CN110926493A

CN110926493A - Navigation method, navigation device, vehicle and computer readable storage medium

Info

Publication number: CN110926493A
Application number: CN201911260326.7A
Authority: CN
Inventors: 唐乾斌; 孙仿逊; 翁志伟
Original assignee: Guangzhou Xiaopeng Motors Technology Co Ltd
Current assignee: Guangzhou Xiaopeng Motors Technology Co Ltd
Priority date: 2019-12-10
Filing date: 2019-12-10
Publication date: 2020-03-27

Abstract

The embodiment of the invention provides a navigation method, a navigation device, a vehicle and a computer readable storage medium, wherein the method comprises the following steps: receiving a navigation request of a user; identifying, from the navigation request, query text for which the user intends to be a multi-location navigation intent; identifying from the query text, a plurality of segmented data corresponding to a single location; analyzing the segmented data of the plurality of corresponding single sites to generate a plurality of single site semantic data; generating multi-place semantic data by adopting the single-place semantic data; and generating a navigation route comprising a plurality of places by adopting the multi-place semantic data. By adopting the navigation method provided by the embodiment of the invention, the navigation route comprising the plurality of places can be generated according to the condition that the user continuously speaks the plurality of places without speaking the plurality of places at intervals, so that the use process of the user is simplified, and the user experience is improved.

Description

Navigation method, navigation device, vehicle and computer readable storage medium

Technical Field

The present invention relates to the field of navigation technologies, and in particular, to a navigation method, a navigation apparatus, a vehicle, and a computer-readable storage medium.

Background

With the development of navigation technology, more and more users are used to adopt navigation systems for navigation, and some navigation systems provide voice navigation functions so that the users can use the navigation systems in the process of driving a vehicle conveniently. The navigation system may identify a navigation request from a user and navigate based on a location identified from the navigation request.

However, voice navigation of the existing navigation system only navigates one location identified from the navigation request, and ignores other locations in the navigation request.

If the user wants to go to multiple places in sequence, the user needs to say that the four sentence navigation systems can determine the navigation route, for example:

statement 1: 'navigation to Beijing university'

A navigation system: "do i find the result, which are you going? "

Statement 2: 'navigation to Beijing amusement park'

A navigation system: "do i find the result, which are you going? "

Statement 3: navigation to capital airport "

A navigation system: "do i find the result, which are you going? "

Statement 4: 'navigation to Beijing west station'

A navigation system: "do i find the result, which are you going? "

After the user speaks the four sentences, the navigation system can identify that the user wants to go to Beijing university, then to Beijing amusement park, then to capital airport, and finally to Beijing Western station, and generates a navigation route.

For users, the mode needs a mode of 'asking one answer at a time' for many times, and is not convenient enough.

In the prior art, a method for generating a navigation route by using a preconfigured sentence pattern template is proposed, where the sentence pattern template has preconfigured slots, and the navigation route including multiple places can be generated by filling contents in a sentence of a user into the matched slots. However, this method depends on maintaining multiple sentence pattern templates in operation, and for users with different language habits, a large number of sentence pattern templates need to be maintained to approximately meet the requirement, and once the matching of the sentence pattern templates fails, a correct navigation route cannot be generated, which affects the use of the user.

Disclosure of Invention

In view of the above, embodiments of the present invention are proposed in order to provide a navigation method, a navigation apparatus, a vehicle, and a computer-readable storage medium that overcome or at least partially solve the above-mentioned problems.

In order to solve the above problem, an embodiment of the present invention discloses a navigation method, including:

receiving a navigation request of a user;

identifying, from the navigation request, query text for which the user intends to be a multi-location navigation intent;

identifying from the query text, a plurality of segmented data corresponding to a single location;

analyzing the segmented data of the plurality of corresponding single sites to generate a plurality of single site semantic data;

generating multi-place semantic data by adopting the single-place semantic data;

and generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

Optionally, the identifying, from the query text, a plurality of segmented data corresponding to a single place, comprises:

identifying a plurality of points of interest from the query text;

according to the interest points, identifying a plurality of sentence segments corresponding to a single place from the query text, and identifying the sequence of the places corresponding to the sentence segments; wherein each sentence segment contains at least one point of interest;

and generating a plurality of segment data corresponding to a single place by adopting the sequence of the plurality of sentence segments and the corresponding places.

Optionally, the identifying, from the query text, a plurality of segmented data corresponding to a single place, further comprises:

identifying anchor words from the query text;

the identifying a plurality of sentence segments corresponding to a single place from the query text according to the plurality of interest points comprises:

determining a point of interest associated with the anchor word;

employing the anchor words, and the associated points of interest, a plurality of sentence segments corresponding to a single place are identified from the query text.

Optionally, the generating, by using the sequence of the plurality of sentence fragments and corresponding places, a plurality of fragment data corresponding to a single place includes:

identifying a plurality of entity types that match points of interest in the plurality of sentence fragments;

generating a plurality of segmented data corresponding to a single location using the interest points in the plurality of sentence segments, the plurality of entity types, the anchor words in the plurality of sentence segments, and the order of the plurality of corresponding locations.

Optionally, generating the segmented data of a plurality of corresponding single locations by using the interest points in the plurality of sentence fragments, the plurality of entity types, the anchor words in the plurality of sentence fragments, and the order of the plurality of corresponding locations comprises:

storing according to a preset structure by adopting interest points in the statement segments, the entity types, anchor words in the statement segments and the sequence of the corresponding places to obtain segmented data of the corresponding single places; the preset structure body comprises an interest point variable, an anchor word variable and a sequence variable.

Optionally, the method further comprises:

and adjusting the sequence of the corresponding places by adopting the anchor words.

Optionally, the adjusting, with the use of the plurality of anchor words, an order of the plurality of corresponding places includes:

and adjusting the sequence of the corresponding places in the segmented data by adopting the sequence relation between the anchor words in the segmented data and a preset anchor word.

Optionally, the parsing the segmented data of the plurality of corresponding single places to generate a plurality of single-place semantic data includes:

acquiring the sequence of a plurality of corresponding places, a plurality of interest points and a plurality of entity types matched with the interest points in the segmented data of the plurality of corresponding single places;

generating a plurality of interest point semantic data aiming at the corresponding places by adopting the plurality of interest points;

determining a plurality of semantic tags aiming at the interest points by adopting the entity types matched with the interest points;

and generating a plurality of single-place semantic data by adopting the plurality of interest point semantic data aiming at the corresponding places, the sequence of the plurality of corresponding places and a plurality of semantic labels aiming at the corresponding places.

Optionally, the method further comprises:

identifying route preference semantic data according to the query text;

the generating a navigation route including a plurality of places by using the multi-place semantic data includes:

merging the multi-place semantic data and the route calculation preference semantic data;

generating a navigation route comprising a plurality of places by adopting the merged multi-place semantic data and route calculation preference semantic data

Optionally, the multi-place semantic data comprises a plurality of point of interest semantic data for a plurality of places, an order of the plurality of places, and a plurality of semantic tags for the places; the generating a navigation route including a plurality of places by using the merged multi-place semantic data and route calculation preference semantic data comprises:

searching a plurality of matched target navigation positions from preset navigation positions by adopting the semantic data of the interest points and the semantic labels aiming at the positions;

determining a starting point, a passing point and an end point by adopting the target navigation positions and the sequence of the positions;

determining a navigation planning strategy by adopting the route calculation preference semantic data;

and generating a navigation route by adopting the starting point, the passing point and the end point of the navigation route and the navigation planning strategy.

The embodiment of the invention also discloses a navigation method, which comprises the following steps:

collecting voice of at least one user;

generating a navigation request by adopting the voice of the at least one user;

sending the navigation request to a server so that the server identifies query text intended by a user as a multi-place navigation intention from the navigation request; identifying from the query text, a plurality of segmented data corresponding to a single location; analyzing the segmented data of the plurality of corresponding single sites to generate a plurality of single site semantic data; generating multi-place semantic data by adopting the single-place semantic data; and generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

Optionally, the generating a navigation request by using the voice of the at least one user includes:

respectively carrying out voice recognition on voices of a plurality of users to obtain corresponding voice recognition results;

performing semantic analysis on the voice recognition results corresponding to the multiple users, and determining a voice recognition result related to navigation;

and generating a navigation request according to the voice recognition result related to navigation.

The embodiment of the invention also discloses a navigation device, which comprises:

the navigation request receiving module is used for receiving a navigation request of a user;

the query text identification module is used for identifying query texts of which the user intentions are multi-place navigation intentions from the navigation requests;

a segmented data identification module for identifying from the query text a plurality of segmented data corresponding to a single location;

the single-place semantic data generation module is used for analyzing the segmented data of the plurality of corresponding single places to generate a plurality of single-place semantic data;

the multi-place semantic data generating module is used for generating multi-place semantic data by adopting the single-place semantic data;

and the navigation route generation module is used for generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

Optionally, the segmented data identification module comprises:

an interest point identification submodule for identifying a plurality of interest points from the query text;

the text segmentation determining sub-module is used for identifying a plurality of sentence segments corresponding to a single place from the query text according to the plurality of interest points and identifying the sequence of the places corresponding to the plurality of sentence segments; wherein each sentence segment contains at least one point of interest;

and the subsection data generation submodule is used for generating a plurality of subsection data corresponding to a single place by adopting the sequences of the plurality of sentence subsections and the corresponding places.

Optionally, the segmented data identification module further comprises:

an anchor word recognition sub-module for recognizing anchor words from the query text;

the text segment determination sub-module includes:

an associated interest point determining unit, configured to determine an interest point associated with the anchor word;

and the text segment identification unit is used for identifying a plurality of sentence segments corresponding to a single place from the query text by adopting the anchor words and the associated interest points.

Optionally, the segmented data generation sub-module includes:

an entity type identification unit, which is used for identifying a plurality of entity types matched with interest points in the plurality of sentence segments;

a single-site segmentation data generation unit, configured to generate a plurality of segmentation data corresponding to a single site by using the interest points in the plurality of sentence segments, the plurality of entity types, the anchor words in the plurality of sentence segments, and the sequence of the plurality of corresponding sites.

Optionally, the single-site segmented data generating unit includes:

a single-site segmented data generation subunit, configured to store, according to a preset structure, the interest points in the multiple sentence segments, the multiple entity types, the anchor words in the multiple sentence segments, and the sequence of the multiple corresponding sites, so as to obtain multiple segmented data corresponding to a single site; the preset structure body comprises an interest point variable, an anchor word variable and a sequence variable.

Optionally, the method further comprises:

and the sequence adjusting module is used for adjusting the sequence of the corresponding places by adopting the anchor words.

Optionally, the order adjustment module includes:

and the sequence adjusting submodule is used for adjusting the sequence of the corresponding places in the segmented data by adopting the sequence relation between the anchor words in the segmented data and a preset anchor word.

Optionally, the single-place semantic data generating module includes:

the segmentation data acquisition sub-module is used for acquiring the sequence of a plurality of corresponding places, a plurality of interest points and a plurality of entity types matched with the interest points in the segmentation data of the plurality of corresponding single places;

the interest point semantic data generating submodule is used for generating a plurality of interest point semantic data aiming at the corresponding places by adopting the plurality of interest points;

the semantic tag determining submodule is used for determining a plurality of semantic tags aiming at the interest points by adopting the entity types matched with the interest points;

and the single-place semantic data generation submodule is used for generating a plurality of single-place semantic data by adopting the interest point semantic data aiming at the corresponding places, the sequence of the corresponding places and the semantic labels aiming at the corresponding places.

Optionally, the method further comprises:

the route calculation preference semantic data identification module is used for identifying route calculation preference semantic data according to the query text;

the navigation route generation module includes:

the semantic data merging submodule is used for merging the multi-place semantic data and the calculation path preference semantic data;

and the navigation route generation sub-module is used for generating a navigation route comprising a plurality of places by adopting the merged multi-place semantic data and route calculation preference semantic data.

Optionally, the multi-place semantic data comprises a plurality of point of interest semantic data for a plurality of places, an order of the plurality of places, and a plurality of semantic tags for the places; the navigation route generation module includes:

the second target navigation location searching sub-module is used for searching a plurality of matched target navigation locations from preset navigation locations by adopting the semantic data of the interest points and the semantic labels aiming at the locations;

the second route and point determining submodule is used for determining a starting point, a passing point and an end point by adopting the target navigation points and the sequence of the points;

the navigation planning strategy determining sub-module is used for determining a navigation planning strategy by adopting the route calculation preference semantic data;

and the second navigation route generation sub-module is used for generating the navigation route by adopting the starting point, the passing point and the end point of the navigation route and the navigation planning strategy.

the user voice acquisition module is used for acquiring the voice of at least one user;

the navigation request generating module is used for generating a navigation request by adopting the voice of the at least one user;

the navigation request sending module is used for sending the navigation request to a server so that the server can identify the query text of which the user intends to be a multi-place navigation intention from the navigation request; identifying from the query text, a plurality of segmented data corresponding to a single location; analyzing the segmented data of the plurality of corresponding single sites to generate a plurality of single site semantic data; generating multi-place semantic data by adopting the single-place semantic data; and generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

Optionally, the navigation request generating module includes:

the voice recognition submodule is used for respectively carrying out voice recognition on the voices of a plurality of users to obtain corresponding voice recognition results;

the navigation result determining submodule is used for performing semantic analysis on the voice recognition results corresponding to the users and determining the voice recognition result related to navigation;

and the navigation request generation sub-module is used for generating a navigation request according to the voice recognition result related to navigation.

The embodiment of the invention also discloses a vehicle, which comprises: a processor, a memory and a computer program stored on the memory and capable of running on the processor, which computer program, when executed by the processor, carries out the steps of the navigation method as described above.

The embodiment of the invention also discloses a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and the computer program realizes the steps of the navigation method when being executed by a processor.

The embodiment of the invention has the following advantages:

the embodiment of the invention receives the navigation request of a user; according to the navigation request, the query text which is intended by the user to be a multi-place navigation intention can be identified from the navigation request; identifying a plurality of segmented data corresponding to a single place from the query text, analyzing the segmented data corresponding to the single place, and generating a plurality of single-place semantic data; generating multi-place semantic data by adopting a plurality of single-place semantic data; and then generating a navigation route comprising a plurality of places by adopting the multi-place semantic data. By adopting the navigation method provided by the embodiment of the invention, the navigation route comprising the plurality of places can be generated according to the condition that the user continuously speaks the plurality of places without speaking the plurality of places at intervals, so that the use process of the user is simplified, and the user experience is improved.

Drawings

FIG. 1 is a flowchart illustrating steps of a first embodiment of a navigation method according to the present invention;

FIG. 2 is a flow diagram of identifying multiple sentence segments from query text in an embodiment of the invention;

FIG. 3 is a flowchart illustrating steps of a second embodiment of a navigation method;

FIG. 4 is a flowchart illustrating the steps of a third embodiment of a navigation method;

FIG. 5 is a block diagram of a navigation device according to a first embodiment of the present invention;

fig. 6 is a block diagram of a second embodiment of a navigation device according to the present invention.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

In an existing navigation method, a user speaks voice containing a place to be visited, a navigation terminal (such as a mobile terminal or a vehicle-mounted terminal) can collect the voice of the user, then a navigation request is generated according to the voice of the user, the navigation request is sent to a navigation server, and a navigation route is generated by the navigation server according to the navigation request. However, the existing navigation server can only identify a place in the navigation request and then generate a navigation route for the place.

When a user wants to go to a plurality of places in sequence, the user needs to speak voices of the plurality of places at intervals, the navigation terminal can generate a plurality of navigation requests by the plurality of voices, and the navigation server generates a plurality of navigation routes according to the plurality of navigation requests.

In contrast, the embodiment of the invention provides a navigation method combined with semantic recognition, which can recognize a query text which is intended by a user as a multi-place navigation intention from a navigation request; identifying a plurality of segmented data corresponding to a single place from the query text, analyzing the segmented data corresponding to the single place, and generating a plurality of single-place semantic data; generating multi-place semantic data by adopting a plurality of single-place semantic data; and then generating a navigation route comprising a plurality of places by adopting the multi-place semantic data. In the case where the user speaks a plurality of places in succession, the navigation system can generate a navigation route including a plurality of places.

Referring to fig. 1, a flowchart illustrating steps of a first embodiment of a navigation method according to the present invention is shown, which may specifically include the following steps:

step 101, receiving a navigation request of a user.

In the actual navigation process, the user sends out voice containing the place to be visited, and the navigation terminal can collect the voice of the user and generate a navigation request according to the voice of the user. The navigation terminal may transmit the navigation request to the navigation server, and the navigation server receives the navigation request.

In an example, the navigation request may be a voice of the user directly as the navigation request, for example, the voice of the user is "hello little P, i want to go to the five mile bridge", after the voice is collected by the navigation terminal, the voice is sent to the navigation server as the navigation request, and the navigation server invokes the recognition engine to recognize the voice, so as to obtain the query text intended by the user as the multi-location navigation intention.

In another example, the navigation terminal may collect a user voice, recognize the voice to obtain a recognition text, and then generate a navigation request using the recognition text.

Step 102, identify from the navigation request a query text intended by the user as a multi-location navigation intent.

The multi-place navigation intention may be a user intention indicating that it is desired to sequentially navigate to a plurality of places, or may be a user intention indicating that a navigation route for navigating to a plurality of places is requested to be generated.

The identification of the navigation request may be performed by an identification engine. If the navigation request is speech, the recognition engine may recognize from the speech a query text that the user intends to be a multi-location navigation intent. If the navigation request is recognition text, the recognition engine may recognize the query text from the recognition text that the user intends to be a multi-location navigation intent.

By identifying a query text for which a user intends to navigate to multiple locations, it may be determined that semantic data for multiple locations needs to be identified when identifying semantic data for the query text.

In one example, the recognition engine can recognize all content in the navigation request, and then recognize whether the corresponding user intention is a multi-location navigation intention, and if so, take all the recognized content as the query text.

In another example, the navigation request may contain a lot of content, some may be navigation related, some may be navigation unrelated, and the navigation related content may not be continuous and may be separated by the navigation unrelated content. The identification engine can identify the navigation-related content in the navigation request during identification, then identify whether the corresponding user intention is a multi-location navigation intention, and if so, take the identified navigation-related content as a query text.

For example, the speech of the navigation request is: ' go to the palace to visit, then go to the southern gong-drum lane to eat, go to the postsea park bar street after eating. The ' wherein the navigation-related content comprises ' go to the home palace ', ' go to the south gong and drum lane ', ' go to the back sea park bar street ', and if the corresponding user intention is the multi-place navigation intention wanting to go to a plurality of places, the obtained query text is ' go to the home palace, go to the south gong and drum lane, and go to the back sea park bar street '.

Step 103, identifying a plurality of segment data corresponding to a single place from the query text;

query text recognition may be employed by the recognition engine to derive a plurality of segmented data corresponding to a single location. The recognition engine can be obtained by collecting a large number of query texts of users and performing machine learning training by using the query texts as training samples.

The query text may contain content corresponding to a plurality of places, and the segmented data may be data generated using content corresponding to a single place.

In the embodiment of the present invention, step 103 may further include the following sub-steps:

a substep S11 of identifying a plurality of points of interest from the query text;

the query text may be segmented using recognition engine recognition to obtain a plurality of terms and labels corresponding to each term. The tags may include a point of interest tag, a word with the point of interest tag may be referred to as a point of interest, and a point of interest (poi) (point of interest) is a term in a geographic information system, and generally refers to all geographic objects that can be abstracted as a point.

It should be noted that the interest points may be classified according to the administrative region level, for example, the administrative region of china is divided into: province-city-district (county) -street (township), one place may correspond to a plurality of interest points of different levels.

For example, the locations are: at Beijing university, the corresponding points of interest may include: beijing, Hai lake, Yihe Yuanlu, three points of interest.

A substep S12, identifying a plurality of sentence segments corresponding to a single place from the query text according to the plurality of interest points, and identifying the order of the places corresponding to the plurality of sentence segments; wherein each sentence segment contains at least one point of interest;

the sentence segments can be identified from at least one point of interest corresponding to the same place.

For example, the query text is "great airport at Beijing university", wherein the points of interest include "great airport" and "Beijing university", and the "great airport" are segmented according to two sentences obtained from the two points of interest. It should be noted that if there is only one interest point in a certain location in the query text, the sentence segmentation may be just one interest point.

The sequence of identifying the places corresponding to the sentence segments from the query text refers to the sequence obtained by sequencing the sentence segments according to the positions of the sentence segments in the query text.

For example, the sentence segments "Beijing university" and "Daxing airport" are sorted according to their positions in the query text "Beijing university Daxing airport", resulting in a sequence of 1 for the sentence segment "Beijing university" and 2 for the sentence segment "Daxing airport". Therefore, the sentence segment "Beijing university" corresponds to the location: the sequence of Beijing university is 1, the sentence is segmented into the corresponding places of the Daxing airport: the grand airport order is 2.

In this embodiment of the present invention, step 103 may further include: identifying anchor words from the query text;

the recognition engine may recognize a plurality of terms in the query text, and a label corresponding to each term. The tags include anchor word tags, and words with anchor word tags may be referred to as anchor words.

The anchor word is a word that anchors a point of interest and can indicate a meaning of arriving somewhere. For example, "navigate to," "then go," "last go," "route," and so on.

In the sub-step S12, the step of identifying a plurality of sentence fragments corresponding to a single place from the query text according to the plurality of interest points may further include:

determining a point of interest associated with the anchor word; a plurality of sentence segments corresponding to a single place are identified from the query text using the anchor word and the associated points of interest.

According to the Chinese grammar rule, in the query text, the next or a plurality of continuous interest point words after the anchor word are the interest points associated with the anchor word.

The anchor word and associated point of interest may indicate a meaning of reaching a location, and thus multiple sentence segments corresponding to a single location may be identified from the query text using the anchor word and associated point of interest.

For example, the query text is "go to the home house then go to the south gong-drum lane then go to the back sea park bar street", wherein the anchor words include "go", "go then", "go again", and the points of interest include "home house", "south gong-drum lane", "back sea park bar street".

The interest point associated with the anchor word "go" is "the home position", the interest point associated with the anchor word "go" is "the south gong and drum lane", and the interest point associated with the anchor word "go" is "the back sea park bar street".

Therefore, the plurality of recognized sentences are segmented into 'home location removal', 'then south gong and drum lane', 'then back sea park bar street', 'the location corresponding to the home location removal' is the home location ',' then south gong and drum lane 'is the south gong and drum lane', and 'then back sea park bar street' is the back sea park bar street.

In order to make the skilled person better understand how to identify a plurality of sentence fragments from the query text, the following is illustrated by a specific example. Referring to FIG. 2, a flow diagram for identifying multiple sentence segments from query text in an embodiment of the present invention is shown.

In particular implementations, the terms of the query text may be stored in an array. For example, the query text "go to the home palace, then go to the south gong-drum lane, then go to the post sea park bar street" is participled to obtain the words: go to the palace/then go/south gong-drum lane/go/back sea park bar street, store into the array as the element of array respectively.

In segmenting the query text, the flow in FIG. 3 is performed from the first element of the array.

It is determined whether there is a next word.

And if the next word exists, acquiring the next word as the current word and acquiring the label of the current word.

If no next word exists, the segmentation process is ended.

For the current words of different labels, different processing modes are respectively adopted:

and if the label of the current word is the anchor word, newly building a segmented structure body as the current segmented structure body, adding the anchor word into the current segmented structure body, and then returning to the step of judging whether the next word exists.

And if the label of the current word is neither the interest point nor the anchor word, not adding the current word to the current segmented structure body, and returning to the step of judging whether the next word exists.

And if the label of the current word is the interest point, judging whether an anchor word exists before the current word.

And if the anchor word does not exist before the current word, not adding the current word to the current segmented structure body, and returning to the step of judging whether the next word exists.

If the anchor word exists before the current word, whether an interest point exists between the current word and the nearest anchor word is judged.

And if no interest point exists between the current word and the nearest anchor word, adding the current word to the current segmented structure, and returning to the step of judging whether the next word exists.

And if the interest point exists between the current word and the nearest anchor word, judging whether the region grades of the interest point between the current word and the nearest anchor word are the same.

And if the region grades of the interest points between the current word and the nearest anchor word are the same, newly building a segmented structural body as the current segmented structural body, adding the nearest anchor word and the current word to the current segmented structural body, and returning to the step of judging whether the next word exists.

And if the region grades of the interest points between the current word and the nearest anchor word are different, adding the current word to the current segmented structure, and returning to the step of judging whether the next word exists.

By the above example, it may be achieved that multiple sentence segments for a single place are identified from the query text according to the labels of the terms.

And a substep S13 of generating a plurality of pieces of segment data corresponding to a single place by using the plurality of sentence segments and the order of the corresponding places.

The sentence segments include the interest points corresponding to the same place, and the segment data corresponding to a single place can be generated by adopting the interest points of the same place and the sequence of the corresponding places.

In an embodiment of the present invention, the sub-step S13 may further include the following sub-steps:

substep S131, identifying a plurality of entity types matching the interest points in the plurality of sentence segments;

the entity type refers to the classification of the entity, and the entity type can be configured according to actual needs.

In the embodiment of the present invention, the entity types for the point of interest may include: location name, location address, administrative area, location type, common address, etc.

For example, the points of interest "Beijing City", "Guangzhou city Tianhe district", "Haihe district", and the matching entity type is administrative region.

The interest points are Beijing university and Wanda square, and the matched entity types are place names.

The interest point 'XX number of the middle street and the village street', and the matched entity type is a place address.

The interest points are scenic spots, banks and gas stations, and the matched entity types are location types.

The type of the matched entity of the interest points of 'home' and 'company' is a common address.

And a substep S132 of generating a plurality of segmented data corresponding to a single location by adopting the sequence of the interest points, the entity types, the anchor words and the corresponding locations in the sentence segments.

For each statement segment, generating segment data by adopting the interest points in the statement segment, the entity types corresponding to the interest points, the anchor words and the sequence of the corresponding places.

Specifically, the interest points in the statement segments, the entity types, the anchor words in the statement segments and the sequence of the corresponding places are adopted and stored according to a preset structure body, so that the segment data of the corresponding single places are obtained; the preset structure body comprises an interest point variable, an anchor word variable and a sequence variable.

The interest points and the entity types may be stored as elements of interest point variables, the anchor words as elements of anchor word variables, and the order of the plurality of corresponding locations as elements of order variables.

For example, the preset structure may be:

"order variable": { "order of places" }

"anchor word variable": { "Anchorage word" }

"point of interest variable": { "Point of interest", "entity type" }

Step 104, analyzing a plurality of segmented data corresponding to a single place to generate a plurality of single-place semantic data;

single-place semantic data refers to semantic data for a single place that is stored in a particular data format, which may include slot positions and slot values for a single place.

In the field of natural language processing, a slot is understood to be a well-defined attribute of an entity, which is an object having a specific meaning in a corpus.

Slot filling is the value that determines the slot position, i.e., the value that extracts the well-defined property of a given entity from the corpus.

For example, for the phrase "departure to airport now," the dialogue semantic representation can be obtained such that the domain is navigation, the intent is to go to airport, two slots are obtained for the departure time and the destination, the slot value for the slot at the departure time is "now," and the slot value for the slot at the destination is "airport.

In an embodiment of the present invention, step 104 may further include the following sub-steps:

a substep S21 of obtaining an order of a plurality of corresponding places, a plurality of interest points, and a plurality of entity types matching the interest points in the segmented data of the plurality of corresponding single places;

a substep S22, generating a plurality of interest point semantic data aiming at the corresponding places by adopting a plurality of interest points;

the point of interest semantic data refers to semantic data for a single point of interest stored in a specific data format, and the point of interest semantic data may include a slot position and a slot value for the single point of interest.

For example, a well-defined attribute of the entity with the interest point being "hai lake region" and the "hai lake region" being the administrative region, so that an administrative region semantic data can be generated, the slot name being "administrative region" and the slot value being "hai lake region".

An example of point of interest semantic data may be:

{

"name": administrative region "

"value": 'Haihe district'

}

Wherein value is the slot value and name is the slot name.

In one example, the slot value may include original text and normalized text.

Original text refers to text that is directly derived from the identified query text. The normalization text refers to a text obtained by normalizing the original text.

In practice, different people call different points of interest, and some calls may not be unified, so that the navigation server cannot find the navigation location. In order to enable the navigation server to more accurately search the matched navigation location, the original text can be normalized by the recognition engine to obtain a normalized text, and both the original text and the normalized text are used as a slot value.

The meaning of the original text in the sentence is the same as or similar to the meaning of the normalized text in the sentence.

For example, for "home", "place of home", "my home", "home" text, the corresponding normalized text may be "home".

For "company", "place to work", the corresponding normalized text may be "company".

An example of point of interest semantic data may be:

{

"name": "Point of interest name"

"value": 'Jia'

"rawvalue": 'my family'

}

Wherein, value is the slot value of the original text, and rawvalue is the slot value of the normalized text.

And a substep S23, determining a plurality of semantic tags for the point of interest by using the entity types matched by the point of interest.

The semantic tags are tags capable of describing types of the interest points, and can distinguish the types of the interest points, so that the navigation server can optimize map search.

For example, the entity type is "common address", and the semantic tag may be set to "common address".

The entity type is 'chain brand, multi-branch organization', and the semantic label can be set as 'location type'.

The entity type is address, and the semantic tag can be set as interest point name.

The entity type is 'administrative region', and the semantic tag can be set as 'interest point name'.

And a substep S24 of generating a plurality of single-place semantic data using the plurality of interest point semantic data for the corresponding places, the order of the plurality of corresponding places, and the plurality of semantic tags for the corresponding places.

And combining the interest point semantic data, the sequence of the corresponding places and the semantic labels of the corresponding places aiming at the same corresponding place to obtain single-place semantic data. The slot value of the single-place semantic data can be interest point semantic data.

For example, an example of single-place semantic data may be:

wherein index is the sequence number of the corresponding place, slot _ tag is the semantic label, and the slot value of the single-place semantic data is the interest point semantic data.

In embodiments of the present invention, segmented data may include anchor words; before performing the sub-step S24, the following steps may be further included:

and adjusting the sequence of the corresponding positions by adopting a plurality of anchor words.

The anchor may represent a meaning of arriving somewhere. And anchor words may also indicate the order of arrival between multiple locations, with different anchor words indicating different meanings for the order of arrival at a location.

Specifically, the step of adjusting the sequence of the plurality of corresponding locations by using the plurality of anchor words may be: and adjusting the sequence of a plurality of corresponding places in the plurality of segmented data by adopting the sequence relation between the anchor words in the plurality of segmented data and the preset anchor words.

In the embodiment of the invention, the precedence order relation implied by the anchor words can be obtained by analyzing the anchor words in a large number of query texts.

For example, for the four anchor words "go first", "go then", "navigate to", "go last", the implicit precedence relationship between them may be: "go first" < "then go" < "navigate to" < "go last". That is, the order of "going first" to the corresponding location precedes the order of "then going" to the corresponding location, the order of "then going" to the corresponding location precedes the order of "navigating to" the corresponding location, and the order of "navigating to" the corresponding location precedes the order of "last going" to the corresponding location.

For the sequence of the corresponding locations in the segmented data, the sequence of each corresponding location can be adjusted by using the anchor words, and then the sequence after the sequence is adjusted is adopted to generate the single-location semantic data.

For example, the query statement is "go to Beijing amusement park first, then to Beijing university and then to Daxing airport", and the sequence of the corresponding places obtained according to the segmented statement may be: the sequence of Beijing amusement park is 1, the sequence of Beijing university is 2, and the sequence of Daxing airport is 3. The data line of the corresponding place after adjustment according to the anchor word may be: the sequence of Beijing amusement park is 1, the sequence of Beijing university is 3, and the sequence of Daxing airport is 2.

In practice, the anchor word in the segmented data may not be the same as the anchor word that implicitly contains the corresponding locality order as determined in advance, but the actual meaning may be the same. Therefore, before the anchor words in the segmented data are adopted and the sequence of the corresponding places is adjusted, normalization processing can be performed on the anchor words to obtain the normalized anchor words which implicitly contain the sequence of the corresponding places.

For example, for "navigate to", "go to", the corresponding normalized anchor word may be "navigate to".

For "go next", "go back", and "go back", the corresponding normalized anchor word may be "go then".

For "last go", "destination", the corresponding normalized anchor may be "last go".

For "go-halfway" and "way", the corresponding normalized anchor may be "way".

For "go first", and "go first", the corresponding normalized anchor may be "go first".

And 105, generating multi-place semantic data by adopting a plurality of single-place semantic data.

The multi-place semantic data refers to semantic data which is stored according to a preset data format and aims at a plurality of places, and the preset data format can be matched with an interface provided by a navigation server, so that parameters in the multi-place semantic data can be adaptively transmitted to the interface provided by the navigation server.

The multi-place semantic data may include slot positions and slot values for multiple places.

Multiple single-place semantic data may be combined to obtain multi-place semantic data.

And 106, generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

Specifically, the navigation server may generate a navigation route including a plurality of locations by using the multi-location semantic data. The multi-point navigation route may be a navigation route including a start point, a pass point, and an end point.

In an embodiment of the present invention, the multi-place semantic data may include a plurality of single-place semantic data corresponding to a single place, the single-place semantic data may further include one or more interest point semantic data corresponding to a single place, an order of the corresponding places, and a plurality of place-specific semantic tags, and step 106 may include:

a substep S31, adopting a plurality of interest point semantic data and a plurality of semantic labels aiming at the places to search a plurality of matched target navigation places from the preset navigation places;

in practice, a plurality of navigation locations are preset in the navigation server, and the navigation locations may be collected and sorted in advance by a service provider of the navigation server.

The semantic data of the interest points can comprise slot positions and slot values, the semantic tags can represent the types of the interest points, and the navigation server can search a plurality of matched target navigation places from preset navigation places by adopting the semantic tags and the slot values of the semantic data of the interest points.

A substep S32 of determining a start point, a pass-through point and an end point using a plurality of target navigation points and an order of the plurality of points;

and determining a starting point, a passing point and an end point of the navigation route by adopting the matched target navigation positions and the sequence of each position.

In sub-step S33, a navigation route is generated using the start point, the route point, and the end point.

The method is different from the existing method for generating the navigation route through the sentence pattern template, the embodiment of the invention identifies a plurality of segmented data corresponding to a single place from the query text, analyzes the segmented data to generate single-place semantic data, then adopts the plurality of single-place semantic data to generate multi-place semantic data, and can analyze the multi-place semantic data from the query text for the query texts of users with different language habits. The embodiment of the invention can not generate a correct navigation route due to the failure of sentence pattern template matching, and does not need to maintain a large number of sentence pattern templates.

Referring to fig. 3, a flowchart illustrating steps of a second embodiment of the navigation method of the present invention is shown, which may specifically include the following steps:

step 301, receiving a navigation request of a user.

The navigation terminal can collect the voice of the user, generate a navigation request by adopting the voice of the user and send the navigation request to the navigation server.

For example, an example of a query request may be:

' query ' navigation to Beijing university, then to Beijing amusement park, then to capital airport, and finally to go home to go unimpeded route ' }

Step 302, identify from the navigation request a query text intended by the user as a multi-location navigation intent.

The navigation server may invoke the recognition engine to recognize from the navigation request the query text intended by the user as a multi-location navigation intent. The recognition engine can be arranged in the navigation server or a special recognition server.

Step 303, identifying a plurality of segment data corresponding to a single place from the query text;

for specific implementation, reference may be made to the above embodiments, which are not described herein again.

Step 304, analyzing a plurality of segmented data corresponding to a single location to generate a plurality of single location semantic data;

305, generating multi-place semantic data by adopting a plurality of single-place semantic data;

And step 306, identifying route preference semantic data according to the query text.

The route calculation preference semantic data refers to semantic data which is stored according to a preset data format and aims at route calculation preference, and the route calculation preference semantic data can comprise slot positions and slot values of the route calculation preference.

The route calculation preference refers to a policy for planning a navigation route. For example, congestion avoidance, high speed priority, no speed, low cost, etc.

The user may also speak route preference when speaking which locations to go to. The navigation request may therefore contain the content of the route calculation preference, and the recognition engine may recognize the route calculation preference semantic data from the query text recognized from the navigation request.

In an embodiment of the present invention, step 306 may include the following sub-steps:

a substep S41 of identifying a routing preference from the query text;

the query text may be segmented by the recognition engine to obtain a plurality of terms and labels corresponding to each term. The labels may include route calculation preference labels, and the words with the route calculation preference labels may be referred to as route calculation preferences.

And a substep S42 of generating the road preference semantic data by using the road preference.

The way calculation preference may be employed to generate way calculation preference semantic data and determine a slot value for the way calculation preference slot, and then the way calculation preference slot and the slot value may be employed to generate way calculation preference semantic data.

For example, the entity of "avoid charging" and "avoid charging" as a well-defined attribute is the route calculation preference, so that a route calculation preference semantic data can be generated, the name of the slot is "route calculation preference", and the value of the slot is "avoid charging".

An example of the way calculation preference semantic data may be:

{

"name": "calculate road preference"

"value": avoiding congestion "

}

In one example, computing the slot value of the route preference semantic data may include: original text and normalized text.

For example, for the texts "avoid toll station", "no charge", "free route", the corresponding normalized text may be "avoid charge".

For "speed of light", "speed of high", first speed of high ", and" best speed of high ", the corresponding normalized text may be" high speed first ".

For "avoid congestion", and "not traffic jam", the corresponding normalized text may be "avoid congestion".

For "unlimited," avoided, unlimited, "no-go, and no-go," the corresponding normalized text may be "evasive.

An example of the way calculation preference semantic data may be:

{

"name": "calculate road preference"

"value": avoiding congestion "

"rawvalue": unblocked "

}

Step 307, merging the multi-place semantic data and the road preference semantic data.

The recognition engine may combine the multi-place semantic data and the route calculation preference semantic data into one total semantic data, and then output the total semantic data to the navigation server.

And 308, generating a navigation route comprising a plurality of places by adopting the merged multi-place semantic data and the route calculation preference semantic data.

The navigation server may generate a navigation route including a plurality of places using the merged total semantic data.

For example, an example of total semantic data may be:

{"data":[

{ "name": route preference ', "raw value": clear', "value": avoid congestion ",

{ "name": location "," raw value ": null", "value": 2

{ "index":0, "slot _ tag": poi _ name "," value ": [ {" name ": passby _ name", "rawvalue": Beijing university "," value ": Beijing university" } ],

{ "index":1, "slot _ tag": poi _ name "," value ": [ {" name ": passby _ name", "rawvalue": Beijing amusement park "," value ": Beijing amusement park" },

{ "index":2, "slot _ tag": poi _ name "," value ": [ {" name ": passby _ name", "rawvalue": first airport "," value ": first airport" },

{ "index":3, "slot _ tag": custom _ address "," value "[ {" name ": custom _ name", "raw value": home "," value ": home" }

]}]]

In an embodiment of the present invention, the multi-place semantic data includes a plurality of interest point semantic data for a plurality of places, an order of the plurality of places, and a plurality of place-specific semantic tags, and step 308 may include the sub-steps of:

a substep S51, adopting a plurality of interest point semantic data and a plurality of semantic labels aiming at the places to search a plurality of matched target navigation places from the preset navigation places;

A substep S52 of determining a start point, a pass-through point and an end point using a plurality of target navigation points and an order of the plurality of points;

and determining a starting point, a passing point and an end point of the navigation route by adopting a plurality of target navigation positions and the sequence of each position.

A substep S53 of determining a navigation planning strategy by using the road calculation preference semantic data;

the route preference semantic data may include slot positions and slot values, and the slot positions and slot values therein may be used to determine a planning strategy for the navigation route.

And a substep S54 of generating a navigation route by using the starting point, the passing point and the end point of the navigation route and the navigation planning strategy.

The navigation server may invoke an interface for generating the navigation route by using the start point, the pass point, and the end point of the navigation route and the navigation planning policy, thereby generating a specific navigation route.

The embodiment of the invention receives the navigation request of a user; identifying, from the navigation request, a query text for which the user intends to be a multi-location navigation intent; and then according to the query text, recognizing the multi-place semantic data and the route calculation preference semantic data, combining the multi-place semantic data and the route calculation preference semantic data, and finally generating a navigation route comprising a plurality of places by adopting the combined multi-place semantic data and the route calculation preference semantic data. By adopting the navigation method provided by the embodiment of the invention, the navigation route comprising a plurality of places can be generated according to the situation that the user continuously speaks a plurality of places and route calculation preferences, the user does not need to speak a plurality of places and route calculation preferences at intervals, the use process of the user is simplified, and the user experience is improved.

Referring to fig. 4, a flowchart illustrating steps of a third embodiment of the navigation method of the present invention is shown, which may specifically include the following steps:

step 401, collecting voice of at least one user;

the embodiment of the invention explains the navigation method from the perspective of the navigation terminal.

The navigation terminal may collect voice of at least one user.

In practice, when a user uses the navigation terminal on a car, the car may have multiple users at the same time, and the navigation terminal may collect voices of the multiple users.

Step 402, generating a navigation request by adopting the voice of at least one user;

the navigation terminal may generate the navigation request using voice of at least one user.

In this embodiment of the present invention, step 402 may include: respectively carrying out voice recognition on voices of a plurality of users to obtain corresponding voice recognition results; performing semantic analysis on voice recognition results corresponding to a plurality of users, and determining a voice recognition result related to navigation; and generating a navigation request according to the voice recognition result related to navigation.

For example, if there are four users in the car, user a, user B, user C and user D,

performing voice recognition on the voice of the user A, wherein the obtained voice recognition result is as follows: "go high speed to site a"; and performing voice recognition on the voice of the user B, wherein the obtained voice recognition result is as follows: "go to site b after reaching site a"; and performing voice recognition on the voice of the user C, wherein the obtained voice recognition result is as follows: "to go to toilet at site b"; and performing voice recognition on the voice of the user D, wherein the obtained voice recognition result is as follows: "last go to location c.

And performing semantic analysis on each voice recognition structure, determining that the voice recognition results of the user A, the user B and the user D are related to navigation, and then generating a navigation request by adopting the voice recognition results of the user A, the user B and the user D.

Further, the step of generating a navigation request according to the speech recognition result related to navigation may include:

determining positions, the sequence of the positions and route calculation preference in the voice recognition result related to navigation; and generating a navigation request by adopting the place, the order of the places and the road calculation preference.

For example, the location of the voice recognition result of the user a is recognized as "location a", the road preference is calculated as "speed of light walking", and the order of the locations a is 1; recognizing the place of the voice recognition result of the user B as a place B, wherein the sequence of the place B is 2; the location of the voice recognition result of the recognition user D is "location c", and the order of the locations c is 3. The following are obtained from the speech recognition results: a navigation request is generated with a point a, a point a in the order of 1, a high speed, a point b in the order of 2, a point c, and a point c in the order of 3.

The navigation terminal collects voices of the three users speaking in sequence, the navigation terminal can splice the voices of the three users, and then the spliced voices are used for generating the navigation request. The navigation request may include the content that the user says in order.

Step 403, sending the navigation request to a server so that the server identifies the query text which is intended by the user as a multi-place navigation intention from the navigation request; identifying a plurality of segmented data corresponding to a single location from the query text; analyzing a plurality of segmented data corresponding to a single place to generate a plurality of single-place semantic data; generating multi-place semantic data by adopting a plurality of single-place semantic data; and generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

The embodiment of the invention adopts the voice of at least one user to generate a navigation request by collecting the voice of at least one user; sending the navigation request to a server so that the server identifies a query text which is intended by the user to be a multi-place navigation intention from the navigation request; identifying a plurality of segmented data corresponding to a single place from the query text, analyzing the segmented data corresponding to the single place, and generating a plurality of single-place semantic data; generating multi-place semantic data by adopting a plurality of single-place semantic data; and then generating a navigation route comprising a plurality of places by adopting the multi-place semantic data. By adopting the navigation method provided by the embodiment of the invention, the navigation route comprising the plurality of places can be generated according to the condition that the user continuously speaks the plurality of places without speaking the plurality of places at intervals, so that the use process of the user is simplified, and the user experience is improved.

It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.

Referring to fig. 5, a block diagram of a first embodiment of a navigation device according to the present invention is shown, which may specifically include the following modules:

a navigation request receiving module 501, configured to receive a navigation request of a user;

a query text recognition module 502 for recognizing a query text intended by the user as a multi-location navigation intention from the navigation request;

a segment data identification module 503 for identifying a plurality of segment data corresponding to a single location from the query text;

a single-site semantic data generation module 504, configured to analyze multiple pieces of segment data corresponding to a single site, and generate multiple pieces of single-site semantic data;

a multi-place semantic data generating module 505, configured to generate multi-place semantic data by using multiple single-place semantic data;

and a navigation route generating module 506, configured to generate a navigation route including multiple locations by using the multi-location semantic data.

In an embodiment of the present invention, the segmented data identifying module 503 may include:

and the segmentation data generation submodule is used for generating a plurality of segmentation data corresponding to a single place by adopting the sequence of the plurality of statement segments and the corresponding places.

In this embodiment of the present invention, the segmented data identifying module 503 may further include:

the text segment determination sub-module includes:

the related interest point determining unit is used for determining interest points related to the anchor words;

In an embodiment of the present invention, the segmentation data generation submodule may include:

an entity type identification unit for identifying a plurality of entity types matched with interest points in a plurality of sentence segments;

and the single-place segmentation data generation unit is used for generating a plurality of pieces of segmentation data corresponding to the single place by adopting the sequence of the interest points in the plurality of sentence segments, the plurality of entity types, the anchor words in the plurality of sentence segments and the plurality of corresponding places.

In an embodiment of the present invention, the single-site segmented data generating unit may include:

the single-site segmented data generation subunit is used for storing the interest points in the statement segments, the entity types, the anchor words in the statement segments and the sequence of the corresponding sites according to a preset structure body to obtain a plurality of segmented data corresponding to the single site; the preset structure body comprises an interest point variable, an anchor word variable and a sequence variable.

In an embodiment of the present invention, the navigation device may further include:

and the sequence adjusting module is used for adjusting the sequence of the corresponding places by adopting a plurality of anchor words.

In the embodiment of the present invention, the sequence adjusting module may include:

and the sequence adjusting submodule is used for adjusting the sequence of a plurality of corresponding places in the plurality of segmented data by adopting the sequence relation between the anchor words in the plurality of segmented data and the preset anchor words.

In this embodiment of the present invention, the single-place semantic data generating module 504 may include:

the segment data acquisition sub-module is used for acquiring the sequence of a plurality of corresponding places in the segment data of a plurality of corresponding single places, a plurality of interest points and a plurality of entity types matched with the interest points;

the interest point semantic data generation submodule is used for generating a plurality of interest point semantic data aiming at the corresponding places by adopting a plurality of interest points;

and the single-place semantic data generation submodule is used for generating a plurality of single-place semantic data by adopting a plurality of interest point semantic data aiming at the corresponding places, the sequence of the corresponding places and a plurality of semantic labels aiming at the corresponding places.

the navigation route generation module 506 may include:

the semantic data merging submodule is used for merging the multi-place semantic data and the route calculation preference semantic data;

and the navigation route generation sub-module is used for generating a navigation route comprising a plurality of places by adopting the merged multi-place semantic data and the route calculation preference semantic data.

In the embodiment of the invention, the multi-place semantic data comprises a plurality of interest point semantic data corresponding to a plurality of places, an order of the plurality of places and a plurality of place-specific semantic tags; the navigation route generation module 506 may include:

a first target navigation location searching sub-module for searching matched multiple target navigation locations from preset navigation locations by adopting multiple interest point semantic data and multiple location-specific semantic labels

The first route point determining submodule is used for determining a starting point, a passing point and an end point by adopting a plurality of target navigation points and the sequence of the plurality of points;

and the first navigation route generation submodule is used for generating a navigation route by adopting a starting point, a passing point and an end point.

In an embodiment of the present invention, the multi-place semantic data includes a plurality of interest point semantic data for a plurality of places, an order of the plurality of places, and a plurality of place-specific semantic tags; the navigation route generation module 506 may include:

the second target navigation location searching sub-module searches a plurality of matched target navigation locations from preset navigation locations by adopting a plurality of interest point semantic data and a plurality of location-specific semantic labels;

the second route and point determining submodule is used for determining a starting point, a passing point and an end point by adopting a plurality of target navigation points and the sequence of the plurality of points;

Referring to fig. 6, a block diagram of a second embodiment of the navigation device of the present invention is shown, which may specifically include the following modules:

a user voice collecting module 601, configured to collect voice of at least one user;

a navigation request generating module 602, configured to generate a navigation request by using voice of at least one user;

a navigation request sending module 603, configured to send a navigation request to a server, so that the server identifies, from the navigation request, a query text intended by a user as a multi-location navigation intention; identifying a plurality of segmented data corresponding to a single location from the query text; analyzing a plurality of segmented data corresponding to a single place to generate a plurality of single-place semantic data; generating multi-place semantic data by adopting a plurality of single-place semantic data; and generating a navigation route comprising a plurality of places by adopting the multi-place semantic data.

In an embodiment of the present invention, the navigation request generating module 602 may include:

For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.

An embodiment of the present invention further provides a vehicle, including:

the navigation method comprises a processor, a memory and a computer program which is stored in the memory and can run on the processor, wherein when the computer program is executed by the processor, the navigation method can realize the processes of the navigation method embodiment, and the same technical effect can be achieved.

The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when being executed by a processor, the computer program implements each process of the navigation method embodiment, and can achieve the same technical effect, and is not described herein again to avoid repetition.

The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.

The present invention provides a navigation method, a navigation device, a vehicle and a computer readable storage medium, which are described in detail above, and the principles and embodiments of the present invention are explained herein by applying specific examples, and the descriptions of the above examples are only used to help understand the method and the core ideas of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. A navigation method, comprising:

receiving a navigation request of a user;

2. The method of claim 1, wherein identifying from the query text segment data for a plurality of corresponding individual places comprises:

identifying a plurality of points of interest from the query text;

3. The method of claim 2, wherein the identifying from the query text the segmented data for a plurality of corresponding individual places further comprises:

identifying anchor words from the query text;

determining a point of interest associated with the anchor word;

4. The method of claim 3, wherein generating a plurality of segment data corresponding to a single location using the plurality of sentence segments and the order of corresponding locations comprises:

5. The method of claim 4, wherein generating segment data for a plurality of corresponding single locations using an order of points of interest in the plurality of sentence fragments, the plurality of entity types, anchor words in the plurality of sentence fragments, and the plurality of corresponding locations comprises:

6. The method of claim 3, further comprising:

7. The method of claim 6, wherein said adjusting the order of said plurality of said corresponding locations using said plurality of anchor words comprises:

8. The method of claim 1, wherein the parsing the segmented data for the plurality of corresponding single places to generate a plurality of single-place semantic data comprises:

9. The method of claim 1, further comprising:

identifying route preference semantic data according to the query text;

and generating a navigation route comprising a plurality of places by adopting the merged multi-place semantic data and the route calculation preference semantic data.

10. The method of claim 9, wherein the multi-place semantic data comprises a plurality of point of interest semantic data for a plurality of places, an order of the plurality of places, and a plurality of semantic tags for the places; the generating a navigation route including a plurality of places by using the merged multi-place semantic data and route calculation preference semantic data comprises:

11. A navigation method, comprising:

collecting voice of at least one user;

generating a navigation request by adopting the voice of the at least one user;

12. The method of claim 11, wherein generating a navigation request using the voice of the at least one user comprises:

13. A navigation device, comprising:

14. A navigation device, comprising:

15. A vehicle, characterized by comprising: processor, memory and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps of the navigation method according to any one of claims 1-10.

16. A vehicle, characterized by comprising: processor, memory and computer program stored on the memory and executable on the processor, which computer program, when being executed by the processor, carries out the steps of the navigation method according to claims 11-12.

17. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the navigation method according to any one of claims 1 to 10.

18. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, carries out the steps of the navigation method according to claims 11-12.