the settlement name. The apparatus subsequently receives a road name S4-9, and searches within its map records for settlement and road name combinations matching the received road name for settlements identified in the candidate location list at step S4-11 to create a refined candidate list location. The apparatus may further receive a name / number S4-15, and search within its map records for settlement, road name and name / number combinations matching the received name / number combinations identified in the refined candidate location list at S4-17 to create a final candidate location list. The resolved reduced set of possible combinations may be one or more branches of the tree structure. This reduced set of combinations then constitutes all or part of the final candidate location list S4-19.

Description

Disambiguation

Field

The present invention relates to disambiguation and in particular, but not exclusively, to disambiguation of location inputs to a navigation system.

Background

In some geographical areas, such as Europe, there are situations where many towns/cities/villages within a given search area have the same name. For a spoken interface, this means that the user will, at some point, need to select which of these settlements to go to when entering a destination for one of these settlement names.

Currently known devices use a visual selection list of towns in combination with some higher-level location classification (e.g. county, state, department). One reference (US2003014255) is known which accepts a nearby geographical identifier, such as a river name or large city, to determine which instance of a settlement name is requested by the user. Another reference (US2006253251) uses a city name and street name input interface to narrow a search for a known point of interest (P01) in order to select a POT as a destination. A further reference (US2003 125870) performs address entry by entering an address one address segment at a time. None of these known methods give a smooth interface to the user, as in many cases the user is requested to provide information that may not be necessary.

The present invention has been made in view of drawbacks of known systems.

Summary

Viewed from a first aspect, the present invention provides an apparatus operable to receive voice input for defining a location and to mask a disambiguation process for disambiguating any ambiguity caused by lack of uniqueness in a part of the location definition information from a user such that additional input is only required in the event that excessive ambiguity remains after a complete location definition input is completed.

Viewed from a first aspect, the present invention can provide an apparatus for providing a mapping result. The apparatus can comprise a receiver operable to receive input for a first tier location identifier, and a processor operable to access a first tier record corresponding to the received first tier location identifier, wherein if the received first tier location identifier corresponds to more than one map entity the first tier record is a disambiguation record associated with the first tier record for each of the more than one map entities. The receiver can be further operable to receive input for a second tier location identifier. The processor can be further operable to use the second tier location identifier to identify one or more second tier records subordinate to the first tier record, wherein in the case where the first tier record is a disambiguation record a second tier record may correspond to one or more of the first tier records associated with the disambiguation record, and to determine whether the identified records provide a candidate location result list having sufficiently little ambiguity therein. The apparatus can further comprise an output operable to provide a mapping result representative of a location result selected from the candidate location result list. Thereby a navigation apparatus can provide a mapping result which is derived by using a minimum of input stages to accept location definition information and to resolve ambiguities in the location definition information by using a forward looking process to use future information to resolve current ambiguities. This results in a smooth and minimally intrusive operation of the device, thereby minimising user attention required to perform the location definition.

In some examples, the first tier location identifier is a settlement name. In some examples, the second tier location identifier is a street name or number. Thus the street name or number can be used to resolve ambiguity in a settlement name by removing from consideration any settlements having the settlement name but for which the street name or number is not valid.

In some examples, the apparatus is further operable to, receive input for a third tier location identifier, and to use the third tier location identifier to identify one or more third tier records subordinate to the first and second tier records, wherein in the case where the first tier record is a disambiguation record a third tier record may correspond to one or more of the first tier records associated with the disambiguation record. Thus further tiers of location information can be used to further resolve ambiguities not entirely resolved by the combination of the first and second tier information.

In some examples, the third tier location identifier is a building identifier. Thus the building identifier can be used to resolve ambiguity in a settlement name and street name or number combination by removing from consideration any settlementlstreet combinations having the settlement name and street name or number but for which the building identifier is not valid.

In some examples, the apparatus can be further operable to prompt for receiving input for a third tier location identifier in the event that it is determined that the identified first tier record and one or more second tier records do not provide a candidate location result list having sufficiently little ambiguity therein. Thereby the apparatus can recognise an excessive ambiguity and prompt for information to resolve that ambiguity.

In other examples fourth or further tiers of location definition information can be requested.

Examples of location definition information types which can be used for a location definition information tier include settlement names, street names/numbers, building identifiers, zip/postal codes, administrative regions, and other geographic regional subdivisions.

In some examples, the apparatus can be further operable to use a location result selected from the candidate location result list as a start point, intermediate point or end point for a navigation calculation. Thus the mapping result can result from a navigation calculation based upon using the define location as a point on a calculated route.

In some examples, the apparatus can be further operable, if the candidate location result list comprises a single entry, to select that candidate as the location result. Thus if all ambiguity is removed by use of the first and second (and as the case may be third or subsequent tiers of information) there is no need for any final selection input.

In some examples, the apparatus can be further operable, if the candidate location result list comprises multiple entries, to receive input select a location result.

Thus is some ambiguity remains after all tiers of location definition information are entered, a final user input for disambiguation from a small list can be requested.

In some examples, the apparatus can be further operable to receive input to select a location result from the candidate location results, if it is determined that the identified first tier record and one or more second tier records provide a candidate location result list having sufficiently little ambiguity therein in that the number of candidate location results in the candidate location result list is lower than a predetermined threshold. Thereby, a user is only asked for a final disambiguation input if the list of candidates is sufficiently small.

In some examples, the apparatus can be further operable to receive additional input to reduce the size of the candidate result list or to provide a mapping result using a different method, if it is determined that the number of candidate location results in the candidate location result list is not lower than a predetermined threshold. Thus a large degree of ambiguity can be resolved by further input, which may be input of information for one or more further location definition tiers and/or alternative location definition type input.

In some examples, the can be operable to: receive, before receiving input for a second tier location identifier, input for a mapping result; provide a candidate location result list based upon the record identified accessed based upon the received first tier location identifier if the resultant candidate location result list has sufficiently little ambiguity therein; and receive input for a second tier location identifier if the resultant candidate location result list has excessive ambiguity therein. Thus if a user requests a mapping result before a second tier input, a,mapping result can be returned if there is no or only a small degree of ambiguity in the first tier information, and second tier information can be requested if some, or a larger degree of ambiguity in the first tier information.

In some examples, the apparatus can be operable to receive location identifiers as voice input. Thus a user can conduct a location definition process with a further reduced level of manual interaction required with the apparatus.

In some examples, the apparatus can be operable to determine whether a disambiguation is required by determining whether a plurality of candidate location identifiers derived from the received location identifier voice input requires disambiguation. Thus ambiguity resolution can be used to resolve speech recognition ambiguities in addition to or instead of the instance ambiguities discussed above.

Viewed from a second aspect, the present invention can provide a navigation apparatus operable to perform location definition by identifying ambiguities in received location information and using later received information to resolve the identified ambiguities as the later received information is entered following the same input conversation sequence as is used if there is no requirement for location disambiguation. Thus the resolution of ambiguities in location definition information can be masked from a user by using future information to resolve current ambiguities.

Viewed from another aspect, the present invention can provide a method of providing a mapping result. The method can comprise: receiving input for a first tier location identifier; accessing a first tier record corresponding to the received first tier location identifier, wherein if the received first tier location identifier corresponds to more than one map entity the first tier record is a disambiguation record associated with the first tier record for each of the more than one map entities; receiving input for a second tier location identifier; using the second tier location identifier to identify one or more second tier records subordinate to the first tier record, wherein in the case where the first tier record is a disambiguation record a second tier record may correspond to one or more of the first tier records associated with the disambiguation record; determining whether the identified records provide a candidate location result list having sufficiently little ambiguity therein; and providing a mapping result representative of a location result selected from the candidate location result list.

Thereby a mapping result can be provided based upon a location definition which is derived by using a minimum of input stages to accept location definition information and to resolve ambiguities in the location definition information by using a forward looking process to use future information to resolve current ambiguities. This results in a smooth and minimally intrusive operation of a navigation device, thereby minimising user attention required to perform the location definition Viewed from a further aspect, the present invention can provide a method for disambiguation of location data received as part of an input conversation which method implicitly disambiguates the location as additional data is entered following the same input conversation sequence as is used if there is no requirement for location disambiguation. Thus the resolution of ambiguities in location definition information can be masked from a user by using future information to resolve current ambiguities.

Viewed from another aspect, the present invention can provide a method of providing a mapping result, the method comprising: receiving input for a settlement name; determining whether, for the received settlement name, a disambiguation is required; if no disambiguation is required, selecting the settlement as a location result; if a disambiguation is required, receiving input for a street name within the settlement; determining whether, for the received settlement and street name combination, a further disambiguation is required; if no further disambiguation is required, selecting the settlement and street combination as a location result; if a further disambiguation is required, receiving input for a location identifier with respect to the street name; determining whether, for the received settlement, street name and location identifier combination, an additional disambiguation is required; if no additional disambiguation is required, selecting the settlement, street and location identifier combination as a location result; if a further disambiguation is required, receiving a disambiguation result to provide a location result; and providing a mapping result representative of the location result. Thereby the use of a street identifier that is to be input in any case can be used to resolve ambiguities in a settlement name and the use of a location identifier for a street that is to be input in any case can be used to resolve ambiguities in a settlementlstreet name combination.

Viewed from another aspect, the present invention can provide a navigation device operable to perform disambiguation of received location data by implicitly disambiguating the location as additional data is entered following a normal input conversation sequence that is used if there is only one instance of the name.

Further and additional aspects and embodiments will be apparent from the following detailed description and the appended claims.

Brief description of the Figures

For a better understanding of the invention and to show how the same may be carried into effect reference is now made by way of example to the accompanying drawings in which: Figure 1 shows a schematic illustration of a navigation device operable to receive input for defining a location; Figure 2 shows a schematic illustration of logical elements of a navigation device; Figure 3 shows a flow chart illustrating steps in a location definition process from the point of view of a user; Figure 4 shows a flow chart illustrating steps in a location definition process from the point of view of a navigation device; Figure 5 shows a flow chart illustrating steps in a location definition process in

a summary schematic manner;

Figure 6 shows a schematic representation of a tree strncture for handling disambiguation in location definition; and Figure 7 shows a flow chart illustrating a generalised set of steps in a location defmition process.

While the invention is susceptible to various modifications and alternative forms, specific embodiments are shown by way of example in the drawings and are herein described in detail. It should be understood, however, that the drawings and detailed description are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents and alternatives falling within the spirit and scope of the present invention as defined in the appended claims.

Specific Description

Figure 1 shows schematically a navigation device 1 operable to receive input for defining a location. The navigation device 1 includes a display 2 on which user interface content can be displayed to a user. The navigation device 1 also includes input elements such as buttons 3 and a microphone 4 for receiving input from a user.

Input elements may also include provision of a display 2 with a touchscreen input capability. The navigation device 1 also includes a loudspeaker from which audible user interface content can be provided to a user. The navigation device also includes a data connectivity port 6 through which the device can receive updated and/or replacement data records and/or user interface instructions. Data connectivity may also or alternatively be provided by a wireless data interface or removable medium (not shown).

In the present examples, the navigation device 1 is intended for use in a vehicle such as a car, van, truck or lorry. In some examples, the navigation device may be additionally or alternatively useable with a straddle type vehicle such as a bicycle, motorbike, moped or scooter, and/or by a pedestrian. To facilitate use in or with a vehicle, the navigation device 1 may be provided with mounting elements (not shown) to mount or secure the device into or onto a vehicle. To facilitate handheld use by a pedestrian, the navigation device 1 may be provided with a handle or grip element (not shown).

Figure 2 shows logical elements within the navigation device 1 which enable the navigation device to carry out location definition processing. A microprocessor or microcontroller 7 is operable to communicate via interface bus 7 (which may be implemented as one or more separate busses) with the screen 2, the buttons 3, the microphone 4, the speaker 5, and the data interface 6. The data interface 6 may have a receiver or removable media reader associated therewith (not shown) in the event that such channels are provided. Also present is a programme memory 9 which stores the operating instructions for the processor/controller 7. In some examples, the programme memory 9 may be integral with the processor/controller 7. Map record data is stored in a map memory 10. The map data includes data describing all individually identifiable points within a mapping area as well as relationship data describing the connections between those points. The map data also includes relationship data describing a hierarchical structure of mapping data classes, such that individual records can be identified as belonging to different classes, such as settlements, streets, buildings, locations etc. A navigation receiver 11 is operable to provide information describing the present location and optionally the present velocity of the navigation device. In the present examples, the navigation receiver is a satellite positioning receiver such as a NAVSTAR GPS receiver. The navigation receiver may have an antenna associated therewith (not shown) which may be internal or external to a body of the navigation device 1.

In the present examples, the navigation device is operable to receive location definition input by user voice input. In this circumstance, the operating instructions for the navigation device include instructions to enable speech recognition. A voice input facility enables the device to be used by a driver of a moving vehicle with significantly lower potential impact to driving safety than in situations where a button press or touchscreen input is used. In other examples, or in other usage modes of the navigation device 1, the device may be operable to receive input via button presses and/or through touchscreen input in addition or alternatively to voice input.

The navigation device 1 of the present examples, is operable to mask ambiguities in location definition input from a user. This is performed by prompting for and receiving input for a location identification tier even if location information input for a higher level tier is insufficient to uniquely identify the location at that higher tier. The received input for the lower tier is combined with the location information from a higher tier to resolve ambiguities in the location without presentation of disambiguation options to the user. If an ambiguity remains after all tiers of location information are received, then a list of options can be provided to the user to make a final selection. This masking of ambiguity resolution provides that the user can define the location with a minimum of additional inputs over and above the standard inputs for each tier of location information.

An example of this process is illustrated from the point of view of a user in Figure 3. The location definition process starts with the user receiving a prompt for a settlement name at step S3 -1. The user provides a settlement name at step S3-3. The user may provide the settlement name via voice input for receipt by microphone 4 or via touch input for receipt by touchscreen 2 or buttons 3. The user then receives a prompt for a road name at step S3-5, in response to which the user provides a road name input at step S3-7. The user then receives a prompt for a location identifier (e.g. a building name/number) for the location at step S3-9, in response to which the user provides a name/number input at step S3-.11. Following entry of the information to define the location, the user receives a prompt at step S3-13 to confirm the location which has been derived from the provided inputs or to select a location from a list of locations which all match the provided inputs. Finally, the user provides confirmation or selection input as required at step S2-15 and the location definition process ends.

As will therefore be understood, the user experience is one of providing an input for both a first and a subsequent location information tier, regardless of whether the input for any given tier results in a unique location identification at that tier, until the user is satisfied that information for the required number of tiers has been entered to define the intended location. At this point the user receives either a single location selection to confirm or, if some ambiguity remains after all tiers have been provided, a request for a selection from a list of locations. Thereby, the ambiguities which will inevitably arise at higher level tiers of location information are masked from the user.

With reference to Figure 4, the operation of the navigation device to provide the process of figure 3 to a user will now be described. The location definition process commences with the navigation device prompting for a settlement name at step S4-1, followed by receiving settlement name input at step S4-3. The navigation device then searches within its map records for settlements matching the received settlement name at step S4-5 to create a candidate location list. In the present example, if a settlement name within the map records is non-unique within the search area (typically a single country, although individual ones or groups of administrative regions such as states, counties or departments could be used) then the map records include or point to a disan-ibiguation record for that settlement name. The disambiguation record may define a tree structure or other maimer of listing or linking to the possible locations which arise from the settlement name.

The navigation device then prompts at step S4-7 for input for a road name, which input is received at step S4-9.

The navigation device then searches within its map records for settlement & road name combinations matching the received road name for settlements identified in the candidate location list at step S4-1 1 to create a refined candidate location list. In the present example, if the candidate location list includes a disambiguation record, then the road name is used to resolve from the disambiguation record a reduced set of possible settlement name & road name combinations. In the example where the disambiguation record is a tree structure, the resolved reduced set of possible combinations may be one or more branches of the tree structure. This reduced set of combinations then constitutes all or part of the refined candidate location list.

The navigation device then prompts at step S4-13 for input for a road name, which input is received at step S4-15.

The navigation device then searches within its map records for settlement & road name & name/number combinations matching the received name/number for settlement & road name combinations identified in the refined candidate location list at step S4-17 to create a final candidate location list. In the present example, if the refined candidate location list includes a disambiguation record, then the name/number is used to resolve from the disambiguation record a reduced set of possible settlement name & road name & name/number combinations. In the example where the disambiguation record is a tree structure, the resolved reduced set of possible combinations may be one or more branches of the tree structure. This reduced set of combinations then constitutes all or part of the final candidate location list.

Having now received all of the standard inputs from the user, the navigation device checks whether the final candidate location list includes more than one entry at step S4-19. If only one entry is present on the final candidate location list, then processing continues at step S4-21 where the navigation device prompts for input to confirm the candidate location as a selected location, which input is received at step S4-23, after which the location is selected and the selection process ends.

If on the other hand, it is determined at step S4-19 that more than one entry exists on the final candidate location list, then processing continues at step S4-25 where the list is presented to the user with a prompt for the user to select one entry from the list. A selection input is received at step S4-27, after which the location is selected and the selection process ends.

Thus an example of location definition with masked disambiguation has now been described. This process enables the user to enter data to define a location in such a way that any ambiguity in the location definition is masked until such time as all requested location definition information has been entered. In this way, if the sum of the location definition information removes any ambiguity in earlier definition levels, then the user is presented with no ambiguity resolution requests at all, and if the sum of the location definition information leaves some ambiguity, then the user can be presented with a single request for ambiguity resolution input at the end of the process once all opportunities for masking the ambiguity have been exhausted.

It will be appreciated that this approach provides for a number of options that would not be possible or would be inconvenient for the user if such a masked disambiguation were not performed.

In some examples, the location information input can be made in any order as any ambiguity caused by entering, for example, a road name before a settlement name can be masked-from the user. Taking an example of common English road names, entering "London Road" without a settlement name might lead to many tens of possible road entries and a road name such as "High Street" or "The Street" without a settlement name might lead to many hundreds of possible road entries. Such ambiguity would be, in practicable terms, unmanageable without the use of masked disambiguation to hide the large list of possible entries until after entry of a settlement name to resolve the ambiguity to a manageable level.

In the present example, the location information can be provided by voice input. In this circumstance, the navigation device operating instructions in the program memory 9 include instructions to enable voice recognition processing by the navigation device to convert voice input to text data for searching against the map records. In such a case, it is possible that additional ambiguity could arise in a location definition process due to uncertainties in the voice recognition process. For example, a user speaking the English settlement name "Lee" could result in possible voice recognition candidates including and "Lee", "Leigh". In another example, a user speaking the Italian settlement name "San Martino" could result in possible voice recognition candidates including "San Marino" and "Santa Marina". Resolution of the ambiguity that this causes can be carried out by presenting recognition candidates to the user to chose between at each data entry stage, or by handling the recognition ambiguity using the masked disambiguation described above. Using a disambiguation structure such as a tree would enable the candidate location list to include tree entries for each of the settlements identified as possible voice recognition candidates, with the road name and name/number inputs being used to carry out the masked disambiguation as described above.

As will be appreciated, one situation in which a voice input is of potential utility is when a user is driving a vehicle that is in motion. As such, it is likely that a location entered by voice input would be a navigation destination (either a final destination or an intermediate destination) in that if a vehicle is already in motion, it is likely that a start location would be its current location. However, voice input could be used to input a start location or any other location is required.

An example of the above described examples in the context of a voice recognition using navigation device will now be presented with reference to Figure 5.

In Figure 5, four columns are shown in parallel to illustrate the processing. In the user's data entry flow, the user starts by entering the city name "San Martino". In the present example, the user is presented with a choice of recognition candidates to choose between. The chosen recognition candidate is then used as the data search key S to search the map records for possible matches. In the present example, the searôb key "San Martino" returns a disambiguation record listing the 62 candidate cities with that name in the present map region (which is assumed in the present example to be Italy).

In order to manage a masked disambiguation search, the disambiguation record combines subsequent location data for all instances of the ambiguous settlement name.

In the present example of a search in Italy, there are a large number of (>60) towns/cities with the name (or common alias) of "San Martino". Thus all street names for all instances of "San Martino" are compiled within the static data into a command list under the single entry for "San Martino". Streets which are non-unique among some or all "San Martino" instances are similarly compiled into a single entry.

Then, the conversation implicitly disambiguates the location as additional data is entered following the normal conversation sequence that is used if there is only one instance of the name. Thus, in the present example, the user next enters a street name of "Via Europa" and chooses the correct recognition candidate from those presented, such that the data search key for the street name is "Via Europa". Using this in combination with the previously identified disambiguation record for the city results in identifying from the disambiguation record the 25 candidate combinations of the entered city name and street name.

The user next enters the building identifier "14" which input is used as as earch key in combination with the 25 candidate combinations for the city and street name to identify the four possibilities for the overall address of "14 Via Europa, San Martino".

These four candidate locations are then presented to the user for a final selection, and the selected entry is then used as a navigation location by the navigation device.

It is therefore apparent how a navigation device can be provided such that a user can be presented with a user interface which avoids presenting disambiguation options to the user until after the user has provided all of the location definition information which a particular location definition conversation normally requires in order to avoid troubling the user unnecessarily for potentially irrelevant input. Thus the location definition process is as short as possible, thereby minimising interaction requirements for a driver of a moving vehicle.

There have been described above, a number of examples in which a disambiguation record is used to provide efficient searching through map records for a region in which a single settlement and/or street name has multiple separate occurrences. An example of a suitable disambiguation record tree structure for the search performed in the example of Figure 5 is illustrated with reference to Figure 6.

The tree structure has as a highest level 30 an identifier to identify the record as relating to the settlement name "San Martino". The next level has identifiers 31 for all possible road names for that city name. This level also has records 32 for a predefined datum point for each city having the city name identified at 30. For each of the possible road name identifiers 31, a further level is provided which includes identifiers 33 for all possible building identifiers (name or number) for each city/road combination. This level also has records 34 for a predefined datum point for each road having the city name identified at 31 and the street name identified at 31. A final level provides the individual map records which the city/road/building combination can correspond to. In some cases, where a given city/street name/building identifier combination has multiple options, there are entries 35 for each of the options, whereas where the given /street name/building identifier combination leads to only a single option, just this single option 36 is present. The individual map records may include some further information (such as alternative location identifiers such as a zip or postcode or regional descriptors) which enable each entry at levels 32, 34 and 35 to be identified from each other one.

Thereby, speech recognition command lists for location entry are compiled within the data, as described above, to include ambiguous names as a single instance.

However, the map-data interface information (for example county or state) is maintained for the map database search and possible small candidate lists. Thus the conversation model can continue as though there is only one instance of each piece, until the point where the candidate list is within acceptable limits.

Accordingly, when map records have been consolidated or linked together to combine the data as described above, when a user enters a town of, for example, "San Martino", the navigation device will then process a received street name against the combined street name list. Then, the navigation device can prompt for a selection from among the choice of street + individual "San Martino" options. Or, if there are still too many possibilities for a smooth speech interface, the navigation device can prompt for entry of a building number or other identifier.

At this point, many instances of "San Martino" are excluded from the candidates because they do not contain the combination of street name and building number given by the user. This way, the navigation device avoids prompting for further information input unnecessarily. The navigation device can prompt for selection from a small list of candidates of number + street + individual "San Martino" options if the list is small enough.

It is possible to show a destination map or a text description of the address (end address entry) by saying a command, such as "I'm finished" or "Start Guidance" after the user has entered either the city name or street name. At this timing, if the number of candidates = 1, the system will show the destination directly. If the number of candidates is over 1 and less than a specified size in the system, then the system will show a candidate list and ask user to choose. If the number of candidates is over the specified size, the system can move to next step of the address entry sequence, without asking the user to select from the list of, for example, "San Martino" entries.

Finally, if there are still too many options for a smooth speech interface after the standard conversation, some other disambiguation method could be used. One example option is to re-start the search from a different point, such as entering a zip or postal code, which will narrow the search quickly and show the user a way to find this location faster next time. Alternatively, the system could request information from a different aspect of the address, such as the postal code, and then automatically take all data to attempt to provide a candidate list to the user. Or, the system could fall back to the other known input methods if all other methods fail It will therefore be appreciated that in the present examples there is no effect on the conversation stages for entering a city name or a street name. The map management method provided in the present examples allows the disambiguation to be done implicitly, i.e. in a hidden or masked manner.

Thus a number of examples for the implementation of a navigation device operable to perform masked disambiguation during location definition by means of a predefined record to facilitate disambiguation processing have been described.

In some examples, the navigation device can be configured to require selection inputs where a voice recognition step receives input relating to a name, number or other free-choice identifier. It is usual in systems that rely on speech recognition for input for each speech recognition based input to be followed by a facility for the recognition candidate(s) to be presented to the user for user selection. This avoids the possibility of a misrecognition at the start of a process leading to a whole process being carried out based on bad input. Thus, if a user provides a voice input (e.g. San Martino), the next stage in a conversation would allow for disambiguation among similar sounding data. For example, the navigation device may find recognition candidates including "San Martino" and "San Marino" and "Santa Marina" and so the recognition confirmation allows for one of these to be selected before the process continues. This additional level of disambiguation is not required in a standard deterministic manual input method. This allows the navigation device to present this minimum of voice recognition disambiguation without separately requesting input to differentiate between different instances of each recognition candidate. Depending upon the way in which the recognition candidates are presented to a user with a selection input prompt, the navigation device may also achieve some instance disambiguation at the same time as achieving voice recognition disambiguation without a second stage of candidate selection being required. It is also noted that in alternative examples, the voice recognition disambiguation could be handled without asking for recognition candidate selection at the recognition stage and instead using all of the recognition candidates as the received respective location tier identifier and using the instance disambiguation based upon further input to handle the recognition ambiguity as well as the instance ambiguity.

Figure 7 shows an abstracted example of a navigation device operation through a location definition process based upon voice input in which instance ambiguity is resolved by use of future location definition input.

Starting at step S78-1, a navigation device receives lS tier location information.

From this received information, a set of recognition candidates are determined at step S7-3. The set may have one or more members based upon the interaction of the received information and the recognition process. A selection is then made between the recognition candidates at step S7-5. This selection may be made by the navigation device by comparing the recognition candidates to map records and eliminating recognition candidates that do not correspond to any entry in the map records. The selection may also include associating a recognition candidate with one or map records which have phonetically similar identifiers. The selection may alternatively or additionally include requesting and receiving input to select one or more recognition candidates from among a number candidates presented by the navigation device for a selection input.

Once one or more tier recognition candidates have been selected, step S7-7 performs a search within the map records to identify map records corresponding to the selected recognition candidates. If any of the selected recognition candidates corresponds to more than one individual map records, a consolidated or disambiguation record may be returned by the search, whereby the consolidated or disambiguation record contains or links to the individual map records corresponding to the searched 1st tier recognition candidate.

Next, at step S 7-9, the navigation device receives tier location information.

From this received information, a set of recognition candidates are detennined at step S7-1 1. The set may have one or more members based upon the interaction of the received information and the recognition process. A selection is then made between the recognition candidates at step S7-13. This selection may be made by the navigation device by comparing the recognition candidates to map records and eliminating recognition candidates that do not correspond to any entry in the map records identified by the search based upon the lS tier recognition candidates. The selection may also include associating a recognition candidate with one or map records which have phonetically similar identifiers. The selection may alternatively or additionally include requesting and receiving input to select one or more recognition candidates from among a number candidates presented by the navigation device for a selection input.

Once one or more 2'' tier recognition candidates have been selected, step S7-performs a search within the map records retuned by the search on the selected tier recognition candidates to identify map records corresponding to the selected 2rn1 tier recognition candidates. If any of the selected 1st tier recognition candidates returned a consolidated or disambiguation record, the selected 2nd tier recognition candidates are used to search within the map records within or linked to by the consolidated or disambiguation record. Thus the 2' tier input is used narrow the pool of potential locations identified in the results of the search against the lS tier input.

The result of this search is one or more location candidates defined by the combination of the and 2nd tier inputs received by the navigation device. In the present example, these candidates are provided to in element that requested a location definition at step S7-17. In other examples, further location information tiers could be processed andlor in the event that the total number of location candidates exceeds 1 after all location information tiers have been processed, the navigation device can request and receive a selection input to select between the location candidates to that a single location candidate is returned by the location definition process.

Accordingly, it will be understood that a variety of implementation options exist for providing a navigation device which uses future location definition input to resolve ambiguities in location definition input and only presents an ambiguity resolution option to a user once all location definition input is received and if any ambiguity remains after processing of all location definition inputs. The ambiguities resolved by this approach can be instance ambiguities where a particular location definition input identifies a location which has multiple instances within a search area and/or can be recognition ambiguities where a speech recognition process returns multiple locations within a search area based upon a single location definition input.

The terms used for different navigation tiers as used herein are interchangeable and any one use of a particular term does not exclude an alternative same tier identifier being applicable in that situation. For example, a settlement tier input may relate to any of a number of different settlement types, such as, for example, a city, town, village, hamlet, community or suburb. Similarly, a road identifier input may relate to any way of identifying a particular thoroughfare, such as, for example, a road, street, track or path name or number. Also, a building identifier input may relate to any way of identifying either a building or a part or a building, such as a building or building part (e.g. flat or apartment) name or number.

Although the above described examples utilise a navigation device based upon a NAVSTAR GPS satellite positioning system receiver, other positioning receivers could be used. In some examples, the navigation receiver could be a receiver for alternative satellite navigation systems such as Galileo, Beidou, COMPASS or GLONASS instead of or in addition to NAVSTAR GPS. In other examples the navigation receiver could be a receiver for alternative radio navigation systems such as the LORAN, DECCA, VLF/Omega or Alpha hyperbolic systems instead of or in addition to a satellite navigation system. In still further examples, the navigation receiver could be based upon use cell positioning information from a cellular radiotelephone network instead of or in addition to a satellite navigation system.

The above described examples are by way of example only and are not intended to be limiting on the scope of the invention. The scope of the present invention is defined in the accompanying claims and encompasses all equivalents and modifications falling within the spirit and scope of the appended claims.

Claims

CLAIMSClaims 1. Apparatus for providing a mapping result, the apparatus comprising: a receiver operable to receive input for a first tier location identifier; a processor operable to: access a first tier record corresponding to the received first tier location identifier, wherein if the received first tier location identifier corresponds to more than one map entity the first tier record is a disambiguation record associated with the first tier record for each of the more than one map entities; the receiver further operable to receive input for a second tier location identifier; the processor further operable to: use the second tier location identifier to identify one or more second tier records subordinate to the first tier record, wherein in the case where the first tier record is a disambiguation record a second tier record may correspond to one or more of the first tier records associated with the disambiguation record; determine whether the identified records provide a candidate location result list having sufficiently little ambiguity therein; and an output operable to provide a mapping result representative of a location result selected from the candidate location result list.
2. The apparatus of claim 1, wherein the first tier location identifier is a settlement name.
3. The apparatus of claim 1 or 2, wherein the second tier location identifier is a street name or number.
4. The apparatus of claim 1, 2 or 3, further operable: to receive input for a third tier location identifier; and to use the third tier location identifier to identify one or more third tier records subordinate to the first and second tier records, wherein in the case where the first tier record is a disambiguation record a third tier record may correspond to one or more of the first tier records associated with the disambiguation record.
5. The apparatus of claim 4, wherein the third tier location identifier is a building identifier.
6. The apparatus of claim 4 or 5, wherein the apparatus is operable to prompt for receiving input for a third tier location identifier in the event that it is determined that the identified first tier record and one or more second tier records do not provide a candidate location result list having sufficiently little ambiguity therein.
7. The apparatus of any preceding claim, further operable to use a location result selected from the candidate location result list as a start point, intermediate point or end point for a navigation calculation.
8. The apparatus of any preceding claim, further operable, if the candidate location result list comprises a single entry, to select that candidate as the location result.
9. The apparatus of any preceding claim, further operable, if the candidate location result list comprises multiple entries, to receive input to select a location result.
10. The apparatus of claim 9, further operable to receive input to select a location result from the candidate location results, if it is determined that the identified first tier record and one or more second tier records provide a candidate location result list having sufficiently little ambiguity therein in that the number of candidate location results in the candidate location result list is lower than a predetermined threshold,.
11. The apparatus of claim 9 or 10, further operable to receive additional input to reduce the size of the candidate result list or to provide a mapping result using a different method, if it is determined that the identified first tier record and one or more second tier records do not provide a candidate location result list having sufficiently little ambiguity therein in that the number of candidate location results in the candidate location result list is not lower than a predetermined threshold,.
12. The apparatus of any preceding claim, further operable to: receive, before receiving input for a second tier location identifier, input for a mapping result; provide a candidate location result list based upon the record identified accessed based upon the received first tier location identifier if the resultant candidate location result list has sufficiently little ambiguity therein; and receive input for a second tier location identifier if the resultant candidate location result list has excessive ambiguity therein.
13. The apparatus of any preceding claim, operable to receive location identifiers as voice input.
14. The apparatus of claim 13, operable to determine whether a disambiguation is required by determining whether a plurality of candidate location identifiers derived from the received location identifier voice input requires disambiguation.
15. Navigation apparatus operable to perform location definition by identifying ambiguities in received location information and using later received information to resolve the identified ambiguities as the later received information is entered following the same input conversation sequence as is used if there is no requirement for location disambiguation.
16. A method of providing a mapping result, the method comprising: receiving input for a first tier location identifier; accessing a first tier record corresponding to the received first tier location identifier, wherein if the received first tier location identifier corresponds to more than one map entity the first tier record is a disambiguation record associated with the first tier record for each of the more than one map entities; receiving input for a second tier location identifier; using the second tier location identifier to identify one or more second tier records subordinate to the first tier record, wherein in the case where the first tier record is a disambiguation record a second tier record may correspond to one or more of the first tier records associated with the disambiguation record; determining whether the identified records provide a candidate location result list having sufficiently little ambiguity therein; and providing a mapping result representative of a location result selected from the candidate location result list.
17. The method of claim 16, wherein the first tier location identifier is a settlement name.
18. The method of claim 16 or 17, wherein the second tier location identifier is a street name or number.
19. The method of claim 16, 17 or 18, further comprising: receiving input for a third tier location identifier; and using the third tier location identifier to identify one or more third tier records subordinate to the first and second tier records, wherein in the case where the first tier record is a disambiguation record a third tier record may correspond to one or more of the first tier records associated with the disambiguation record.
20. The method of claim 19, wherein the third tier location identifier is building identifier.
21. The method of claim 19 or 20, further comprising prompting for receiving input for a third tier location identifier in the event that it is determined that the identified first tier record and one or more second tier records do not provide a candidate location result list having sufficiently little ambiguity therein.
22. The method of any of claims 16 to 21, further comprising using a location result selected from the candidate location result list as a start point, intermediate point or end point for a navigation calculation.
23. The method of any of claims 16 to 22, wherein if the candidate location result list comprises a single entry, selecting that candidate as the location result.
24. The method of any of claims 16 to 23, wherein if the candidate location result list comprises multiple entries, receiving input to select a location result.
25. The method of claim 24, further comprising receiving input to select a location result from the candidate location results, if it is determined that the identified first tier record and one or more second tier records provide a candidate location result list having sufficiently little ambiguity therein in that the number of candidate location results in the candidate location result list is lower than a predetermined threshold.
26. The method of claim 24 or 25, further comprising receiving additional input to reduce the size of the candidate result list or to provide a mapping result using a different method, if it is determined that the identified first tier record and one or more second tier records do not provide a candidate location result list having sufficiently little ambiguity therein in that the number of candidate location results in the candidate location result list is not lower than a predetermined threshold.
27. The method of any of claims 16 to 26, further comprising: receiving, before receiving input for a second tier location identifier, input for a mapping result; providing a candidate location result list based upon the record identified accessed based upon the received first tier location identifier if the resultant candidate location result list has sufficiently little ambiguity therein; and receiving input for a second tier location identifier if the resultant candidate location result list has excessive ambiguity therein.
28. The method of any of claims 16 to 27, wherein received location identifiers are received as voice input.
29. The method of claim 28, wherein the determining whether a disambiguation is required comprises determining whether a plurality of candidate location identifiers derived from the received location identifier voice input requires disambiguation.
30. A method for disambiguation of location data received as part of an input conversation which method implicitly disambiguates the location as additional data is entered following the same input conversation sequence as is used if there is no requirement for location disambiguation.
31. Apparatus substantially as hereinbefore described.
32. A method substantially as hereinbefore described.