Car navigation system and voice guide control method

Abstract

To perform a route guide by voice guidance including language representation that is more easily recognized by a user.SOLUTION: According the present invention, a car navigation system provides a user with a route guide. The system includes: an object recognition part for recognizing an object in an image frame on the basis of imaging data; an attribute recognition part for recognizing attributes of the object, and calculating the degree of certainty of each attribute value of the attributes; a commonization part for commonizing the attribute value on the basis of the degree of certainty; and a voice guide control part for generating prescribed voice guidance according to position information of a vehicle. The commonization part includes: an attribute value extraction part for extracting two or more attribute values having the degree of certainty larger than a prescribed threshold among the recognized attribute values; and a common parameter determination part for determining a common parameter showing commonality among the two or more attribute values. The voice guide control part generates voice guidance related to the object on the basis of the common parameter.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a car navigation system and a voice guidance control method. [Background technology]

[0002] Car navigation systems that electronically guide the driver (user) to the destination while showing the current location of the vehicle are widely used. Car navigation systems generally provide voice guidance to the user while driving the vehicle at a timing appropriate to the driving situation, so as to keep the user's eyes on the road as much as possible.

[0003] For example, Patent Document 1 below discloses a navigation device and method that provides voice guidance at a timing that the user wants to check while driving a vehicle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2015-59811 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the above navigation device and method, when the user's utterance is determined to be a confirmation utterance, the target object at the guide point is determined, the time for the vehicle to reach the guide point is calculated, and the guidance voice content is generated according to the target object and the arrival time, which causes a delay in the user's confirmation. For example, when a guidance voice guide for a target object is given to a user who is driving, if the user wants to confirm the content, it is assumed that the vehicle will reach the target object during the conversation.

[0006] SUMMARY OF THE PRESENT DISCLOSURE An object of the present invention is to provide a car navigation system that can provide route guidance using voice guidance that includes language expressions that are easy for the user to understand.

[0007] More specifically, the present invention aims to provide a car navigation system that enables human-like guidance (HLG) in voice guidance that can be more intuitively and immediately understood by the user while using dynamic landmarks within the user's field of vision. [Means for solving the problem]

[0008] The present invention, which aims to solve the above problems, includes the following invention-specific matters and technical features.

[0009] The present invention according to one aspect is a car navigation system that provides route guidance to a user who drives a vehicle. The car navigation system includes an object recognition unit that recognizes an object in an image frame based on acquired imaging data, an attribute recognition unit that recognizes attributes of the recognized object and calculates a certainty factor of each of a plurality of attribute values ​​of the attributes, a commonization unit that commonizes the attribute values ​​based on the calculated certainty factor, and a voice guidance control unit that generates a predetermined voice guidance according to position information of the vehicle. The commonization unit includes an attribute value extraction unit that extracts at least two or more attribute values ​​from the recognized plurality of attribute values, the certainty factor of which is greater than a predetermined threshold value, and a common parameter determination unit that determines a common parameter indicating a commonality between the at least two or more extracted attribute values. The voice guidance control unit generates a voice guidance related to the object based on the determined common parameter.

[0010] The present invention also provides a voice control method for the car navigation system, a computer program for executing the method, and a recording medium for non-temporarily recording the computer program. Effect of the Invention

[0011] According to the present invention, route guidance can be provided by voice guidance including linguistic expressions that are easier for the user to recognize. In particular, according to the present invention, it is possible to provide human-like voice guidance including expressions that the user can intuitively and immediately understand while using dynamic landmarks within the user's field of view.

[0012] Other technical features, objects, and operational effects or advantages of the present invention will become apparent from the following embodiments described with reference to the accompanying drawings. The effects described in this specification are merely examples and are not limited thereto, and other effects may also be present. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing an example of a functional model of a car navigation system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a graph showing an example of the confidence level for attribute values ​​in the attributes of an object. [Diagram 3] FIG. 3 is a diagram showing an example of a data configuration of the attribute information database in the car navigation system according to one embodiment of the present invention. [Figure 4] FIG. 4 is a block diagram showing an example of a specific configuration of a common unit of a car navigation system according to an embodiment of the present invention. [Diagram 5] FIG. 5 is a block diagram showing an example of a specific configuration of a voice guidance control unit of a car navigation system according to an embodiment of the present invention. [Figure 6] FIG. 6 is a flowchart illustrating an example of an object attribute recognition process in the car navigation system according to an embodiment of the present invention. [Figure 7A] FIG. 7A is a flowchart illustrating an example of a route guidance process in the car navigation system according to one embodiment of the present invention. [Figure 7B] FIG. BA is a flowchart illustrating an example of a route guidance process in a car navigation system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. However, the embodiment described below is merely an example, and there is no intention to exclude various modifications and application of techniques not specified below. The present invention can be implemented by various modifications (for example, combining the respective embodiments) within the scope of the gist of the present invention. In addition, in the following description of the drawings, the same or similar parts are represented by the same or similar reference numerals. The drawings are schematic and do not necessarily correspond to actual dimensions, ratios, etc. The drawings may also include parts with different dimensional relationships and ratios.

[0015] 1 is a block diagram showing an example of a functional model of a car navigation system according to an embodiment of the present invention. As shown in the figure, the car navigation system 1 includes, for example, a location information acquisition unit 11, a route guidance control unit 12, a road map information database 13, a screen control unit 14, a voice guidance database 15, a voice guidance control unit 16, and a user interface unit 17. In addition, in this disclosure, the car navigation system 1 further includes an imaging data acquisition unit 21, an object recognition unit 22, an attribute recognition unit 23, an attribute information database 24, and a common unit 25.

[0016] The car navigation system 1 is a device that searches for a route (driving route) from the current position of the vehicle to a destination desired by the user (driver) based on road map information, displays a navigation screen on a display so that the vehicle can travel along the searched route, and provides voice guidance at predetermined guidance points. In the present disclosure, the car navigation system 1 can provide voice guidance (non-standard voice guidance) including expressions related to dynamic (moving) objects such as a vehicle in the user's field of view, as described later.

[0017] The position information acquisition unit 11 acquires the current position information of the vehicle. Typically, the position information acquisition unit 11 calculates the current position information of the vehicle based on a GPS signal received from a GPS (Global Positioning System) and output signals of various sensors (not shown) such as a mileage sensor and a gyro sensor mounted on the vehicle.

[0018] The route guidance control unit 12 controls the processing required for route guidance to the user while referring to the road map information database 13 based on the acquired position information. The route guidance control unit 12 may include a route search engine 122. The route search engine 122 searches for and proposes an optimal route from the current position to the destination based on the road map information database 13. The route guidance control unit 12 controls so that a navigation screen is displayed on the display of the user interface unit 17 according to the route searched and proposed by the route search engine 122. The navigation screen is, for example, a screen in which the vehicle and the proposed route are superimposed on a road map. In addition, the route guidance control unit 12 controls the voice guidance control unit 16 so that appropriate voice guidance is provided at multiple guidance points before approaching a passing target point through which the vehicle is scheduled to pass. In the present disclosure, the route guidance control unit 12 provides non-standard voice guidance at a specific guidance point among multiple guidance points, and provides standard voice guidance at guidance points other than the specific point. For example, the route guidance control unit 12 controls the voice guidance control unit 16 so that typical voice guidance and non-typical voice guidance are generated alternately depending on the distance to the passing target point. For example, the route guidance control unit 12 performs control so that typical voice guidance is generated 300 meters before an intersection where a right turn (or left turn) is planned, non-typical voice guidance is generated 150 meters, and typical voice guidance is generated 50 meters. This makes it possible to provide a varied voice guidance to a user approaching a passing target point. In particular, by providing non-typical voice guidance using dynamic objects such as vehicles in the voice guidance, the user can grasp multifaceted guidance information in the flow of traffic.

[0019] The road map information database 13 stores information about road maps (hereinafter referred to as "road map information"). The road map information may include information necessary for searching for a route to a destination, such as road network information, road-related information, and environmental information. The road network information is information consisting of nodes (intersections) and links (roads). The road-related information includes, for example, traffic regulations such as speed limits, one-way streets, no right or left turns, no parking or stopping, and road closures, road types such as community roads and main roads (including lanes, etc.), and information about control of traffic lights at intersections. In addition, the road-related information may include, for example, the situation between buildings and roads (for example, sidewalk width, the presence or absence of accessories such as guardrails and shrubbery, the degree of visibility, etc.). The environmental information includes, for example, the situation of parked and stopped vehicles and pedestrians at each point by time period, the frequency of right and left turns at intersections, the frequency of parking and stopping, and the frequency of traffic accidents.

[0020] Under the control of the route guidance control unit 12, the screen control unit 14 refers to the road map information database 13 and controls the user interface unit 17 so that an icon indicating the current position of the vehicle is displayed on the display of the user interface unit 17 together with a road map.

[0021] The voice guidance database 15 stores, for example, voice guidance data for providing voice guidance to a user on a route along which a vehicle should travel. In the present disclosure, the voice guidance database 15 includes a voice guidance template database 151 and a vocabulary database 152 (see FIG. 5). The voice guidance template database 151 stores voice guidance templates using predetermined reference targets such as intersections and junctions. In addition, the vocabulary database 152 is used to generate expressions that allow humans to intuitively understand dynamic objects such as people and vehicles, as will be described later.

[0022] The voice guidance control unit 16, under the control of the route guidance control unit 12, refers to the voice guidance database 15 and generates voice guidance for the user. In the present disclosure, as described later, the voice guidance control unit 16 generates either a standard voice guidance using a predefined reference target or an unstandard voice guidance using a dynamic object as a reference target.

[0023] The user interface unit 17 interactively provides route guidance information to the user. Although not shown, for example, the user interface unit 17 includes a display that displays a navigation screen based on a road map, an input panel that accepts user input operations, and a speaker that outputs voice guidance. The display may be configured with a touch panel that accepts user input operations. The user interface unit 17 may also include a microphone that accepts user voice input.

[0024] The imaging data acquisition unit 21 captures images of the area in front of the vehicle (i.e., the user's line of sight) in real time using an imaging camera (not shown), thereby acquiring imaging data. Typically, the imaging data is video data that is output every moment. The imaging camera includes, but is not limited to, a visible light camera or an infrared distance measuring camera (depth camera), for example. The imaging data acquisition unit 21 passes the acquired imaging data to the object recognition unit 22.

[0025] The object recognition unit 22 recognizes objects in an image frame based on the captured image data. For example, the object recognition unit 22 recognizes buildings, signs, road accessories, people, vehicles, and the like in the image frame as objects. For example, buildings, signs, and road accessories are recognized as static objects, and people and vehicles are recognized as dynamic objects. This enables non-standard voice guidance using dynamic objects as described below. The object recognition unit 22 may be configured to include an AI model in which machine learning is performed using, for example, feature amounts based on data strings of pixels in an image frame as explanatory variables and objects as objective variables.

[0026] The attribute recognition unit 23 acquires object attribute related information for the recognized dynamic object. That is, the attribute recognition unit 23 recognizes the attributes of the recognized dynamic object, and calculates the confidence level for each of a plurality of attribute values ​​of the recognized attributes. The attribute recognition unit 23 associates the recognized object with the attributes, the attribute values, and the confidence levels, and stores and updates the object attribute related information in a register (not shown).

[0027] For example, a vehicle recognized as an object has various attributes such as color, shape, size, and type. Among these, for example, the color attribute can have many attribute values ​​such as red, yellow, green, blue, purple, white, and black in the color space. For example, a bluish purple vehicle can be represented as a graph showing the confidence level for each attribute value as shown in FIG. 2. In addition, the type attribute can have attribute values ​​such as sedan, van, SUV, bus, and truck. The attribute recognition unit 23 calculates the confidence level for each of these attribute values ​​to recognize the attribute of the object. The attribute recognition unit 23 may be configured to include an AI model in which machine learning is performed using the attribute of the object as an explanatory variable and the attribute value as a target variable.

[0028] The attribute information database 24 is a database that stores common parameters defined according to information on commonality between attribute values ​​in each attribute. When an attribute value is represented by a plurality of parameters, the attribute information database 24 stores a parameter common between the attribute values ​​as a common parameter. For example, assuming that the attribute value of a color attribute is represented by parameters [R, G, B], the attribute value "blue" has parameters [0, 0, 255], and the attribute value "purple" has parameters [255, 0, 255]. In this case, the common parameter is (B: 255). Therefore, the attribute information database 24 associates a set of attribute values ​​(blue, purple) with a common parameter (B, 255) and stores them, for example, as shown in FIG. 3(a). By using such an attribute information database 24, it is possible to determine a common parameter between predetermined attribute values.

[0029] Furthermore, for example, when it is appropriate for an attribute value to be represented by a category, the attribute information database 24 stores a parameter of a superordinate concept that encompasses the attribute values ​​as a common parameter. In other words, when an attribute value has a single parameter, the attribute value is defined as being represented by a category. For example, as shown in FIG. 3(b), when the attribute values ​​of the type attribute are "microbus" and "large bus," "bus" is defined as the parameter of the superordinate concept (i.e., the common parameter). By using such an attribute information database 24, even when an attribute value is represented by a category, the superordinate concept can be used as a common parameter.

[0030] The commonization unit 25 refers to the attribute information database 24 and determines common parameters for the attributes of the recognized object in order to convey the recognized object in an easily understandable voice. That is, the recognized object is expressed using attributes that focus on the commonality between specific attribute values. For example, for a vehicle with high certainty of being "blue" and "purple" as shown in FIG. 2, it is possible to express it as a "blueish car" which has commonality, rather than an expression such as "purple car" which has a wide range of interpretations depending on the person. Therefore, the commonization unit 25 determines common parameters for expressing attributes of the recognized object that people can understand more intuitively and immediately, as described below. The commonization unit 25 passes the determined common parameters to the voice guidance control unit 16.

[0031] 4 is a block diagram showing an example of a specific configuration of a common unit of a car navigation system according to an embodiment of the present invention. As shown in the figure, the common unit 25 includes, for example, an attribute value extraction unit 251 and a common parameter determination unit 252.

[0032] The attribute value extraction unit 251 compares the certainty of each of the multiple attribute values ​​recognized by the attribute recognition unit 23 with a predetermined threshold value, and extracts at least two or more attribute values ​​whose certainty is greater than the predetermined threshold value. For example, in the example shown in Fig. 2, the attribute values ​​"blue" and "purple" are extracted. The attribute value extraction unit 251 passes the extracted attribute values ​​to the common parameter determination unit 252.

[0033] The common parameter determination unit 252 refers to the attribute information database 24 and determines a common parameter indicating commonality between two or more extracted attribute values. In the example shown in FIG. 2, the common parameter determination unit 252 specifies "blue" as the common parameter for the extracted attribute values ​​"blue" and "purple." In addition, the common parameter determination unit 252 determines "bus" as the common parameter for the extracted attribute values ​​"microbus" and "large bus." The common parameter determination unit 252 passes the determined common parameter to the voice guidance control unit 16.

[0034] Furthermore, when three or more attribute values ​​having a higher certainty than a predetermined threshold value are extracted, the common parameter determination unit 252 extracts three attribute values ​​in descending order of certainty. When the first attribute value and the second attribute value and the second attribute value and the third attribute value have commonality among these three attribute values, and the first attribute value and the third attribute value do not have commonality, the common parameter determination unit 252 determines the second attribute value as a common parameter. For example, when the first attribute value of a certain attribute has a parameter representing the properties of A and B, the second attribute value has a parameter representing the properties of B, and the third attribute value has a parameter representing the properties of B and C, the parameter of the second attribute value is determined as a common parameter. In this way, even when multiple attribute values ​​are applicable as those indicating the attributes of an object, their common parameters are determined.

[0035] 5 is a block diagram for explaining details of a voice guidance control unit of a car navigation system according to an embodiment of the present invention. As shown in the figure, the voice guidance control unit 16 includes, for example, a fixed voice guidance generating unit 161, a vocabulary database 152, and an unfixed voice guidance generating unit 162. As described above, the voice guidance control unit 16 refers to the voice guidance database 15 to generate voice guidance for the user. In the present disclosure, the voice guidance database 15 includes the voice guidance template database 151 and the vocabulary database 152, as described above.

[0036] The voice guidance template database 151 stores standard voice guidance templates. For example, the voice guidance template database 151 stores voice templates such as "Turn left at the intersection DDD meters ahead" (where DDD is a variable) and "Turn right here."

[0037] The vocabulary database 152 is a corpus that stores structured words and sentences. For example, an expression "bluish" is defined for a common parameter of "blue" and "purple." This allows an expression that is more intuitive and easy for the user to understand to be derived for a certain word or sentence.

[0038] The standard voice guidance generating unit 161 generates standard voice guidance under the control of the route guidance control unit 12, with reference to the voice guidance template database 151. For example, when the route guidance control unit 12 determines based on the acquired current position information that the vehicle is about 300 meters away from an intersection on the route where the vehicle should turn left, it instructs the standard voice guidance generating unit 161 to provide voice guidance. In response to this, the standard voice guidance generating unit 161 generates voice data such as "Please turn left at the intersection 300 meters ahead" with reference to the voice guidance template database 151, and outputs the voice data to the user interface unit 17. This causes the standard voice guidance to be played from the speaker.

[0039] The non-standard voice guidance generating unit 162 generates non-standard voice guidance using an expression related to the attribute of the object, referring to the vocabulary database 152 based on the common parameter. For example, when the route guidance control unit 12 determines, based on the acquired current position information, that the vehicle is about 150 meters away from an intersection on the route where the vehicle should turn left, it instructs the non-standard voice guidance generating unit 162 to provide voice guidance. In response to this, the non-standard voice guidance generating unit 162 refers to the vocabulary database 152 based on the common parameters related to the currently recognized object and its attribute, specifies an expression related to the attribute of the object, and further refers to the voice guidance template database 151 to generate voice guidance using the specified expression. For example, in a situation where a purple vehicle is recognized as an object, if the common parameter between the attribute values ​​of the color attribute of the vehicle is "blue", the non-standard voice guidance generating unit 162 specifies the expression "bluish car" using the vocabulary database 152, generates voice data of "Please follow the bluish car ahead," and outputs it to the user interface unit 17. This allows a non-standard voice guidance that focuses on the attributes of a dynamic object to be played from the speaker.

[0040] Fig. 6 and Figs. 7A and 7B are flowcharts for explaining an example of route guidance processing in a car navigation system according to an embodiment of the present invention. Specifically, Fig. 6 is a flowchart for explaining an example of object attribute recognition processing in a car navigation system, and Figs. 7A and 6B are flowcharts for explaining an example of route guidance processing linked to object attribute recognition processing in a car navigation system. Such processing is realized, for example, by the car navigation system 1 of the vehicle executing a predetermined navigation program under the control of a processor in cooperation with predetermined hardware resources. Note that, prior to such processing, it is assumed that the car navigation system 1 has already searched for a route to the destination and is providing route guidance to the user along the searched route (i.e., a proposed route).

[0041] 6, the car navigation system 1 acquires imaging data output from an imaging camera (S601), and then recognizes various objects in an image frame based on the acquired imaging data (S602). At this time, the car navigation system 1 recognizes dynamic objects and attributes such as vehicles, people, and motorcycles.

[0042] Next, the car navigation system 1 recognizes the attributes of the recognized dynamic object based on the recognized attribute (S603). The car navigation system 1 then calculates the confidence level for each of a plurality of attribute values ​​in the recognized attribute (S604). For example, when the car navigation system 1 recognizes a vehicle traveling ahead as a dynamic object, it calculates the confidence level for each attribute value in the color attribute. The recognized object is associated with attributes, attribute values, and confidence levels, and is stored and updated in a register (not shown) as object attribute related information, and is referred to as necessary.

[0043] The car navigation system 1 repeats the above-mentioned process while providing route guidance to the user, and obtains and stores object attribute related information of the object recognized based on the acquired imaging data.

[0044] 7A, the car navigation system 1 acquires the current position information of the vehicle in real time (S701), and monitors whether the vehicle has reached a guide point for a passing target point on the proposed route based on the acquired position information (S702). The guide point is, for example, a point 300 meters, 150 meters, and 50 meters before the intersection where the vehicle is to turn right (or left), which is the passing target point.

[0045] When the car navigation system 1 determines that the vehicle has reached a guide point for a passing destination on the proposed route (Yes in S702), it determines whether to provide typical voice guidance or non-typical voice guidance based on the current guide point (S703). For example, the car navigation system 1 determines to provide typical voice guidance at the most distant and nearest guide points among multiple guide points for a passing destination, and determines to provide non-typical voice guidance at intermediate guide points.

[0046] If the car navigation system 1 determines that the current guide point is a guide point for which standard voice guidance is to be provided (Yes in S703), it refers to the voice guidance template database 151, extracts an appropriate voice guidance template, and generates voice data based on the extracted voice guidance template (S704). Then, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (S705).

[0047] On the other hand, if the car navigation system 1 determines that the current guidance point is not a guidance point for which standard voice guidance is provided (i.e., a guidance point for which non-standard voice guidance is provided) (No in S703), it refers to the register to obtain object attribute-related information for the currently recognized object (S706).

[0048] Next, the car navigation system 1 judges whether the certainty of the attribute value indicated by the acquired object attribute related information is greater than a predetermined threshold value (S707). When the car navigation system 1 judges that the certainty of the attribute value is not equal to or greater than the predetermined threshold value (No in S607B), the car navigation system 1 refers to the voice guidance template database 151 to extract an appropriate voice guidance template and generates voice data based on the extracted voice guidance template (S704) to provide typical voice guidance, and then the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (S705). In other words, when there is no appropriate object attribute to be used for non-typical voice guidance for an object recognized based on the imaging data, the car navigation system 1 provides typical voice guidance.

[0049] On the other hand, if the car navigation system 1 determines that the certainty of the attribute value is greater than or equal to the predetermined threshold (Yes in S707), then the car navigation system 1 branches to each of the following processes depending on the number of attribute values ​​whose certainty is greater than the predetermined threshold (S708B).

[0050] That is, if the car navigation system 1 determines that the number of attribute values ​​whose certainty is greater than a predetermined threshold is one (terminal 1), the car navigation system 1 then determines whether the attribute value of the recognized attribute is made up of multiple parameters (S709 in Figure 7B).

[0051] When the car navigation system 1 determines that the attribute value is made up of multiple parameters (Yes in S709), the car navigation system 1 applies the attribute value of the recognized attribute to a voice guidance template to generate voice data (S710). In other words, if there is one attribute value whose certainty is greater than a predetermined threshold value and the attribute value is made up of multiple parameters, common parameters between the attribute values ​​cannot be identified, and voice data is generated using the attribute value. Next, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (terminal R / S705).

[0052] On the other hand, if the car navigation system 1 determines that the attribute value is not composed of multiple parameters (i.e., the attribute value is composed of one parameter) (No in S709), the car navigation system 1 refers to the attribute information database 24 and determines the attribute value that indicates a superordinate concept of the attribute value as the common parameter (S711). For example, if the attribute value is "large bus", the car navigation system 1 determines "bus" as the common parameter. Next, the car navigation system 1 generates voice data based on the determined common parameter (S712). Next, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (terminal R / S705). For example, the car navigation system 1 provides voice guidance such as "Please follow the bus ahead."

[0053] Furthermore, if the car navigation system 1 determines that the number of attribute values ​​whose confidence level is greater than a predetermined threshold value is two (terminal 2), the car navigation system 1 then determines whether these two attribute values ​​consist of multiple parameters (S713 in FIG. 7B). When the car navigation system 1 determines that the attribute value is composed of multiple parameters (Yes in S713), the car navigation system 1 refers to the attribute information database 24 and determines a common parameter based on the multiple parameters of the attribute value (S714). For example, when the attribute values ​​are "blue" and "purple", the car navigation system 1 determines "blue" as the common parameter. Next, the car navigation system 1 generates voice data for voice guidance based on the determined common parameter (S712). At this time, the car navigation system 1 derives an expression "blueish" by referring to the vocabulary database 152 based on the common parameter (for example, a parameter indicating "blue"). Next, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (terminal R / S705). For example, the car navigation system 1 performs voice guidance such as "Please follow the blueish vehicle ahead."

[0054] On the other hand, when the car navigation system 1 determines that the attribute value is not composed of multiple parameters (i.e., the attribute value is composed of one parameter) (No in S713), the car navigation system 1 refers to the attribute information database 24 and determines the attribute value indicating the superordinate concept of the attribute value as the common parameter (S711). For example, when the attribute values ​​are "microbus" and "large bus", the car navigation system 1 determines "bus" as the common parameter. Next, the car navigation system 1 generates voice data based on the determined common parameter (S712). Next, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (terminal R / S705). For example, the car navigation system 1 gives voice guidance such as "Please follow the bus ahead."

[0055] Furthermore, when the car navigation system 1 determines that the number of attribute values ​​whose confidence level is greater than a predetermined threshold value is three (terminal 3), it determines a parameter common to the three attribute values ​​as a common parameter (S715). When the first attribute value and the second attribute value, and the second attribute value and the third attribute value among the three attribute values ​​have commonality, and the first attribute value and the third attribute value do not have commonality, it specifies the second attribute value as the common parameter. Next, the car navigation system 1 generates voice data based on the determined common parameter as described above (S712). Next, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (terminal R / S705).

[0056] Furthermore, if the car navigation system 1 determines that the number of attribute values ​​whose certainty is greater than a predetermined threshold is greater than three (terminal 3'), it selects three attribute values ​​from among those attribute values ​​in descending order of certainty (S716). Next, the car navigation system 1 similarly determines parameters common to the three attribute values ​​as common parameters (S715), and generates voice data based on the determined common parameters (S712). Next, the car navigation system 1 outputs voice guidance from the speaker based on the generated voice data (terminal R / S705).

[0057] As described above, according to this embodiment, the car navigation system 1 can provide route guidance using voice guidance including linguistic expressions that are easier for the user to recognize. In particular, according to the present invention, it is possible to provide human-like voice guidance including expressions that the user can intuitively and immediately understand, using dynamic landmarks within the user's field of vision.

[0058] The above-described embodiments are merely examples for explaining the present invention, and are not intended to limit the present invention to these embodiments. The present invention can be embodied in various forms without departing from the spirit of the present invention.

[0059] For example, in the voice guidance control method disclosed in this specification, steps, operations, or functions may be performed in parallel or in a different order, as long as the results are not inconsistent. The steps, operations, and functions described are provided merely as examples, and some of the steps, operations, and functions may be omitted or may be combined into one without departing from the spirit of the invention, or other steps, operations, or functions may be added.

[0060] In addition, although various embodiments are disclosed in this specification, specific features (technical matters) in one embodiment can be added to or replaced with specific features in other embodiments, with appropriate modifications, and such forms are also included in the spirit of the present invention. [Explanation of symbols]

[0061] 1. Car navigation system 11...Location information acquisition unit 12...Route guidance control unit 13…Road map information database 14...Screen control section 15…Voice guidance database 151…Voice guidance template database 152… Vocabulary Database 16...Voice guidance control unit 161…Standard voice guidance generating unit 162...Non-standard voice guidance generating unit 17...User interface section 21...Imaging data acquisition unit 22…Object recognition unit 23...Attribute recognition section 24…Attribute information database 25…Commonization department 251...Attribute value extraction unit 252...Common parameter determination unit

Claims

1. A car navigation system that provides route guidance to a user who drives a vehicle, an object recognition unit that recognizes an object in an image frame based on the acquired imaging data; an attribute recognition unit that recognizes attributes of the recognized object and calculates a certainty factor for each of a plurality of attribute values ​​of the attributes; a commonization unit that commonizes the attribute values ​​based on the calculated certainty; a voice guidance control unit that generates a predetermined voice guidance in accordance with the vehicle position information, the commonization unit includes an attribute value extraction unit that extracts at least two attribute values ​​having a certainty level greater than a predetermined threshold value from among the recognized attribute values; a common parameter determination unit that determines a common parameter indicating a commonality between the at least two or more extracted attribute values, The voice guidance control unit generates voice guidance related to the object based on the determined common parameters. Car navigation system.

2. The object recognition unit recognizes the dynamic object in the image frame based on the imaging data.

2. The car navigation system according to claim 1.

3. the common parameter determination unit determines the second attribute value as the common parameter when, in a case where three or more attribute values ​​are extracted by the attribute value extraction unit, a first attribute value and a second attribute value and a second attribute value and a third attribute value among the three or more attribute values ​​have commonality with each other, and the first attribute value and the third attribute value do not have commonality with each other; 3. The car navigation system according to claim 2.

4. An attribute information database that stores the common parameters is further provided, the common parameter determination unit determines the common parameters between the attribute values ​​based on the attribute information database; 3. The car navigation system according to claim 2.

5. The common parameters are defined according to the commonality of the parameters of the attribute values.

5. The car navigation system according to claim 4.

6. The common parameters are defined based on a superordinate concept of the attribute values.

5. The car navigation system according to claim 4.

7. a route guidance control unit that controls whether or not to generate the voice guidance related to the object in accordance with position information of the vehicle; 3. The car navigation system according to claim 2.

8. the route guidance control unit controls the voice guidance control unit to generate non-standard voice guidance related to a dynamic object at a specific guidance point among a plurality of guidance points before approaching a passing target point through which the vehicle is scheduled to pass; 8. The car navigation system according to claim 7.

9. the route guidance control unit controls the voice guidance control unit to generate standard voice guidance related to a static object associated with road map information other than the dynamic object at a guidance point other than the specific guidance point among the plurality of guidance points.

9. The car navigation system according to claim 8.

10. the route guidance control unit controls the voice guidance control unit so that the typical voice guidance associated with the static object and the non-typical voice guidance associated with the dynamic object are alternately generated at the plurality of guidance points.

10. The car navigation system according to claim 9.

11. A voice guidance control method for a car navigation system that provides route guidance to a user who drives a vehicle, comprising: Recognizing an object within the image frame based on the acquired imaging data; recognizing attributes of the recognized object; calculating a confidence level for each of a plurality of attribute values ​​of the attribute; performing commonalization of the attribute values ​​based on the calculated certainty; generating a predetermined voice guidance in accordance with the vehicle position information; The commonalization includes extracting at least two or more attribute values ​​having a certainty degree greater than a predetermined threshold value from among the recognized attribute values; determining a common parameter indicating a commonality between the at least two or more extracted attribute values; and generating audio guidance based on the determined common parameters and generating audio guidance associated with the object. A voice guidance control method.

Images