Vehicle position estimation device, vehicle position estimation method, and computer-readable storage medium

Abstract

The present invention relates to a vehicle position estimation device, a method thereof, and a program. The vehicle position estimation device includes: a longitudinal position estimation unit that estimates the longitudinal position of the vehicle based on the output of a vehicle state detection device; a lane line detection unit that detects the actual lane lines on the left and right sides of the vehicle; a lateral position estimation unit that compares the lane lines on the map at the longitudinal position of the vehicle with the actual lane lines and estimates the lateral position of the vehicle based on the relative position relationship between the actual lane lines and the vehicle; and a travel path determination unit that determines the travel path of the vehicle. The lateral position estimation unit calculates the amount of change in the lateral position with respect to the travel path for each of the lane lines on the left and right sides of the travel path in a predetermined range before and after the longitudinal position of the vehicle according to the map information, and determines the lane lines on the map to be compared based on the amount of change in the lateral position. The decrease in the estimation accuracy of the vehicle lateral position due to the estimation error of the vehicle longitudinal position is suppressed.

Description

Technical Field

[0001] The present invention relates to a vehicle position estimation device, a vehicle position estimation method, and a computer program for vehicle position estimation. Background Art

[0002] Conventionally, there has been known a technique for estimating the own position of a vehicle in order to make the vehicle travel along a desired travel path (for example, Patent Documents 1 to 4). In the own position estimation device described in Patent Document 1, the left and right lane markings of the travel lane included in the road surface image captured by the camera are compared with the left and right lane markings of the travel lane included in the map information, thereby estimating the lateral position of the vehicle.

[0003] Prior Art Documents

[0004] Patent Documents

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2019-132762

[0006] Patent Document 2: International Publication No. 2019 / 189098

[0007] Patent Document 3: Japanese Patent No. 6342104

[0008] Patent Document 4: Japanese Unexamined Patent Application Publication No. 2020-26985 Summary of the Invention

[0009] However, in the case where an estimation error occurs in the longitudinal position of the vehicle, as the lane markings on the map to be compared, the lane markings at a position different from the actual position of the vehicle are used. As a result, the estimation accuracy of the lateral position of the vehicle may decrease.

[0010] Therefore, in view of the above problems, an object of the present invention is to suppress a decrease in the estimation accuracy of the lateral position of the vehicle due to an estimation error in the longitudinal position of the vehicle.

[0011] The gist of the present disclosure is as follows.

[0012] (1) A vehicle position estimation device includes: a longitudinal position estimation unit that estimates the longitudinal position of the vehicle based on the output of a vehicle state detection device that detects the state quantity of the vehicle; a lane line detection unit that detects the actual lane lines on the left and right sides of the vehicle; a lateral position estimation unit that compares the lane lines on the map at the longitudinal position of the vehicle with the actual lane lines, and estimates the lateral position of the vehicle based on the relative position relationship between the actual lane lines and the vehicle; and a travel path determination unit that determines the travel path of the vehicle. The lateral position estimation unit calculates the change amount of the lateral position with respect to the travel path for each of the lane lines on the left and right sides of the travel path within a predetermined range before and after the longitudinal position of the vehicle according to the map information, and determines the lane lines on the map to be compared based on this change amount.

[0013] (2) In the vehicle position estimation device according to (1) above, when the change amount with respect to the lane line on one side of the travel path is less than a predetermined threshold value and the change amount with respect to the lane line on the other side of the travel path is equal to or greater than the threshold value, the lateral position estimation unit uses the lane lines on the map on this one side as the comparison object.

[0014] (3) In the vehicle position estimation device according to (1) or (2) above, when the actual lane lines detected by the lane line detection unit change in the lateral position with respect to the travel path at a certain location, the longitudinal position estimation unit corrects the longitudinal position of the vehicle based on the relative position relationship between this location and the vehicle.

[0015] (4) A vehicle position estimation method includes: estimating the longitudinal position of the vehicle based on the output of a vehicle state detection device that detects the state quantity of the vehicle; detecting the actual lane lines on the left and right sides of the vehicle; comparing the lane lines on the map at the longitudinal position of the vehicle with the actual lane lines, and estimating the lateral position of the vehicle based on the relative position relationship between the actual lane lines and the vehicle; and determining the travel path of the vehicle. Estimating the lateral position of the vehicle includes: calculating the change amount of the lateral position with respect to the travel path for each of the lane lines on the left and right sides of the travel path within a predetermined range before and after the longitudinal position of the vehicle according to the map information, and determining the lane lines on the map to be compared based on this change amount.

[0016] (5) A computer program for estimating a vehicle position causes a computer to perform: estimating a longitudinal position of the vehicle based on an output of a vehicle state detection device that detects a state quantity of the vehicle; detecting actual lane lines on both left and right sides of the vehicle; comparing the lane lines on a map at the longitudinal position of the vehicle with the actual lane lines, and estimating a lateral position of the vehicle based on a relative positional relationship between the actual lane lines and the vehicle; and determining a driving path of the vehicle. Estimating the lateral position of the vehicle includes: calculating a change amount of the lateral position with respect to the driving path for each of the lane lines on both left and right sides of the driving path within a predetermined range before and after the longitudinal position of the vehicle based on map information, and determining the lane lines on the map to be compared based on the change amount.

[0017] According to the present invention, it is possible to suppress a decrease in the estimation accuracy of the lateral position of the vehicle due to an estimation error in the longitudinal position of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 FIG. schematically shows a configuration of an autonomous driving system including a vehicle position estimation device according to a first embodiment of the present invention.

[0019] Figure 2 FIG. shows Figure 1 a specific example of a target detection device.

[0020] Figure 3 FIG. is Figure 1 a functional block diagram of an ECU.

[0021] Figure 4 FIG. shows a map of a branch point on an expressway for automobiles.

[0022] Figure 5 FIG. shows a map of a merging point on an expressway for automobiles.

[0023] Figure 6 FIG. shows a map of a road having discontinuous lane lines.

[0024] Figure 7 FIG. is a flowchart showing a control routine of a lateral position estimation process in a first embodiment of the present invention.

[0025] Figure 8 FIG. shows an example of a lateral position of a lane line within a predetermined range.

[0026] Figure 9 FIG. is a flowchart showing a control routine of a longitudinal position correction process in a second embodiment of the present invention.

[0027] REFERENCE SIGNS

[0028] 10: Electronic Control Unit (ECU); 15: Longitudinal Position Estimation Unit; 16: Lateral Position Estimation Unit; 17: Lane Line Detection Unit; 18: Travel Path Determination Unit; 19: Vehicle Control Unit; 20: Vehicle. Detailed Embodiment

[0029] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, the same reference numerals are assigned to the same components.

[0030] <First Embodiment>

[0031] First, with reference to Figures 1 to 8 , the first embodiment of the present invention will be described.

[0032] <Structure of Autonomous Driving System>

[0033] Figure 1 FIG. schematically shows the structure of an autonomous driving system 1 equipped with a vehicle position estimation device according to the first embodiment of the present invention. The autonomous driving system 1 is mounted on a vehicle and implements autonomous driving of the vehicle. Under autonomous driving of the vehicle, acceleration, steering, and deceleration (braking) of the vehicle are automatically controlled in part or in whole to achieve autonomous driving of the vehicle.

[0034] As Figure 1 shown, the autonomous driving system 1 includes an object detection device 2, a GNSS receiver 3, a vehicle speed sensor 4, a yaw rate sensor 5, a map database 6, a Human Machine Interface (HMI) 7, a navigation device 8, an actuator 9, and an Electronic Control Unit (ECU) 10. The object detection device 2, the GNSS receiver 3, the vehicle speed sensor 4, the yaw rate sensor 5, the map database 6, the HMI 7, the navigation device 8, and the actuator 9 are communicably connected to the ECU 10 via an in-vehicle network conforming to a standard such as CAN (Controller Area Network).

[0035] The object detection device 2 detects objects (other vehicles, lane lines, signs, falling objects, etc.) existing around the vehicle (this vehicle). Specifically, the object detection device 2 detects the presence or absence of an object around the vehicle 20, the distance from the vehicle 20 to the object, and the relative speed between the vehicle 20 and the object. The object detection device 2 includes, for example, a camera, a LIDAR (Laser Imaging Detection And Ranging), a millimeter wave radar, an ultrasonic sensor (sonar), etc. The output of the object detection device 2 is sent to the ECU 10.

[0036] Figure 2 is a diagram showing Figure 1 a specific example of the target detection device 2. In Figure 2 this example, as the target detection device 2, an in-vehicle camera 21, a lidar 22, a millimeter-wave radar 23, and an ultrasonic sensor (sonar) 24 are provided on the vehicle 20.

[0037] The in-vehicle camera 21 captures the surroundings of the vehicle 20 and generates an image of the surroundings of the vehicle 20. For example, the in-vehicle camera 21 is disposed in front of the vehicle 20 (e.g., the back of the interior mirror inside the vehicle, the front bumper, etc.) to capture the front of the vehicle 20. In addition, the in-vehicle camera 21 may also be a stereo camera capable of detecting the distance from the vehicle 20 to the target.

[0038] The lidar 22 irradiates laser light on the surroundings of the vehicle 20 and receives the reflected light of the laser. Thereby, the lidar 22 detects the presence or absence of a target in the surroundings of the vehicle 20, the distance from the vehicle 20 to the target, and the relative speed between the vehicle 20 and the target. For example, the lidar 22 is disposed at the front and rear of the vehicle 20 (e.g., the front bumper and the rear bumper of the vehicle 20).

[0039] The millimeter-wave radar 23 emits millimeter waves to the surroundings of the vehicle 20 and receives the reflected waves of the millimeter waves. Thereby, the millimeter-wave radar 23 can detect the presence or absence of a target in the surroundings of the vehicle 20, the distance from the vehicle 20 to the target, and the relative speed between the vehicle 20 and the target. For example, the millimeter-wave radar 23 is disposed at the front and rear of the vehicle 20 (e.g., the front bumper and the rear bumper of the vehicle 20).

[0040] The ultrasonic sensor 24 emits ultrasonic waves to the surroundings of the vehicle 20 and receives the reflected waves of the ultrasonic waves. Thereby, the ultrasonic sensor 24 can detect the presence or absence of a target in the surroundings of the vehicle 20, the distance from the vehicle 20 to the target, and the relative speed between the vehicle 20 and the target. For example, the ultrasonic sensor 24 is disposed on both sides of the vehicle 20 (e.g., the left and right front fenders of the vehicle 20).

[0041] In addition, the positions and numbers of the in-vehicle camera 21, the lidar 22, the millimeter-wave radar 23, and the ultrasonic sensor 24 are not limited to the above. In addition, a part of them may be omitted.

[0042] The GNSS receiver 3 detects the current position of the vehicle 20 (e.g., the latitude and longitude of the vehicle 20) based on the positioning information obtained from multiple (e.g., three or more) positioning satellites. Specifically, the GNSS receiver 3 captures multiple positioning satellites and receives the radio waves transmitted from the positioning satellites. Then, the GNSS receiver 3 calculates the distance to the positioning satellite based on the difference between the transmission time and the reception time of the radio wave, and detects the current position of the vehicle 20 based on the distance to the positioning satellite and the position (orbital information) of the positioning satellite. The output of the GNSS receiver 3 is sent to the ECU 10. The GNSS receiver 3 is an example of a vehicle state detection device that detects a state quantity of the vehicle 20 (in this case, the position of the vehicle 20).

[0043] In addition, GNSS (Global Navigation Satellite System) is a general term for satellite positioning systems such as the GPS in the United States, GLONASS in Russia, Galileo in Europe, QZSS in Japan, BeiDou in China, and IRNSS in India. Therefore, as the GNSS receiver 3, a GPS receiver is included.

[0044] The vehicle speed sensor 4 detects the speed of the vehicle 20. For example, the vehicle speed sensor 4 detects the speed of the vehicle 20 by detecting the rotational speed of the wheels of the vehicle 20. The output of the vehicle speed sensor 4 is output to the ECU 10. The vehicle speed sensor 4 is an example of a vehicle state detection device that detects a state quantity of the vehicle 20 (in this case, the speed of the vehicle 20).

[0045] The yaw rate sensor 5 detects the yaw rate, which is the rotational angular velocity about the vertical axis passing through the center of gravity of the vehicle 20. For example, a gyro sensor is used as the yaw rate sensor 5. The output of the yaw rate sensor 5 is output to the ECU 10. The yaw rate sensor 5 is an example of a vehicle state detection device that detects a state quantity of the vehicle 20 (in this case, the yaw rate of the vehicle 20).

[0046] The map database 6 stores three-dimensional map information such as road surface information, lane information, and the position information of buildings. The map stored in the map database 6 is a so-called high-precision map. The ECU 10 acquires the map information from the map database 6. In addition, the map information stored in the map database 6 can also be updated regularly using communication with the outside of the vehicle 20, SLAM (Simultaneous Localization and Mapping) technology, etc. Additionally, the map database 6 can also be provided in a server outside the vehicle 20.

[0047] The HMI 7 performs input and output of information between the driver and the vehicle 20. The HMI 7 includes, for example, a display for displaying information, a speaker for emitting sound, operation buttons, operation switches, or a touch screen for the driver to perform input operations, a microphone for receiving the driver's voice, and the like. The output of the ECU 10 is transmitted to the driver via the HMI 7, and the input from the driver is sent to the ECU 10 via the HMI 7.

[0048] The navigation device 8 sets the driving route of the vehicle 20 to the destination based on the current position of the vehicle 20 detected by the GNSS receiver 3, the map information of the map database 6, the input made by the driver, and the like. The driving route set by the navigation device 8 is sent to the ECU 10.

[0049] The actuator 9 makes the vehicle 20 operate. For example, the actuator 9 includes a driving device (at least one of an engine and a motor) for accelerating the vehicle 20, a brake actuator for braking the vehicle 20, a steering motor for steering the vehicle 20, and the like. The ECU 10 controls the actuator 9 when implementing the autonomous driving of the vehicle 20.

[0050] The ECU 10 performs various controls of the vehicle. As Figure 1 shown, the ECU 10 includes a communication interface 11, a memory 12, and a processor 13. The communication interface 11 and the memory 12 are connected to the processor 13 via signal lines. In addition, in the present embodiment, one ECU 10 is provided, but multiple ECUs may also be provided for each function.

[0051] The communication interface 11 has an interface circuit for connecting the ECU 10 to the in-vehicle network. The ECU 10 communicates with the other in-vehicle devices as described above via the communication interface 11.

[0052] The memory 12 has, for example, a volatile semiconductor memory and a non-volatile semiconductor memory. The memory 12 stores computer programs, data, etc. used when the processor 13 performs various processes.

[0053] The processor 13 has one or more CPUs (Central Processing Unit) and their peripheral circuits. In addition, the processor 13 may also have an arithmetic circuit for numerical operations, an arithmetic circuit for graphics processing, an arithmetic circuit for logical operations, and the like.

[0054] <Vehicle Position Estimation Device>

[0055] In the present embodiment, the ECU 10 functions as a vehicle position estimation device that estimates the own position of the vehicle 20 (this vehicle). Figure 3 is Figure 1Functional block diagram of the ECU 10. In the present embodiment, the ECU 10 includes a longitudinal position estimation unit 15, a lateral position estimation unit 16, a lane marking detection unit 17, a travel path determination unit 18, and a vehicle control unit 19. The longitudinal position estimation unit 15, the lateral position estimation unit 16, the lane marking detection unit 17, the travel path determination unit 18, and the vehicle control unit 19 are functional modules implemented by a computer program stored in the memory 12 of the ECU 10 and executed by the processor 13 of the ECU 10. In addition, each of these functional modules may also be a dedicated arithmetic circuit provided in the processor 13.

[0056] The longitudinal position estimation unit 15 estimates the longitudinal position of the vehicle 20, and the lateral position estimation unit 16 estimates the lateral position of the vehicle 20. In addition, the longitudinal position of the vehicle means the position of the vehicle (such as the center or the center of gravity of the vehicle) in the extending direction of the driving lane of the vehicle, and the lateral position of the vehicle means the position of the vehicle (such as the center or the center of gravity of the vehicle) in the width direction of the driving lane of the vehicle.

[0057] The lane marking detection unit 17 detects the actual lane markings on the left and right sides of the vehicle 20. Specifically, the lane marking detection unit 17 detects the actual lane markings according to the output of the target detection device 2 (especially the external camera 21 or the lidar 22) by a known object recognition method (such as line detection based on the Hough transform, etc.). The position of the actual lane markings detected by the lane marking detection unit 17 is represented, for example, by the XY coordinates in the vehicle coordinate system with the position of the vehicle 20 (such as the center or the center of gravity of the vehicle 20) as the origin.

[0058] The travel path determination unit 18 determines the travel path of the vehicle 20. In addition, the travel path of the vehicle 20 means the path on the lane through which the vehicle 20 (the center of the vehicle 20) passes on the travel route preset in advance by the navigation device 8 or the like when implementing the autonomous driving of the vehicle 20. For example, the travel path determination unit 18 determines the travel path of the vehicle 20 in such a way that the vehicle 20 travels in the center of the lane. In this case, the center line of the lane is selected as the travel path of the vehicle 20.

[0059] When implementing the autonomous driving of the vehicle 20, the vehicle control unit 19 controls the acceleration, steering, and deceleration (braking) of the vehicle 20 using the actuator 9. For example, the vehicle control unit 19 controls the steering of the vehicle 20 using the actuator 9 (specifically, the steering motor) in such a way that the vehicle 20 travels on the travel path determined by the travel path determination unit 18.

[0060] Hereinafter, the method for estimating the longitudinal position and the lateral position of the vehicle 20 will be described in more detail. The longitudinal position estimation unit 15 estimates the longitudinal position of the vehicle 20 based on the output of the vehicle state detection device that detects the state quantity of the vehicle 20. For example, the longitudinal position estimation unit 15 estimates the longitudinal position of the vehicle 20 based on the output of the GNSS receiver 3.

[0061] In addition, the longitudinal position estimation unit 15 may also estimate the longitudinal position of the vehicle 20 by known dead reckoning (self-navigation) based on the outputs of the vehicle speed sensor 4 and the yaw rate sensor 5. In this case, the longitudinal position estimation unit 15 repeatedly calculates the moving distance and the moving direction of the vehicle 20 based on the outputs of the vehicle speed sensor 4 and the yaw rate sensor 5, thereby estimating the longitudinal position of the vehicle 20. The longitudinal position of the vehicle 20 estimated by the longitudinal position estimation unit 15 is represented, for example, as latitude and longitude or coordinate values in the world coordinate system. Thereby, the longitudinal position of the vehicle 20 on the map is determined.

[0062] On the other hand, the lateral position estimation unit 16 compares the lane lines on the map at the longitudinal position of the vehicle 20 estimated by the longitudinal position estimation unit 15 with the actual lane lines detected by the lane line detection unit 17, and estimates the lateral position of the vehicle 20 based on the relative positional relationship between the actual lane lines and the vehicle 20. Specifically, the lateral position estimation unit 16 obtains information on the lane lines on the map at the longitudinal position of the vehicle 20 from the map database 6, and aligns the actual lane lines with the lane lines on the map. Then, the lateral position estimation unit 16 estimates the lateral position of the vehicle 20 on the map based on the position of the vehicle 20 relative to the actual lane lines arranged on the map in a manner consistent with the lane lines on the map. The lateral position of the vehicle 20 estimated by the lateral position estimation unit 16 is represented, for example, as latitude and longitude or coordinate values in the world coordinate system. Thereby, the lateral position of the vehicle 20 on the map is determined.

[0063] However, when there is an estimation error in the longitudinal position of the vehicle 20, lane lines at a position different from the actual position of the vehicle are used as the lane lines on the map to be compared. As a result, the estimation accuracy of the lateral position of the vehicle 20 may decrease.

[0064] In particular, as Figures 4 to 6 shown, when the lateral position of the lane lines on the map changes with respect to the driving path of the vehicle 20, the influence caused by the estimation error of the longitudinal position becomes significant. Figure 4 It is a diagram showing a map of a branch point on an expressway for automobiles.

[0065] Figure 5 It is a diagram showing a map of a merging point on an expressway for automobiles. Figure 6A figure showing a map of a road with discontinuous dashed lines.

[0066] At Figure 4 the branch point shown and Figure 5 the merging point shown, as dashed lines, white lines and zebra areas (diversion strips) are provided on the road. In the map information stored in the map database 6, a line connecting the centers of white lines of a predetermined width is registered as a dashed line corresponding to the white line, and a line connecting the centers of the white lines at the boundary of the zebra area is registered as a dashed line corresponding to the zebra area.

[0067] From Figure 4 and Figure 5 it can be seen that in the white line area WA and the zebra area ZA, the lateral positions of the left dashed lines relative to the driving path are different. Therefore, when the distance between the actual dashed line and the vehicle 20 is D, a deviation occurs between the lateral position of the vehicle 20 inferred with the left dashed line in the white line area WA as the comparison object and the lateral position of the vehicle 20 inferred with the left dashed line in the zebra area ZA as the comparison object. Thus, when the inferred longitudinal position of the vehicle 20 is in the zebra area WA and the actual longitudinal position of the vehicle 20 is in the white line area WA, or vice versa, the inference accuracy of the lateral position of the vehicle 20 decreases.

[0068] In addition, in Figure 6 the example, the extending direction of the right white line changes discontinuously. Therefore, in the straight area DA where the white line extends in the traveling direction of the vehicle 20 and the inclined area TA where the white line extends in a direction different from the traveling direction of the vehicle 20, the lateral positions of the right dashed lines relative to the driving path are different. Therefore, when the distance between the actual dashed line and the vehicle 20 is D, a deviation occurs between the lateral position of the vehicle 20 inferred with the right dashed line in the straight area DA as the comparison object and the lateral position of the vehicle 20 inferred with the right dashed line in the inclined area TA as the comparison object. Thus, when the inferred longitudinal position of the vehicle 20 is in the inclined area TA and the actual longitudinal position of the vehicle 20 is in the straight area DA, or vice versa, the inference accuracy of the lateral position of the vehicle 20 decreases.

[0069] On the other hand, in Figure 4 and Figure 5 the example, the lateral position of the right dashed line relative to the driving path is constant. Therefore, when inferring the lateral position of the vehicle 20 with the right dashed line as the comparison object, no deviation in the lateral position of the vehicle 20 caused by the inference error of the longitudinal position of the vehicle 20 occurs. In addition, in Figure 6In the example, the lateral position of the left dashed line is constant with respect to the driving path. Therefore, when the left dashed line is used as a comparison object to estimate the lateral position of the vehicle 20, there is no deviation in the lateral position of the vehicle 20 caused by the estimation error of the longitudinal position of the vehicle 20.

[0070] Thus, in the present embodiment, the lateral position estimation unit 16 calculates the change amount of the lateral position with respect to the driving path for each of the dashed lines on the left and right sides of the driving path of the vehicle 20 in a predetermined range before and after the longitudinal position of the vehicle 20 estimated by the longitudinal position estimation unit 15 based on the map information, and determines the dashed line on the map as the comparison object according to the change amount of the lateral position. Thereby, it is possible to suppress the decrease in the estimation accuracy of the lateral position of the vehicle 20 due to the estimation error of the longitudinal position of the vehicle 20.

[0071] For example, the lateral position estimation unit 16 determines the dashed line on the map as the comparison object by comparing the change amount of the lateral position for each of the dashed lines on the left and right sides of the driving path with a threshold value. Specifically, when the change amount of the lateral position of the dashed line on one side of the driving path is less than the threshold value and the change amount of the lateral position of the dashed line on the other side of the driving path is equal to or greater than the threshold value, the lateral position estimation unit 16 uses the dashed line on the map on the above-mentioned one side as the comparison object. On the other hand, when the change amount of the lateral position of the dashed lines on both sides of the driving path is less than the threshold value or the change amount of the lateral position of the dashed lines on both sides of the driving path is equal to or greater than the threshold value, the lateral position estimation unit 16 uses the dashed lines on the map on both sides as the comparison object.

[0072] The vehicle control unit 19 controls the steering of the vehicle 20 using the actuator 9 (specifically, the steering motor) so that the lateral position of the vehicle 20 estimated by the lateral position estimation unit 16 is located on the driving path.

[0073] <Lateral Position Estimation Process>

[0074] Hereinafter, with reference to Figure 7 the flowchart, the control for estimating the lateral position of the vehicle 20 will be described in detail. Figure 7 is a flowchart showing a control routine of the lateral position estimation process in the first embodiment of the present invention. This control routine is repeatedly executed by the ECU 10 at a predetermined execution interval. The predetermined execution interval is, for example, the interval at which the output of the target detection device 2 is sent to the ECU 10.

[0075] First, in step S101, the longitudinal position estimation unit 15 estimates the longitudinal position of the vehicle 20 based on the output of the vehicle state detection device that detects the state quantity of the vehicle 20. Next, in step S102, the lane marking detection unit 17 detects the actual lane markings on the left and right sides of the vehicle 20 based on the output of the target detection device 2 (specifically, the in-vehicle camera 21 or the lidar 22).

[0076] Next, in step S103, the lateral position estimation unit 16 obtains the position information of the lane markings on the left and right sides of the driving path of the vehicle 20 in a predetermined range before and after the longitudinal position of the vehicle 20 estimated by the longitudinal position estimation unit 15 from the map database 6. The predetermined range is determined in advance considering the estimation error of the longitudinal position of the vehicle 20 and the like.

[0077] Next, in step S104, the lateral position estimation unit 16 calculates the lateral position relative to the driving path for each of the lane markings on the left and right sides of the driving path in the above-mentioned predetermined range. The lateral position of the lane marking relative to the driving path is calculated, for example, as the coordinate value of the axis orthogonal to the driving path based on the driving path and the position information of the lane marking. At this time, for example, when the lane marking is on the right side of the driving path, the lateral position of the lane marking is calculated as a positive value, and when the lane marking is on the left side of the driving path, the lateral position of the lane marking is calculated as a negative value.

[0078] Figure 8 is a diagram showing an example of the lateral position of the lane markings in the predetermined range. In Figure 8 it shows Figure 4 the lateral position of the lane markings when the longitudinal position of the vehicle 20 at the branch point shown is estimated to be a position in front of the zebra area. In Figure 8 this example, in front of the estimated longitudinal position of the vehicle 20 (the boundary position between the white line area WA and the zebra area ZA), the lateral position of the left lane marking changes significantly.

[0079] Next, in step S105, the lateral position estimation unit 16 calculates the change amount of the lateral position relative to the driving path for each of the lane markings on the left and right sides of the driving path in the above-mentioned predetermined range. For example, the lateral position estimation unit 16 divides the above-mentioned predetermined range into a plurality of intervals and calculates the change amount of the lateral position in each interval. In addition, the lateral position estimation unit 16 may calculate, for each of the lane markings on the left and right sides, the change amount of the lateral position as the difference between the maximum value and the minimum value of the lateral position in the above-mentioned predetermined range.

[0080] Next, in step S106, the lateral position estimation unit 16 determines whether the change amount of the lateral position of the lane line on the left side of the driving path is equal to or greater than a threshold value. The threshold value is determined in advance considering the allowable range of the error in the lateral position and the like. Further, in step S105, when calculating the change amount of the lateral position for each of the plurality of divided sections, the maximum value of the change amount of the lateral position is compared with the threshold value.

[0081] When it is determined in step S106 that the change amount of the lateral position of the lane line on the left side of the driving path is equal to or greater than the threshold value, this control routine proceeds to step S107. In step S107, the lateral position estimation unit 16 determines whether the change amount of the lateral position of the lane line on the right side of the driving path is equal to or greater than the threshold value. The threshold value in step S107 is the same as the threshold value in step S106. Further, similarly to step S106, in step S105, when calculating the change amount of the lateral position for each of the plurality of divided sections, the maximum value of the change amount of the lateral position is compared with the threshold value.

[0082] When it is determined in step S107 that the change amount of the lateral position of the lane line on the right side of the driving path is less than the threshold value, this control routine proceeds to step S108. In step S108, the lateral position estimation unit 16 uses the lane line on the right side of the map as the comparison object to estimate the lateral position of the vehicle 20. That is, the lateral position estimation unit 16 compares the lane line on the right side of the map with the actual lane line on the right side, and estimates the lateral position of the vehicle 20 based on the relative position relationship between the actual lane line on the right side and the vehicle 20. After step S108, this control routine ends.

[0083] On the other hand, when it is determined in step S106 that the change amount of the lateral position of the lane line on the left side of the driving path is less than the threshold value, this control routine proceeds to step S109. In step S109, similarly to step S107, the lateral position estimation unit 16 determines whether the change amount of the lateral position of the lane line on the right side of the driving path is equal to or greater than the threshold value.

[0084] When it is determined in step S109 that the change amount of the lateral position of the lane line on the right side of the driving path is less than the threshold value, this control routine proceeds to step S110. In step S110, the lateral position estimation unit 16 uses the lane line on the left side of the map as the comparison object to estimate the lateral position of the vehicle 20. That is, the lateral position estimation unit 16 compares the lane line on the left side of the map with the actual lane line on the left side, and estimates the lateral position of the vehicle 20 based on the relative position relationship between the actual lane line on the left side and the vehicle 20. After step S110, this control routine ends.

[0085] In addition, when it is determined in step S107 that the change amount of the lateral position of the lane line on the right side of the driving path is equal to or greater than the threshold value, or when it is determined in step S109 that the change amount of the lateral position of the lane line on the right side of the driving path is less than the threshold value, this control routine proceeds to step S111. In step S111, the lateral position estimation unit 16 estimates the lateral position of the vehicle 20 with the lane lines on both sides of the map as the comparison objects. In this case, the lateral position estimation unit 16 calculates the first lateral position of the vehicle 20 using the right lane line in the same manner as in step S108, and calculates the second lateral position of the vehicle 20 using the left lane line in the same manner as in step S110. Then, the lateral position estimation unit 16 calculates the average position of the first lateral position and the second lateral position as the estimated lateral position of the vehicle 20. After step S111, this control routine ends.

[0086] <Second Embodiment>

[0087] The vehicle estimation device according to the second embodiment is basically the same in structure and control as the vehicle estimation device according to the first embodiment except for the points described below. Therefore, hereinafter, the second embodiment of the present invention will be described centering on the parts different from the first embodiment.

[0088] As described above, at a location as shown in Figures 4 to 6 , the lateral position of the lane line with respect to the driving path of the vehicle 20 changes. As long as the change in the lateral position can be detected, the location where the lateral position changes can be used as an index to correct the longitudinal position of the vehicle 20.

[0089] Therefore, in the second embodiment, when the longitudinal position estimation unit 15 detects a location where the lateral position of the actual lane line with respect to the driving path of the vehicle 20 changes by the lane line detection unit 17, the longitudinal position of the vehicle 20 is corrected based on the detected relative positional relationship between the location and the vehicle 20. Thereby, the estimation error of the longitudinal position of the vehicle 20 can be reduced.

[0090] <Longitudinal Position Correction Process>

[0091] Hereinafter, with reference to the Figure 9 flowchart, the control for correcting the longitudinal position of the vehicle 20 will be described in detail. Figure 9 is a flowchart showing a control routine of the longitudinal position correction process in the second embodiment of the present invention. This control routine is repeatedly executed by the ECU 10 at a predetermined execution interval. The predetermined execution interval is, for example, the interval at which the output of the target detection device 2 is sent to the ECU 10.

[0092] First, in step S201, the longitudinal position estimation unit 15 determines whether a location where the lateral position of the actual lane line detected by the lane line detection unit 17 changes with respect to the traveling path of the vehicle 20 (hereinafter referred to as "lateral position change location") is detected. As an example of the lateral position change location, the boundary between a white line and a zebra crossing ( Figure 4 and Figure 5 ), a location where the extending direction of the white line changes discontinuously ( Figure 6 ), etc. can be cited. The lane line detection unit 17 detects the lateral position change location based on the output of the target detection device 2 (specifically, the in-vehicle camera 21 or the lidar 22).

[0093] When it is determined in step S201 that no lateral position change location is detected, this control routine ends. On the other hand, when it is determined in step S201 that a lateral position change location is detected, this control routine proceeds to step S202.

[0094] In step S202, the longitudinal position estimation unit 15 corrects the longitudinal position of the vehicle 20 based on the relative positional relationship between the lateral position change location detected by the lane line detection unit 17 and the vehicle 20. Specifically, the longitudinal position estimation unit 15 determines the position on the map of the lateral position change location detected by the lane line detection unit 17 based on the longitudinal position of the vehicle 20 estimated from the output of the vehicle state detection device. Then, the longitudinal position estimation unit 15 corrects the longitudinal position of the vehicle 20 estimated from the output of the vehicle state detection device to the position on the map of the lateral position change location. After step S202, this control routine ends.

[0095] As described above, the preferred embodiments of the present invention have been explained, but the present invention is not limited to these embodiments, and various modifications and changes can be made within the scope of the claims.

[0096] For example, the lateral position estimation unit 16 may also calculate an initial value of the lateral position of the vehicle 20 based on the output of the vehicle state detection device, and estimate the lateral position of the vehicle 20 by correcting this initial value. In this case, the lateral position estimation unit 16 compares the lane lines on the map with the actual lane lines, and corrects the initial value of the lateral position based on the relative positional relationship between the actual lane lines and the vehicle 20, thereby estimating the lateral position of the vehicle 20. In addition, in this case, when the amount of change in the lateral position of the lane lines on both sides of the traveling path is equal to or greater than a threshold value ( Figure 7 in the case where the determination in step S107 is affirmative), the initial value of the lateral position may also be used as the estimated position of the lateral position of the vehicle 20.

[0097] In addition, the lateral position estimation unit 16 may use the lane line on one side of the driving path as the comparison target when the amount of change in the lateral position of the lane line on one side of the driving path is smaller than the amount of change in the lateral position of the lane line on the other side of the driving path. In addition, the lateral position estimation unit 16 may use the lane line on the side with the smaller amount of change as the comparison target when the difference between the amount of change in the lateral position of the lane line on one side of the driving path and the amount of change in the lateral position of the lane line on the other side of the driving path is equal to or greater than a predetermined value.

[0098] In addition, a computer program for causing a computer to implement the functions of the respective parts of the processor of the vehicle position estimation device according to the above-described embodiment may also be provided in a form stored in a computer-readable recording medium. Examples of the computer-readable recording medium include a magnetic recording medium, an optical recording medium, or a semiconductor memory.

Claims

1. A vehicle position estimation device, comprising: A longitudinal position estimation unit that estimates the longitudinal position of the vehicle based on the output of a vehicle state detection device that detects the state quantity of the vehicle; A lane line detection unit that detects the actual lane lines on the left and right sides of the vehicle; A lateral position estimation unit that compares the lane lines on the map at the longitudinal position of the vehicle with the actual lane lines, and estimates the lateral position of the vehicle based on the relative position relationship between the actual lane lines and the vehicle; and A travel path determination unit that determines the travel path of the vehicle, The lateral position estimation unit calculates the change amount of the lateral position with respect to the travel path for each of the lane lines on the left and right sides of the travel path within a predetermined range before and after the longitudinal position of the vehicle based on the map information, and determines the lane lines on the map to be compared based on the change amount; When the change amount with respect to the lane line on one side of the travel path is less than a predetermined threshold and the change amount with respect to the lane line on the other side of the travel path is equal to or greater than the threshold, the lateral position estimation unit uses the lane lines on the map on that one side as the comparison object.

2. The vehicle position estimation device according to claim 1, wherein, When the longitudinal position estimation unit detects a location where the lateral position of the actual lane line with respect to the driving path changes, the longitudinal position of the vehicle is corrected based on the relative positional relationship between this location and the vehicle.

3. A vehicle position estimation method, comprising: Estimate the longitudinal position of the vehicle based on the output of a vehicle state detection device that detects the state quantity of the vehicle; Detect the actual lane lines on the left and right sides of the vehicle; Compare the lane line on the map at the longitudinal position of the vehicle with the actual lane line, and estimate the lateral position of the vehicle based on the relative positional relationship between the actual lane line and the vehicle; And Determine the driving path of the vehicle, Estimating the lateral position of the vehicle includes: Based on the map information, for each lane line on the left and right sides of the driving path within a predetermined range before and after the longitudinal position of the vehicle, calculate the change amount of the lateral position with respect to the driving path, and determine the lane line on the map to be compared based on this change amount. When the change amount of the lane line on one side of the driving path is less than a predetermined threshold and the change amount of the lane line on the other side of the driving path is equal to or greater than the threshold, use the lane line on the map of this side as the comparison object.

4. A computer-readable storage medium storing a computer program for vehicle position estimation, the program causing a computer to execute: Estimate the longitudinal position of the vehicle based on the output of a vehicle state detection device that detects the state quantity of the vehicle; Detect the actual lane lines on the left and right sides of the vehicle; Compare the lane line on the map at the longitudinal position of the vehicle with the actual lane line, and estimate the lateral position of the vehicle based on the relative positional relationship between the actual lane line and the vehicle; And Determine the driving path of the vehicle, Wherein, Estimating the lateral position of the vehicle includes: Based on the map information, for each lane line on the left and right sides of the driving path within a predetermined range before and after the longitudinal position of the vehicle, calculate the change amount of the lateral position with respect to the driving path, and determine the lane line on the map to be compared based on this change amount. When the change amount of the lane line on one side of the driving path is less than a predetermined threshold and the change amount of the lane line on the other side of the driving path is equal to or greater than the threshold, use the lane line on the map of this side as the comparison object.

Images