JP2019215177A - Track setting device - Google Patents

Abstract

To provide a technique which improves accuracy of measurement of a position of an own vehicle without the need of including position data of a stationary target in map data.SOLUTION: A relative position calculation unit acquires a setting route set using a node and a link included in map data and calculates an exit relative position representing a relative position of an exit node Oj to an entrance node Ii at a specific intersection, the specific intersection being a next intersection at which a right or left turn is made, in the setting route. A range calculation unit calculates, on the basis of a position of the entrance node, a candidate range U1 which is a range within which a position of a real exit as an actual exit may be present. A likelihood calculation unit calculates an exit distribution which is a distribution of an exit likelihood representing a likelihood of the exit within the candidate range. An exit setting unit sets as the real exit P2 a spot within the candidate range corresponding to a point having the maximum exit likelihood. A track calculation unit calculates a running track A1 from a position of an own vehicle toward the real exit.SELECTED DRAWING: Figure 9

Description

  The present disclosure relates to a trajectory setting device.

  In order to perform automatic driving of a vehicle, it is necessary to set the position of the own vehicle and the trajectory on which the own vehicle runs. Particularly at intersections, moving targets such as vehicles and pedestrians are crowded because vehicles make right or left turns. For this reason, the accuracy of position information used for performing automatic driving requires higher measurement accuracy than the accuracy of position information used for general vehicle navigation.

  Here, in Patent Document 1, a stationary target (hereinafter referred to as a stationary target) such as a landmark or an infrastructure around the own vehicle such as a traffic light or a sign provided on the road is detected by a camera or the like, and the stationary target is detected. A technique is disclosed in which a position at which a target is detected is compared with a position of the detected stationary target on map data, and the position of the vehicle is corrected using the verified position data.

JP 2017-9553 A

However, as a result of detailed studies by the inventor, the following problems were found.
That is, in the above-described method of setting the position of the own vehicle, it is necessary to include the position data of the stationary target on the map data. However, since there are many stationary targets on the road, it is impossible to include the position data of all the stationary targets in the map data. Further, when the position data of the stationary target does not exist in the map data around the own vehicle, the position of the own vehicle cannot be verified, and the position of the own vehicle cannot be corrected.

  One aspect of the present disclosure is to provide a technique for improving the measurement accuracy of the position of a vehicle even when map information does not include position information of a stationary target.

  One embodiment of the present disclosure is a track setting device mounted on a vehicle, and includes a map data unit (30), a relative position calculation unit (S320), a position acquisition unit (S210), and a range calculation unit (S120). And a target acquisition unit (S420), a likelihood calculation unit (S420), an exit setting unit (S430), and a trajectory calculation unit (S440). In the map data section, a traffic network is represented using nodes representing intersections and dead ends of roads, and links connecting the nodes, and each intersection represents an entrance or exit of the intersection, and each lane represents a traffic lane. Is stored in the map data represented by the plurality of nodes set in. The relative position calculation unit obtains a set route set using the node and the link, and determines, in the obtained set route, a nearest intersection where the vehicle turns right or left as a specific intersection, and among a plurality of nodes belonging to the specific intersection, A node that allows vehicles to pass when entering a specific intersection is referred to as an entry node (Ii), a node that allows vehicles to pass when exiting from a specific intersection is an exit node (Oj), and the relative position of the exit node with respect to the entrance node. The exit relative position representing the position is calculated. The position acquisition unit acquires the position of the vehicle, which is the position of the vehicle. The range calculation unit may use the map data and the upper limit value of the error that may be included in the position of the own vehicle to locate the actual exit (P2), which is the actual exit, based on the position of the entrance node. A candidate range (U1), which is a range, is calculated. The target acquisition unit acquires target information including a relative position of the stationary target detected in the candidate range with respect to the vehicle. The likelihood calculation unit calculates an exit distribution, which is a distribution of exit likelihoods indicating the likelihood of exit in the candidate range, using the target information. The exit setting unit sets a point in the candidate range corresponding to the point at which the exit likelihood is maximum in the exit distribution as an actual exit. The trajectory calculation unit calculates a trajectory (A1) that is a trajectory for passing the vehicle from the own vehicle position toward the actual exit.

  According to such a configuration, a candidate range that is a range in which an actual exit at an intersection can exist is calculated, and a relative positional relationship with the actual exit that exists in the calculated candidate range is determined in advance. A stationary target is acquired, and an exit likelihood, which is a likelihood that an exit exists, is calculated using the positional relationship with the acquired position of the stationary target. Then, the position of the actual exit is set based on the calculated likelihood, and the traveling trajectory is calculated based on the set position of the actual exit. For this reason, the measurement accuracy of the own vehicle position can be improved without requiring the position data of the stationary target on the map data.

  Note that the reference numerals in parentheses described in this column and in the claims indicate the correspondence with the specific means described in the embodiment described later as one aspect, and the technical scope of the present disclosure It is not limited.

It is a block diagram showing composition of an automatic operation system. It is a flowchart of an automatic operation process. It is a flowchart of an intersection specification process. It is a flowchart of a range calculation process. It is a flowchart of a traveling control process. It is a figure showing the intersection included in map data. It is a figure showing the error range in an approach position and an actual exit. It is a figure showing the candidate range which added the error range. FIG. 9 is a diagram illustrating an actual exit and a candidate range calculated in a coordinate-transformed captured image. It is a figure showing the running track calculated for exiting the intersection.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[1. Constitution]
As shown in FIG. 1, the automatic driving system 1 includes a GNSS receiver 10, a sensor unit 20, a map memory 30, a navigation system 40, a track setting device 100, and a vehicle control device 200. Hereinafter, a vehicle equipped with the track setting device 100 is also referred to as a host vehicle.

  The GNSS receiver 10 is a device that receives a transmission radio wave from a GNSS artificial satellite via an antenna and measures the position of the own vehicle. GNSS is an abbreviation for Global Navigation Satellite System.

  The sensor unit 20 detects information about various objects existing in a peripheral area including the front of the own vehicle, a road marking such as a lane marking drawn on a road surface, and a target including at least a stationary target such as unevenness of a road surface shape. Includes cameras and radar sensors. As the radar sensor, for example, a millimeter wave radar, a laser radar, an ultrasonic radar, or the like can be used. Further, the sensor unit 20 recognizes a stationary target among the targets acquired by the camera and the radar sensor.

The recognition of the stationary target may be performed, for example, by acquiring the speed of the own vehicle and recognizing a target whose relative speed to the own vehicle matches the speed of the own vehicle as the stationary target.
The map memory 30 is a memory that stores map data.

The map data includes node information and link information of a road on which the vehicle travels.
The node information is information indicating the positions of a plurality of nodes, and the nodes represent an intersection and a dead end point which is a point where the road is at a dead end. The link information is information indicating the positions of a plurality of links, and the links connect nodes and are set for each road on which the vehicle travels. Note that the nodes and links are not limited to those set for each road. For example, when a road has a plurality of lanes, the nodes and links may be set for each lane. Hereinafter, description will be made assuming that nodes and links are set for each lane.

  As illustrated in FIG. 6, the node information includes an entrance node Ii, an exit node Oj, and an intersection link Lij. The entrance node Ii is a node that passes a vehicle when entering an intersection. The exit node Oj is a node that allows vehicles to pass when exiting from an intersection. The intersection link Lij is a link connecting the entrance node Ii and the exit node Oj, and in the intersection, in the vehicle width direction of the lane connecting the entrance node Ii to the exit node Oj along the lane in the intersection. Represents the center position of.

  Here, i and j each represent a number assigned to each lane connected to the intersection. For example, as shown in FIG. 6, the entrance node Ii and the exit node Oj are respectively represented as entrance nodes I1-4 and exit nodes O1-4 for each lane connected to the intersection. Further, for example, the intersection link Lij from the entrance node I1 to the exit node O2 is also described as an intersection link L12, and the intersection link Lij from the entrance node I3 to the exit node O4 is also described as an intersection link L34.

The navigation system 40 has a function of setting at least a route on a map from a current position of the own vehicle to a destination set by a driver of the own vehicle as a set route.
Here, the set route is set using nodes and links included in the map data stored in the map memory 30.

The set route includes data of the entrance node Ii and the exit node Oj of the intersection where the vehicle turns right or left to the destination.
The trajectory setting device 100 sets the trajectory on which the vehicle travels based on information acquired from the GNSS receiver 10, the sensor unit 20, the map memory 30, and the navigation system 40.

  The trajectory setting device 100 includes a microcomputer having a CPU 110 and a semiconductor memory (hereinafter, a memory 120) such as a RAM or a ROM. Each function of the trajectory setting device 100 is realized by the CPU 110 executing a program stored in a non-transitional substantial recording medium. In this example, the memory 120 corresponds to a non-transitional substantial recording medium storing a program. Also, by executing this program, a method corresponding to the program is executed. The trajectory setting device 100 may include one microcomputer, or may include a plurality of microcomputers.

  The method for realizing the various functions of the trajectory setting device 100 is not limited to software, and some or all of the elements may be realized using one or a plurality of hardware. For example, when the above functions are implemented by an electronic circuit that is hardware, the electronic circuit may be implemented by a digital circuit, an analog circuit, or a combination thereof.

  The vehicle control device 200 realizes automatic driving by controlling the accelerator, brake, and steering of the own vehicle according to the traveling trajectory on which the own vehicle travels set by the trajectory setting device 100.

Note that the map memory 30 corresponds to a map data section.
[2. processing]
<Automatic operation processing>
Next, the automatic driving process executed by the CPU 110 will be described with reference to the flowchart of FIG. The automatic driving process is executed, for example, when the track setting device 100 is started.

  In S110, the CPU 110 obtains the position of the host vehicle on the map and the set route of the host vehicle set in the navigation system 40, and specifies an intersection where the host vehicle turns right or left next as a specific intersection. Perform specific processing.

  In S120, the CPU 110 calculates a candidate range that is a range in which the exit position indicated by the exit node Oj can exist with respect to the entrance position indicated by the entrance node Ii of the specific intersection in the map data. I do.

  In S130, CPU 110 determines whether or not the vehicle has entered the specific intersection, using entrance node Ii and the position of the vehicle. Specifically, when the position of the vehicle acquired by the GNSS receiver 10 is in an intersection area that is a predetermined area between the entrance node Ii and the exit node Oj included in the specific intersection, It is determined that the vehicle has entered the specific intersection. On the other hand, when the position of the vehicle acquired by the GNSS receiver 10 is not in the intersection area included in the specific intersection, it is determined that the vehicle has not entered the specific intersection. Here, for example, the intersection area may be set to have a range extended along the road surface by an amount corresponding to the road width around an intersection link Lij from the entrance node Ii to the exit node Oj. That is, the intersection area is set as an area where the vehicle is considered to be passing through the intersection.

If CPU 110 determines in S130 that the vehicle has entered the specific intersection, the process proceeds to S140.
In S140, the CPU 110 travels along the intersection link Lij from the entrance position indicated by the entrance node Ii of the specific intersection to the exit position indicated by the exit node Oj so as to exit from the specific intersection that has entered. A traveling control process for performing the control is performed, and the process returns to S110 to perform the subsequent processes.

On the other hand, if the CPU 110 determines in step S130 that the vehicle has not entered the specific intersection, the process proceeds to step S150.
In S150, CPU 110 acquires the set route set by navigation system 40.

  In S160, the CPU 110 determines whether or not to change the specific intersection by determining whether the setting route acquired in S110 and the setting route acquired in S150 are the same.

  Specifically, if the setting route acquired in S110 is the same as the setting route acquired in S150, CPU 110 shifts the processing to S130. On the other hand, if the setting route acquired in S110 is different from the setting route acquired in S150, the process proceeds to S110.

  That is, when the set route is changed by the navigation system 40, the process returns to S110, and the specific intersection is reset on the newly set route. On the other hand, if the set route has not been changed by the navigation system 40, it is determined that there is no need to reset the specific intersection, and it is continuously determined whether or not the specific intersection has been entered.

Note that the processing in S120 corresponds to processing as a range calculation unit.
<Intersection identification processing>
Next, the intersection specifying process performed by the CPU 110 in S110 will be described with reference to FIG.

In S210, CPU 110 obtains the own vehicle position using GNSS receiver 10.
In S220, CPU 110 acquires map data using map memory 30.
In S230, CPU 110 obtains the set route from navigation system 40.

In S240, when the vehicle travels on the set route acquired in S230, the CPU 110 specifies an intersection to turn right or left next, and sets it as a specific intersection.
Note that the processing in S210 corresponds to processing as a position acquisition unit.
<Range calculation processing>
Next, the range calculation process performed by the CPU 110 in S120 will be described with reference to FIG.

In S310, the CPU 110 extracts the entrance node Ii and the exit node Oj of the specific intersection specified by the intersection specification processing of S110.
In S320, the CPU 110 calculates an exit relative position indicating the relative position of the exit node Oj with respect to the entrance node Ii of the specific intersection extracted in S310. The exit relative position may be obtained, for example, from the positions represented by the entrance node Ii and the exit node Oj. In S330, CPU 110 obtains the upper limit of the error that can be included in the position of the vehicle and the upper limit of the error that can be included in the position of exit node Oj at the specific intersection. The upper limit value of the error that can be included in the own vehicle position refers to, for example, the upper limit value of an error that occurs when the GNSS receiver 10 measures the own vehicle position. The upper limit value of the error that can be included in the position of the exit node Oj at the specific intersection is, for example, an error generated between the position of the exit node Oj represented by the map data included in the map memory 30 and the actual exit position. Means the upper limit. When the error σ1 of the own vehicle position and the error σ2 of the position of the exit node Oj are shown in FIG. 7, they are represented by, for example, a substantially circular or elliptical shape centered on the own vehicle position and the position represented by the exit node Oj, respectively.

  In S340, CPU 110 calculates error range σ3 of the exit from the own vehicle position. The error range σ3 with respect to the own vehicle position is calculated by adding the error σ1 of the own vehicle position and the error σ2 of the position of the exit node Oj with the position of the exit node Oj as the center, as shown in FIG. .

That is, the error range σ3 of the exit node Oj with respect to the own vehicle position corresponds to error data which is an error generated when displaying the position data of the vehicle on the road map represented by the map data.
In S350, the CPU 110 performs coordinate conversion of the error range σ3 of the actual exit P2 with respect to the approach position so as to be a range in front of the vehicle, which is the detection range of the sensor unit 20, as shown in FIG. 9, and calculates the candidate range U1. .

S320 corresponds to processing as a relative position calculation unit.
<Driving control processing>
Next, the traveling control process performed by the CPU 110 in S160 will be described with reference to FIG.

In S410, the CPU 110 acquires the candidate range U1 calculated by the range calculation process in S120.
In S420, the CPU 110 calculates an exit distribution which is a distribution of exit likelihoods in the candidate range U1. Here, the exit likelihood refers to the likelihood that is the position where the actual exit P2 exists in the candidate range U1. That is, the exit likelihood indicates the degree of exit-likeness of an intersection.

Here, the exit likelihood is calculated based on the position and shape of the target existing on the road.
For example, in a case where the lane marking information, which is information corresponding to the positions of the pair of lane markings P21 and P22 representing the outer end of the lane in the candidate range, is acquired using the sensor unit 20, the pair of lane markings is obtained. The position of the center in the width direction of P21 and P22 may be calculated such that the likelihood of the actual exit P2 increases. At this time, for example, a target that exists in the candidate range and matches the shape of the predetermined lane marking may be acquired as the lane marking.

  Also, instead of the lane markings representing the outside of the lane, if the unevenness is acquired and irregularities large enough to represent a curb at the end of the road are detected, the position is determined by the outer boundary of the lane. May be calculated such that the likelihood of the actual exit P2 at the position of the center of the lane becomes high, with the position moved in the width direction by half the width of the lane from the position of the curb as the center of the lane. .

  The unevenness threshold is set to the size of the unevenness such that the vehicle can pass over it, and the position that is equal to or less than the unevenness threshold is defined as the flat position, and the exit likelihood of the flat position is higher than the position other than the flat position. May be calculated to be higher.

  When the pedestrian crossing information, which is information corresponding to the position of the pedestrian crossing P10 in the candidate range, is acquired using the sensor unit 20, only a predetermined distance from the pedestrian crossing outward from the crossing center when viewed from the center of the intersection. The distant position may be calculated so that the exit likelihood increases.

In S430, the CPU 110 sets, as the actual exit P2, a point where the exit likelihood is maximum in the exit distribution of the candidate range acquired in S420.
That is, in S410 to S430, the actual exit P2 within the candidate range U1 is extracted as the actual exit position in the range detected using the sensor unit 20.

  In S440, the CPU 110 sets the traveling trajectory A1 between the actual exit P2 acquired in S430 and the own vehicle position as shown in FIG. The shape of the traveling trajectory A1 to be set is such that, for example, a path is set such that the start point is the vehicle position, the actual exit P2 is the end point, and the actual exit P2 is in the traveling direction of the route represented by the set route. Is also good. The trajectory of the traveling trajectory A1 to be set is not limited to the one set by such a method, and may be set by various methods.

In S450, CPU 110 causes vehicle control device 200 to control the own vehicle so as to travel along traveling trajectory A1 set in S440.
In S460, CPU 110 obtains the vehicle position from GNSS receiver 10 and determines whether or not the exit at the specific intersection is completed by determining whether or not the vehicle is included in the intersection area of the specific intersection. That is, when the own vehicle position is included in the intersection area of the specific intersection, it is determined that the exit from the specific intersection is not completed, and when the own vehicle position is not included in the intersection area of the specific intersection, the exit from the specific intersection is performed. Is determined to be completed.

If the CPU 110 determines in S470 that the exit at the specific intersection has been completed, the traveling control process ends.
On the other hand, if the CPU 110 determines in S470 that the exit from the specific intersection is not completed, the process proceeds to S410.

That is, the travel control process is repeatedly performed until the vehicle exits from the specific intersection, and the travel control process ends when the vehicle exits from the specific intersection.
Note that the processing in S420 corresponds to processing as a target acquisition unit, unevenness acquisition unit, unevenness calculation unit, and likelihood calculation unit, the processing in S430 corresponds to processing as an exit setting unit, and the processing in S440. The processing corresponds to the processing as a trajectory calculation unit.

[3. effect]
According to the embodiment described in detail above, the following effects can be obtained.
(1) According to the above embodiment, the candidate range is set in advance from the map data, and the position of the exit is estimated using a stationary target whose relative positional relationship with the exit is known. By setting the vehicle to proceed toward the exit, automatic driving at the intersection is realized.

  Therefore, the position data of the stationary target itself does not need to be stored in the map data in advance, and the accuracy of estimating the position of the own vehicle can be improved. Then, the traveling trajectory A1 can be calculated based on the position of the own vehicle with the improved estimation accuracy.

  (2) According to the above-described embodiment, the position from the position of the own vehicle to the actual exit P2 is repeatedly calculated until the vehicle exits the intersection, so that the travel of the vehicle according to the previous calculation result is directed to the exit. Even if it deviates, it can be corrected by recalculation.

  (3) The position of the actual exit P2 is repeatedly calculated until the vehicle exits the intersection. Therefore, even if the position of the actual exit P2 is blocked by the shield from the position where the vehicle has entered the intersection, if the positional relationship with the shield changes while the vehicle is traveling in the specific intersection, The travel trajectory A1 is calculated based on the position of the actual exit P2 detected again. Accuracy of calculating the traveling trajectory A1 can be improved.

[4. Other Embodiments]
As mentioned above, although embodiment of this indication was described, this indication is not limited to the above-mentioned embodiment, and can carry out various modifications.

  (1) In the above embodiment, the CPU 110 sets the travel path after setting the actual exit P2. However, the timing of setting the traveling trajectory is not limited to the timing set between the host vehicle position and the actual exit P2 after the actual exit P2 is set. For example, at the stage where the candidate range is set, a temporary traveling trajectory connecting the center of the candidate range and the position of the own vehicle is set, and the set temporary traveling trajectory is corrected at the timing when the actual exit P2 is set. May be used to set the traveling trajectory.

  (2) In the above-described embodiment, the own vehicle position is measured by the GNSS receiver 10, but the device for measuring the own vehicle position is not limited to the GNSS receiver 10. For example, the position of the host vehicle may be measured using a speed sensor, an acceleration sensor, a yaw rate sensor, or the like instead of or in addition to the GNSS receiver 10.

  (3) In the above embodiment, the map data is acquired from the map memory 30. However, restriction information that is information that restricts a range in which the vehicle can travel in the map information may be further acquired. Here, the restriction information may be, for example, information on a place where construction is being performed. Further, when the restriction information is acquired, the candidate range U1 may be set wider. Thereby, even when a road sign cannot be detected, it is possible to control automatic driving.

  Furthermore, the restriction information may further include information on a non-traveling range, which is a range in which vehicles on the road cannot travel. In this case, the candidate range U1 may be set so as to be a range excluding the non-traveling range.

  (4) In the above embodiment, the sensor unit 20 is used to determine the road markings representing the road edges of the candidate range and the irregularities such as curbs, and the exit likelihood is set with the relative position from the road edges as the center of the road. The exit likelihood is calculated so that the exit probability becomes higher, but the calculation of the exit likelihood is not limited to such a method. For example, when a road sign that is likely to be at the center of the road such as “stop” is detected, the exit likelihood at that position may be calculated to be higher, and the detected object may be detected in the image. The degree of flatness of the target may be obtained using the sensor unit 20, and the maximum flatness position may be calculated so that the exit likelihood is higher than other non-flat parts.

  (5) In the above-described embodiment, in the travel control process, the process shifts to S460 after the travel control in S450. However, after S450, the process may shift to S460 after a standby time for one iteration has elapsed. Here, one iteration refers to an interval at which the GNSS receiver 10 measures the position of the own vehicle. Further, the standby time is not limited to such a time, and may be, for example, an interval at which the CPU 110 acquires the own vehicle position from the GNSS receiver 10.

  (6) A plurality of functions of one component in the above embodiment may be realized by a plurality of components, or one function of one component may be realized by a plurality of components. . Further, a plurality of functions possessed by a plurality of constituent elements may be realized by one constituent element, or one function realized by a plurality of constituent elements may be realized by one constituent element. Moreover, you may abbreviate | omit a part of structure of the said embodiment. In addition, at least a part of the configuration of the above embodiment may be added to or replaced with the configuration of the other embodiment. In addition, all the aspects included in the technical idea specified from the wording described in the claims are embodiments of the present disclosure.

  (7) In addition to the trajectory setting device 100 described above, a system including the trajectory setting device 100 as a component, a program for causing a computer to function as the trajectory setting device 100, and a non-transitional device such as a semiconductor memory storing the program. The present disclosure can be realized in various forms, such as an actual recording medium and a trajectory setting method.

  DESCRIPTION OF SYMBOLS 1 ... Automatic driving system, 10 ... Receiver, 20 ... Sensor part, 30 ... Map memory, 40 ... Navigation system, 100 ... Track setting device, 110 ... CPU, 120 ... Memory, 200 ... Vehicle control device, A1: Running track , Lij ... intersection link, Ii ... entrance node, Oj ... exit node, P10 ... crosswalk, P2 ... actual exit, P21 ... compartment line, U1 ... candidate range.

Claims (8)

A track setting device mounted on a vehicle,
A traffic network is represented by using nodes representing intersections and dead ends of roads, and links connecting the nodes, and the intersections each represent an entrance or exit of the intersection and are set for each lane. A map data unit (30) for storing map data represented by the plurality of nodes;
A set route obtained by using the node and the link is obtained, and in the obtained set route, a nearest intersection where the vehicle turns right or left is a specific intersection, and among the plurality of nodes belonging to the specific intersection, A node that allows the vehicle to pass when entering the specific intersection is referred to as an entry node (Ii); a node that allows the vehicle to pass when exiting from the specific intersection is an exit node (Oj); A relative position calculator (S320) configured to calculate an exit relative position indicating a relative position of the exit node;
A position acquisition unit (S210) configured to acquire the position of the host vehicle that is the position of the vehicle;
Using the map data and the upper limit value of the error that can be included in the position of the vehicle, a position where the actual exit (P2), which is an actual exit, can exist based on the position of the entrance node. A range calculator (S120) configured to calculate a certain candidate range (U1);
A target acquisition unit (S420) configured to acquire target information including a relative position of the stationary target detected in the candidate range with respect to the vehicle;
A likelihood calculation unit (S420) configured to calculate an exit distribution, which is a distribution of exit likelihoods representing the likelihood of exit in the candidate range, using the target information;
An exit setting unit (S430) configured to set a point in the candidate range corresponding to a point at which the exit likelihood is maximum in the exit distribution as the actual exit;
A track calculating unit (S440) configured to calculate a running track (A1) that is a track for passing the vehicle from the host vehicle position toward the actual exit;
A trajectory setting device comprising: The trajectory setting device according to claim 1,
The target acquisition unit is configured to acquire information representing the position of at least one pair of lane markings representing the position of the end of the lane as the target information,
The trajectory setting device, wherein the likelihood calculating unit calculates the likelihood of the exit at the center of a range sandwiched by the pair of lane markings acquired by the target acquiring unit so as to increase the likelihood. The trajectory setting device according to claim 1 or 2, wherein:
The target acquisition unit is configured to acquire, as the target information, pedestrian crossing information that is information indicating a position of a pedestrian crossing (P10) included in the specific intersection,
The likelihood calculation unit is configured to correspond to the position of the pedestrian crossing and the position of the exit represented by the exit node based on the position of the pedestrian crossing represented by the pedestrian crossing information acquired by the target acquisition unit. A trajectory setting device that calculates the likelihood of the exit at a position calculated according to a relationship. 4. The trajectory setting device according to claim 1, wherein the target acquiring unit is configured to acquire irregularities of a road surface shape in the candidate range (S <b> 420). )
The likelihood calculation unit includes an unevenness calculation unit (S420) that calculates an exit likelihood based on the unevenness of the road surface shape acquired by the unevenness acquisition unit.
A trajectory setting device comprising: The trajectory setting device according to claim 4, wherein
The unevenness acquisition unit acquires unevenness in the candidate range,
The unevenness calculation unit, the exit likelihood at a flat position that is equal to or less than an unevenness threshold value set to a predetermined unevenness size is higher than the exit likelihood at a position other than the flat position. Calculated so that
The trajectory setting device, wherein the unevenness threshold is set to one of a size of unevenness enough to allow a vehicle to pass and a size of unevenness at the center of a road. The trajectory setting device according to claim 4, wherein
The unevenness acquisition unit is configured to acquire at least unevenness of a curb size installed at an end of a road,
The trajectory setting device, wherein the unevenness calculating unit calculates the exit likelihood at the center of a range sandwiched by the unevenness of the curb size acquired by the unevenness acquiring unit so as to increase. The trajectory setting device according to any one of claims 1 to 6, wherein
The target acquisition unit, the likelihood calculation unit, the exit setting unit, and the trajectory calculation unit may include a case where the vehicle position is included between a position represented by the entrance node and a position represented by the exit node. A trajectory setting device configured to execute a process assigned to each configuration. The trajectory setting device according to any one of claims 1 to 7, wherein:
The target acquisition unit, the likelihood calculation unit, the exit setting unit, and the trajectory calculation unit are configured such that the own vehicle position is included in a position represented by the entrance node from a position represented by the entrance node. A trajectory setting device configured to repeatedly execute a process assigned to the trajectory.

Images