Lane change detection device, driving assistance device, and lane change detection method

Abstract

The objective of the present disclosure is to detect a lane change by a vehicle without using a camera. A lane change detection device (12) comprises: an autonomous navigation unit (127) that creates a history of DR positions and DR orientations as a position history; and a lane change detection unit (128) that uses the position history to detect a lane change by an object vehicle (200). The lane change detection unit (128) extracts a first section (S1) in which the DR orientations in the position history change by an amount exceeding a predetermined threshold as well as a second section (S2) and a third section (S3) which come before and after the first section (S1), uses an approximate curve of the DR positions in at least the second section (S2) and the third section (S3) to detect a lane change by the object vehicle (200) in the first section (S1), and extracts DR positions in the first section (S1) where a lane change is detected as well as in the second section (S2) and the third section (S3) adjacent to the first section (S1) as a path of travel at the time of the lane change.

Description

[Technical Field]

[0001] The present disclosure relates to a technique for detecting lane changes in a vehicle. [Background technology]

[0002] Vehicles are equipped with a variety of driving assistance functions to enable drivers to drive safely and comfortably in various traffic conditions and to prevent accidents. Driving assistance functions include functions to enhance awareness of surrounding conditions, functions to warn or control the distance to prevent collisions with vehicles ahead, functions to warn or prevent departure from the driving lane, functions to assist in safe lane changes, and functions to adjust the beam axis of the headlights.

[0003] Furthermore, devices capable of highly accurate vehicle positioning have begun to be used for driver assistance. In the automotive field, inexpensive GNSS receivers and antennas compatible with dual-frequency signals (L1 and L2C; both civilian signals, L5 is not supported) from multi-GNSS (Global Navigation Satellite System) satellites, such as GPS, GLONASS, Galileo, Beidou, and QZSS, have come into use in recent years. Furthermore, in Japan, positioning augmentation information for precise point positioning (PPP-RTK method) is now distributed free of charge from quasi-zenith satellites, making high-precision positioning of float and fixed solutions available to positioning devices. This has enabled autonomous navigation and hybrid positioning using vehicle speed (wheel speed) pulses and an inertial measurement unit (IMU) to achieve the positioning accuracy required to update the vehicle's position in the driving lane, enabling highly accurate positioning information to be utilized by driver assistance systems.

[0004] Patent Document 1 discloses a technique for determining whether or not a vehicle has crossed a lane marking based on a captured image.

[0005] Patent Document 2 discloses a technique for recognizing a marking line from a captured image and determining the distance between the vehicle and the marking line. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-162552 [Patent Document 2] Patent No. 4876147 Summary of the Invention [Problem to be solved by the invention]

[0007] According to the technologies disclosed in Patent Documents 1 and 2, a camera that measures the distance between the vehicle and the white line of the driving lane is essential for determining whether to change lanes. Therefore, if the camera does not perform as expected, a functional safety issue arises in that the expected driving assistance cannot be provided to the driver.

[0008] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a technology for detecting lane changes of a vehicle without using a camera. [Means for solving the problem]

[0009] The lane change detection device of the present disclosure includes an autonomous navigation unit that creates a position history that includes a history of DR positions and DR orientations, which are the positions and orientations of a target vehicle measured by autonomous navigation, and a lane change detection unit that detects lane changes of the target vehicle using the position history. The lane change detection unit extracts a first section in which a change in DR orientation exceeds a predetermined threshold in the position history, and a second section and a third section that are a predetermined number of sections before and after the first section, detects lane changes of the target vehicle in the first section using an approximation curve of the DR positions in at least the second and third sections, extracts the DR positions in the first section in which the lane change was detected, and the second and third sections adjacent to the first section, as a driving trajectory during the lane change, and uses the driving trajectory during the lane change for driving assistance of the target vehicle. [Effects of the Invention]

[0010] The lane change detection device of the present disclosure can detect lane changes without using a camera based on the position history of a target vehicle obtained by autonomous navigation. Objects, features, aspects, and advantages of the present disclosure will become more apparent from the following detailed description and the accompanying drawings. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of a positioning system according to a first embodiment. [Figure 2] 4 is a flowchart showing the operation of the positioning system according to the first embodiment. [Figure 3] 3 is a diagram showing timings for determining a lane change by the lane change detection device according to the first embodiment; FIG. [Figure 4] 3 is a diagram showing a method for determining a lane change by the lane change detection device according to the first embodiment. FIG. [Figure 5] 3 is a diagram showing a method of determining whether a vehicle is overtaking by the lane change detection device according to the first embodiment. FIG. [Figure 6] FIG. 10 is a diagram illustrating a method for determining lane changes in a curved road section. [Figure 7] FIG. 10 is a diagram illustrating a method for determining lane changes in a curved road section. [Figure 8] FIG. 10 is a block diagram showing the configuration of a driving assistance device according to a second embodiment. [Figure 9] FIG. 2 is a diagram showing the mounting positions and detection ranges of a stereo camera and a millimeter-wave radar when the target vehicle is viewed from the right side. [Figure 10] 1 is a diagram showing the mounting positions and detection ranges of a stereo camera and a millimeter-wave radar when the target vehicle is viewed from above. FIG. [Figure 11] 6 is a flowchart showing the operation of the driving assistance device according to the second embodiment. [Figure 12] FIG. 10 is a diagram illustrating an example of determining whether or not a lane change is possible, taking into consideration a preceding vehicle. [Figure 13] FIG. 10 is a diagram illustrating an example of determining whether or not a lane change is possible, taking into consideration a preceding vehicle. [Figure 14]FIG. 10 is a diagram showing a state immediately after the target vehicle starts lane change assistance. [Figure 15] FIG. 10 is a diagram showing a state in which a vehicle ahead is out of the measurement range during lane change assistance for a target vehicle. [Figure 16] FIG. 10 is a diagram showing a state in which lane change assistance for a target vehicle according to a recommended driving path has been completed. [Figure 17] 10A and 10B are diagrams illustrating timings at which forgetting to operate a turn signal when changing lanes is detected. [Figure 18] 10A and 10B are diagrams illustrating timings at which a driver forgets to check the mirrors when changing lanes. [Figure 19] FIG. 10 is a diagram illustrating an example of determining whether or not a lane change is possible, taking into consideration a vehicle behind; [Figure 20] FIG. 10 is a diagram showing a situation in which a nearby vehicle traveling parallel to the target vehicle in an adjacent lane changes lanes in front of the target vehicle. [Figure 21] FIG. 10 is a diagram showing a situation in which the target vehicle must pass a parked vehicle by straying into an adjacent lane. [Figure 22] FIG. 10 is a diagram showing a situation in which a target vehicle stops before a recommended driving path. [Figure 23] FIG. 10 is a diagram showing a situation in which the distance between the vehicle and the rear vehicle is sufficiently long to determine whether the vehicle can overtake the rear vehicle. [Figure 24] FIG. 10 is a block diagram showing the configuration of a driving assistance device according to a third embodiment. [Figure 25] 10 is a flowchart showing a driving assistance process performed by a driving assistance device according to a third embodiment. [Figure 26] FIG. 10 is a diagram showing the relationship between the low beam illumination range and the recommended driving path. [Figure 27] FIG. 10 is a diagram showing the relationship between the high beam illumination range and the recommended driving path. [Figure 28] FIG. 10 is a diagram showing a state in which the optical axis direction of the headlights is adjusted toward an adjacent lane. [Figure 29] FIG. 10 is a diagram showing a long recommended driving path illuminated with high-intensity high beams. [Figure 30] 10 is a diagram showing an example of a display by a driving assistance device according to a third embodiment. FIG. [Figure 31] It is a diagram showing the hardware configuration of a lane change detection system and a driving support device. [Figure 32] It is a diagram showing the hardware configuration of a lane change detection system and a driving support device.

Embodiments for Carrying Out the Invention

[0012] <A. Embodiment 1> <A-1. Configuration> FIG. 1 is a block diagram showing the configuration of a lane change detection system 10 according to Embodiment 1. The lane change detection system 1 Ten is configured to include a satellite positioning device 11, a lane change detection device 12, and high-precision map data 13. The lane change detection system 10 detects a lane change of a vehicle and extracts a travel trajectory at the time of lane change. In the present specification, lane change includes overtaking. Overtaking means changing lanes to pass a vehicle ahead and then returning to the original lane. In the present specification, the vehicle for which the lane change detection system 10 detects a lane change is referred to as the target vehicle.

[0013] The satellite positioning device 11 is configured to include a GNSS receiver 111, a positioning augmentation signal receiver 112, and a satellite positioning unit 113.

[0014] The GNSS receiver 111 includes a GNSS antenna and receives radio waves in a predetermined frequency band broadcast from a plurality of GNSS satellites existing above the target vehicle, and outputs time data, observation data, and orbit data. The plurality of GNSS satellites include at least GPS satellites. The radio waves received by the GNSS receiver 111 are, for example, two-frequency positioning signals of the L1 signal and the L2C signal. The observation data includes the pseudo range, carrier phase, and Doppler shift frequency for each satellite signal. The orbit data is the data of the broadcast ephemeris necessary for calculating the position of the satellite.

[0015] The positioning augmentation signal receiver 112 receives a PPP-RTK (Precise Point Positioning-Real Time Kinematic) positioning augmentation signal (L6 signal) broadcast to the ground by a quasi-zenith satellite using a GNSS antenna, and outputs positioning augmentation data.

[0016] The satellite positioning unit 113 first uses time data and observation data and orbit data of the satellites used for standalone positioning to determine the position of the target vehicle (standalone positioning solution) and the error between the built-in clock of the GNSS receiver 111. Next, as described in International Publication No. 2022 / 259365, the satellite positioning unit 113 determines the position of the target vehicle (float solution, fixed solution) and carrier phase bias using the observation data and positioning augmentation data of the satellites that are the target of the positioning augmentation data, and predicts the positioning error of each positioning solution of the satellite positioning. This completes the configuration of the satellite positioning device 11.

[0017] The lane change detection device 12 includes a speed sensor 121, an angular velocity sensor 124, a distance measurement unit 122, a speed sensor correction unit 123, a yaw angle measurement unit 125, an angular velocity sensor correction unit 126, an autonomous navigation unit 127, a lane change detection unit 128, and a positioning error prediction unit 129.

[0018] The speed sensor 121 is a sensor for autonomous navigation, and outputs a pulse signal according to the distance traveled by the vehicle.

[0019] The distance measurement unit 122 calculates the moving distance and the speed from the number of pulses of the speed sensor 121 measured at predetermined intervals.

[0020] The speed sensor correction unit 123 calculates an SF coefficient (scale factor) that indicates the distance per pulse output by the speed sensor 121.

[0021] The angular velocity sensor 124 adds a signal corresponding to the angular velocity (yaw rate) to the zero point output, with the vertical direction of the device housing as the sensor detection axis, and outputs the result.

[0022] The yaw angle measurement unit 125 calculates the yaw angle from the output of the angular velocity sensor 124 measured at predetermined timings.

[0023] The angular velocity sensor correction unit 126 obtains the zero output of the angular velocity sensor 124.

[0024] The dead reckoning unit 127 updates the position (DR position), speed, and orientation (DR orientation) of the target vehicle using the moving distance calculated by the distance measurement unit 122 and the yaw angle calculated by the yaw angle measurement unit 125, and creates a history of the most recent position (DR position) of the target vehicle.

[0025] The lane change detection unit 128 obtains the error of dead reckoning based on the positioning result and prediction error in the satellite positioning unit 113, and corrects the position, speed, and orientation of the target vehicle obtained by the dead reckoning unit 127. Further, the lane change detection unit 128 extracts the traveling trajectory at the time of lane change by analyzing the position history.

[0026] The positioning error prediction unit 129 predicts the errors in the speed and orientation of the target vehicle.

[0027] The high-precision map data 13 has information such as three-dimensional shape information for each lane, three-dimensional shape information of the road shoulder, their longitudinal gradients, transverse gradients, and road elevation, which are produced with an absolute accuracy of less than 50 cm.

[0028] <A-2. Operation> FIG. 2 is a flowchart of the lane change detection system 10 regarding the extraction of the traveling trajectory at the time of lane change. FIG. 3 shows the determination timing of lane change by the lane change detection device 12. FIGS. 4 and 5 show the determination method of lane change by the lane change detection device 12. Hereinafter, the processing of the lane change detection device 12 will be described with appropriate reference to FIGS. 2 to 5.

[0029] First, in step S101, the lane change detection device 12 initializes the processing related to integrated positioning. Here, for example, the position of the target vehicle measured last time is restored.

[0030] Next, in step S102, the distance measurement unit 122 calculates the distance traveled by multiplying the number of pulses of the speed sensor 121 measured at each predetermined period by an SF coefficient, and calculates the speed using a value obtained by passing the number of pulses at each predetermined period through a low-pass filter.

[0031] Then, in step S103, the angular velocity sensor correction unit 126 determines whether the target vehicle has stopped based on the distance traveled by the distance measurement unit 122, using the methods described in Japanese Patent No. 3137784 and Japanese Patent No. 3751513, calculates the average value of the output of the angular velocity sensor 124 while the target vehicle is stopped, and sets the average value as the output bias of the angular velocity sensor 124.

[0032] Next, in step S104, the yaw angle measurement unit 125 calculates the yaw angle by removing the output bias from the output of the angular velocity sensor 124 measured at each predetermined timing.

[0033] Then, in step S105, the autonomous navigation unit 127 calculates a movement vector for each predetermined period using the movement distance and yaw angle, and updates the position of the target vehicle (DR position) by adding the movement vector to the previously measured position of the target vehicle.

[0034] Next, in step S106, the autonomous navigation unit 127 creates a recent history of the target vehicle's position (DR position) and orientation (DR orientation). This history is referred to as the position history. The recent position history is a history of sufficient length to include not only lane changes but also the time before and after the changes, and the number of histories is predetermined from that perspective. As shown in FIG. 3, the recent position history includes the history of a first section S1 in which the amount of change in the target vehicle's orientation (DR orientation) exceeds a predetermined reference value, and a second section S2 and a third section S3 in which the yaw angle remains equal to or less than a predetermined value in the sections before and after the first section S1.

[0035] The first section S1 includes, for example, a section from when the yaw angle swings in a certain direction by a predetermined threshold or more to when the yaw angle swings back in the opposite direction by a predetermined threshold or more. The second section S2 and the third section S3 are each a section of a predetermined time when the target vehicle is traveling at a predetermined speed or more.

[0036] Then, in step S107, the lane change detection unit 128 detects a first section S1 in which the amount of change in the DR orientation of the target vehicle exceeds a predetermined threshold from the most recent position history of the target vehicle. After that, when the position histories of the second section S2 and the third section S3 before and after the first section S1 are collected, the lane change detection unit 128 determines whether or not a lane change has occurred from the approximate polynomial of the trajectory of the DR position in each section.

[0037] For example, as shown in Fig. 4, the lane change detection unit 128 performs polynomial approximation on the locus of the DR positions in the second section S2 and the third section S3. The locus in the second section S2 is expressed as y2 = a2x 2 +b2x+c2, and the running trajectory of the third section S3 is y3=a3x 2 +b3x+c3 The lane change detection unit 128 determines whether the difference between these two polynomial approximation curves is due to a lane change.

[0038] Specifically, the lane change detection unit 128 determines that a lane change other than overtaking has occurred if the deviation d between the two polynomial approximation curves in the lane width direction perpendicular to the direction of travel of the target vehicle 200, i.e., the lateral direction, is greater than or equal to the width of the driving lane.

[0039] On the other hand, if the target vehicle 200 temporarily overtakes and moves into an adjacent lane, its travel trajectory will be as shown in Fig. 5, and the lateral deviation d of the polynomial approximation curves in the second section S2 and the third section S3 will be less than the travel lane width. In this case, the lane change detection unit 128 compares the lateral movement amount dmax in the first section S1 with the lane width, and if the movement amount dmax exceeds the lane width, determines that the target vehicle 200 has changed lanes and overtaken.

[0040] Thereafter, in step S108, the lane change detection unit 128 determines whether or not a lane change has occurred in the most recent position history. If a lane change has occurred, the process of the lane change detection device 12 proceeds to step S109, and if a lane change has not occurred, the process of the lane change detection device 12 proceeds to step S111.

[0041] In step S109, the lane change detection unit 128 extracts the history of the DR positions in the second section S2 where the lane change was confirmed and the first section S1 and third section S3 before and after it as a travel trajectory at the time of the lane change.

[0042] Thereafter, in step S110, the lane change detection unit 128 learns driving tendencies during lane changes through statistical processing. The driving tendencies learned here include at least the driving trajectory. Specifically, the lane change detection unit 128 organizes not only lane changes by time, day of the week, or road, but also the speed and driving trajectory characteristics of the sections before and after the lane changes. Here, time is distinguished as daytime, evening, and nighttime. Furthermore, day of the week is distinguished as weekdays, holidays, and weekends. Roads are distinguished by whether they are expressways or general roads, whether they are straight or curved, flat or hilly, and the number of lanes.

[0043] After step S110, or if there is no lane change in step S108, the lane change detection unit 128 sets the DR position as the initial value of the composite position of the target vehicle 200 in step S111.

[0044] Next, in step S112, the lane change detection device 12 checks whether the positioning dimension of the satellite positioning cycle is non-positioning. If the positioning dimension of the satellite positioning cycle is not non-positioning, the processing of the lane change detection system 10 proceeds to step S113, and if it is non-positioning, the processing of the lane change detection system 10 ends.

[0045] In step S113, the satellite positioning unit 113 performs satellite positioning of the target vehicle 200. Specifically, as described in Japanese Patent No. 4988028, Japanese Patent No. 6482720, and International Publication No. 2022 / 259365, the satellite positioning unit 113 compares the pseudorange with the Doppler shift frequency or the carrier phase to calculate a pseudorange residual and select a satellite to be used for positioning, and then determines the position of the target vehicle 200 (standalone positioning solution) and the error of the clock built into the GNSS receiver 111 using time data and observation data and orbit data of the satellites used for standalone positioning. Next, as described in International Publication No. 2022 / 259365, the satellite positioning unit 113 uses the observation data of the satellite that is the target of the positioning augmentation data and the positioning augmentation data to determine the position (float solution, fixed solution) and carrier phase bias of the target vehicle, and predicts the positioning error of each positioning solution of the satellite positioning.

[0046] Next, in step S114, the lane change detection unit 128 corrects the composite position of the target vehicle using satellite positioning data (observation values ​​of satellites used for positioning, positioning augmentation data, and point solution), as described in Japanese Patent Nos. 5855249, 6482720, 6877854, and 6929492. This enables the lane change detection unit 128 to calculate a composite positioning solution that matches the driving trajectory even in the presence of local radio wave obstruction or multipath. Furthermore, when precise point positioning using a positioning augmentation signal is performed, as described in International Publication No. 2022 / 259365, the lane change detection unit 128 corrects the error in the target vehicle's position for composite positioning based on the satellite positioning data (float solution or fixed solution and their prediction error). This allows the lane change detection unit 128 to update the position of the target vehicle within the driving lane.

[0047] Next, in step S115, the speed sensor correction unit 123 corrects the SF coefficient of the pulse signal based on the method described in Japanese Patent No. 5606656.

[0048] Thereafter, in step S116, while the target vehicle is in motion, the angular velocity sensor correction unit 126 corrects the output bias of the angular velocity sensor 124 from the difference between the azimuth obtained by integrating the yaw angle at each moment with the azimuth of the target vehicle at an arbitrary time as the initial value and the azimuth of the target vehicle for composite positioning of the lane change detection unit 128, as disclosed in Japanese Patent No. 3321096 and Japanese Patent No. 3727489.

[0049] <A-3. Modified Example> Even in the following cases where there are some differences from the content described in the first embodiment, it exhibits remarkable effects not found in the prior art.

[0050] In the above description, the autonomous navigation unit 127 created a history of the DR position of the most recent target vehicle in step S106. However, in step S106, the autonomous navigation unit 127 may also create a history of the operation signals of the direction indicator in the same time zone. Then, in step S109, the lane change detection unit 128 extracts the operation history of the direction indicator in the same time zone as the driving trajectory, and in step S110, learns (statistical processing) the driving tendency (direction indicator operation) at the time of lane change, and may associate the presence or absence and timing of the operation of the direction indicator when organizing the characteristics of the speed or driving trajectory required for lane change. Thereby, the driving tendency at the time of lane change can be better grasped and utilized for driving support.

[0051] In the above description, the lane change detection unit 128 used a polynomial approximation formula to express the position history by a mathematical formula in step S107. However, any approximation formula other than the polynomial approximation formula may be used as long as it can appropriately approximate the position history. Also, the creation conditions of the position history used for the calculation of the approximation formula may be changed.

[0052] Regarding lane change detection, as described above, the lane change detection unit 128 detects a first section S1 in which the target vehicle 200 experiences a large change in direction, as shown in Figures 4 and 5. After the position histories of the second and third sections S2 and S3 before and after the first section S1 are collected, the lane change detection unit 128 performs polynomial approximation on the position histories of the second and third sections S2 and S3, and determines whether or not a lane change has occurred based on the deviation between the two polynomial approximation curves. However, in road sections where the lanes before and after a lane change are not parallel, a different method may be used. Figures 6 and 7 show a method for determining whether or not a lane change has occurred in a curved road section.

[0053] That is, the lane change detection unit 128 performs polynomial approximation of the first travel path (P1-P20) which is the actual travel path in the first section S1, the second section S2, and the third section S3. The first curve which is a polynomial approximation curve of the first travel path (P1-P8, P14-P20) is expressed as y1=a1x 3 +b1x 2 It is expressed as +c1x+d1.

[0054] Furthermore, the lane change detection unit 128 creates a first parallel running coordinate group (P2'-P8') that runs parallel to the actual running trajectory (P1-P8) in the second section S2 in an adjacent lane, and creates a second parallel running coordinate group (P15'-P20') that runs parallel to the actual running trajectory (P14-P20) in the third section S3 in an adjacent lane. The first parallel running coordinate group (P2'-P8') is parallel to the actual running trajectory (P1-P8), and the second parallel running coordinate group (P15'-P20') is parallel to the actual running trajectory (P14-P20). The first parallel running coordinate group (P2'-P8') is a coordinate group in the adjacent lane in the direction in which the DR orientation changed in the first section. The second parallel running coordinate group (P15'-P20') is a coordinate group in the adjacent lane in the opposite direction to the direction in which the DR orientation changed in the first section.

[0055] The lane change detection unit 128 performs polynomial approximation of a second traveling trajectory (P1-P8, P15'-P20') that connects the actual traveling trajectory (P1-P8) in the second section S2 and the second parallel traveling coordinate group (P15'-P20') in the third section S3. The second curve, which is a polynomial approximation curve of the second traveling trajectory (P1-P8, P15'-P20'), is expressed as y2=a2x 3 +b2x 2 It is expressed as +c2x+d2.

[0056] Furthermore, the lane change detection unit 128 performs polynomial approximation of a third traveling trajectory (P2'-P8', P14-P20) that connects the first parallel traveling coordinate group (P2'-P8') in the second section S2 and the actual traveling trajectory (P14-P20) in the third section S3. The third curve, which is a polynomial approximation curve of the third traveling trajectory (P2'-P8', P14-P20), is expressed as y3=a3x 3 +b3x 2 It is expressed as +c3x+d3.

[0057] The lane change detection unit 128 detects the first curve y1=a1x 3 +b1x 2 +c1x+d1 and the second curve y2=a2x 3 +b2x 2 +c2x+d2, and if there is a significant difference in the parameters (a1, a2, b1, b2, c1, c2, d1, d2) of these two curves, it is determined that a lane change has occurred in the first section S1. 3 +b1x 2 +c1x+d1 and the third curve y3=a3x 3 +b3x 2 +c3x+d3, and if there is a significant difference in the parameters (a1, a3, b1, b3, c1, c3, d1, d3) of the two curves, it may be determined that a lane change has occurred in the first section S1. Alternatively, the lane change detection unit 128 may determine that a lane change has occurred in the first section S1 if there is a significant difference in the parameters of the polynomial approximation curves between the first and second traveling trajectories and if there is a significant difference in the parameters of the polynomial approximation curves between the first and third traveling trajectories.

[0058] <A-4. Effect> The lane change detection device 12 according to Embodiment 1 includes an autonomous navigation unit 127 and a lane change detection unit 128. The autonomous navigation unit 127 creates a position history of the DR position and DR orientation, which are the position and orientation of the target vehicle measured by autonomous navigation, as a position history. The lane change detection unit 128 detects a lane change of the target vehicle using the position history. More specifically, the lane change detection unit 128 extracts a first section in which the change amount of the DR orientation in the position history exceeds a predetermined threshold value, and second and third sections that are sections of a predetermined number of histories before and after the first section. Further, the lane change detection unit 128 detects a lane change of the target vehicle in the first section using at least an approximate curve of the DR position in the second and third sections. Also, the lane change detection unit 128 extracts the DR positions in the first section where a lane change is detected and the second and third sections adjacent to the first section as the driving trajectory at the time of lane change. The driving trajectory at the time of lane change is used for driving support of the target vehicle. With the above configuration, according to the lane change detection device 12, since the lane change of the target vehicle 200 can be detected without using a camera, the lane change can be detected even when the performance of the camera cannot be exhibited.

[0059] <B. Embodiment 2> <B-1. Configuration> FIG. 8 is a block diagram showing the configuration of the driving support device 20 according to Embodiment 2. The driving support device 20 performs driving support for the target vehicle 200 using the processing result of the lane change detection system 10. In the present embodiment, other vehicles traveling around the target vehicle 200 are referred to as surrounding vehicles.

[0060] In addition to the lane change detection system 10 according to Embodiment 1, the driving support device 20 includes a surrounding measurement unit 21 that measures or detects the surroundings of the target vehicle 200 using a sensor provided in the target vehicle 200, an operation input unit 31, a driving support control unit 32, and a display control unit 33.

[0061] The sensors provided on the target vehicle 200 include a stereo camera and a millimeter-wave radar. The stereo camera is a pair of cameras provided on the left and right sides of the target vehicle 200, and captures images of obstacles such as nearby vehicles or pedestrians around the target vehicle 200, as well as road markers such as white or yellow lines. The periphery measurement unit 21 measures the three-dimensional position, size, and shape of the obstacle based on the displacement (parallax) of the obstacle in the image captured by the stereo camera. The periphery measurement unit 21 also detects road markers such as white or yellow lines from the image captured by the stereo camera, and measures the lateral positions of the target vehicle and the road markers.

[0062] Stereo cameras have the advantage of being able to detect the movement of obstacles crossing the detection direction. However, their detection ability is reduced when the windshield in front of the camera lens becomes dirty or foggy, or when the camera is placed in bad weather such as heavy rain or in a backlit environment. Stereo cameras also have performance limitations, such as being unable to detect vehicles with unlit lights at night or in tunnels.

[0063] Millimeter-wave radar emits millimeter waves (electromagnetic waves) at a predetermined detection angle and detects the distance to an obstacle based on the time it takes for the millimeter waves to reflect off the obstacle and return. Millimeter-wave radar has the advantage of being excellent at measuring long distances and being able to ensure distance measurement performance regardless of sunlight conditions, brightness, or weather (rain, fog). However, millimeter-wave radar has performance limitations in that it has difficulty detecting obstacles with low reflectivity and cannot detect the movement of obstacles that cross the radar's detection direction.

[0064] For this reason, the driving assistance device 20 monitors the periphery of the target vehicle by combining a stereo camera and a millimeter wave radar, which have different characteristics.

[0065] The stereo camera that photographs the area in front of the target vehicle 200 is referred to as a stereo camera (front) 22, and the stereo camera that photographs the area behind the target vehicle 200 is referred to as a stereo camera (rear) 24.

[0066] A millimeter wave radar whose detection direction is in front of the target vehicle 200 is referred to as millimeter wave radar (front) 23. A millimeter wave radar whose detection direction is to the left front of the target vehicle 200 is referred to as millimeter wave radar (left front) 25. A millimeter wave radar whose detection direction is to the right front of the target vehicle 200 is referred to as millimeter wave radar (right front) 26. A millimeter wave radar whose detection direction is to the left rear of the target vehicle 200 is referred to as millimeter wave radar (left rear) 27. A millimeter wave radar whose detection direction is to the right rear of the target vehicle 200 is referred to as millimeter wave radar (right rear) 28.

[0067] Fig. 9 shows the mounting positions and detection ranges of the stereo camera and millimeter-wave radar when the target vehicle 200 is viewed from the right side. Note that the detection ranges of the millimeter-wave radar (right front) 26 and the millimeter-wave radar (right rear) 28 are omitted from Fig. 9. Fig. 10 shows the mounting positions and detection ranges of the stereo camera and millimeter-wave radar when the target vehicle 200 is viewed from above.

[0068] The stereo camera (front) 22 is installed at the top of the windshield. In Fig. 10, the detection range of the stereo camera (front) 22 is indicated by the symbol 22A. The same applies to the other sensors, and the detection range is indicated by a symbol obtained by adding an A to the symbol of that sensor. The detection angle of the stereo camera (front) 22 is 40 degrees forward, and the detection distance is 100 m. The stereo camera (rear) 24 is installed at the top of the rear windshield, and the detection angle is 40 degrees backward, and the detection distance is 40 m.

[0069] The millimeter-wave radar (front) 23 is installed at the center of the front bumper, with a detection angle of 20 degrees in the front and a detection distance of 200 m. The millimeter-wave radar (left front) 25, the millimeter-wave radar (right front) 26, the millimeter-wave radar (left rear) 27, and the millimeter-wave radar (right rear) 28 are installed at the left corner of the front bumper, the right corner of the front bumper, the left corner of the rear bumper, and the right corner of the rear bumper respectively, with a detection angle of 120 degrees and a detection distance of 30 m. Note that the installation of the sensors shown in FIGS. 9 and 10 is an example. For example, the stereo camera (rear) 24, the millimeter-wave radar (left rear) 27, and the millimeter-wave radar (right rear) 28 may not be provided, and the surrounding measurement unit 21 may be configured to be able to measure only in front of the target vehicle 200.

[0070] The surrounding measurement unit 21 integrally manages the presence and behavior of obstacles detected by the stereo camera and the millimeter-wave radar respectively.

[0071] <B-2. Operation> FIG. 11 is a flowchart showing the operation of the driving support device 20. FIGS. 12 and 13 are diagrams showing examples of determination of whether lane change is possible considering the vehicle ahead. Hereinafter, the operation of the driving support device 20 will be described while appropriately referring to these figures.

[0072] In step S201 of FIG. 十一年11, the driving support device 20 initializes the related processing.

[0073] Next, in step S202, the operation input unit 31 receives an operation input from the driver indicating a desire for driving support processing. When the driver intends to change lanes, for example, the driver operates the operation lever of the direction indicator. This operation is the operation in this step. Alternatively, when the display device is configured as a touch panel, the driver may instruct the driving support device 20 to start the driving support processing by operating the touch panel. Also, both the operation lever of the direction indicator and the operation on the display device may be performed in this step.

[0074] In the subsequent step S203, composite positioning is performed for the target vehicle 200. The processing of the satellite positioning device 11 and the lane change detection device 12 in this step is the same as in the first embodiment, and therefore a description thereof will be omitted.

[0075] Next, in step S204, the periphery measurement unit 21 measures the periphery of the target vehicle 200. In this step, the periphery measurement unit 21 detects surrounding vehicles (hereinafter, "forward vehicles") ahead of the target vehicle 200 using the stereo camera (forward) 22 and the millimeter-wave radar (forward) 23, as described in International Publication No. 2022 / 259365. Because the stereo camera has a wide detection angle, it can detect not only the lane in which the target vehicle 200 is traveling (hereinafter, "traveling lane"), but also a forward vehicle in the lane immediately to the left of the traveling lane. However, if the distance to the forward vehicle exceeds the detection distance, the stereo camera cannot perform detection. For traveling lanes with good visibility, the stereo camera can detect whether the white lines on the left and right of the traveling lane are broken or solid, and can also measure the lateral distance between the white lines and the camera.

[0076] In contrast, millimeter-wave radar has a narrow detection angle, so it may not be able to detect vehicles ahead in the lane to the left of the lane you are driving in. However, because millimeter-wave radar has a long detection distance, it can detect vehicles ahead at a distance in the lane you are driving in that cannot be measured by a stereo camera.

[0077] Similarly, the periphery measurement unit 21 detects surrounding vehicles behind and to the sides (left lane, right lane) that exist within each measurement range using a stereo camera (rear) 24, a millimeter wave radar (left front) 25, a millimeter wave radar (right front) 26, a millimeter wave radar (left rear) 27, and a millimeter wave radar (right rear) 28, and measures the positional relationship between the host vehicle and the surrounding vehicles. Based on these detection results, the periphery measurement unit 21 recognizes the behavior of the surrounding vehicles, the relative positions of the target vehicle 200 and the surrounding vehicles, and the position of the target vehicle 200 on the driving lane, as described in International Publication No. 2022 / 259365.

[0078] Next, in step S205, the driving assistance control unit 32 performs driving assistance processing. For example, as shown in FIG. 12 , when the target vehicle 200 attempts to change lanes to the lane immediately to the left, the driving assistance control unit 32 generates a recommended driving path 202 from the driving trajectory at the time of the lane change, which the lane change detection unit 128 has learned as a driving tendency. For example, when the current conditions are “daytime, weekday, general, straight, flat, two-lane,” the driving assistance control unit 32 extracts driving trajectories from past lane changes under the same conditions from the lane change detection unit 128 of the lane change detection device 12, and creates a recommended driving path 202 from the extracted driving trajectory. When one driving trajectory is extracted, the driving assistance control unit 32 may use the extracted driving trajectory as the recommended driving path 202 as is. When multiple driving trajectories are extracted, the driving assistance control unit 32 may use, for example, an average of the multiple driving trajectories as the recommended driving path 202, or the longest driving trajectory as the recommended driving path 202.

[0079] It is assumed that the stereo camera (front) 22 detects that the white line in the lane immediately to the left of the driving lane is a broken line. Accordingly, the driving assistance control unit 32 determines that the current position of the target vehicle 200 is a position where a lane change to the immediately left lane is permitted. The millimeter wave radar (front) 23 measures the positional relationship between the target vehicle 200 and the forward vehicle 201 traveling in the immediately left lane. The driving assistance control unit 32 calculates the inter-vehicle distance L between the target vehicle 200 and the forward vehicle 201. B1 However, the total length of the recommended driving path 202 is L A1 If the distance is longer than the distance obtained by adding the margin distance α to the B1 >L A1 +α), it determines that it is possible to change lanes.

[0080] In the example in Figure 12, L B1 <L A1 Therefore, the driving assistance control unit 32 determines that the lane change is not possible. B1 If the distance is sufficiently secured, the driving assistance control unit 32 determines that a lane change is possible.

[0081] After step S205, in step S206, the display control unit 33 performs display control processing. The display control unit 33 causes the display device mounted on the target vehicle 200 to perform a display such as that shown in FIG. 12 or FIG. 13. This display includes, for example, the driving road, the driving lane and the position of the target vehicle 200 on the driving lane, the dividing lines or white lines of the plurality of lanes constituting the driving road, and the display of the positions of surrounding vehicles measured by the surrounding measurement unit 21. Further, the recommended driving path 202 may be displayed. Thereby, the driver can grasp the driving tendency when changing lanes by himself / herself. Further, the driver can determine whether or not he / she can change lanes according to his / her own driving characteristics based on the positional relationship between the recommended driving path 202 and the surrounding vehicles. Thus, the display control unit 33 functions as a presenting unit that presents the recommended driving path 202 and the like to the driver.

[0082] Further, the display control unit 33 may cause the display device to display the result of the determination as to whether or not lane change is possible by the driving support control unit 32. For example, when the driving support control unit 32 determines that lane change is not possible, a message such as "Lane change is not possible. Please maintain the current driving lane." is displayed on the display device. When the driving support control unit 32 determines that lane change is possible, a message such as "Lane change is possible." is displayed on the display device.

[0083] <B-3. Modified Example> In the first embodiment, it has been described that the lane change detection unit 128 of the lane change detection device 12 learns the driving tendency of the driver. However, the driving support control unit 32 may perform learning of the driving tendency.

[0084] FIG. 14 shows the state immediately after the target vehicle 200 starts to change lanes. The positions of the target vehicle 200 and the preceding vehicle 201 at this time are P A3 , P B3 respectively. The inter-vehicle distance between the two vehicles is L B3 and the remaining length L of the recommended driving path 202 A3While the target vehicle 200 is changing lanes in accordance with the recommended driving path 202, the remaining length of the recommended driving path 202 becomes shorter as the target vehicle 200 travels. Even during the lane change, the periphery measurement unit 21 continually measures the inter-vehicle distance to the preceding vehicle 201, and the driving assistance control unit 32 confirms that the inter-vehicle distance is sufficient for the recommended driving path 202.

[0085] 15 shows a state in which the preceding vehicle 201 is out of the measurement range of the periphery measurement unit 21 while the target vehicle 200 is changing lanes. Here, the preceding vehicle 201 is out of the detection range 22A of the stereo camera (forward) 22 and the detection range 23A of the millimeter wave radar (forward) 23. In such a case, the periphery measurement unit 21 detects the position P B3 From the state of motion of the vehicle, the latest position P B4 and confirms that the inter-vehicle distance is sufficient for the remaining portion of the recommended driving path 202. If the inter-vehicle distance is no longer sufficient for the remaining portion of the recommended driving path 202, the display control unit 33 may display a warning to that effect.

[0086] 16 shows a state in which the target vehicle 200 has completed a lane change in accordance with the recommended driving path 202. At this time, the target vehicle 200 and the preceding vehicle 201 are positioned at P A5 ,P B5 and the distance between the two vehicles is L B5 is.

[0087] 14 to 16 show the positional relationship between the target vehicle 200 and the preceding vehicle 201, the road to be traveled, the lane to be traveled, the measurement range by the periphery measurement unit 21, and the recommended driving path, but the display control unit 33 may cause the display device to display such information. This allows the driver to receive lane change assistance safely along the recommended driving path 202 while checking the surrounding conditions.

[0088] The autonomous navigation unit 127 may create a history of turn signal operation while the target vehicle 200 is changing lanes and include this in the position history. In this case, the driving assistance control unit 32 can detect whether the turn signal has been forgotten to be turned on or off from the position history and warn the driver of this fact via the display control unit 33. In FIG. 17 , if the turn signal has not been turned on even when the target vehicle 200 reaches position C2 between the time lane change assistance is started at position C1 and the time lane change assistance is ended at position C4, a warning may be issued to the driver. Also, if the turn signal has not been turned off at position C3, a warning may be issued to the driver. Also, if the turn signal operation is not updated even after the warning, the driving assistance control unit 32 may switch the turn signal on or off.

[0089] Furthermore, the autonomous navigation unit 127 may create a confirmation history of whether or not the driver performed a safety confirmation action, such as checking the rearview mirror or door mirrors, while the target vehicle 200 was changing lanes, and include this in the position history. The driver's safety confirmation action can be detected by a DMS (Driver Monitoring System) provided in the vehicle cabin. In this case, the driving assistance control unit 32 can detect insufficient safety confirmation from the position history and warn the driver to that effect using the display control unit 33. In FIG. 18, if the target vehicle 200 starts changing lanes at position C1 and then does not check the mirrors even when it reaches position C5, a warning may be issued to the driver.

[0090] The driving assistance control unit 32 and the display control unit 33 may provide driving assistance when changing lanes, taking into account the distance to the rear vehicle. Fig. 19 shows a situation in which the target vehicle 200 is about to change lanes to the right lane, and the rear vehicle 203 is traveling behind the target vehicle 200 in the right lane.

[0091] At this time, the driving assistance control unit 32 determines whether a lane change is possible based on the relationship between the inter-vehicle distance between the target vehicle 200 and the rear vehicle 203 and the recommended driving path 202 that matches the driving tendency of the driver of the target vehicle 200. For example, the driving assistance control unit 32 calculates the required driving time for the recommended driving path 202 from the speed of the target vehicle 200. The driving assistance control unit 32 also estimates the future position of the rear vehicle 203 from the inter-vehicle distance and the speed of the rear vehicle 203. The driving assistance control unit 32 may then determine that a lane change is not possible if the rear vehicle 203 overlaps with the recommended driving path 202 within the required driving time. Furthermore, when the target vehicle 200 changes lanes in such a situation, the driving assistance control unit 32 may determine that the driver of the target vehicle 200 has insufficiently confirmed safety based on the inter-vehicle distance between the target vehicle 200 and the rear vehicle 203 at the time of the lane change, and notify the driver to that effect.

[0092] 20 illustrates a situation in which a nearby vehicle 204 traveling parallel to the target vehicle 200 in an adjacent lane changes lanes ahead of the target vehicle 200. When the target vehicle 200 exhibits panicked behavior, such as braking or steering to avoid the lane change caused by the nearby vehicle's lane change, the lane change detection unit 128 learns the panicked behavior of the target vehicle 200. After learning, if the nearby vehicle 204 traveling parallel to the target vehicle in an adjacent lane is about to enter the target vehicle's driving lane, for example, the driving assistance control unit 32 may issue a warning in advance to increase the inter-vehicle distance when the nearby vehicle crosses lanes. Furthermore, before the nearby vehicle 204 changes lanes, the DMS may check whether the driver of the target vehicle 200 is visually observing the nearby vehicle 204 to the side, and if not, a warning may be issued to the driver.

[0093] Figure 21 shows a situation where there is a parked vehicle 206 in front of the driving lane of the target vehicle 200, and the target vehicle 200 has to protrude into the adjacent lane and overtake the parked vehicle 206. In this case, the driving support control unit 32 extracts the past driving trajectory at the time of overtaking from the position history of the target vehicle 200, and creates a recommended driving path 202 from the extracted driving trajectory. Then, the driving support control unit 32 determines whether it is safe to overtake according to the driving tendency based on the positional relationship between the target vehicle 200 and the surrounding vehicles 205 measured by the surrounding measurement unit 21 and the recommended driving path 202. For example, the driving support control unit 32 calculates the required driving time of the recommended driving path 202 from the speed of the target vehicle 200. Also, the driving support control unit 32 estimates the future position of the surrounding vehicle 205 from the inter-vehicle distance with the surrounding vehicle 205 and the speed of the surrounding vehicle 205. And when the surrounding vehicle 205 overlaps the recommended driving path 202 within the required driving time, the driving support control unit 32 may determine that overtaking is not possible.

[0094] When the driving support control unit 32 determines that overtaking is not possible, as shown in Figure 22, the driver may be instructed to stop the target vehicle 200 in front of the recommended driving path 202. This instruction is, for example, given by the display control unit 33 causing the display device to display the stop position. After that, when the driving support control unit 32 determines that overtaking is possible, it gives the driver an instruction to start overtaking. This instruction is, for example, given by the display control unit 33 causing the display device to display text such as "Please start overtaking." on the display device. Figure 23 shows a situation where the surrounding vehicle 205 has passed by and there is another surrounding vehicle 207 outside the measurement range of the surrounding measurement unit 21, that is, the inter-vehicle distance is sufficiently long, so the driving support control unit 32 determines that overtaking is possible.

[0095] <B-4. Effect> The driving support device 20 according to Embodiment 2 includes the lane change detection system 10 according to Embodiment 1, a surrounding measurement unit 21 that measures the positional relationship between the target vehicle 200 and surrounding vehicles, and a driving tendency including the driving trajectory during a lane change. By learning for each situation of the target vehicle during a lane change that at least includes time, day of the week, and road type, a driving support control unit 32 generates a recommended driving path 202 during a lane change according to the driving tendency and the situation of the target vehicle 200, and a display control unit 33 that causes the positional relationship between the target vehicle 200 and surrounding vehicles and the recommended driving path 202 to be displayed on a display device provided in the target vehicle 200. Therefore, the driver of the target vehicle 200 can grasp whether or not it is possible to perform a lane change according to his / her own driving tendency.

[0096] <C. Embodiment 3> <C-1. Configuration> FIG. 24 is a block diagram showing the configuration of the driving support device 30 according to Embodiment 3. The driving support device 30 is obtained by adding a lighting control unit 34 to the configuration of the driving support device 20 according to Embodiment 2. The lighting control unit 34 controls the lighting of an external lighting device mounted on the target vehicle 200 and illuminating the periphery of the target vehicle 200. The external lighting device includes a direction indicator 35, a headlight 36, a side irradiation lamp 37, and a tail lamp 38. The lighting control unit 34 controls the headlight 36 in a dark environment such as at night, in the evening, or inside a tunnel so that the entire recommended driving path 202 is irradiated, enabling the driver to visually confirm the recommended driving path 202.

[0097] The direction indicator 35 blinks when the driver operates an operation lever, informing surrounding vehicles or pedestrians, etc. of the direction in which the target vehicle 200 turns.

[0098] The headlight 36 has a low beam and a high beam, and either one can be selectively lit. The low beam can irradiate, for example, the road surface up to 40 m ahead of the target vehicle 200, and the high beam can irradiate, for example, the road surface up to 100 m ahead of the target vehicle 200.

[0099] The side illumination lamp 37 irradiates the road surface up to 40 m ahead on the side of the target vehicle 200. The tail lamp 38 can confirm lighting from surrounding vehicles or pedestrians at a predetermined distance behind.

[0100] <C-2. Operation> FIG. 25 is a flowchart showing the driving support process by the driving support device 30. This driving support process is repeatedly performed at any time such as at a fixed cycle.

[0101] In step S301, the driving support device 30 initializes related processes.

[0102] Since the processes from step S302 to step S305 after that are the same as the processes from step S202 to step S205 in FIG. 11 described in Embodiment 2, the description thereof is omitted.

[0103] After step S305, in step S306, the lighting control unit 34 performs lighting control processing of the headlamp 36. Here, the lighting control unit 34 controls the irradiation range of the headlamp 36 so that the entire recommended driving path 202 at the time of lane change is irradiated and visible to the driver.

[0104] [[ID=二十二]]FIG. 26 shows the relationship between the irradiation range 36A of the headlamp 36 before lighting control by the lighting control unit 34 and the recommended driving path 202. In addition, FIG. 26 also shows the irradiation range 37A of the side illumination lamp 37 and the irradiation range 38A of the tail lamp 38. Here, the headlamp 36 irradiates the front of the target vehicle 200 with a low beam. However, since the irradiation distance of the low beam is short, it may not be able to irradiate the entire recommended driving path 202. For example, if the length of the recommended driving path 202 is 100 m and the irradiation distance of the low beam is 40 m, the low beam cannot irradiate the entire recommended driving path 202.

[0105] In such a case, as shown in FIG. 27, the lighting control unit 34 switches the lit portion of the headlamp 36 from low beam to high beam. The irradiation distance of the high beam is, for example, 100 m, and it can irradiate farther than the low beam. However, since the irradiation angle of the high beam is smaller than that of the low beam, there may be cases where the adjacent lane cannot be illuminated. Since the recommended driving path 202 exists across from the current driving lane of the target vehicle 200 to the adjacent lane, if the adjacent lane cannot be irradiated with the high beam, the entire recommended driving path 202 cannot be irradiated.

[0106] Therefore, as shown in FIG. 28, the lighting control unit 34 adjusts the optical axis orientation of the headlamp 36 toward the adjacent lane side where a lane change is to be made. Thereby, the entire recommended driving path 202 extending from the driving lane to the adjacent lane can be irradiated.

[0107] FIG. 29 shows a recommended driving path 202 that is longer than that shown in FIG. 28. When the entire recommended driving path 202 cannot be irradiated even with the high beam, the lighting control unit 34 may increase the luminance of the high beam to lengthen the irradiation distance of the high beam so that the entire recommended driving path 202 can be irradiated. In addition, even by the lighting control process of the lighting control unit 34 described above, when the headlamp 36 can irradiate only a part of the recommended driving path 202, the recommended driving path 202 will be irradiated step by step as the target vehicle 200 moves.

[0108] [[ID=1十二]]Next, in step S307, display control processing by the display control unit 33 is performed. Here, in addition to the display control processing described in step S206 of FIG. 11, the irradiation range of the vehicle exterior lighting device is displayed. FIG. 30 shows an example of the display in this step. In the example of FIG. 30, the positional relationship between the driving road of the target vehicle 200, the target vehicle 200 on the driving road, the surrounding vehicles 207, 208, and the recommended driving path 202 is displayed in terms of lanes. In addition, the detection ranges 22A, 23A, 24A, 25A, 26A, 27A, 28A of the sensors used by the surrounding measurement unit 21 and the irradiation ranges 36A, 37A, 38A of the vehicle exterior lighting devices are displayed.

[0109] <C-3. Effect> The driving support device 30 according to Embodiment 3 includes a lighting control unit 34 that is mounted on the target vehicle 200 and controls an external lighting device that illuminates the outside of the target vehicle 200. When the lighting control unit 34 receives an input from the driver indicating a desire for lane change support, the lighting control unit 34 causes the external lighting device to illuminate the recommended driving path 202. As a result, the driver can easily visually confirm the recommended driving path 202 and perform a lane change safely.

[0110] <D. Hardware Configuration> Each component of the above-described lane change detection system 10 and the driving support devices 20 and 30 including the lane change detection system 10 is realized by a processing circuit 81 shown in FIG. 31. That is, the processing circuit 81 includes each component of the lane change detection system 10 and the driving support devices 20 and 30 including the lane change detection system 10. A dedicated hardware may be applied to the processing circuit 81, or a processor that executes a program stored in a memory may be applied. The processor is, for example, a central processing unit, a processing device, an arithmetic device, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), or the like.

[0111] When the processing circuit 81 is dedicated hardware, the processing circuit 81 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof. Each function of each component may be realized by a plurality of processing circuits 81, or the functions of each part may be realized by one processing circuit collectively.

[0112] When the processing circuit 81 is a processor, the functions of each component of the lane change detection system 10 and the driving assistance devices 20, 30 are realized by a combination of software, etc. (software, firmware, or software and firmware). The software, etc. is written as a program and stored in a memory. As shown in FIG. 32, the processor 82 applied to the processing circuit 81 realizes the functions of each part by reading and executing a program stored in a memory 83. That is, the lane change detection system 10 and the driving assistance devices 20, 30 include a memory 83 for storing a program that, when executed by the processing circuit 81, results in the function of each component being performed. In other words, the program can be said to cause a computer to execute the procedure or method of each component. Here, the memory 83 may be, for example, a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), HDD (Hard Disk Drive), magnetic disk, flexible disk, optical disk, compact disk, mini disk, DVD (Digital Versatile Disk) and its drive device, or any storage medium that will be used in the future.

[0113] The above describes a configuration in which the functions of each component are realized either by hardware or software, etc. However, the present invention is not limited to this, and a configuration in which some of the components are realized by dedicated hardware and other parts are realized by software, etc.

[0114] As described above, the processing circuit can realize each of the above-described functions by hardware, software, or a combination of these.

[0115] It should be noted that the embodiments can be freely combined, and each embodiment can be modified or omitted as appropriate. The above description is an example in all respects. It is understood that countless variations not illustrated can be envisioned. [Explanation of symbols]

[0116] 10 Lane change detection system, 11 Satellite positioning device, 12 Lane change detection device, 13 High-precision map data, 20 Driving assistance device, 21 Periphery measurement unit, 22A Detection range, 23A, 24A, 25A, 26A, 27A, 28A Detection range, 30 Driving assistance device, 31 Operation input unit, 32 Driving assistance control unit, 33 Display control unit, 34 Light control unit, 35 Turn signal, 36 Headlight, 36A, 37A, 38A Irradiation range, 37 Side illumination light, 38 Taillight, 81 Processing circuit, 82 Processor, 83 Memory, 111 GNSS receiver, 112 Positioning augmentation signal receiver, 113 Satellite positioning unit, 121 Speed ​​sensor, 122 Distance measurement unit, 123 Speed ​​sensor correction unit, 124 Angular velocity sensor, 125 Yaw angle measurement unit, 126 Angular velocity sensor correction unit, 127 autonomous navigation unit, 128 lane change detection unit, 129 positioning error prediction unit, 200 target vehicle, 201 forward vehicle, 202 recommended driving path, 203 rear vehicle, 204, 205, 207, 208 surrounding vehicles, 206 parked vehicle, LB1 inter-vehicle distance, S1 first section, S2 second section, S3 third section.

Claims

1. an autonomous navigation unit that creates a position history that is a history of a DR position and a DR orientation, which are the position and orientation of a target vehicle measured by autonomous navigation; a lane change detection unit that detects a lane change of the target vehicle using the position history, The lane change detection unit extracting a first section in which the amount of change in the DR orientation exceeds a predetermined threshold in the position history, and a second section and a third section which are sections of a predetermined number of histories before and after the first section; detecting a lane change of the target vehicle in the first section using an approximation curve of the DR position in at least the second section and the third section; extracting the DR positions in the first section where the lane change was detected and the second section and the third section adjacent to the first section as a travel trajectory at the time of the lane change; The travel trajectory at the time of the lane change is used for driving assistance of the target vehicle. Lane change detection device.

2. the lane change detection unit detects a lane change of the target vehicle in the first section when a deviation amount in a lateral direction perpendicular to the traveling direction of the target vehicle between curves that respectively approximate the DR positions in the second section and the third section is equal to or greater than a travel lane width of the target vehicle. The lane change detection device according to claim 1 .

3. the lane change detection unit detects the overtaking of the target vehicle in the first section when a deviation amount in a lateral direction perpendicular to the traveling direction of the target vehicle between the coordinates obtained by curve approximating the DR positions in the second section and the third section is less than a driving lane width of the target vehicle and a movement amount in the lateral direction of the DR position in the first section is greater than a predetermined reference value; The lane change detection device according to claim 1 .

4. The lane change detection unit a first group of parallel coordinates running parallel to the DR position in a lane adjacent to the lane to which the DR position belongs in the second section in a direction in which the DR orientation changed in the first section; and a second group of parallel coordinates running parallel to the DR position in a lane adjacent to the lane to which the DR position belongs in the third section in a direction opposite to the direction in which the DR orientation changed in the first section. creating a first curve approximating a locus connecting the DR positions in the first section, the second section, and the third section, a second curve approximating a locus connecting the DR position in the second section and the second parallel coordinate group, and a third curve approximating a locus connecting the first parallel coordinate group and the DR position in the third section; detecting a lane change of the target vehicle in the first section based on a difference between the first curve and at least one of the second curve and the third curve; The lane change detection device according to claim 1 .

5. The lane change detection device according to claim 1 ; a peripheral measurement unit that measures the positional relationship between the target vehicle and peripheral vehicles; a driving assistance control unit that learns driving tendencies, including a driving trajectory when changing lanes, for each situation of the target vehicle when changing lanes, including at least a time, a day of the week, and a road type, and thereby generates a recommended driving path when changing lanes according to the driving tendencies and the situation of the target vehicle; a presentation unit that presents the positional relationship between the target vehicle and the surrounding vehicles and the recommended driving path to a driver; Driving assistance device.

6. The driving assistance control unit determines whether or not the target vehicle can change lanes by comparing the distance between the target vehicle and the surrounding vehicles with the length of the recommended driving path, and causes the presentation unit to present the result of the determination of whether or not the target vehicle can change lanes. The driving assistance device according to claim 5.

7. the autonomous navigation unit links an operation history of a turn signal of the target vehicle to the position history; The driving assistance control unit determines whether the driver of the target vehicle has insufficient safety confirmation based on the operation history linked to the driving trajectory at the time of the lane change, and causes the presentation unit to present the determination. The driving assistance device according to claim 5.

8. the autonomous navigation unit links a driver's confirmation history of the door mirror or the rearview mirror of the target vehicle to the position history; The driving assistance control unit determines whether the driver of the target vehicle has insufficient safety confirmation based on the confirmation history linked to the driving trajectory at the time of the lane change, and causes the presentation unit to present the determination. The driving assistance device according to claim 5.

9. The driving assistance control unit determines whether the driver of the target vehicle has insufficient safety confirmation based on a vehicle-to-vehicle distance between the target vehicle and the surrounding vehicle at the time of the lane change, and causes the presentation unit to present the determination. The driving assistance device according to claim 5.

10. The driving assistance control unit causes the presentation unit to present a message to the effect that the vehicle-to-vehicle distance between the surrounding vehicle and the target vehicle should be increased when the surrounding vehicle is traveling alongside the target vehicle, based on a driving tendency of the target vehicle when the surrounding vehicle previously changed lanes to a lane ahead of the target vehicle. The driving assistance device according to claim 5.

11. a light control unit that is mounted on the target vehicle and controls a headlight that illuminates the outside of the target vehicle; When receiving an input from the driver indicating a desire for lane change assistance, the lighting control unit causes the headlights to illuminate the recommended driving path. The driving assistance device according to claim 5.

12. The headlights include low beams and high beams; When the low beam cannot illuminate the entire recommended traveling path, the light control unit switches the lit portion of the headlight from the low beam to the high beam. The driving assistance device according to claim 11.

13. the light control unit adjusts the orientation of the optical axis of the headlight so that the headlight illuminates the recommended driving path on a lane adjacent to the driving lane of the target vehicle. The driving assistance device according to claim 11.

14. The light control unit increases the brightness of the high beam when the current brightness of the high beam is not enough to illuminate the entire recommended driving path. The driving assistance device according to claim 12.

15. A history of the DR position and DR orientation, which are the position and orientation of the target vehicle measured by autonomous navigation, is created as a position history; Detecting a lane change of the target vehicle using the location history; extracting a first section in which the DR orientation exceeds a predetermined threshold in the position history, and a second section and a third section which are sections of a predetermined number of histories before and after the first section; detecting a lane change of the target vehicle in the first section using an approximation curve of the DR position in at least the second section and the third section; extracting the DR positions in the first section where the lane change was detected and the second section and the third section adjacent to the first section as a travel trajectory at the time of the lane change; The travel trajectory at the time of the lane change is used for driving assistance of the target vehicle. Lane change detection method.

Images