JP7149082B2 - Driving support method for driving support device and driving support device - Google Patents

Description

TECHNICAL FIELD The present invention relates to a driving support method and a driving support device for a driving support device.

Conventionally, when white lines cannot be recognized, lane information including the road shape of the own lane is obtained from the road map information (map), and the movement trajectory of the preceding vehicle is obtained from the sensor provided in the own vehicle, and then obtained from the map. A technique is disclosed for estimating a movable area in front of the own vehicle based on lane information and a movement trajectory obtained by a sensor.

JP 2017-016403 A

However, in the above technology, the structure of lanes on the map and the movement trajectory detected by the vehicle are integrated based on the position of the vehicle obtained using GPS. There is a problem that the structure of the lane above and the movement trajectory detected by the sensor of the own vehicle cannot be integrated at an accurate position.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and an object of the present invention is to provide a driving assistance method and a driving assistance device capable of accurately integrating road structure data detected by sensors of own vehicle and lane boundary data on a map. is to provide

A driving support method according to one aspect of the present invention includes setting a route on a map indicating a lane along which the vehicle is scheduled to travel to a destination, detecting the traveling position of the vehicle on the route set on the map, The road structure data detected by the vehicle is converted into a map coordinate system based on the traveling position on the route . Then, integrated road structure data is generated by integrating the road structure data converted into the map coordinate system and the lane boundary data. Then, the integrated road structure data is used to assist the running of the vehicle.

ADVANTAGE OF THE INVENTION According to this invention, the road structure data detected by the sensor of the own vehicle and the lane boundary data on a map can be integrated correctly.

FIG. 1 is a functional block diagram showing a schematic configuration of a driving support device according to the first embodiment. FIG. 2 is a flow chart of the driving support method in the first embodiment. FIG. 3 is a diagram showing a part of the map and an example of the route RT. FIG. 4 is a diagram showing an example of an actual route T actually traveled and detection data. FIG. 5 is a diagram showing how vehicle 100 travels on a four-lane road. FIG. 6 is a diagram showing an example of a route RT and a travel locus P set on a map. FIG. 7 is a diagram showing how integrated road structure data is generated on a map. FIG. 8 is a diagram showing how the own vehicle 100 stops before the stop line ST. FIG. 9 is a functional block diagram showing a schematic configuration of a driving support device according to the second embodiment. FIG. 10 is a flow chart showing processing performed in the second embodiment in place of step S1. FIG. 11 is a diagram showing an example of the first own vehicle position P1 and the second own vehicle position P2. FIG. 12 is a diagram showing how the own vehicle 100 turns. FIG. 13 is a diagram showing the relationship between the turning amount S, the turning state, and the host vehicle position. FIG. 14 is a diagram showing the relationship between the coefficient of friction μ, the turning state, and the vehicle position. FIG. 15 is a diagram showing the relationship between elapsed time T, turning amount S, and host vehicle position. FIG. 16 is a diagram showing the relationship between the elapsed time T, the coefficient of friction μ, and the vehicle position. FIG. 17A is a diagram showing lane boundary lines D11 and D12 detected using the first host vehicle position P1. FIG. 17B is a diagram showing lane boundary lines D21 and D22 detected using the second host vehicle position P2. FIG. 17C is a diagram showing lane boundary lines D11 and D12 and actual lane boundary lines D31 and D32. FIG. 18A is a diagram showing how vehicle 100 travels on a road with two or more lanes. FIG. 18B is a diagram showing lane boundary lines D41 and D42 recognized as host vehicle 100. FIG. FIG. 19 is a diagram showing a state in which host vehicle 100 travels on lane L41 and other vehicle 120 travels on lane L42 facing lane L41. FIG. 20 is a diagram showing a state in which host vehicle 100 runs on lane L51 and another vehicle 120 runs on lane L52 in the same direction.

Embodiments will be described with reference to the drawings. In the description of the drawings, the same parts are denoted by the same reference numerals, and the description thereof is omitted.

(First embodiment)
FIG. 1 shows a schematic configuration of a driving support device in a first embodiment. The driving support device is mounted on a vehicle (hereinafter referred to as own vehicle), and includes a map database 1, an own vehicle position estimation unit 2, a route setting unit 3, a target lane identification unit 4, a road structure data acquisition unit 5, and integrated road structure data. A generation unit 6 and a driving support unit 7 are provided.

The driving support device is a general-purpose microcomputer having a CPU (central processing unit), memory, and input/output unit. A computer program (driving assistance program) for functioning as a driving assistance device is installed in the microcomputer. By executing the computer program, the microcomputer functions as a plurality of information processing circuits (2 to 7) included in the driving assistance device. Here, an example of realizing a plurality of information processing circuits (2 to 7) provided in the driving support device by software is shown, but of course, dedicated hardware for executing each information processing shown below is prepared. It is also possible to configure the information processing circuits (2 to 7) by Also, the plurality of information processing circuits (2 to 7) may be configured by individual hardware. Furthermore, the information processing circuits (2 to 7) may also be used as an electronic control unit (ECU) used for other controls related to the own vehicle.

Map database 1 is the memory of a microcomputer storing maps with lane boundary data on roads. The lane boundary data includes coordinate data indicating the boundaries between lanes and their surroundings. Boundaries include white or yellow continuous lines or broken lines (lane boundary lines) at the boundary between two lanes, guardrails, curbs, roadside strips, and the like provided at the boundary between lanes and sidewalks. The lane boundary data also includes attribute data indicating the color of the lane boundary line, line type, and the like.

In addition, the map includes information (lane attributes) such as the width of the lane and the traveling direction of the vehicle in the lane. The map also stores coordinate data of stop lines indicating positions at which vehicles should stop, traffic signals at intersections, and signs at intersections and roadsides.

The map is, for example, a so-called HD map (High Definition Map). In this case, the map data is generated from an image or the like obtained while the vehicle is running using a vehicle sensor, and is not necessarily accurate. For example, it may be generated from images taken on different vehicles, etc., which also contributes to the inaccuracy.

Therefore, if a travel locus is set to a lane on an inaccurate map and the vehicle is controlled to travel along the travel locus, there is a possibility that the vehicle will deviate from the actual lane.

Therefore, it is preferable to generate data (integrated road structure data, which will be described later) indicating the actual position of the lane, and use the integrated road structure data to support the driving of the own vehicle. As a result, lane deviation can be prevented.
Also, the integrated road structure data may be generated on a map. This allows the map to be updated accurately. For example, discontinuous lane boundaries on the map can be made continuous to compensate for continuity.

The own vehicle position estimator 2 estimates (calculates) the position of the own vehicle (hereinafter referred to as the own vehicle position) using either the running locus model or GPS (Global Positioning System). As time elapses, the estimation of the vehicle position is repeated, and the old vehicle position is updated with the latest vehicle position. A method of estimating the position of the vehicle using the running locus model is known as self-position estimation using odometry. The own vehicle position includes errors and does not necessarily match the position of the own vehicle on the map.

The route setting unit 3 calculates the roads on which the vehicle is scheduled to travel from the position of the vehicle on the map to the destination, calculates the lanes on which the vehicle is scheduled to travel from the calculated roads, and displays the calculated lanes on the map. Set as root RT.

The target lane identification unit 4 identifies the lane in which the vehicle is currently traveling (hereinafter referred to as the target lane) from the route RT set on the map based on the vehicle position.
The road structure data acquisition unit 5 is a sensor mounted on the own vehicle and acquires road structure data around the vehicle. The road structure data acquisition unit 5 can be configured with a laser radar, a millimeter wave radar, a camera, or the like.

The integrated road structure data generator 6 converts the road structure data detected by the own vehicle into a map coordinate system using the route RT, and generates the road structure data converted into the map coordinate system and the lane boundary data on the map. Integrate to generate integrated road structure data. Integrated road structure data may be generated separately from the map, or may be generated in the map.

The driving support unit 7 uses the integrated road structure data to support driving of the own vehicle. For example, using the integrated road structure data, the travel locus of the own vehicle is calculated, and the own vehicle is controlled so as to travel along the travel locus.

FIG. 2 is a flow chart of the driving support method in the first embodiment.

(Step S1) First, the own vehicle position estimation unit 2 estimates (calculates) the own vehicle position.
(Step S3) Next, the route setting section 3 calculates a road to be traveled from the position of the vehicle on the map to the destination.
(Step S5) Next, the route setting unit 3 calculates the lane along which the vehicle is to travel to the destination from the calculated road, and sets the calculated lane as the route RT on the map.

FIG. 3 shows a portion of the map. The route RT is the route RT set on the map. The own vehicle runs along this route RT. First, the own vehicle runs on the right lane of the road R1 with two lanes on one side toward the intersection K1 and turns right at the intersection K1. Then, drive on the road R2 with one lane on one side and turn left at the intersection K2. Subsequently, the own vehicle travels in the left lane of the two-lane road R3 and changes lanes to the right lane. Subsequently, the own vehicle turns right at the intersection K3 and travels in the left lane of the two-lane road R4. Route RT does not include intersection K4, road R5, or road R6.

Also, as described above, the map is not always accurate, and may include discontinuous roads and lanes, such as road R7. That is, the continuity of the lane boundaries is not compensated for on the road R7 or the like.

Returning to FIG. 2, the description is continued.
(Step S7) The target lane identification unit 4 identifies the target lane in which the vehicle is currently traveling from the route RT set on the map based on the vehicle position, and acquires lane boundary data of the target lane from the map. Since the own vehicle position is updated as the own vehicle travels, the latest own vehicle position is used here. Then, the position of the own vehicle on the map is obtained from the position of the own vehicle, and the lane of the position of the own vehicle is selected from the route RT and set as the target lane. Then, the lane boundary data of the target lane is acquired from the map.

(Step S9) Next, the road structure data acquisition unit 5 (sensor) detects the structure around the vehicle. For example, the surroundings of the own vehicle are photographed by a camera mounted on the own vehicle. Alternatively, structures around the vehicle are irradiated with laser radar or millimeter wave radar.

(Step S11) Next, the road structure data acquisition unit 5 acquires road structure data around the own vehicle based on the detection result of step S9. The road structure data includes road structure data in the traveling direction of the vehicle and in the opposite direction (hereinafter collectively referred to as the front-back direction), and road structure data in the right and left directions in the traveling direction of the vehicle (hereinafter collectively referred to as the left-right direction). ) road structure data. Both the front-rear direction and the left-right direction are directions toward the outside of the host vehicle. Specifically, the road structure data acquiring unit 5 calculates coordinate data (road structure data) of the surrounding structures in a relative coordinate system with the position of the vehicle as the origin.

The road structure data in the left-right direction of the own vehicle includes coordinate data of lane boundaries in the left-right direction of the own vehicle, that is, lane boundary information. The lane boundary information is coordinate data of lane boundary lines, guardrails and curbs provided at the boundaries between roads and sidewalks, roadside strips, and the like. In addition, the road structure data in the horizontal direction includes attribute information such as the color of the lane boundary line.

The road structure data in the front-rear direction of the own vehicle includes coordinate data of stop lines indicating positions where the vehicle should stop in the front-rear direction of the own vehicle, traffic signals at intersections, and the like.

(Step S13) Next, using the route RT, the road structure data acquisition unit 5 converts the road structure data calculated in the relative coordinate system into road structure data in the map coordinate system used in the map. Here, the travel position of the own vehicle within the lane is specified from the road structure data, and the road structure data is converted into the road structure data of the map coordinate system using the travel position and the route RT. Since the road structure data acquisition unit 5 acquires the coordinate data of the lane boundary line, that is, the road structure data, based on the position of the vehicle, it means that the running position of the vehicle within the lane is specified.

Then, assuming that the position of the own vehicle on the route RT is the same as the traveling position, the road structure data is converted into the road structure data of the map coordinate system. Therefore, using the travel position and the route RT, the road structure data is converted into the road structure data of the map coordinate system. Similarly, the road structure data acquisition unit 5 converts the lane boundary information included in the road structure data into lane boundary information in the map coordinate system.

FIG. 4 shows the route actually traveled (referred to as actual route T). Also, FIG. 4 shows the road structure data converted into a structure (lane boundary lines, etc.) for convenience. As shown in FIG. 4, road structure data is obtained for roads R1, R2, R3, R4 and intersections K1, K2, K3 included in the route T. FIG. On the other hand, road structure data cannot be obtained for the intersection K4, the road R5, and the road R6 in FIG. 3, which are not included in the route T.

Returning to FIG. 2, the description is continued.
(Step S15) Next, the integrated road structure data generator 6 determines whether or not it is time to integrate the road structure data and the lane boundary data. This determination will be explained later. If it is determined that it is not the time to integrate (NO), the process returns to step S7.

(Step S17) When the integrated road structure data generator 6 determines that it is time to integrate (S15: YES), the road structure data obtained in step S13 and the lane boundary data obtained in step S7 are integrated. to generate integrated road structure data. Integrated road structure data may be generated separately from the map, or may be generated in the map. As a concrete example of integrating road structure data and lane boundary data, when generating integrated road structure data on a map, the lane boundary data at the position corresponding to the road structure data is replaced with the road structure data to create the integrated road structure data. can be generated. Further, when the integrated road structure data is generated in the map, it is preferable to update the lane attributes (lane width, etc.) stored in the map based on the road structure data.

(Step S19) Next, the driving support unit 7 uses the integrated road structure data to support driving of the own vehicle, and the process ends.
For example, the driving support unit 7 uses the integrated road structure data to calculate the travel locus of the own vehicle, and controls the own vehicle to travel along the travel locus. The travel support unit 7 controls the accelerator, brake, and steering of the vehicle to cause the vehicle to travel along the travel locus.

Here, a case where there is a predetermined divergence between the road structure data and the lane boundary data will be described. For example, if the positional deviation between the road structure data and the lane boundary data is greater than or equal to a predetermined value (for example, 50 cm), it is determined that there is a deviation. Further, if the deviation of the representative point, the deviation of the center point, or the average value of the deviation of each point is equal to or greater than a predetermined value, it may be determined that there is deviation.

If there is a deviation, there is a possibility that the lane indicated by the road structure data and the lane indicated by the lane boundary data are different. Therefore, integrating these may reduce the accuracy of the integrated road structure data. If this integrated road structure data is used to assist the vehicle in driving, the possibility of lane deviation increases.

Since the road structure data is obtained by actual detection, it can be estimated to be more accurate than the lane boundary data. Therefore, when there is a predetermined discrepancy between the road structure data and the lane boundary data, the driving support unit 7 uses the road structure data instead of the integrated road structure data to control the driving of the own vehicle as described above. Assist. As a result, lane deviation can be prevented.

By generating the integrated road structure data on the map, the accuracy of the map can be improved. As a result, when the position of the own vehicle is indicated on the map, the possibility that the own vehicle is shown at a position different from the actual position is reduced. That is, it is possible to improve the accuracy of the position of the host vehicle shown on the map.

Also, by driving the own vehicle on discontinuous roads and lanes like road R7 in FIG. can. That is, it is possible to compensate for the continuity of lane boundaries.

(Comparative example)
Here, as a comparative example, a case where the route RT is the route RT at the road level will be cited and compared with the first embodiment.

As shown in FIG. 5, it is assumed that the own vehicle 100 is traveling on a one-way four-lane road. In the case of the comparative example, since the lanes L1 to L4 that make up the road cannot be distinguished, the vehicle 100 actually runs in the lane L2, but the vehicle 100 runs in a different lane on the map. likely to become The greater the number of lanes, the greater the likelihood of such lane misidentification.

On the other hand, in the first embodiment, it is possible to specify the target lane in which the vehicle 100 is scheduled to travel to the destination, so the possibility of traveling in a different lane can be reduced. Even if the number of lanes is large, one of them (target lane) can be specified, so the possibility of lane misrecognition can be reduced regardless of the number of lanes.

As shown in FIG. 6, when the route RT is set on the road R13, which is actually a straight road but has a crank shape due to the inaccuracy of the map, the route RT is curved. Therefore, by setting a travel locus P for the straight road R12 and causing the vehicle to travel along the travel locus P, the own vehicle can travel straight on the straight road R12.

In this way, the road structure data around the own vehicle and the lane boundary data on the road are integrated to generate integrated road structure data, and the integrated road structure data is used to calculate the travel trajectory of the vehicle. By controlling the own vehicle so that it travels along the trajectory, it is possible to calculate an appropriate travel trajectory in line with the actual travel situation, and to perform stable travel assistance.

As described above, according to the first embodiment, the route RT indicating the lane along which the vehicle will travel to the destination is set on the map (S5), and the road structure data detected by the sensor of the vehicle is set. is transformed into the map coordinate system using the route RT (S13). Then, the road structure data converted into the map coordinate system and the lane boundary data on the map are integrated to generate integrated road structure data (S17). (S19).
The own vehicle is scheduled to travel along the set route RT. In particular, when the self-vehicle is performing autonomous driving in which at least the steering control is performed autonomously among the travel control of the vehicle, it is planned to travel along the set route RT. The road structure data detected by the sensor of the own vehicle traveling along the route RT is converted into the map coordinate system using the route RT. Thereby, the road structure data detected by the sensor of the own vehicle and the lane boundary data on the map can be accurately integrated. That is, it is possible to reduce the error between the HD map and the actual road structure.

Generally, when detecting road structure data with a sensor, only short-distance data can be detected, but the road structure can be detected with high accuracy. On the other hand, map data may contain errors, but the road structure at a position distant from the vehicle can be detected from the map. By integrating these, it is possible to generate integrated data (integrated road structure data) that makes up for the merits and demerits. Therefore, when the travel locus is calculated, it is possible to make a travel plan with high accuracy at a position close to the own vehicle and to a distant position. In other words, it is possible to calculate a travel locus that straddles between the road structure data around the vehicle and the lane boundary data far from the vehicle.

Incidentally, at a position far from the own vehicle, there is a little more time to travel than at a position at a short distance from the own vehicle. Even if the accuracy is low, it is possible to make a driving plan with an eye on the distant position by making a driving plan to a position far away from the own vehicle, and to suppress the sense of discomfort that the vehicle's behavior suddenly changes. be able to.

In addition, the sensor (5) acquires road structure data including lane boundary information around the vehicle (S11), identifies the driving position within the lane from the road structure data, and uses the driving position and the route RT, The road structure data (including lane boundary information) is converted into the map coordinate system (S13). Therefore, the lane boundary information in the integrated road structure data is also accurate, and the continuity of lane boundary lines can be compensated.

Also, using the integrated road structure data, the travel locus of the own vehicle is calculated, and the own vehicle is controlled so as to travel along the travel locus (S19). Thus, as described above, it is possible to make a travel plan to a position distant from the own vehicle with high accuracy, and to make the own vehicle travel according to the travel plan.

Further, by replacing the lane boundary data at the position corresponding to the road structure data with the road structure data to generate the integrated road structure data (S17), the accuracy of the map can be improved. Also, it is possible to compensate for the continuity of lane boundaries on the map.

(Embodiment in the first embodiment)
Here, an embodiment of the first embodiment will be described.

(Embodiment A1)
In embodiment A1, integrated road structure data is generated with priority given to the lateral direction of the vehicle over the longitudinal direction. For example, integrated road structure data is generated by integrating road structure data and lane boundary data only in the lateral direction of the own vehicle. For example, the integrated road structure data in the longitudinal direction is generated only when a predetermined condition is satisfied. For example, the conditions for generating the lateral integrated road structure data are made looser than the conditions for generating the longitudinal integrated road structure data.

In the first embodiment, the lane boundary information (coordinate data of the lane boundary line, etc.) is included in the road structure data in the lateral direction of the vehicle, since it is possible to specify the target lane in which the vehicle is to travel.

On the other hand, the road structure data in the longitudinal direction of the host vehicle includes stop line data. A stop line is less likely to be detected while the vehicle is running than a lane boundary line. In other words, the stop line may not be detected unless the vehicle approaches by about 10 m.

Compared to stop lines and the like, lane boundaries and the like are more likely to be detected during driving. In other words, the lane boundary lines and the like always exist in the left and right direction of the own vehicle while traveling in the lane. Therefore, by prioritizing the generation of the integrated road structure data using the road structure data in the lateral direction of the own vehicle, the accuracy of the integrated road structure data and the map can be further improved. Also, by giving priority to the left-right direction of the own vehicle, it is possible to compensate for the continuity of the lane boundaries.

In addition, by preferentially generating the integrated road structure data in the horizontal direction on the map, it is possible to improve the accuracy of the map also in the longitudinal direction of the own vehicle. Specifically, the yaw angle (turning angle) of the vehicle changes by approximately 90 degrees when the vehicle turns left or right while traveling along the route RT. As a result, the left-right direction and the front-rear direction of the own vehicle before turning change to the front-rear direction and the left-right direction of the own vehicle after turning. Therefore, by creating integrated road structure data in the left-right direction on the map throughout the turn, it is possible to improve the accuracy of the map not only in the left-right direction but also in the front-rear direction.

As shown in FIG. 7, it is assumed that the position of road R11 is shifted to the left of the vehicle traveling on road R11 on the map. However, by generating integrated road structure data in the left-right direction on the map, the road R11 is updated to the road R111 in the correct position. As a result, the position where the road R12 that intersects the road R111 connects to the intersection is correct. That is, it is possible to improve the accuracy of the map in the longitudinal direction of the vehicle traveling on the road R12.

(Embodiment A2)
In embodiment A2, it is determined whether or not the road structure data in the left-right direction of the own vehicle has been acquired, and when the road structure data in the left-right direction of the own vehicle has been acquired, the road structure data converted into the map coordinate system and , with lane boundary data on the map. In other words, if the road structure data in the lateral direction of the own vehicle has not been acquired, the integrated road structure data is not generated. As a result, the processing load required for generating integrated road structure data can be reduced.

(Embodiment A3)
In embodiment A3, a position for generating integrated road structure data in the longitudinal direction of the vehicle is set on the route RT, and integrated road structure data in the longitudinal direction is generated when the vehicle reaches the set position. do. For example, when the own vehicle reaches 10 meters before the stop line, integrated road structure data is generated. Also, the arrival may be determined three seconds before stopping, or the arrival may be determined by other arrival determination methods.
For example, a position such as a stop line is set in advance on the route RT. Then, the map is read, and when the own vehicle reaches a position such as a preset stop line, integrated road structure data in the longitudinal direction of the vehicle is generated.

As shown in FIG. 8, for example, when the own vehicle 100 stops before the stop line ST, the road structure data in the longitudinal direction of the own vehicle 100 is acquired with respect to the stop line ST, and the longitudinal direction of the own vehicle 100 is integrated. Generate road structure data. As a result, the stop line ST does not always need to be detected while the vehicle is running, and the position of the stop line ST on the map can be correctly corrected with the minimum necessary processing. Note that the position for generating integrated road structure data in the front-rear direction may be a position (position of a pedestrian crossing, etc.) in front of a right or left turn, other than the stop line.

(Embodiment A4)
As embodiment A4, when the vehicle speed of the own vehicle is equal to or higher than a predetermined value, the integrated road structure data of the left and right direction of the own vehicle is generated, and when the vehicle speed of the own vehicle is less than the predetermined value, Generate longitudinal integrated road structure data. Note that when the vehicle speed is equal to or higher than a predetermined value, the integrated road structure data in the horizontal direction is generated, and when the vehicle speed is less than the predetermined value, the generation of the integrated road structure data need not be specified.

The distance to the vehicle, such as a lane boundary line in the left-right direction of the vehicle, rarely varies depending on the position in the traveling direction of the vehicle. In other words, even if the own vehicle is running, the distance to the lane boundary is almost the same unless the lane is changed. Therefore, even if the vehicle speed is relatively high, the lane boundary can be detected.

On the other hand, the distance from the own vehicle often varies depending on the position in the traveling direction of the vehicle, such as a stop line in the front-rear direction. In other words, when the vehicle travels, the distance between the vehicle and the stop line changes greatly, so when the vehicle speed is high, it is difficult to detect the stop line.

Therefore, as described above, it is preferable to generate the integrated road structure data in the horizontal direction when the vehicle speed is equal to or higher than the predetermined value, and to generate the integrated road structure data in the horizontal direction and the longitudinal direction when the vehicle speed is less than the predetermined value. Thereby, the accuracy of the integrated road structure data can be further improved.

(Embodiment A5)
In embodiment A5, the route setting unit 3 sets the route RT so that the number of lane changes to the destination is minimized. For example, when estimating the position of the own vehicle using a running locus model (odometry), the position of the own vehicle is more accurate when the own vehicle is traveling straight ahead than when the own vehicle is turning. Therefore, when the number of lane changes is small, the time for the vehicle to go straight increases, and the vehicle position becomes accurate. For example, an inaccurate ego vehicle position will reduce the accuracy of the integrated road structure data. Therefore, by minimizing the number of changes, the position of the vehicle becomes accurate, and the accuracy of the integrated road structure data can be improved.

(Embodiment A6)
In embodiment A6, when the road on which the vehicle is to be traveled to the destination includes roads with a plurality of lanes, the route setting unit 3 selects a lane on the low speed side rather than a lane on the high speed side to set the route RT. . For example, when estimating the position of the vehicle using GPS, GPS signals are transmitted at regular intervals, so the distance between GPS signal reception points increases in the high-speed lane where the vehicle speed is high. In other words, the number of points where the correct vehicle position can be obtained is reduced, making it difficult to improve the accuracy of the integrated road structure data. On the other hand, in the low-speed lane where the vehicle speed is low, the distance between GPS signal reception points is short. In other words, the number of points where the correct vehicle position can be obtained increases, and the accuracy of the integrated road structure data can be improved. That is, by setting the route RT with priority given to lanes on the low-speed side, it is possible to improve the accuracy of the integrated road structure data.

In addition, as described in embodiment A4, at low speeds, integrated road structure data can be generated for both the longitudinal direction and the lateral direction of the own vehicle. Accuracy can be improved

(About step S15)
Here, step S15 will be described. In the following description, a plurality of criteria for determining whether to return to step S7 or proceed to step S17 will be presented. Either one may be implemented, or multiple criteria may be combined.

(Determination Criteria B1)
In step S15, if the vehicle is running through an intersection stored in the map, the process returns to step S7, and if the vehicle is not running through the intersection, the process proceeds to step S17. Therefore, integrated road structure data is not generated when the host vehicle is traveling through an intersection.

Generating integrated road structure data can reduce the accuracy of integrated road structure data, as intersections often lack suitable structures (such as lane boundaries) to obtain accurate integrated road structure data. There is Therefore, by not generating the integrated road structure data at intersections, it is possible to prevent a decrease in the accuracy of the integrated road structure data. Whether or not the vehicle is traveling through an intersection may be determined from the map and the route RT set on the map, or may be determined from the road structure data. In addition, by not generating integrated road structure data at intersections, the processing load and system resources required for generating integrated road structure data can be reduced.

(Determination Criteria B2)
In step S15, if the vehicle is not traveling in the lane stored in the map, the process returns to step S7, and if the vehicle is traveling in the lane stored in the map, the process proceeds to step S17. Therefore, when the own vehicle is not traveling in the lane stored in the map, integrated road structure data is not generated.

Vacant lots are places other than lanes where vehicles can run. In vacant lots, like intersections, there are few structures suitable for obtaining accurate integrated road structure data, so integrated road structure data is generated. do not do. Therefore, it is possible to prevent a decrease in accuracy in the integrated road structure data. Whether or not the vehicle is traveling on a vacant lot or the like may be determined from the map and the route RT set on the map, or may be determined from the road structure data. Also, by not generating the integrated road structure data, the processing load and system resources required for generating the integrated road structure data can be reduced.

(Determination Criteria B3)
In step S15, if the host vehicle has changed lanes, the process returns to step S7, and if the vehicle has not changed lanes, the process proceeds to step S17. Therefore, when changing lanes, integrated road structure data is not generated.

For example, the position of the lane change is stored in the route RT set on the map, and the integrated road structure data is not generated at the position of the lane change. In practice, it is not possible to change the lane at the lane change position, and the timing of the lane change may be delayed. Alternatively, the integrated road structure data is not generated even when changing lanes during automated driving.

As described in embodiment A5, the vehicle position is more accurate when the vehicle is traveling straight without changing lanes than when the vehicle is turning after changing lanes. Therefore, by not generating the integrated road structure data when changing lanes, it is possible to prevent a decrease in the accuracy of the integrated road structure data. Whether or not the vehicle is changing lanes may be determined from the map and the route RT set on the map, or may be determined from the road structure data. Also, by not generating the integrated road structure data, the processing load and system resources required for generating the integrated road structure data can be reduced.

(Determination Criteria B4)
In step S15, if the host vehicle is automatically driving and the driver of the host vehicle is overriding, the process returns to step S7, and if not, the process proceeds to step S17. Therefore, integrated road structure data is not generated when the driver of the self-vehicle during automatic driving is overriding.

Override means that the driver operates the accelerator, brake, or steering during automatic driving. Overriding during automatic driving may cause an error in the position of the vehicle. Therefore, by not generating the integrated road structure data during the override, it is possible to prevent a decrease in the accuracy of the integrated road structure data. Whether or not there is an override during automatic operation can be determined from various flags used in automatic operation control. Also, by not generating the integrated road structure data, the processing load and system resources required for generating the integrated road structure data can be reduced.

(Second embodiment)
Next, a second embodiment will be described. Here, the differences from the first embodiment will be mainly described, and descriptions of the same or similar contents will be omitted because they are redundant.

As shown in FIG. 9, the driving support system of the second embodiment is obtained by adding an own vehicle position evaluation unit 8 to the driving support system of the first embodiment. Further, the own vehicle position estimation unit 2 can estimate the own vehicle position (second own vehicle position) using the running locus model, and can also estimate the own vehicle position (first own vehicle position) by GPS. ing.

The own vehicle position evaluation unit 8 determines whether or not the turning state of the own vehicle (details will be described later) satisfies a predetermined condition. If the turning state satisfies the conditions, it is determined that the vehicle position estimated using the traveling locus model (second vehicle position) is more accurate than the vehicle position estimated by GPS (first vehicle position). do. On the other hand, if the turning state does not satisfy the condition, it is determined that the first vehicle position estimated by GPS is more accurate than the second vehicle position estimated by using the running locus model.

FIG. 10 is a flow chart showing the process performed in the second embodiment in place of step S1 in FIG. 2. After this process is finished, the process from step S3 onward in FIG. 2 is executed.

(Steps S21 and S23) First, the own vehicle position estimator 2 receives GPS signals from GPS satellites, and estimates (calculates) the own vehicle position with emphasis on the position of the own vehicle based on the GPS signals. For example, the position of the vehicle based on GPS signals is assumed to be the position of the vehicle. Hereinafter, this host vehicle position is referred to as a first host vehicle position P1.

(Step S25) Further, the own vehicle position estimating section 2 obtains the position of the own vehicle using the running locus model, and estimates the own vehicle position by placing importance on this position. For example, the position of the own vehicle obtained using the running locus model is set as the own vehicle position. Hereinafter, this host vehicle position will be referred to as a second host vehicle position P2.

As in the first embodiment, the own vehicle position estimating unit 2 repeats the estimation of the first own vehicle position P1 and the second own vehicle position P2 as time elapses. Update position.

(Step S27) Next, the own vehicle position evaluation unit 8 determines whether or not the turning state of the own vehicle satisfies a predetermined condition. The determination in step S27 will be described later.

(Step S29) When the turning state satisfies the condition, the vehicle position evaluation unit 8 estimates the second vehicle position P2 estimated using the running locus model more accurately than the first vehicle position P1 estimated using GPS. is high, the second host vehicle position P2 is selected, and the process ends.

(Step S31) On the other hand, if the turning state does not satisfy the conditions, the vehicle position evaluation unit 8 determines that the first vehicle position P1 estimated by GPS is the second vehicle position estimated by using the running locus model. It determines that the accuracy is higher than P2, selects the first host vehicle position P1, and terminates the process.

When this process is finished, the process proceeds to step S3 in FIG. 2. As the own vehicle position, the own vehicle position (first own vehicle position P1 or second own vehicle position P2) selected in step S29 or step S31 is used. .
That is, in step S13, the road structure data is converted into the map coordinate system based on the selected vehicle position, and in step S17, integrated road structure data is generated from the road structure data converted based on the selected vehicle position. do.

(Description of first vehicle position P1 and second vehicle position P2)
Here, the first vehicle position P1 estimated by GPS and the second vehicle position P2 estimated by using the running locus model will be described.

As shown in FIG. 11, the first host vehicle position P1 changes as the host vehicle travels. Similarly, the second own vehicle position P2 also changes as the own vehicle travels. Since the first vehicle position P1 and the second vehicle position P2 contain an error, they do not necessarily match as shown in FIG.

For example, even if the GPS positioning is good and the accuracy of the first vehicle position P1 is high, the friction coefficient of the road surface is low and the vehicle slips, so the accuracy of the second vehicle position P2 is reduced. becomes lower, and the first host vehicle position P1 and the second host vehicle position P2 are different.

As shown in FIG. 12, when the own vehicle 100 slips, the own vehicle 100 turns around the center of gravity 101 of the own vehicle. Therefore, the state in which the vehicle 100 turns (referred to as a turning state) changes according to the degree of slip.

In step S27 of FIG. 10, when the state of positioning by GPS is better than a predetermined state, the process proceeds to step S29 or S31 depending on the turning state of the host vehicle. The positioning state can be determined, for example, based on the reception level of GPS signals.

(When the state of positioning by GPS is better than the predetermined state)
In step S27, for example, the turning amount S of the host vehicle is used as the turning state.
As shown in FIG. 13, when the turning amount S is equal to or less than a predetermined turning amount threshold value Sth (S≤Sth), the turning state satisfies the condition. Therefore, in this case, the second own vehicle position P2 estimated by the running locus model is selected. Conversely, if the turning amount S is greater than the turning amount threshold value Sth (S>Sth), the turning state does not satisfy the condition. Therefore, in this case, the first host vehicle position P1 estimated by GPS is selected.

Further, in step S27, for example, the friction coefficient μ of the road surface on which the host vehicle travels can be used as the turning state. As shown in FIG. 14, when the friction coefficient μ is greater than a predetermined friction coefficient threshold value μth (μ>μth), the turning state satisfies the condition. Therefore, in this case, the second own vehicle position P2 estimated by the running locus model is selected. Conversely, if the friction coefficient μ is less than or equal to the friction coefficient threshold value μth (μ≦μth), the turning state does not satisfy the condition. Therefore, in this case, the first host vehicle position P1 estimated by GPS is selected.

For example, the coefficient of friction μ is a flag indicating whether or not the anti-lock brake system of the vehicle is operating, a flag indicating whether or not the system that controls slippage of the vehicle is operating, and whether or not the system that controls the traction of the vehicle is operating. flag indicating whether the wiper of the own vehicle is operating, the recognition result of the road surface by the camera of the own vehicle, or probe information obtained from another vehicle other than the own vehicle. It is possible.

(When the state of positioning by GPS is the same as or worse than the predetermined state)
On the other hand, if the state of positioning by GPS is the same as or worse than the predetermined state, the process proceeds to step S29. Therefore, in this case, the second own vehicle position P2 estimated by the running locus model is selected.

Even if the state of positioning by GPS is the same as or worse than the predetermined state, as shown in FIG. If the turning amount S is equal to or less than the threshold value Tth (T≦Tth) and the turning amount S is greater than the turning amount threshold value Sth (S>Sth), the process proceeds to step S31 exceptionally. Therefore, in this case, the first host vehicle position P1 estimated by GPS is selected.

That is, when the elapsed time T is equal to or less than the elapsed time threshold Tth (T≤Tth), the accuracy of the first host vehicle position P1 is not extremely low. On the other hand, when the turning amount S is greater than the turning amount threshold value Sth (S>Sth), the accuracy of the second vehicle position P2 is relatively low, so the first vehicle position P1 with relatively high accuracy is selected.

Further, even if the state of positioning by GPS is the same as or worse than the predetermined state, as shown in FIG. If it is equal to or less than the threshold value Tth (T≦Tth) and the friction coefficient μ is equal to or less than the friction coefficient μth (μ≦μth), the process proceeds to step S31 exceptionally. Therefore, in this case, the first host vehicle position P1 estimated by GPS is selected.

That is, when the elapsed time T is equal to or less than the elapsed time threshold Tth, the accuracy of the first host vehicle position P1 is not extremely low. On the other hand, when the coefficient of friction μ is equal to or less than the coefficient of friction μth, the accuracy of the second vehicle position P2 is relatively low, so the first vehicle position P1 with relatively high accuracy is selected. Therefore, the first host vehicle position P1 estimated by GPS with relatively high accuracy can be selected, and the accuracy of the integrated road structure data can be improved.

As shown in FIG. 17A, in the second embodiment, when the lane boundary lines D11 and D12 are detected using the highly accurate first vehicle position P1, the accuracy of the position of the lane boundary lines D11 and D12 is high. .

On the other hand, as shown in FIG. 17B, when the lane boundary lines D21 and D22 are detected using the second own vehicle position P2 with low accuracy, the accuracy of the position of the lane boundary lines D21 and D22 is also low.

In the second embodiment, the second vehicle position P2 with low accuracy is not used, and the first vehicle position P1 with high accuracy is used. Therefore, as shown in FIG. lane boundary lines D31 and D32. Therefore, by generating the integrated road structure data in step S17, the accuracy of the integrated road structure data can be improved more than when using the second own vehicle position P2 with low accuracy.

Conversely, when the second vehicle position P2 is more accurate, the second vehicle position P2 is used, so the lane boundary lines D21 and D22 are very close to the actual lane boundary lines D1 and D2. Become. Therefore, the accuracy of the integrated road structure data can be improved as compared with the case of using the first host vehicle position P1 with low accuracy.

Also, when the first vehicle position P1 or the second vehicle position P2 is selected, it is stored which of the first vehicle position P1 or the second vehicle position P2 is selected in association with the position of the vehicle. do. Then, when the own vehicle position is detected again at that position, the stored own vehicle position is detected. Therefore, it is unnecessary to determine the turning state of the own vehicle, and the processing load and system resources required for generating integrated road structure data can be reduced.

(About step S15)
Here, step S15 in the second embodiment will be described. In the following description, a plurality of determination criteria will be shown for determining whether to return to step S7 or proceed to step S17. Either one may be performed, or multiple criteria may be combined

(Judgment Criteria C1)
In step S15, it is determined whether or not it is time to generate integrated road structure data in step S17, based on the accuracy of the own vehicle position. If it is not the generation timing, the process returns to step S7, and if it is the generation timing, the process proceeds to step S17.

For example, when the accuracy of the vehicle position is higher than a predetermined accuracy, the integrated road structure data is generated, and when the accuracy of the vehicle position is less than the predetermined accuracy, it is not generated. Therefore, it is possible to prevent the accuracy of the integrated road structure data from deteriorating due to the use of the vehicle position with low accuracy.

(Determination Criteria C2)
In step S15, based on the recognition accuracy of the lane boundary line information included in the road structure data acquired by the road structure data acquisition unit 5, it is determined whether it is time to generate integrated road structure data. If it is not the generation timing, the process returns to step S7, and if it is the generation timing, the process proceeds to step S17.

For example, when a camera is used to detect lane boundaries, it is often the case that the visibility is good in the daytime on fine weather and the recognition accuracy is higher than a predetermined accuracy. In this case, the integrated road structure data is generated. On the other hand, visibility is poor in rainy weather or at night, and recognition accuracy is often lower than a predetermined accuracy. In this case, integrated road structure data is not generated. Therefore, it is possible to prevent the accuracy of the integrated road structure data from deteriorating due to the use of lane boundaries with low recognition accuracy.

(Judgment Criteria C3)
Further, when the accuracy of the own vehicle position is equal to or less than a predetermined accuracy, it is possible to determine whether to generate the integrated road structure data based on the number of lanes of the road on which the vehicle travels. For example, the number of lanes is stored in the map and the number of lanes is obtained from the map.

As shown in FIG. 18A, when the own vehicle 100 is traveling on a road with two or more lanes, when the own vehicle recognizes the lane boundary line, as shown in FIG. In some cases, only the lane boundary lines D41 and D42 are recognized. Therefore, the lane boundary lines D41 and D42 cannot identify any of the actual lane boundary lines D41, D42 and D43 in FIG. 18A. In other words, it is not possible to accurately identify the lane in which the vehicle is traveling only from the lane boundary line, and there is a possibility that the lane may be misidentified, so integrated road structure data is not generated. That is, it returns to step S7 assuming that it is not the time to generate. For example, lane misidentification can reduce the accuracy of the integrated road structure data.

On the other hand, if the number of lanes is 1, the vehicle is traveling in the only lane, so integrated road structure data is generated. That is, it is determined that it is time to generate the data, and the process proceeds to step S17. Therefore, it is possible to prevent the accuracy of the integrated road structure data from deteriorating due to misidentification of lanes.

(Criteria C4 and C5)
Further, when the accuracy of the own vehicle position is equal to or less than a predetermined accuracy, it is possible to determine whether to generate integrated road structure data based on attributes (attribute data) of lanes and lane boundary lines.

(Determination Criteria C4)
For example, as shown in FIG. 19, assume that a road R15 on which the vehicle 100 travels is a two-lane road (L41, L42) facing each other. The own vehicle 100 runs on the lane L41, and the other vehicle 120 runs on the lane L42 opposite to the lane L41. Note that the traveling direction of the vehicle (lane attribute) in the lanes L41 and L42 can be obtained from the map.
Since the own vehicle 100 is substantially traveling on a one-lane (L41) road, the road detected by the own vehicle 100 (lane L41 ) and the lane boundary data for the lane (lane L41) of the two lanes in the same direction as the traveling direction of the own vehicle on the map are integrated to generate integrated road structure data. That is, if such a condition is satisfied, the process proceeds to step S17, and if not satisfied, the process returns to step S7. The traveling direction of the own vehicle can be obtained from the transition of the position of the own vehicle over time.

Therefore, even if the accuracy of the own vehicle position is less than the predetermined accuracy, the accuracy of the integrated road structure data can be improved.

(Judgment Criteria C5)
Also, as shown in FIG. 20, assume that the road R16 on which the vehicle 100 travels is a road having a plurality of lanes (L51, L52) in the same direction. The host vehicle 100 is traveling on the lane L51, and the other vehicle 120 is traveling on the lane L52. Note that the traveling direction of the vehicle (lane attribute) in the lanes L51 and L52 can be obtained from the map. It can be determined from the map and the position of the vehicle that the vehicle 100 is traveling on a road having a plurality of lanes (L51, L52) in the same direction. The traveling direction of the own vehicle can be obtained from the transition of the position of the own vehicle over time.

In the case of FIG. 20, integrated road structure data is generated based on the attribute of the lane boundary line and the traveling direction of the own vehicle. For example, the color (attribute) of the lane boundary line D51 on the boundary between the lanes L51 and L52 is different from the color (attribute) of the left lane boundary line D52 of the lane L51, and the color (attribute) of the lane boundary line D53 on the right side of the lane L52. (attribute) may also be different. In this case, road structure data for the road (lane L51) detected by the vehicle 100 and lane boundary data for the lane L51 on the map are generated based on the traveling direction of the vehicle and the color of the lane boundary line (attribute data). Integrate to generate integrated road structure data. This is because the colors of the left and right lane boundary lines are different, so the possibility of erroneously identifying lane L51 as lane L52 is low. That is, if such a condition is satisfied, the process proceeds to step S17, and if not satisfied, the process returns to step S7.

Therefore, even if the accuracy of the own vehicle position is less than the predetermined accuracy, the accuracy of the integrated road structure data can be improved.

As described above, according to the second embodiment, the position of the vehicle is estimated using the traveling locus model, the position of the vehicle is estimated using the GPS, and the position of the vehicle with the higher accuracy of the two vehicle positions is determined. A position is selected, and based on the selected vehicle position, the road structure data is converted into a map coordinate system to generate integrated road structure data. Therefore, the position of the vehicle with the higher accuracy can be selected, and the accuracy of the integrated road structure data can be improved.

If there is an error in the vehicle's position, the position of the preceding vehicle cannot be detected accurately, that is, it is shifted by one lane and merged with the map, resulting in accurate control along the actual road. cannot be executed. Therefore, in the second embodiment, as described above, by selecting the position of the vehicle with the higher accuracy, the accuracy of the integrated road structure data can be improved. be able to run

Further, when the state of positioning by GPS is better than a predetermined state, which of the first vehicle position P1 and the second vehicle position P2 is more accurate depends on the situation. By selecting the vehicle position, the vehicle position with the higher accuracy can be selected and the accuracy of the integrated road structure data can be increased.

In addition, by using the amount of turning of the vehicle as the turning state as described above, it is possible to select a highly accurate position based on the amount of turning, thereby improving the accuracy of the integrated road structure data.

Further, when the turning amount is less than or equal to the predetermined turning amount threshold value (R≦Sth) as described above, the second host vehicle position estimated by the running locus model is selected, and the turning amount is greater than the predetermined turning amount threshold value ( If S>Sth), by selecting the first host vehicle position estimated by GPS, a highly accurate position can be selected from the turning amount, and the accuracy of the integrated road structure data can be improved.

In addition, by using the coefficient of friction of the road surface on which the vehicle travels as the turning state as described above, it is possible to select a highly accurate position based on the coefficient of friction, thereby increasing the accuracy of the integrated road structure data.

Further, when the friction coefficient is larger than the predetermined friction coefficient threshold value (μ>μth) as described above, the second host vehicle position estimated by the running locus model is selected, and the friction coefficient is equal to or less than the predetermined friction coefficient threshold value (μ .ltoreq..mu.th), by selecting the first host vehicle position estimated by GPS, it is possible to select a highly accurate position based on the coefficient of friction, thereby increasing the accuracy of the integrated road structure data.

Further, as described above, a flag indicating whether or not the anti-lock brake system of the vehicle is operating, a flag indicating whether or not the system controlling the slip of the vehicle is operating, and a flag indicating whether or not the system controlling the traction of the vehicle is operating. , whether or not the wiper of the vehicle is operating, the recognition result of the road surface by the vehicle camera, and the probe information obtained from other vehicles other than the vehicle are used to estimate the friction coefficient. A highly accurate position can be selected from the friction coefficient estimated from the vehicle information, and the accuracy of the integrated road structure data can be improved.

Further, as described above, the elapsed time from the time when the state of positioning by GPS is the same as or worse than the predetermined state and the state of positioning by GPS is better than the predetermined state is equal to or less than a predetermined elapsed time threshold (T ≤ Tth ) and the turning amount of the vehicle is greater than a predetermined turning amount threshold value (S>Sth), the first host vehicle position estimated by GPS is selected. Therefore, the first host vehicle position estimated by GPS with high accuracy can be selected, and the accuracy of integrated road structure data can be improved.

Further, as described above, the elapsed time from the time when the state of positioning by GPS is the same as or worse than the predetermined state and the state of positioning by GPS is better than the predetermined state is equal to or less than a predetermined elapsed time threshold (T ≤ Tth ) and the friction coefficient of the road surface on which the vehicle travels is equal to or less than a predetermined friction coefficient threshold value (μ≦μth), the first host vehicle position estimated by GPS is selected. Therefore, the first host vehicle position estimated by GPS with high accuracy can be selected, and the accuracy of integrated road structure data can be improved.

In addition, in this embodiment, the vehicle is equipped with a driving support device. However, it is possible to install a driving support device on a server device that can communicate with the own vehicle or on another vehicle that is not the own vehicle, and transmit and receive necessary information and instructions by communication between the server device or the other vehicle and the own vehicle. , the driving assistance method may be remotely applied to a similar self-vehicle. Communication between the server device and the own vehicle can be performed by wireless communication or road-to-vehicle communication. Communication between other vehicles and own vehicle can be performed by so-called inter-vehicle communication.

Although embodiments of the present invention have been described above, the statements and drawings forming part of this disclosure should not be construed as limiting the present invention. Various alternative embodiments, implementations and operational techniques will become apparent to those skilled in the art from this disclosure.

Each function shown in each of the above embodiments may be implemented by one or more processing circuits. Processing circuitry includes programmed processing devices, such as processing devices that include electrical circuitry. Processing devices also include devices such as application specific integrated circuits (ASICs) and conventional circuit components arranged to perform the functions described in the embodiments.

1 Map database (memory)
2 Own vehicle position estimation unit 3 Route setting unit 4 Target lane identification unit 5 Road structure data acquisition unit 6 Integrated road structure data generation unit 7 Driving support unit 8 Own vehicle position evaluation unit 100 Own vehicle (vehicle)
101 Center of gravity 120 Other vehicles D11, D12, D21, D22, D31, D32, D41, D42, D43, D51, D52, D53 Lane boundaries K1, K2, K3 Intersections L1 to L4, L41, L42, L51, L52
P Traveling locus P1 First vehicle position P2 Second vehicle position T Actual route actually traveled RT Route set on the map R1, R2, R3, R4, R5, R6, R7, R11, R12, R13, R15, R16, R111 Road S Turning amount Sth Turning amount threshold T Elapsed time Tth Elapsed time threshold μ Friction coefficient μth Friction coefficient threshold

Claims (24)

A sensor that acquires road structure data around the vehicle and a memory that stores a map having lane boundary data on the road, and uses the road structure data acquired by the sensor and the lane boundary data on the map. A driving support method for a driving support device for supporting driving of the vehicle,
setting a route on the map indicating a lane that the vehicle is scheduled to travel to the destination;
detecting the traveling position of the vehicle on the route set on the map;
converting the road structure data detected by the vehicle into a map coordinate system based on the travel position on the route;
generating integrated road structure data by integrating the road structure data converted into the map coordinate system and the lane boundary data on the map;
A travel assistance method, comprising: assisting travel of the vehicle using the integrated road structure data. Acquiring the road structure data including lane boundary information around the vehicle by the sensor, and identifying the traveling position of the vehicle in the lane from the road structure data,
The driving support method according to claim 1, wherein the road structure data is converted into a map coordinate system using the driving position within the lane and the driving position on the route. calculating a travel locus of the vehicle using the integrated road structure data;
The driving support method according to claim 1 or 2 , further comprising: controlling the vehicle to travel along the travel locus. The traveling according to any one of claims 1 to 3 , wherein the lane boundary data at a position corresponding to the road structure data is replaced with the road structure data to generate the integrated road structure data. how to help. The driving support method according to any one of claims 1 to 4 , wherein the integrated road structure data is generated with priority given to the lateral direction of the vehicle over the longitudinal direction of the vehicle. Determining whether or not road structure data in the left-right direction of the vehicle has been acquired,
3. When the road structure data in the lateral direction of the vehicle is obtained, the road structure data converted into the map coordinate system and the lane boundary data on the map are integrated. The driving support method described in . setting a position on the route for generating integrated road structure data in the longitudinal direction of the vehicle;
The driving support method according to any one of claims 1 to 6 , further comprising generating integrated road structure data in the longitudinal direction of the vehicle when the vehicle reaches the position. generating the integrated road structure data in the lateral direction of the vehicle when the vehicle speed of the vehicle is equal to or higher than a predetermined value;
The driving support method according to any one of claims 1 to 7 , wherein the integrated road structure data for the left-right direction and the front-rear direction of the vehicle is generated when the vehicle speed of the vehicle is less than a predetermined value. . The driving support method according to any one of claims 1 to 8 , wherein the route is set so that the number of lane changes to the destination is minimized. 10. When the road on which the vehicle is to be traveled to the destination includes a road with multiple lanes, the route is set by selecting a lane on the low speed side rather than a lane on the high speed side. The driving support method according to any one of 1. When there is a predetermined discrepancy between the road structure data detected by the vehicle and the lane boundary data on the map, driving of the vehicle is supported based on the road structure data on the map. The driving support method according to any one of claims 1 to 10 , characterized by: The driving support method according to any one of claims 1 to 11 , wherein the integrated road structure data is not generated when the vehicle is traveling through an intersection stored in the map. The driving support method according to any one of claims 1 to 12 , wherein the integrated road structure data is not generated when the vehicle is not traveling in the lane stored in the map. The driving support method according to any one of claims 1 to 13 , wherein the integrated road structure data is not generated when the vehicle is changing lanes. The driving support method according to any one of claims 1 to 14 , wherein the integrated road structure data is not generated when the driver of the vehicle under automatic driving is overriding. estimating a first own vehicle position, which is the position of the vehicle, by GPS;
estimating a second host vehicle position, which is the position of the vehicle, using a running trajectory model;
selecting the position of the vehicle with higher accuracy from the first vehicle position and the second vehicle position;
The driving support according to any one of claims 1 to 15 , wherein the road structure data is converted into a map coordinate system based on the position of the selected vehicle to generate the integrated road structure data. Method. 17. The method according to claim 16 , wherein whether or not to generate said integrated road structure data is determined based on the estimated positional accuracy of said vehicle or recognition accuracy of lane boundary information included in said road structure data. driving support method. 18. The driving support method according to claim 17 , wherein the integrated road structure data is not generated when the recognition accuracy is equal to or less than a predetermined accuracy. 18. The method according to claim 17 , wherein when the positional accuracy of the vehicle is equal to or less than a predetermined accuracy, whether or not to generate the integrated road structure data is determined based on the number of lanes of the road on which the vehicle travels. driving support method. The driving support method according to claim 19 , wherein the integrated road structure data is generated when the number of lanes is one. 18. The method according to claim 17 , wherein when the accuracy of the position of the vehicle is equal to or less than a predetermined accuracy, whether or not to generate the integrated road structure data is determined based on attributes of the lanes and lane boundary lines. Driving support method. When the road on which the vehicle travels is a two-lane road facing each other, the road structure data and the lane in the same direction as the traveling direction of the vehicle, regardless of the attribute of the lane boundary line. 22. The driving support method according to claim 21 , wherein the lane boundary data of the two are integrated to generate the integrated road structure data. When the road on which the vehicle travels has a plurality of lanes in the same direction, the integrated road structure data is generated based on the attribute of the lane boundary line and the traveling direction of the vehicle. The driving support method according to claim 21 . A sensor that acquires road structure data around the vehicle and a memory that stores a map having lane boundary data on the road, and uses the road structure data acquired by the sensor and the lane boundary data on the map. A driving support device that supports driving of the vehicle,
a route setting unit that sets a route on the map indicating a lane along which the vehicle is scheduled to travel to the destination;
a target lane identifying unit that detects the traveling position of the vehicle on the route set on the map;
The road structure data detected by the vehicle is converted into a map coordinate system based on the traveling position on the route , and the road structure data converted into the map coordinate system and the lane boundary data on the map are stored. an integrated road structure data generation unit that integrates and generates integrated road structure data;
A driving support device comprising: a driving support unit that supports driving of the vehicle using the integrated road structure data.

Images