JP2023146579A - Map generation device - Google Patents

Abstract

To accurately generate a map that includes position information for a division line even when the division line is discontinuous in the middle.SOLUTION: A map generation device 20 comprises: a camera 1a that detects an environment situation around a self vehicle; a map generation unit 17 that generates a map including position information for a division line defining a traveling traffic lane of the self vehicle on the basis of the detected environment situation; a division line estimation unit 21 that estimates a position of the division line which is predicted to be detected at a second point of time, on the basis of information on the environment situation detected at a first point of time; and a determination unit 22 that determines the presence/absence of a traffic lane change by the self vehicle on the basis of a difference between the estimated position of the division line and the position of the division line detected by the camera 1a at the second point of time. The map generation unit 17 generates a map on the basis of a result of this determination.SELECTED DRAWING: Figure 3

Description

The present invention relates to a map generation device that generates a map based on data indicating external conditions obtained while driving.

Conventionally, this type of device uses images captured by a camera mounted on a vehicle to recognize road markings and detect lane changes, thereby estimating the lane in which the vehicle is not driving. 2. Description of the Related Art A device that updates map data is known (for example, see Patent Document 1 mentioned above). The device described in Patent Document 1 detects a lane change when the area where the road marking line is located in the camera image gradually moves in the left-right direction.

Japanese Patent Application Publication No. 2007-241470

However, in the device described in Patent Document 1, road marking lines need to be detected continuously in order to detect a lane change. Therefore, if the lane markings are interrupted in the middle, a lane change cannot be detected, and it is difficult to accurately generate a map that includes positional information of the lane markings.

A map generation device that is one aspect of the present invention includes an external world detection unit that detects the external environment surrounding the host vehicle, and defines a driving lane for the host vehicle based on information about the external world situation detected by the external world detection unit. a map generation unit that generates a map including position information of the lane markings; and a map generation unit that generates a map that includes position information of the lane markings; a parking line estimating unit that estimates the position of the parking line that is predicted to be detected; the position of the parking line estimated by the parking line estimating unit; and the position of the partition line detected by the external world detection unit at a second time point; and a determination unit that determines whether or not the host vehicle changes lanes based on the difference between the two. The map generation section generates a map based on the determination result by the determination section.

According to the present invention, even in a case where the lane markings are interrupted in the middle, it is possible to accurately generate a map including positional information of the lane markings.

1 is a block diagram schematically showing the overall configuration of a vehicle control system having a map generation device according to an embodiment of the present invention. 1 is a diagram showing an example of a driving scene to which a map generation device according to an embodiment of the present invention is applied. FIG. 1 is a block diagram showing the main configuration of a map generation device according to an embodiment of the present invention. FIG. 3 is a diagram showing an example of an approximate curve calculated by the map generation device according to the embodiment of the present invention. FIG. 3 is a diagram illustrating an example of an operation related to lane change determination by the map generation device according to the embodiment of the present invention. FIG. 6 is a diagram illustrating another example of the operation related to lane change determination by the map generation device according to the embodiment of the present invention. FIG. 6 is a diagram illustrating yet another example of the operation related to lane change determination by the map generation device according to the embodiment of the present invention. 4 is a flowchart illustrating an example of a process executed by the controller in FIG. 3. FIG.

Embodiments of the present invention will be described below with reference to FIGS. 1 to 6. A map generation device according to an embodiment of the present invention is configured to generate a map (environmental map described below) used when, for example, a vehicle having an automatic driving function (automatic driving vehicle) travels. Note that the vehicle provided with the map generation device according to this embodiment may be referred to as the own vehicle to distinguish it from other vehicles.

The map generation device generates a map when the driver manually drives the vehicle. Therefore, the map generation device can be provided in a vehicle (manually driven vehicle) that does not have an automatic driving function. Note that the map generation device can be provided not only in manually driven vehicles but also in automatic driving vehicles that can switch from an automatic driving mode that does not require a driver's driving operation to a manual driving mode that requires a driver's driving operation. In the following, a map generation device will be described assuming that the map generation device is installed in an automatic driving vehicle.

First, the configuration of the automated driving vehicle will be explained. The host vehicle may be an engine vehicle having an internal combustion engine as a driving source, an electric vehicle having a traveling motor as a driving source, or a hybrid vehicle having an engine and a driving motor as a driving source. FIG. 1 is a block diagram schematically showing the overall configuration of a vehicle control system 100 having a map generation device according to an embodiment of the present invention.

As shown in FIG. 1, the vehicle control system 100 includes a controller 10, an external sensor group 1, an internal sensor group 2, and an input/output device 3, each of which is communicably connected to the controller 10 via a CAN communication line or the like. It mainly includes a positioning unit 4, a map database 5, a navigation device 6, a communication unit 7, and a driving actuator AC.

The external sensor group 1 is a general term for a plurality of sensors (external sensors) that detect external conditions that are information around the own vehicle. For example, external sensor group 1 includes a lidar that detects the position (distance and direction from the own vehicle) of objects around the own vehicle by emitting laser light and detecting reflected light, and a lidar that detects the position (distance and direction from the own vehicle) of objects around the own vehicle by emitting laser light and detecting reflected light. These include a radar that detects the position of objects around the vehicle, and a camera that has an imaging device such as a CCD or CMOS and captures images of the surroundings of the vehicle (front, rear, and sides).

The internal sensor group 2 is a general term for a plurality of sensors (internal sensors) that detect the driving state of the host vehicle. For example, the internal sensor group 2 includes a vehicle speed sensor that detects the vehicle speed of the host vehicle, an acceleration sensor that detects the longitudinal and lateral acceleration of the host vehicle, a rotation speed sensor that detects the rotation speed of the travel drive source, and the like. . The internal sensor group 2 also includes sensors that detect driving operations by the driver in the manual driving mode, such as accelerator pedal operations, brake pedal operations, steering wheel operations, and the like.

The input/output device 3 is a general term for devices through which commands are input from the driver and information is output to the driver. For example, the input/output device 3 includes various switches through which the driver inputs various commands by operating operating members, a microphone through which the driver inputs commands by voice, a display which provides information to the driver via displayed images, and a display that provides information to the driver via a displayed image. This includes speakers that provide information.

The positioning unit (GNSS unit) 4 has a positioning sensor that receives positioning signals transmitted from positioning satellites. A positioning sensor can also be included in the internal sensor group 2. Positioning satellites are artificial satellites such as GPS satellites and quasi-zenith satellites. The positioning unit 4 uses the positioning information received by the positioning sensor to measure the current position (latitude, longitude, altitude) of the own vehicle.

The map database 5 is a device that stores general map information used in the navigation device 6, and is composed of, for example, a hard disk or a semiconductor device. The map information includes road position information, road shape information (curvature, etc.), and position information of intersections and branch points. Note that the map information stored in the map database 5 is different from the highly accurate map information stored in the storage unit 12 of the controller 10.

The navigation device 6 is a device that searches for a target route on a road to a destination input by the driver and provides guidance along the target route. Input of the destination and guidance along the target route are performed via the input/output device 3. The target route is calculated based on the current position of the own vehicle measured by the positioning unit 4 and map information stored in the map database 5. The current position of the own vehicle can also be measured using the detection values of the external sensor group 1, and the target route is calculated based on this current position and the highly accurate map information stored in the storage unit 12. Good too.

The communication unit 7 communicates with various servers (not shown) via networks including wireless communication networks such as the Internet and mobile phone networks, and periodically or arbitrarily transmits map information, driving history information, traffic information, etc. Obtained from the server at the timing of Networks include not only public wireless communication networks but also closed communication networks established for each predetermined management area, such as wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), and the like. The acquired map information is output to the map database 5 and storage unit 12, and the map information is updated. It is also possible to communicate with other vehicles via the communication unit 7.

Actuator AC is a travel actuator for controlling travel of the own vehicle. When the traveling driving source is an engine, the actuator AC includes a throttle actuator that adjusts the opening degree of a throttle valve of the engine (throttle opening degree). When the travel drive source is a travel motor, the travel motor is included in the actuator AC. The actuator AC also includes a brake actuator that operates the braking device of the host vehicle and a steering actuator that drives the steering device.

The controller 10 is composed of an electronic control unit (ECU). More specifically, the controller 10 includes a computer having an arithmetic unit 11 such as a CPU (microprocessor), a storage unit 12 such as a ROM or RAM, and other peripheral circuits (not shown) such as an I/O interface. Consists of. Although a plurality of ECUs with different functions such as an engine control ECU, a travel motor control ECU, and a braking device ECU can be provided separately, in FIG. 1, the controller 10 is shown as a collection of these ECUs for convenience. .

The storage unit 12 stores highly accurate road map information. This road map information includes road position information, road shape (curvature, etc.) information, road slope information, intersection and branch point position information, number of lanes, lane width and position information for each lane ( This information includes information on the center position of the lane (lane center position and boundary line of the lane position), position information on landmarks (traffic lights, signs, buildings, etc.) on the map, and information on the road surface profile such as road surface irregularities. The map information stored in the storage unit 12 includes map information acquired from outside the own vehicle through the communication unit 7 and detection values of the external sensor group 1 or the detection values of the external sensor group 1 and the internal sensor group 2. This includes map information created by the own vehicle itself using detected values. The storage unit 12 also stores travel history information consisting of detection values of the external sensor group 1 and the internal sensor group 2 in association with the map information.

The calculation unit 11 has a vehicle position recognition unit 13, an external world recognition unit 14, an action plan generation unit 15, a travel control unit 16, and a map generation unit 17 as functional components.

The own vehicle position recognition unit 13 recognizes the position of the own vehicle on the map (own vehicle position) based on the position information of the own vehicle obtained by the positioning unit 4 and the map information of the map database 5. The position of the own vehicle may be recognized using the map information stored in the storage unit 12 and the surrounding information of the own vehicle detected by the external sensor group 1, and thereby the position of the own vehicle can be recognized with high accuracy. can. Note that when the vehicle position can be measured by a sensor installed outside on the road or beside the road, the vehicle position can also be recognized by communicating with the sensor via the communication unit 7.

The external world recognition unit 14 recognizes the external situation around the host vehicle based on signals from the external sensor group 1 such as lidar, radar, and cameras. For example, the position, speed, and acceleration of surrounding vehicles (vehicles in front and behind) driving around the own vehicle, the positions of surrounding vehicles stopped or parked around the own vehicle, and the positions and conditions of other objects, etc. recognize. Other objects include signs, traffic lights, markings such as road division lines and stop lines, buildings, guardrails, utility poles, signboards, pedestrians, bicycles, and the like. Other object states include the color of traffic lights (red, blue, yellow), the speed and direction of pedestrians and bicycles, etc.

The action plan generation unit 15 uses, for example, the target route calculated by the navigation device 6, the map information stored in the storage unit 12, the own vehicle position recognized by the own vehicle position recognition unit 13, and the external world recognition unit 14. Based on the recognized external situation, a travel trajectory (target trajectory) of the own vehicle from the current moment to a predetermined time ahead is generated. When there are multiple trajectories that are candidates for the target trajectory on the target route, the action plan generation unit 15 selects the optimal trajectory from among them that complies with laws and regulations and satisfies criteria such as efficient and safe driving. Then, the selected trajectory is set as the target trajectory. Then, the action plan generation unit 15 generates an action plan according to the generated target trajectory. The action plan generation unit 15 performs overtaking driving to overtake a preceding vehicle, lane changing driving to change the driving lane, following driving to follow the preceding vehicle, lane keeping driving to maintain the driving lane so as not to deviate from the driving lane, and deceleration driving. Alternatively, various action plans corresponding to accelerated driving etc. are generated. When generating a target trajectory, the action plan generation unit 15 first determines a running mode, and generates a target trajectory based on the running mode.

The travel control unit 16 controls each actuator AC so that the host vehicle travels along the target trajectory generated by the action plan generation unit 15 in the automatic driving mode. More specifically, the driving control unit 16 generates the required driving force to obtain the target acceleration per unit time calculated by the action plan generating unit 15 in consideration of the driving resistance determined by the road gradient etc. in the automatic driving mode. Calculate. Then, for example, the actuator AC is feedback-controlled so that the actual acceleration detected by the internal sensor group 2 becomes the target acceleration. That is, the actuator AC is controlled so that the host vehicle travels at the target vehicle speed and target acceleration. Note that when the driving mode is the manual driving mode, the driving control unit 16 controls each actuator AC according to driving commands (steering operation, etc.) from the driver acquired by the internal sensor group 2.

The map generation unit 17 generates an environmental map consisting of three-dimensional point cloud data using detection values detected by the external sensor group 1 while the vehicle is traveling in manual driving mode. Specifically, from a camera image acquired by a camera, edges indicating the outline of an object are extracted based on information on the brightness and color of each pixel, and feature points are extracted using the edge information. A feature point is, for example, a point on an edge or an intersection of edges, and corresponds to a division line on a road surface, a corner of a building, a corner of a road sign, etc. The map generation unit 17 calculates the distance to the extracted feature points and sequentially plots the feature points on the environmental map, thereby generating an environmental map of the area around the road on which the host vehicle has traveled. Instead of a camera, data acquired by radar or lidar may be used to extract feature points of objects around the host vehicle to generate an environmental map.

The own vehicle position recognition unit 13 performs a position estimation process of the own vehicle in parallel with the map generation process by the map generation unit 17. That is, the position of the vehicle is estimated based on changes in the position of the feature points over time. The map creation process and the position estimation process are performed simultaneously according to a SLAM (Simultaneous Localization and Mapping) algorithm using, for example, signals from a camera or a lidar. The map generation unit 17 can generate an environmental map not only when the vehicle is traveling in the manual driving mode but also when the vehicle is traveling in the automatic driving mode. If an environmental map has already been generated and stored in the storage unit 12, the map generating unit 17 may update the environmental map using newly obtained feature points.

Next, a description will be given of the map generation device according to this embodiment, that is, the configuration of the vehicle control system 100 as a map generation device. FIG. 2 is a diagram showing an example of a road 200 to which the map generation device according to the present embodiment is applied. FIG. 2 shows an example in which the host vehicle 101 changes lanes. That is, the own vehicle 101 changes lanes from the first lane LN1 defined by the left and right lane markings 201 and 202 to the second lane LN2 defined by the left and right lane lanes 202 and 203, as shown by arrow A. . The host vehicle 101 travels on the first lane LN1 at the first time t1 (state B1), and then travels on the second lane LN2 at the second time t2 (state B2).

A camera 1a is mounted on the front of the host vehicle 101. The camera 1a is capable of photographing a substantially fan-shaped photographing area AR centered on the camera 1a, which is determined by a predetermined viewing angle θ. The photographing area AR of the host vehicle 101 at the first time t1 includes the marking lines 201 and 202, and the photographing area AR at the second time t2 includes the marking lines 202 and 203. Therefore, by extracting edge points from the camera image, the partition lines 201 and 202 can be detected at the first time t1, and the partition lines 202 and 203 can be detected at the second time t2, respectively. By sequentially connecting the lane markings (specifically, feature points corresponding to the lane markings) detected by the camera images in chronological order in this way, a map including the lane markings 201 to 203 can be generated.

However, when changing lanes from the first lane LN1 to the second lane LN2, if the lane markings 202 are interrupted as shown in FIG. 2, the lane markings 202 may not be continuously detected. Even if the marking line 202 is not interrupted, the marking line 202 may be closed even if there is an obstacle (for example, another vehicle) that obstructs the field of view of the camera 1a, or if the detection accuracy of the camera 1a is reduced due to the influence of the weather. 202 may not be detected. As a result, the partition line 201 detected at the first time t1, the partition line 202 detected at the second time t2, and the partition line 202 detected at the first time t1 and the partition line detected at the second time t2, 203 may be connected to each other, and an incorrect map may be generated. Therefore, in this embodiment, a map generation device is configured as follows so that a map can be generated accurately even when there is a lane change.

FIG. 3 is a block diagram showing the main configuration of the map generation device 20 according to this embodiment. Map generation device 20 is included in vehicle control system 100 in FIG. As shown in FIG. 3, the map generation device 20 includes a camera 1a, a sensor 2a, and a controller 10.

The camera 1a is a monocular camera having an image sensor such as a CCD or CMOS, and forms part of the external sensor group 1 in FIG. The camera 1a may be a stereo camera. As shown in FIG. 2, the camera 1a is attached to a predetermined position at the front of the own vehicle 101, and continuously images the space in front of the own vehicle 101 to obtain an image of an object (camera image). The objects include lane markings 201 to 203 on the road. Note that the object may be detected by a lidar or the like instead of or together with the camera 1a.

The sensor 2a is a detector used to calculate the amount and direction of movement of the own vehicle 101. The sensor 2a is part of the internal sensor group 2, and includes, for example, a vehicle speed sensor and a yaw rate sensor. That is, the controller 10 (vehicle position recognition unit 13) integrates the vehicle speed detected by the vehicle speed sensor to calculate the amount of movement of the own vehicle 101, and also integrates the yaw rate detected by the yaw rate sensor to calculate the yaw angle. The position of the own vehicle 101 is estimated by odometry when calculating and creating a map. Note that the configuration of the sensor 2a is not limited to this, and the vehicle position may be estimated using information from other sensors.

The controller 10 in FIG. 3 includes a marking line estimating section 21 and a determining section 22 in addition to the storage section 12 and the map generating section 17, as functional components handled by the calculating section 11 (FIG. 1). Note that the lane marking estimation section 21 and the determination section 22 also have a map generation function, so they can also be included in the map generation section 17.

The storage unit 12 stores map information. The stored map information includes map information acquired from outside the host vehicle 101 via the communication unit 7 (referred to as external map information), and map information created by the host vehicle itself (referred to as internal map information). is included. The external map information is, for example, information on a map acquired via a cloud server (referred to as a cloud map), and the internal map information is, for example, information on a map (called a cloud map) made of point cloud data generated by mapping using technology such as SLAM. This information is called an environmental map. The external map information is shared between the host vehicle 101 and other vehicles, whereas the internal map information is map information unique to the host vehicle 101 (for example, map information that the host vehicle has independently). The storage unit 12 also stores information regarding various control programs, threshold values used in the programs, and the like.

The marking line estimating unit 21 estimates the position of the marking line that is predicted to be detected by the camera 1a at the current point in time (or at the future point in time), based on the camera image acquired at the past point in time (or at the current point in time). Specifically, feature points corresponding to the partition lines 201 and 202 included in the camera image are extracted from a camera image acquired at a past point in time (for example, the first time point t1 in FIG. 2), and a line passing through the feature points is extracted. Calculate the approximate curve. The approximate curve may be calculated not only from the past time but also from feature points before the past time, for example, from feature points for several frames before the past time. FIG. 4 is a diagram showing an example of approximate curves L1 and L2. In the figure, the horizontal and vertical axes are the X and Y coordinates of the feature point P when an arbitrary point is set as the coordinate origin. The X and Y coordinates are coordinates taken on the road surface. The feature points P are a feature point group P1 (referred to as a left feature point group) corresponding to one side (for example, the left side) of a pair of left and right partition lines, and a feature point group P2 (referred to as a right feature point group) corresponding to the other (for example, the right side) of a pair of left and right partition lines. (called point clouds).

The approximate curves L1 and L2 are expressed, for example, by polynomials that pass through the plurality of feature points P detected at the first time point t1, that is, polynomials such that the value of the Y coordinate is a function of the value of the X coordinate. Note that the approximate curves L1 and L2 can also be calculated by smoothly extending curves obtained by sequentially and smoothly connecting feature points along the vehicle traveling direction. The approximate curves L1 and L2 match or almost match the partition lines 201 and 202 detected by the camera image within the dotted line area Ra including the feature point P. The calculated approximate curves L1 and L2 are stored in the storage unit 12.

When a new feature point P corresponding to a marking line is detected as the host vehicle 101 moves, the marking line estimation unit 21 uses the new feature point P to calculate approximate curves L1 and L2. As a result, the approximate curves L1 and L2 are updated as time passes. By averaging the approximate curves calculated at a plurality of consecutive points in time (tn, tn-1, tn-2, . . . ), a representative approximate curve at time tn is obtained, and this is stored in the storage unit 12. It is also possible to store the information in .

If a new feature point corresponding to the partition line is not detected due to a break in the partition line, etc., the approximate curves L1 and L2 are not updated. Therefore, as shown in FIG. 2, when the own vehicle 101 changes lanes from the first lane LN1 to the second lane LN2 at a point where the marking line 202 is not detected, immediately after the lane change (second time t2), the lane The approximate curves L1 and L2 (approximate curves corresponding to the partition lines 201 and 202) calculated immediately before the change (first time point t1) remain stored in the storage unit 12. The approximate curves L1 and L2 calculated a predetermined time ago from the current time point may be stored in the storage unit 12. Approximate curves L1 and L2 calculated at a point a predetermined distance before the current position of the host vehicle 101 may be stored in the storage unit 12. In other words, the approximate curves L1 and L2 stored in the storage unit 12 may be updated after a predetermined time has elapsed from the current point (for example, after several seconds have elapsed), and after a predetermined distance has passed from the current position (for example, after a predetermined distance has passed from the current position) The approximate curves L1 and L2 stored in the storage unit 12 may be updated after the vehicle has traveled several meters.

The determining unit 22 compares the positions of the approximate curves L1 and L2 estimated by the lane marking estimation unit 21 and stored in the storage unit 12 with the position of the feature point corresponding to the lane marking detected from the camera image at the current time. Calculate the difference. Then, based on the magnitude of the difference, it is determined whether or not the own vehicle 101 should change lanes. FIGS. 5A to 5C are diagrams showing an example of the relationship between approximate curves L1 and L2 calculated at past times and feature points P detected at the current time. That is, after the approximate curves L1 and L2 are calculated based on the feature points P at the past point in time as shown in FIG. This is an example where the feature point P is detected again at the present time while the approximate curves L1 and L2 are still stored.

In FIG. 5A, as shown in the enlarged view of part a, the difference between the approximate curve L1 corresponding to the left partition line and the left feature point group P1 (for example, the difference in Y coordinate Δy1) is less than the predetermined value Δy1a. Note that the difference Δy between the approximate curve L2 corresponding to the right partition line and the right feature point group P2 is also less than the predetermined value Δy1a. The predetermined value Δy1a is set to a value less than or equal to 1/2 the lane width (for example, about 1 m). In this way, when the difference Δy1 between the approximate curve L1 and the left feature point group P1, which are on the same side in the left-right direction, is less than or equal to the predetermined value Δy1a, the determination unit 22 determines whether the own vehicle 101 has changed lanes between the past time and the present time. In other words, it is determined that the vehicle is maintaining its lane.

In FIG. 5B, as shown in the enlarged view of part a, the difference Δy2 between the approximate curve L2 corresponding to the right partition line and the left feature point group P1 is less than or equal to the predetermined value Δy2a. The predetermined value Δy2a is a value different from or the same as the predetermined value Δy1a, and is set to a value equal to or less than half the lane width. In this manner, when the difference Δy2 between the right-hand approximate curve L2 and the left-hand feature point group P1 is less than or equal to the predetermined value Δy2a, the determination unit 22 determines that the host vehicle 101 has changed lanes to the right lane. On the other hand, in FIG. 5C, as shown in the enlarged view of part a, the difference Δy3 between the approximate curve L1 corresponding to the left partition line and the right feature point group P2 is less than or equal to the predetermined value Δy3a. The predetermined value Δy3a is, for example, the same value as the predetermined value Δy2a, and is set to a value equal to or less than 1/2 the lane width. In this manner, when the difference Δy3 between the left-hand approximate curve L1 and the right-hand feature point group P2 is less than or equal to the predetermined value Δy3a, the determination unit 22 determines that the host vehicle 101 has changed lanes to the left lane.

In addition, in FIGS. 5A to 5C, the determination unit 22 determines whether the Y-coordinate differences Δy1, Δy2, and Δy3 between the approximate curves L1 and L2 and the feature point groups P1 and P2 are equal to or less than predetermined values Δy1a, Δy2a, and Δy3a, respectively. Instead of determining whether this is the case, it may be determined whether the shortest distance from the approximate curves L1, L2 to the feature point groups P1, P2 is less than or equal to a predetermined value. Then, depending on the determination result, it may be determined that the vehicle maintains the lane, changes to the right lane, or changes to the left lane.

The map generation unit 17 generates an environmental map of the point where the marking line was not recognized based on the determination result by the determination unit 22. That is, when the determination unit 22 determines that the vehicle 101 has not changed lanes (FIG. 5A), the left and right marking lines of the vehicle 101 recognized from the camera image at the past time and the lane markings recognized from the camera image at the current time are compared. The left and right marking lines of the host vehicle 101 that have been marked are connected.

On the other hand, when the determination unit 22 determines that the vehicle 101 has changed lanes to the right (FIG. 5B), the marking line on the right side of the vehicle 101 recognized from the camera image at the past time and the lane mark recognized from the camera image at the current time and the marking line on the left side of the host vehicle 101 that has been marked. When the determination unit 22 determines that the vehicle 101 has changed lanes to the left (FIG. 5C), the left marking line of the vehicle 101 recognized from the camera image at the past point in time and the marking line on the left side of the vehicle 101 recognized from the camera image at the current time are determined. It is connected to the right side marking line of the own vehicle 101. This allows the location of the lane to be specified. The lane whose position has been specified is stored in the storage unit 12 as part of the map information.

FIG. 6 is a flowchart showing an example of a process executed by the controller 10 of FIG. 3 according to a predetermined program. The process shown in this flowchart is started, for example, when the own vehicle 101 is running in manual driving mode to generate an environmental map, and is repeated at a predetermined period.

As shown in FIG. 6, first, in step S1, signals from the camera 1a and the sensor 2a are read. Next, in step S2, it is determined based on the camera image whether a pair of left and right lane markings defining the driving lane (own lane) of the own vehicle 101 has been detected. If the result in step S2 is negative, the process proceeds to step S9, and the flag is set to 1. Note that the flag is 0 in the initial state, and is set to 1 when at least one of the left and right partition lines is not detected. Next, in step S8, an environmental map is generated and stored in the storage unit 12. In this case, information on the environmental map in which the vehicle's own lane is not specified is stored.

If the result in step S2 is affirmative, the process proceeds to step S3, where it is determined whether the flag is 1 or not. That is, it is determined whether or not the marking line has just started to be detected again. If the result in step S3 is negative, the process proceeds to step S10, where a pair of left and right approximate curves are created that pass through the plurality of feature points based on the marking line detected in step S2, more precisely, the plurality of feature points corresponding to the marking line. Calculate L1 and L2. Next, in step S8, an environmental map is generated and stored in the storage unit 12. In this state, the left and right lane markings continue to be detected, and map information including the position information of the vehicle's own lane is stored. In this state where the left and right lane markings are continuously being detected, it is possible to determine whether or not the vehicle 101 has changed lanes by determining whether or not the vehicle 101 has crossed the lane marking. In step S8, the approximate curves L1 and L2 calculated in step S10 are also stored.

If the determination in step S3 is affirmative, that is, if it is determined that the marking line has just started to be detected again, the process proceeds to step S4, and the left approximate curve L1 stored in the storage unit 12 and the step S2 It is determined whether the difference Δy1 from the left feature point group P1 detected in is less than or equal to a predetermined value Δy1a. If the answer is affirmative in step S4, the process proceeds to step S11, where it is determined that the lane is to be maintained, and the process proceeds to step S8. In step S8, the left and right lane markings detected at the past time point, that is, the lane markings stored in the process from step S10 to step S8, and the left and right lane markings detected at the current time point (step S2) are connected to each other. A map is generated and stored in the storage unit 12. That is, in an area where no lane markings have been detected, a map is generated in which the lane markings before and after the lane are connected to each other. This makes it possible to generate an environmental map that includes lane position information.

On the other hand, if the result in step S4 is negative, the process proceeds to step S5, where the difference Δy2 between the right approximate curve L2 stored in the storage unit 12 and the left feature point group P1 detected in step S2 is less than or equal to the predetermined value Δy2a. Determine whether or not. If the result in step S5 is affirmative, the process proceeds to step S12, where it is determined that the host vehicle 101 has changed lanes to the right, and the process proceeds to step S8. In step S8, the right side lot line detected at the past point in time, that is, the right side lot line among the left and right lot lines stored in the process from step S10 to step S8, and the right side lot line detected at the current point in time (step S2). A map that connects the left lane markings to each other is generated and stored in the storage unit 12. As a result, even if lane markings are not clearly detected when changing lanes, an environmental map including lane position information can be generated.

On the other hand, if the result in step S5 is negative, the process proceeds to step S6, where the difference Δy3 between the left approximate curve L1 stored in the storage unit 12 and the right feature point group P2 detected in step S2 is less than or equal to the predetermined value Δy3a. Determine whether or not. If the result in step S6 is affirmative, the process proceeds to step S7, where it is determined that the host vehicle 101 has changed lanes to the left, and the process proceeds to step S8. In step S8, the left side lot line detected at the past point in time, that is, the left side lot line among the left and right lot lines stored in the process from step S10 to step S8, and the left side lot line detected at the current point in time (step S2). A map that connects the right lane markings is generated and stored in the storage unit 12. This makes it possible to generate an environmental map that includes lane position information regardless of whether the vehicle changes lanes to the left or right.

If the result in step S6 is negative, the process advances to step S8. In this case, whether or not to change lanes is undetermined. Therefore, a map is generated without connecting the lane markings in areas where lane markings have not been detected, and is stored in the storage unit 12. Thereafter, when it is determined through repeated processing whether or not there is a lane change, marking lines are connected in areas where marking lines have not been detected, and the map information is updated.

The operation of the map generation device 50 according to this embodiment will be explained in more detail. While the host vehicle 101 is traveling in manual driving mode, an environmental map of the surroundings of the host vehicle 101 is generated based on camera images. This environmental map includes positional information of a pair of left and right lane markings that define the vehicle's own lane. As shown in FIG. 2, when the host vehicle 101 changes lanes from the first lane LN1 to the second lane LN2, the left and right marking lines 201 and 202 before the lane change (first time t1) and the left and right marking lines 201 and 202 after the lane change (second time point t1) If the left and right lane markings 202 and 203 at time t2) continue to be detected, the vehicle 101 changes lanes by detecting that the vehicle 101 has crossed the lane marking 202 based on the camera image. It can be determined that As a result, map information including lane position information can be generated. At this time, approximate curves L1 and L2 along the partition line are calculated at any time using the latest feature points detected from the camera image, and are stored in the storage unit 12 (step S10→step S8).

On the other hand, as shown in FIG. 2, when the host vehicle 101 changes lanes from the first lane LN1 to the second lane LN2, if it is not detected that the host vehicle 101 has crossed the lane marking 202, the lane change is performed using an approximate curve. The presence or absence of is determined. That is, in this case, the difference Δy2 between the right approximate curve L2 along the partition line 202 calculated at the first time t1 and the left feature point P1 detected at the second time t2 is equal to or less than the predetermined value Δy2a. , Accordingly, a lane change to the right lane is determined (step S12). As a result, it is possible to prevent the right lane marking line 202 before the lane change from being erroneously connected to the right lane marking line 203 after the lane change. Therefore, it is possible to generate an accurate environmental map in which identical lane markings are connected.

According to this embodiment, the following effects can be achieved.
(1) The map generation device 20 uses a camera 1a that detects the external environment around the own vehicle 101, and a marking line 201 that defines the driving lane of the own vehicle 101 based on information about the external environment detected by the camera 1a. A map generation unit 17 that generates a map including position information of 203 and 203; A marking line estimating unit 21 estimates the positions of the marking lines 201 to 203 predicted to be detected by the marking line estimation unit 1a, that is, approximate curves L1 and L2; A determination unit 22 is provided that determines whether or not the own vehicle 101 changes lanes based on the difference from the position of the lane marking detected by the camera at time t2 (FIG. 3). The map generation unit 17 generates a map based on the determination result by the determination unit 22.

Thereby, even if the own vehicle 101 is traveling on a road without map information and the lane markings are interrupted, it is possible to accurately determine whether or not a lane change is to be made. Therefore, it is possible to accurately generate a map that includes position information of the lane markings. As a result, for example, it is possible to set an optimal target route in autonomous driving mode, and to provide driving support technology that further improves traffic safety and convenience and contributes to the development of a sustainable transportation system. I can do it. Since lane changes are determined based only on camera images, the map generation device 20 can be constructed at low cost. When generating an environmental map using recognition information, it is necessary to connect sequentially recognized lane markings to form a continuous lane. It is necessary to correctly determine whether the vehicle is maintaining the same lane or changing lanes. In this regard, in this embodiment, since it is possible to correctly determine whether or not a lane change is to be made, a map including positional information of lane markings can be generated with high accuracy.

(2) The marking line estimating unit 21 calculates approximate curves L1 and L2 that pass through the marking lines 201 and 202 based on the position information of the marking lines 201 and 202 detected by the camera 1a at the first time t1. Then, based on the approximate curves L1 and L2, the positions of the lane markings 201 and 202 that are predicted to be detected by the camera 1a at the second time point t2, that is, the positions of the lane markings 201 and 202 when lane maintenance is assumed, are determined. Estimate (Figure 4). As a result, even if there are points where the lane markings 201 and 202 cannot be detected, the positions of the lane markings can be estimated well, and robustness is improved.

(3) The camera 1a is configured to detect a marking line 201 (first marking line) on one side in the left-right direction and a marking line 202 (second marking line) on the other side in the left-right direction, which define the driving lane of the own vehicle 101. (Figure 2). The determining unit 22 determines whether the lane will change to the right lane based on the difference Δy2 between the position of the marking line 202 estimated by the marking line estimating unit 21 (approximate curve L2) and the position of the marking line 202 detected by the camera 1a. The presence or absence of changes is determined (FIG. 6). As a result, it is possible to prevent positional changes in the vehicle width direction when traveling in the same lane from being erroneously determined as a lane change, and it is possible to satisfactorily determine whether a lane change has occurred.

(4) The second time point t2 is the current time point, and the map generation unit 17 generates a map including position information of the lane markings detected by the camera 1a while the host vehicle 101 is traveling. Thereby, the host vehicle 101 can immediately generate a highly accurate environmental map according to the road shape while driving on a road for which there is no map information.

The above embodiment can be modified in various forms. In the above embodiment, the external sensor group 1 such as the camera 1a is used to detect the external environment around the host vehicle 101, but the external environment may also be detected using a lidar or the like. The configuration is not limited to that described above. In the embodiment described above, the map generation unit 25 generates the environmental map while driving in the manual driving mode, but it may also generate the environmental map while driving in the automatic driving mode. In the above embodiment, the environmental map is generated based on the camera image, but feature points of objects around the own vehicle 101 may be extracted using data acquired by radar or lidar instead of the camera 1a. , an environmental map may be generated. Therefore, the configuration of the map generation section is not limited to that described above.

In the embodiment described above, approximate curves L1 and L2 that pass through the lane markings are calculated based on the position information of the lane markings detected by the camera 1a at the first time point t1, and based on the approximate curves L1 and L2, , the position of the marking line that is predicted to be detected by the camera at the second time point t2 when the own vehicle 101 is traveling while maintaining the lane is estimated, but the configuration of the marking line estimating unit 21 is as described above. Not limited to things. In other words, the position of the lot line that is predicted to be detected at the second time point may be estimated based on the information on the external environment detected at the first time point, without using an approximate curve. The structure of the section is not limited to that described above. Therefore, the determining unit 22 determines whether the own vehicle 101 changes lanes based on the difference between the position of the marking line estimated by the marking line estimating unit 21 and the position of the marking line detected at the second time t2. The configuration is also not limited to that described above.

In the above embodiment, the map generation unit 17 generates the environmental map while the own vehicle 101 is running, but data obtained from camera images while the own vehicle 101 is running is stored in the storage unit 12, After the vehicle 101 completes traveling, the environmental map may be generated using the stored data. Therefore, it is not necessary to generate a map while driving.

In the above embodiment, an example has been described in which the own vehicle 101 having the automatic driving function functions as the map generation device 20, but the own vehicle 101 without the automatic driving function may function as the map generating device. In this case, the map information generated by the map generation device 20 may be shared with other vehicles, and the map information may be used to support the driving of the other vehicle (for example, an automatic driving vehicle). That is, the host vehicle 101 may have only the function of the map generation device 20.

The above description is merely an example, and the present invention is not limited to the above-described embodiments and modifications as long as the characteristics of the present invention are not impaired. It is also possible to arbitrarily combine the above embodiment and one or more of the modifications, and it is also possible to combine the modifications.

1a Camera, 10 Controller, 17 Map generation unit, 20 Map generation device, 21 Compartment line estimation unit, 22 Determination unit, 201, 202 Compartment line, L1, L2 Approximate curve

Claims (4)

an outside world detection unit that detects outside world conditions around the host vehicle;
a map generation unit that generates a map including positional information of lane markings defining the driving lane of the host vehicle based on information on the outside world situation detected by the outside world detection unit;
Estimating the position of a marking line that is predicted to be detected by the outside world detection unit at a second time point after the first time point, based on information on the outside world situation detected by the outside world detection unit at a first time point. a lot line estimating unit,
A determination of whether or not the own vehicle changes lanes based on a difference between the position of the marking line estimated by the marking line estimating unit and the position of the marking line detected by the outside world detecting unit at the second time point. and,
The map generation device is characterized in that the map generation section generates a map based on a determination result by the determination section. The map generation device according to claim 1,
The marking line estimating unit calculates an approximate curve that passes through the marking line based on the position information of the marking line detected by the external world detecting unit at the first time point, and calculates an approximate curve passing through the marking line based on the approximate curve. A map generation device that estimates the position of a lane marking that is predicted to be detected by the outside world detection unit at a second time point. The map generation device according to claim 1 or 2,
The external world detection unit is provided to detect a first marking line on one side in the left-right direction and a second marking line on the other side in the left-right direction, which define a driving lane of the own vehicle,
The determination unit determines whether or not to change lanes based on the difference between the position of the first lane marking estimated by the lane marking estimation unit and the position of the second lane marking detected by the external world detection unit. A map generation device characterized by determining. The map generation device according to any one of claims 1 to 3,
The second point in time is the current point in time,
The map generation device is characterized in that the map generation unit generates a map including positional information of lane markings detected by the outside world detection unit while the own vehicle is traveling.

Images