JP2018086928A - Vehicle guiding device and vehicle guiding method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To guide a vehicle by a suitable method in accordance with a situation.SOLUTION: A vehicle guiding device includes: sensor information acquiring means for acquiring sensor information from a sensor provided in a vehicle; storage means for storing track data that is data showing a traveling track when the vehicle moved in a predetermined place in the past in association with a situation when the vehicle moved; and guiding means for guiding the vehicle by using the track data corresponding to a current situation among the stored track data and the sensor information.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an apparatus for guiding a vehicle.

  Research is underway on technologies to support drivers in situations that require a certain level of driving skills, such as garage parking. For example, Patent Document 1 discloses a parking assistance device that uses an image processing technique to calculate a vehicle guidance route and displays the guidance route superimposed on the overhead video. Patent Document 2 discloses a parking assistance device that selects an appropriate mode from a plurality of parking modes according to the positional relationship between the host vehicle and a parking space, and determines a guidance route. According to this technique, the guidance route can be dynamically generated according to the positional relationship between the vehicle and the parking space.

JP 2010-208483 A JP 2014-125196 A JP 2006-298115 A JP 2014-141216 A JP 2014-125195 A

  However, the device as described above provides high accuracy in an environment where parking spaces are separated by white lines and the entrance of the vehicle can be easily identified, but the approach route of the vehicle is difficult to distinguish on the image. In a difficult environment, sufficient accuracy cannot be obtained.

  On the other hand, when the pattern of entering the vehicle is the same every time as in a garage at home, there is a method of guiding the vehicle based on past driving operation history instead of dynamically generating the guidance route of the vehicle. is there. For example, in the driving support device described in Patent Document 3, driving operation information when parking in the past is stored as a history, and the driver is instructed to perform the same operation to guide the vehicle.

However, depending on the environment, it may not always be appropriate to enter the warehouse in the same manner as in past driving. For example, when an obstacle such as a luggage or a bicycle is placed in a part of a garage, it is necessary to enter the vehicle while avoiding the obstacle. In addition, for example, when the garage is narrow and it is necessary to get in, the entry route to the garage may change depending on the presence or absence of the passenger. In addition, for example, in the case of rain, it may be necessary to adjust the parking position by bringing the opening of the trunk under the roof to prevent the luggage from getting wet.
As described above, in the existing technology, there is a problem that it is difficult to guide the vehicle through an optimal route according to the situation.

  The present invention has been made in consideration of the above-described problems, and an object thereof is to guide a vehicle by an appropriate method according to the situation.

A vehicle guidance device according to the present invention includes:
A situation in which the sensor information acquisition means for acquiring the sensor information from the sensor provided in the vehicle and the trajectory data, which is data representing the travel trajectory when the vehicle has moved in the past at a predetermined location, are moved. Storage means for storing in association with the situation identifier representing the vehicle, and the guidance means for guiding the vehicle using the sensor information and the trajectory data corresponding to the current situation among the stored trajectory data. It is characterized by that.

  The sensor may be of any type as long as it is for grasping the position of the vehicle in the space. For example, an ultrasonic distance sensor, a laser distance sensor, a LIDAR, a high-precision GPS device, or the like may be used. Moreover, the camera (front camera, side camera, back camera, these combination etc.) etc. which were attached to the vehicle body may be sufficient. In this case, the image itself may be sensor information, or information obtained by analyzing the image may be sensor information. The information acquired by the camera may be a visible light image, an infrared image, a distance image, or the like.

The trajectory data is data representing a travel trajectory when the vehicle has moved in the past at a predetermined location (a location where the vehicle is guided). The trajectory data may be, for example, data representing a tire trajectory, or data representing a vehicle trajectory (for example, the entire vehicle body or a representative point trajectory). Further, the trajectory data does not necessarily represent the trajectory itself as long as it is data for calculating a continuous vehicle position. For example, it may be a set of images periodically acquired by a camera.
The trajectory data is stored in association with an identifier (situation identifier) that represents the situation when the vehicle has moved. The situation represented by the situation identifier is, for example, a place where guidance is performed (for example, an identifier of a parking lot or a parking space), a driver identifier, the number of passengers, a passenger seating position, a passenger identifier, weather, cargo The presence / absence, the amount of cargo, the type of cargo, and combinations thereof may be any as long as they can affect the guidance route of the vehicle.

  The guiding means is means for selecting a vehicle corresponding to the current situation from a plurality of stored trajectory data and guiding the vehicle. The situation comparison may be performed automatically or may be performed by the driver. For example, the stored trajectory data may be displayed side by side on the screen together with a description of the associated situation identifier, and may be selected by the driver. Alternatively, the current situation may be sensed and an appropriate one may be automatically selected. Thereby, suitable locus data can be selected according to the situation.

  Note that the situation identifier associated with the trajectory data used by the guiding means does not necessarily match the current situation. For example, when the driver is a beginner, the trajectory data generated by driving by an expert may be more preferable than the trajectory data generated by driving himself.

  The guidance of the vehicle may be performed by teaching a specific driving operation to the driver, or may be performed by presenting an amount of deviation from the trajectory data. Moreover, when the automatic driving device is mounted on the vehicle, the operation may be performed by issuing a command to the automatic driving device.

  Further, the guiding means may teach a driving operation necessary for the driver of the vehicle based on a result of comparing the sensor information and the trajectory data.

  For example, it is possible to intuitively transmit a method for adjusting the position of the vehicle by teaching a necessary amount of steering operation, acceleration / deceleration of an accelerator or a brake, and the like.

The trajectory data further includes information on an object around the vehicle detected when the vehicle moves, and the guidance unit includes the trajectory data as a result of comparing the sensor information with the trajectory data. When an object that cannot be detected is detected, whether or not the vehicle can proceed may be confirmed with the driver of the vehicle.

  Information related to objects around the vehicle may be information acquired by a distance sensor or the like, or may be information acquired by analyzing an image. Moreover, the image itself may be sufficient. In general, it is relatively easy for a machine to detect an object that is not included in data acquired in the past, but it is not possible to determine whether it is an object that affects the progress of the vehicle. It's not easy. For example, when a person is shown in the image acquired by the back camera, a special process is required to determine whether there is a person on the path of the vehicle or a person in a safe place. Also, it is not easy to determine whether the detected object is a human or a life-size poster. However, they are often obvious to the driver. Therefore, when a difference is detected, safety can be ensured without incurring cost by confirming with the driver whether the vehicle can proceed.

  In addition, the vehicle guidance device according to the present invention may further include an extracting unit that extracts the trajectory data having a related situation identifier similar to the current situation from the plurality of stored trajectory data. .

  Extraction may be performed automatically. For example, by obtaining the situation such as the location where guidance is performed, the driver's identifier, the number of passengers, the passenger's seating position, the passenger's identifier, the weather, the presence or absence of cargo, the amount of cargo, the type of cargo, etc. The trajectory data most suitable for the current situation can be automatically selected. Of course, the current situation may be input to the driver, and extraction may be performed semi-automatically.

  In addition, the extracting unit may cause the driver of the vehicle to select trajectory data used for guidance from the extracted trajectory data.

  Thus, extraction may be performed partially automatically, and the user may select the trajectory data to be finally used. For example, when the situation identifier is defined by a plurality of items representing the situation, filtering may be performed using only the matching items, and the driver may be finally selected which one to use.

  The extraction means may extract the trajectory data further based on an evaluation performed by the vehicle driver on the trajectory data.

  The evaluation may be performed by a score or may be performed using a sentence. Moreover, when making a driver | operator select locus | trajectory data, the content of evaluation may be shown and you may sort and filter based on the content of evaluation. According to this configuration, the vehicle can be guided in accordance with the driver's preference.

  In addition, the vehicle guidance device according to the present invention acquires trajectory data relating to the movement and a situation when the movement is performed when the vehicle moves at the predetermined location, and stores the acquired data in the storage unit. Further, it may be characterized by further comprising update means for determining whether or not the storage means stores new trajectory data on the basis of the similarity to the recorded data.

For example, when the vehicle is moved in an environment where the situation does not match any of the stored data, new trajectory data may be registered. Further, when the locus does not match any of the stored data (including a substantially coincidence), new locus data may be registered.
Further, when the vehicle is moved in an environment where the situation matches any of the stored data and the locus is similar, the matching data may be updated (updated).
According to such a configuration, data for guiding the vehicle can be accumulated with high accuracy.

The situation identifier may represent at least one of a location where the vehicle is guided, an identifier of the driver, weather, presence / absence of a passenger, and presence / absence of cargo.
This is because these factors may affect the guidance route of the vehicle.

  In addition, the guidance unit may perform the teaching in consideration of a time lag of a driving operation performed by the driver.

  There is a time lag between the teaching performed by the device and the operation performed by the driver. Therefore, if a simple difference between the current vehicle state and the trajectory data is taught, the trajectory of the vehicle may be vibrated and high-precision guidance cannot be performed. Therefore, it is preferable to teach in consideration of the time lag of the driving operation. For example, the position of the vehicle may be predicted in advance and control for correcting the teaching content may be performed.

  In addition, this invention can be specified as a vehicle guidance apparatus containing at least one part of the said means. It can also be specified as a vehicle guidance method performed by the vehicle guidance device. The above processes and means can be freely combined and implemented as long as no technical contradiction occurs.

  According to the present invention, the vehicle can be guided by an appropriate method according to the situation.

The figure explaining the path | route of the vehicle which enters into a parking space. 1 is a system configuration diagram of a vehicle guidance device 100 according to a first embodiment. The example of the guidance information table in 1st embodiment. The process flowchart figure of the vehicle guidance apparatus 100 which concerns on 1st embodiment. The example of the screen shown to a driver | operator by step S11. The process flowchart figure of the vehicle guidance apparatus 100 which concerns on 1st embodiment. An example of an interface for teaching a driver of necessary driving operations. The figure for demonstrating the guidance route of a vehicle. The example of the guidance information table in 2nd embodiment. The process flowchart figure of the vehicle guidance apparatus 100 which concerns on 2nd embodiment. The example of the screen which confirms whether guidance is continued when an unknown object is detected. The example of the guidance information table in 4th embodiment.

(First embodiment)
<System overview>
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
The vehicle guidance device according to the first embodiment is a device that gives a driving operation instruction to a driver so that a vehicle entering the parking space can follow an appropriate route (trajectory).

  First, the route of the vehicle to be stored will be described. In FIG. 1, reference numeral 1001 indicates a vehicle to be guided, and reference numeral 1002 indicates a parking space. The dotted line represents the route of the vehicle to be stored. In this example, a scene in which the vehicle 1001 is stored in the parking space 1002 provided in front of the house 1003 will be described as an example.

As shown in FIG. 1, when the vehicle 1001 is stored, it is necessary to reverse the direction of the vehicle in advance and store the vehicle 1001 while moving backward (in the illustrated example, it is necessary to switch back three times).
. However, if the approach route to the parking space deviates, there is a risk that the vehicle may protrude from the road or touch the entrance porch or building. In such a case, conventionally, the driver had to memorize the position to switch back and had to perform an operation based on experience.

The vehicle guidance device according to the first embodiment stores a movement route when a vehicle has entered in the past, and issues a driving operation instruction to the driver to trace the movement route. Solve. Moreover, the vehicle guidance device according to the present embodiment stores a plurality of movement routes, and has a feature of selecting an appropriate movement route according to the situation. As a result, for example, it is possible to flexibly respond such as “get on and off at a place with a roof on a rainy day” and “stop at a position where a passage can be secured if there is a load”.
Hereinafter, the vehicle guidance device according to the present embodiment will be described.

<System configuration>
FIG. 2 is a system configuration diagram of the vehicle guidance device 100 according to the present embodiment. The vehicle guidance device 100 according to the present embodiment includes an image acquisition unit 101, a control unit 102, a storage unit 103, an input / output unit 104, a situation acquisition unit 105, and an outside camera 110. The vehicle outside camera 110 is installed in a vehicle and includes a plurality of cameras that capture the outside of the vehicle.

  The vehicle guidance device 100 can be configured as an information processing device having a CPU, a main storage device, and an auxiliary storage device. Functions described below are realized by loading a program stored in the auxiliary storage device into the main storage device and executing it by the CPU. Note that all or part of the illustrated functions may be executed using a circuit designed exclusively (such as a semiconductor integrated circuit).

The image acquisition unit 101 is a means for acquiring an image from the connected external camera 110.
The vehicle exterior camera 110 includes a plurality of cameras (for example, a front camera, a back camera, a side camera, etc.) installed in the vehicle. In the present embodiment, in order to estimate the position of the vehicle based on the image acquired by the camera outside the vehicle, it is preferable to have at least two types of cameras facing the traveling direction and the lateral direction. In the present specification, for the sake of simplicity of explanation, only the images acquired by the back camera are used for illustration.

  The control unit 102 is a unit that controls the entire vehicle guidance device 100. Specifically, (1) a process of acquiring data representing the trajectory of a vehicle that has been stored (hereinafter referred to as trajectory data) and storing the data in association with the current situation in a storage unit 103 described later; and (2) the current situation. The trajectory data corresponding to is acquired from the storage unit 103, and the vehicle is guided. Specific processing contents will be described later.

The “situation” in this specification refers to, for example, an element that affects the trajectory when a vehicle enters, such as an identifier of a parking space, an identifier of a driver, the presence or absence of a passenger, the presence or absence of cargo, and the weather. Represents a set of elements.
In addition, the “trajectory” in the present specification represents a moving route of the vehicle, and is not necessarily a wheel trajectory. The trajectory may be, for example, a movement path of a point representing the vehicle body or a trajectory of the entire vehicle body region. In the present embodiment, the trajectory is a movement trajectory (a set of points) of points representing the vehicle body.

  The storage unit 103 is means for temporarily or permanently storing data used by the apparatus. The storage unit 103 is preferably a storage medium that can read and write at high speed and has a large capacity. For example, a flash memory can be suitably used. In the storage unit 103, trajectory data representing the trajectory when the vehicle is received is stored as a guidance information table in association with the situation when the vehicle is received.

  Here, data stored in the storage unit 103 will be described. FIG. 3 is an example of a guidance information table stored in the storage unit 103. The guidance information table is a table in which trajectory data is associated with a situation corresponding to the trajectory data.

In the example of FIG. 3, “garage ID”, “driver”, “passenger presence / absence”, “presence / absence of baggage”, and “weather” are affected by the entry method (for example, the approach route from the road and the stop position). Is defined as an element that provides the five elements. Data represented by the five elements is a situation identifier in the present invention.
The guidance information table is used for guidance of the vehicle, and is updated under a predetermined condition every time when the entry of the vehicle is completed (a specific method will be described later).

  The input / output unit 104 is a unit that displays various information such as an image generated by the vehicle guidance device 100 and acquires an instruction from the user. The input / output unit 104 is typically a touch panel display, but may be other than this.

  The situation acquisition unit 105 is a means for acquiring a situation at the time of vehicle entry based on a sensor (not shown) provided in the vehicle or information acquired from an external information source. The situation acquired by the situation acquisition unit 105 matches the situation recorded in the guidance information table. In this example, garage ID, driver identifier, presence / absence of passengers, presence / absence of luggage, and weather are five acquisition targets.

  The garage ID is an identifier of the parking space. The garage ID can be acquired based on position information obtained from a GPS device provided in the vehicle, for example. In addition, when position information cannot be acquired, you may number garage ID based on the image which the image acquisition part 101 acquired. For example, an image is acquired every time a vehicle is completed, compared with images acquired in the past, and when a similarity equal to or greater than a predetermined value is obtained, it is determined that the parking spaces are the same, In other cases, it may be determined that it is a new parking space.

  For example, the driver identifier can be obtained by wireless communication by preparing an intelligent key for turning on the ignition of the vehicle for each driver. In addition, the determination may be made based on the result of sensing the driver's biological information.

  The presence or absence of a passenger can be determined by, for example, a seating sensor provided on the seat. Alternatively, it may be determined by recording whether doors other than the driver's seat are opened or closed before the vehicle departs. In addition to this, it may be estimated that the passenger has boarded the vehicle based on the travel route of the vehicle.

  The presence / absence of luggage can be determined based on, for example, the result of sensing a luggage compartment. Alternatively, it may be determined by recording whether the rear gate or the like has been opened or closed before the vehicle departs. In addition to this, it may be estimated that the load is loaded based on the travel route of the vehicle (for example, whether the departure point of the vehicle is a parking lot of a retail store).

  The weather can be determined by, for example, a sensor (rainfall sensor or the like) provided in the vehicle. Moreover, you may determine by acquiring weather information through a network.

  In addition, in this embodiment, although the situation was represented using five elements, as long as it is an element which affects the warehousing method, you may use what was illustrated. Moreover, the sensor does not necessarily need to acquire a physical quantity. For example, the situation may be acquired using an image acquired by a camera.

<Processing at first goods receipt>
Next, a method for generating trajectory data will be described.
When the vehicle on which the vehicle guidance device 100 is mounted arrives in front of the parking space, the driver activates the device, whereby the flowchart shown in FIG. 4 is started.
Note that when the vehicle enters the vehicle, the driver selects one of the following two patterns.
(1) Guidance is performed using data stored in the storage unit 103. (2) Guidance is not performed, and only trajectory data is recorded. First, a case where data is not registered in the guidance information table is illustrated in FIG. This will be described with reference to FIG.

  First, in step S <b> 11, the control unit 102 causes the driver to select which of the data stored in the storage unit 103 is used for guidance via the input / output unit 104. FIG. 5 is an example of a screen for performing selection. On the other hand, if a driver inputs, a process will progress to step S12. In addition, when data is not memorize | stored, it is assumed that the driver | operator selected "do not use guidance" (step S12-No).

When the guidance is not used, the control unit 102 periodically acquires an image around the vehicle via the image acquisition unit 101, analyzes the acquired image, acquires the vehicle position information, and thereby the trajectory data. Is generated (step S13). The trajectory can be represented, for example, by a set of points representing the vehicle. The generated information is temporarily stored by the control unit 102.
When the driver performs an operation indicating that the warehousing has been completed, the process transitions to FIG. 6 (step S21). The completion of warehousing may be notified to the device by the driver via the input / output unit 104, or the device itself may determine using information that can be acquired from the vehicle (for example, parking brake information).

FIG. 6 is a flowchart of processing executed at the timing when warehousing is completed.
First, in step S <b> 21, the control unit 102 acquires trajectory data for the current warehousing and acquires the current situation via the situation acquisition unit 105.
Next, in step S22, it is confirmed whether or not existing data exists in the guidance information table. Here, when no data record exists, the process proceeds to step S25, and a new record is added.

Returning to FIG. 4, the processing when performing the second and subsequent warehousing will be described.
In the second and subsequent warehousing, since a record is recorded in the storage unit 103, data corresponding to the record is displayed on the screen (FIG. 5) provided in step S11. Again, the driver selects whether to receive guidance using any of the trajectory data or not to use guidance. For example, when it is desired to receive guidance based on recorded data, any data may be selected.

On the other hand, if the situation associated with the recorded trajectory data is different from the current situation, and the recorded data does not provide sufficient guidance, the driver selects “Do not use guidance” can do. For example, the last time was sunny and no passengers, but this time it was raining and there were passengers, so the parking position needs to be adjusted.
In such a case, the trajectory data is generated by the above-described processing, and a record corresponding to the situation “rainy and passengers” is newly created.

<Vehicle guidance method>
Next, a method for guiding the vehicle using the stored trajectory data will be described.
In step S11, if selection to receive guidance using specific trajectory data is performed (step S12-Yes), the control unit 102 starts guidance of the vehicle using the selected trajectory data.

  First, in step S14, the control unit 102 compares the trajectory data (that is, the trajectory to be traced) and the current position of the vehicle, and calculates a deviation amount and an operation amount necessary to correct the deviation.

  Next, in step S <b> 15, the control unit 102 issues an operation instruction to the driver via the input / output unit 104. FIG. 7 is an example of an interface for instructing a driver about a necessary driving operation. Note that the teaching content may be only a steering operation, or may include a throttle or a brake operation. Moreover, you may guide the distance to the next stop position.

  In this example, the contents of the driving operation are displayed with graphics and characters. However, a figure overlooking the vehicle is generated, and the relationship between the locus used for guidance and the current vehicle may be graphically displayed. Good. In addition, the direction of the vehicle to be corrected may be taught by a method other than the exemplified method. For example, the direction to be corrected may be superimposed and displayed on the image acquired by the camera. Further, voice input / output means may be added to guide the operation by voice.

  In step S16, as in step S13, the control unit 102 generates trajectory data.

  Since the trajectory data is a set of points, there is a risk that the vehicle body will fluctuate as shown in FIG. 8A if the trajectory is traced accurately. Therefore, the trajectory may be given a certain margin and guided so as to be within the margin range. For example, as shown in FIG. 8B, a certain amount of width may be given to the left and right, and the same locus may be regarded as being within the range.

  By repeating the processes of steps S14 to S16, the vehicle can be guided so that the vehicle can be entered along the same locus as the selected locus data.

When the warehousing is completed, the process proceeds to step S21 in FIG.
In the above description (first warehousing), a negative determination is made in step S22. However, in the second and subsequent warehousing, since the trajectory data is stored in the storage unit 103, an affirmative determination is made in step S22. .

Steps S23 and S24 are steps for determining whether or not the trajectory and situation at the most recent warehousing are similar to the trajectory and situation at the past warehousing.
Specifically, first, in step S23, it is determined whether or not the degree of similarity between the trajectory in the most recent warehousing and any of the trajectory data stored in the storage unit 103 is greater than or equal to a predetermined value. . Here, if none of the trajectory data is similar, the process proceeds to step S25, and a new record is registered.

  Note that the degree of similarity between trajectories can be determined by a known method. For example, as shown in FIG. 8C, an average value of distances connecting corresponding points on the trajectory can be used. At this time, the set of corresponding points on the trajectory may be specified based on the similarity of images acquired by the camera.

On the other hand, if it is determined that the similarity with any of the trajectory data is greater than or equal to a predetermined value, whether or not the similarity between the situation associated with the trajectory data and the current situation is greater than or equal to a predetermined value. judge. As a result, if the situation is not similar, the process proceeds to step S25, and a new record is registered.
In the present embodiment, the situation is composed of a plurality of elements. For this reason, after obtaining the similarity between the elements, the overall similarity may be calculated by a known method.

  If it is determined that both the trajectory and the situation are similar, it means that the garage has been placed under the same conditions as the trajectory data that has already been stored, so that no data is added and the process ends.

  As described above, the vehicle guidance device 100 according to the first embodiment accumulates the trajectory when the vehicle is received and uses it in the next and subsequent guidance. In addition, a plurality of trajectories are stored in a state associated with the situation when the vehicle enters, and is selected each time by the driver. That is, an optimal locus can be used according to the situation when the vehicle enters.

  In the first embodiment, the position of the vehicle is determined based only on the image acquired by the outside camera, but other sensing means may be used in combination. For example, the position of the vehicle can be acquired more accurately by using a plurality of distance sensors and high-precision GPS together. Further, the position of the vehicle may be detected mainly using these sensor means, and the image acquired by the outside camera may be used supplementarily (for example, for providing information to the driver).

(Modification of the first embodiment)
In the first embodiment, if the determination in step S24 is affirmative, the processing is terminated as it is, but the stored trajectory data may be updated using the trajectory data in the latest warehousing. For example, if it is determined that the trajectory in the current warehousing and the stored trajectory data are similar, the margin associated with the stored trajectory data may be expanded so that both are regarded as the same trajectory. Good. Further, the margin may be raised to an overall value or may be enlarged only for a specific place.

  In the first embodiment, the difference between the current vehicle state and the trajectory data is taught. However, according to such a method, the trajectory of the vehicle may be vibrated due to the time lag of the driving operation. . Therefore, teaching may be performed in consideration of the time lag of the driving operation. For example, the position of the vehicle may be predicted in advance by a known method, and the teaching content may be corrected.

(Second embodiment)
In the first embodiment, the vehicle is guided according to the stored trajectory data. However, even if the selected situation is the same, the actual parking space situation is not necessarily the same. For example, if there is no room on the parking lot site, there is a possibility that an obstacle such as a bicycle is placed. However, although an obstacle can be detected relatively easily, it is costly to let the machine determine whether or not the object is an object that affects the progress of the vehicle.
The second embodiment corresponds to such a problem, and in the case where a difference is detected between the stored data and the sensed data, an embodiment for confirming with the driver whether or not the vehicle is advancing. It is.

  In the second embodiment, as shown in FIG. 9, a plurality of images acquired by the camera outside the vehicle are recorded in association with the trajectory data.

FIG. 10 is a flowchart for explaining differences in the second embodiment among the processes shown in FIG. In the second embodiment, after executing step S16, an image corresponding to the current position is acquired, and the acquired image is compared with the current image acquired by the vehicle camera, and whether there is a difference. Is determined (step S17). Here, when it is determined that the degree of difference is greater than the predetermined threshold, the difference is presented to the driver, and it is confirmed whether or not to continue the guidance (step S18). For example, a screen as shown in FIG. 11 is presented, and guidance is interrupted until confirmation is performed. As a result, if there is a continuation instruction, the process returns to step S14 to continue the guidance. If there is an interruption instruction, the process ends and the guidance is interrupted.
When a continuation instruction is given, it is preferable to continue tracking an object detected as a difference and ignore it in the subsequent loops.

  According to the second embodiment, the device only detects the difference between the sensing result acquired in the past and the current sensing result, and the driver determines whether or not the vehicle guidance may be continued. According to this configuration, both low cost and safety can be achieved.

  In the second embodiment, the object is detected using the image, but the object may be detected using other sensing means.

(Third embodiment)
The vehicle guidance device 100 according to the first and second embodiments determines the situation every time when entering the garage, and automatically creates new trajectory data when a situation different from the past situation is detected. Remember. Thereby, trajectory data suitable for each situation can be generated. On the other hand, when such a method is used, the number of trajectory data stored in the device increases, making it difficult for the driver to select. In order to cope with this, the third embodiment is an embodiment in which the trajectory data used for guidance is filtered and provided to the driver.

  In the third embodiment, when selecting trajectory data in step S11, the controller 102 considers that the controller 102 is appropriate after the situation acquisition unit 105 acquires the current situation, rather than displaying all the data. Extract only the data that can be displayed. For example, when it is known that there is a passenger, only ID = 2, 3, and 4 are extracted and displayed from the data shown in FIG. If it is known that the current weather is rain, only ID = 2, 4 are extracted and displayed from the data shown in FIG. In this embodiment, in this way, filtering is automatically performed based on the current situation grasped by the device, and the trajectory data to be finally used is selected by the driver.

In this example, filtering is automatically performed according to the current situation, but part or all of the filtering may be performed manually. When filtering is performed manually, the result of filtering performed in the past may be stored and used.
In addition, when screen display is performed, only items for which filtering has not been performed may be displayed. According to this configuration, the driver can easily understand what characteristics the displayed trajectory data has.

(Fourth embodiment)
In the fourth embodiment, in addition to the third embodiment, the trajectory data is further evaluated.
In the first to third embodiments, even if the situation is the same, if the trajectories are different, they are stored as different trajectory data. In such a case, a lot of similar locus data is extracted in step S11, which makes it difficult to discriminate.
Therefore, in the fourth embodiment, the trajectory data is evaluated based on two criteria of “self-evaluation by the driver” and “time required for warehousing”, and the display is performed in the order of evaluation scores in step S11.

FIG. 12 is an example of the guidance information table in the fourth embodiment.
Self-evaluation by the driver can be performed at the timing when the warehousing is completed. The self-evaluation can be performed through the input / output unit 104, for example.
In the fourth embodiment, a score is assigned to the trajectory data based on the self-evaluation and the required time, and the data is sorted and displayed in descending order of the score. According to such a configuration, it is possible to enter more smoothly and preferentially provide trajectory data that the driver likes.

  Note that the self-evaluation may be an evaluation of the driver's own driving or an evaluation of the trajectory data itself. Also, when inputting self-assessment, a label to be added to the trajectory data (eg, “Recommended for shopping on rainy days”) is input, and the label is displayed on the screen of FIG. Good.

(Modification)
The above embodiment is merely an example, and the present invention can be implemented with appropriate modifications within a range not departing from the gist thereof.
For example, in the description of the embodiment, warehousing into a parking space has been illustrated, but the vehicle guidance device according to the present invention can also be used for guidance of vehicles in other places. For example, you may utilize as an apparatus which assists the passage of the vehicle on a narrow road.

  In the description of the embodiment, the vehicle guidance device 100 is an in-vehicle device. However, a part of the functions of the vehicle guidance device 100 may be provided to another device. For example, the function of the storage unit 103 may be provided in a network storage (cloud service or other server device). Moreover, you may mount the vehicle guidance apparatus 100 as an application which operate | moves in a portable computer (smartphone etc.).

  Further, in the description of the embodiment, as shown in FIG. 6, it is determined whether or not new data is added based on the locus and the similarity of the situation. You may do it. In this case, when selecting data, data having similar trajectories and situations may be merged and provided.

  In the description of the embodiment, an example is given in which a driver is instructed about a necessary driving operation. However, the vehicle may be guided by issuing a command to the automatic driving device.

DESCRIPTION OF SYMBOLS 100 Vehicle guidance device 101 Image acquisition part 102 Control part 103 Storage part 104 Input / output part 105 Situation acquisition part 110 Outside camera

Claims (10)

Sensor information acquisition means for acquiring sensor information from a sensor provided in the vehicle;
Storage means for storing trajectory data, which is data representing a travel trajectory when the vehicle has moved in the past at a predetermined location, in association with a situation identifier representing a situation when the movement is performed;
Guidance means for guiding the vehicle using the trajectory data corresponding to the current situation among the stored trajectory data and the sensor information;
A vehicle guidance device. The guiding means teaches a driving operation necessary for a driver of the vehicle based on a result of comparing the sensor information and the trajectory data.
The vehicle guidance device according to claim 1. The trajectory data further includes information related to objects around the vehicle detected when the vehicle moves,
The guidance means confirms whether or not the vehicle can proceed when the object not included in the trajectory data is detected as a result of comparing the sensor information and the trajectory data.
The vehicle guidance device according to claim 2. An extraction means for extracting the trajectory data whose associated situation identifier is similar to the current situation from the plurality of stored trajectory data;
The vehicle guidance device according to any one of claims 1 to 3. The extraction means causes the driver of the vehicle to select trajectory data used for guidance from the extracted trajectory data.
The vehicle guidance device according to claim 4. The extraction means extracts the trajectory data further based on an evaluation performed by the vehicle driver on the trajectory data;
The vehicle guidance device according to claim 4 or 5. When the vehicle moves at the predetermined place, the trajectory data related to the movement and the situation when the movement is performed are acquired, and based on the similarity between the data stored in the storage unit , Further comprising update means for determining whether or not to store as new trajectory data in the storage means,
The vehicle guidance device according to any one of claims 1 to 6. The situation identifier represents at least one of the location where the vehicle is guided, the driver's identifier, the weather, the presence or absence of a passenger, and the presence or absence of cargo.
The vehicle guidance device according to any one of claims 1 to 7. The guidance means performs the teaching in consideration of a time lag of a driving operation performed by the driver.
The vehicle guidance device according to claim 2. A vehicle guidance method performed by a vehicle guidance device,
A sensor information acquisition step of acquiring sensor information from a sensor provided in the vehicle;
Corresponding to the current situation from a storage device that stores trajectory data, which is data representing a travel trajectory when the vehicle has moved in the past at a predetermined location, in association with a situation identifier representing the situation at the time of the movement. A data acquisition step for acquiring trajectory data to be performed;
A guidance step of guiding the vehicle using the acquired trajectory data and the sensor information;
A vehicle guidance method including:

Images