Travel path recognition device, travel path recognition method, and computer program

Abstract

To recognize a road by using a detection result of a sensor from the initial stage of a turning operation.SOLUTION: Travel path recognition devices 100, 100a each comprise: a turning identification unit 11 that determines whether or not a vehicle V1 is performing a turning operation; a detection result acquisition unit 12 that acquires a detection result of a sensor 200; and a boundary line identification unit 13 that identifies left and right boundary lines BLL, BLR of a planned travel path C1 by using the detection result. The boundary line identification unit, when it is identified that the vehicle is performing the turning operation, (i) identifies the outer boundary line BLL being a boundary line on the outside in a turning direction, by using outer detection portions OP1, OP1a being at least a part of the outer boundary line included in the detection result, and (ii) identifies the inner boundary line BLR being a boundary line on the inside in the turning direction, by extending and estimating portions IP1, IP1a of the inner boundary line included in the detection result by using the outer detection portions.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a lane recognition device, a lane recognition method, and a computer program.

Background Art

[0002] Conventionally, a lane recognition device has been proposed that detects the environment in front of a vehicle using sensors such as a stereo camera or a millimeter-wave radar, and recognizes the planned driving lane of the vehicle using the detection result (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a conventional lane recognition device, since the detection range of the sensor is in front of the vehicle, when performing a turning operation such as a right turn or a left turn, the environment outside the turning direction can be detected by the sensor, while the environment inside the turning direction cannot be detected by the sensor at the beginning of the turning operation. For example, when making a right turn, the environment on the left side of the vehicle corresponding to the outside of the turning direction can be detected by the sensor, while the environment on the right side of the vehicle corresponding to the inside of the turning direction cannot be detected by the sensor at the beginning of the turning operation. For this reason, at the beginning of the turning operation, there has been a problem that the left and right boundary lines of the road cannot be specified and the lane cannot be recognized. Therefore, a technique that can recognize the road using the detection result of the sensor from the beginning of the turning operation is desired.

Means for Solving the Problems

[0005] As one aspect of the present disclosure, a lane recognition device that recognizes a planned driving lane of a vehicle is provided. This lane recognition device (100, 100a) includes a turning identification unit (11) that identifies whether the vehicle (V1) is performing a turning operation, a detection result acquisition unit (12) that acquires the detection result of a sensor (200) mounted on the vehicle and detecting the environment around the vehicle, and a boundary line identification unit (13) that uses the detection result to identify the left and right boundary lines (BLL, BLR) of the planned driving lane (C1). When it is identified that the vehicle is performing a turning operation, the boundary line identification unit (i) identifies the outer boundary line (BLL), which is the boundary line on the outer side of the turning direction among the left and right boundary lines, using at least a part of the outer boundary line, i.e., the outer detection part (OP1, OP1a), included in the detection result, and (ii) identifies the inner boundary line (BLR), which is the boundary line on the inner side of the turning direction among the left and right boundary lines, by extending and estimating a part (IP1, IP1a) of the inner boundary line included in the detection result using the outer detection part.

[0006] According to the lane recognition device of the above aspect, among the left and right boundary lines of the planned driving lane, the inner boundary line, which is the boundary line on the inner side of the turning direction, is identified by extending and estimating a part of the inner boundary line included in the detection result using the outer detection part. Therefore, the inner boundary line can be accurately identified from the beginning of the turning operation, and the planned driving lane can be recognized.

[0007] The present disclosure can also be realized in various forms. For example, it can be realized in the form of a lane recognition method, a lane recognition device, a computer program for realizing the lane recognition method, a non-transitory recording medium on which such a computer program is recorded, and the like.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Embodiments for Carrying Out the Invention

[0009] A. First Embodiment: A1. Device Configuration: The travel path recognition device 100 of the first embodiment shown in FIG. 1 is mounted on the vehicle V1 and recognizes the travel path that the vehicle V1 plans to travel. In this embodiment, the vehicle V1 is configured as a vehicle capable of autonomous driving. As will be described later, based on the travel path planned to be traveled recognized by the travel path recognition device 100, the coordinates (target coordinates) that the vehicle V1 should pass through are determined, and the driving is automatically controlled so as to pass through such coordinates. The vehicle V1 may be configured as any type of vehicle, such as an engine vehicle, a hybrid vehicle (HEV), a plug-in hybrid vehicle (PHEV), an electric vehicle (EV), a fuel cell vehicle (FCV, FCHV).

[0010] In the vehicle V1, in addition to the lane recognition device 100, there are provided a sensor 200, a yaw rate sensor 300, a vehicle control device 400, a drive device 410, a steering device 420, and a braking device 430.

[0011] The sensor 200 detects the environment around the vehicle V1. The sensor 200 includes a front sensor 210 and a surrounding sensor 220.

[0012] The front sensor 210 detects the environment in front of the vehicle V1. In the present embodiment, the front sensor 210 includes an imaging camera and a radar sensor. The imaging camera images the front of the vehicle V1 and, by performing image processing on the obtained imaging image data, detects information in front of the vehicle V1, for example, obstacles such as other vehicles, pedestrians, and structures such as utility poles, and information such as white lines. The viewing angle of the imaging camera included in the front sensor 210 is narrow, and it can detect distant obstacles with high resolution and high accuracy compared to the surrounding sensor 220. The radar sensor emits electromagnetic waves (radio waves or light) of a predetermined wavelength and uses the received reflected waves to detect obstacles in front of the vehicle V1. As the radar sensor, for example, a millimeter-wave radar or LiDAR (Light Detection and Ranging) can be used. In the case of LiDAR, light of a predetermined wavelength is emitted while scanning, the reflected light is received, the direction of the obstacle is specified from the position where the peak of the reflected wave is detected, and the distance to the obstacle can be specified using the time (TOF: Time Of Flight) from the emission of the light to the reception of the reflected wave.

[0013] The surrounding sensor 220 detects the environment around the vehicle V1. In the present embodiment, it detects the environment on the side and rear of the vehicle V1. Note that the surrounding sensor 220 may also detect the environment in the front in addition to the side and rear. Further, the surrounding sensor 220 may detect only the environment on either one of the side and rear. That is, the surrounding sensor 220 may detect at least the environment on the side or rear of the vehicle V1. In the present embodiment, the surrounding sensor 220 includes an imaging camera and a radar sensor, similar to the front sensor 210. In the present embodiment, a total of five surrounding sensors 220 are installed on the side and rear portions of the vehicle body of the vehicle V1. Each surrounding sensor 220 needs to detect objects in a relatively wide range. For this reason, the angle of view of the imaging camera included in each surrounding sensor 220 is wide, and obstacles can be detected in a wide-angle range compared to the front sensor 210. However, each surrounding sensor 220 can detect distant obstacles with lower accuracy compared to the front sensor 210, and can detect only short-distance obstacles with high resolution and high accuracy.

[0014] As shown in FIG. 2, the detection range Sa1 of the front sensor 210 includes positions at a relatively small angle and at a relatively long distance from the vehicle V1. On the other hand, the detection ranges Sa11, Sa12, Sa13, Sa14, and Sa15 of the five surrounding sensors 220 include only positions at a relatively wide angle and at a relatively short distance from the vehicle V1. Hereinafter, the range obtained by combining the detection ranges Sa11, Sa12, Sa13, Sa14, and Sa15 of the five surrounding sensors 220 is also referred to as the "detection range S10".

[0015] The yaw rate sensor 300 shown in FIG. 1 detects the yaw rate of the vehicle V1. When the vehicle V1 turns right or left, a turning operation is executed, and the yaw rate detected by the yaw rate sensor 300 becomes equal to or greater than a predetermined threshold value.

[0016] The vehicle control device 400 shown in FIG. 1 sets the above-described target coordinates and the speed at the time of passing based on the planned driving path recognized by the path recognition device 100, and controls the drive device 410, the steering device 420, and the braking device 430 so as to achieve such target coordinates and speed. For example, the vehicle control device 400 controls these devices 410 to 430 to execute an automatic parking process. The automatic parking process is a process of searching for an empty parking space in a parking lot or the like and automatically parking the vehicle V1 in the found parking space.

[0017] The drive device 410 is a group of devices for driving the vehicle V1. The drive device 410 includes a device that generates a driving force such as an engine or a motor generator, various actuators for driving the engine or the motor generator, and an ECU (Electronic Control Unit) for controlling such actuators. The steering device 420 is a group of devices for steering the vehicle V1. The steering device 420 includes a steering wheel, a steering angle sensor, an assist hydraulic device, an actuator for generating hydraulic pressure, and an ECU for controlling steering. The braking device 430 is a group of devices for generating a braking force on the vehicle V1. The braking device 430 includes a disk rotor, a brake pad, a brake hydraulic device, an actuator for generating hydraulic pressure, and an ECU for controlling braking. Each of the above-described drive device 410, steering device 420, and braking device 430 communicates with the vehicle control device 400. At this time, the vehicle control device 400 transmits a control signal to these devices 410 to 430, and on the other hand, receives a signal indicating the operating state of each device from the devices 410 to 430. The above-described "signal indicating the operating state" includes signals indicating the vehicle speed of the vehicle V1, the presence or absence of brake operation, the presence or absence of turn signal operation, the amount of wheel rotation, the amount of steering, and the like.

[0018] In this embodiment, the walking recognition device 100 is configured by an ECU in which a CPU 10, a ROM 20, and a RAM 30 can communicate with each other via an internal bus 90. The ROM 20 includes a non-volatile memory, for example, an EEPROM, and stores a control program in advance. The CPU 10 functions as a turning identification unit 11, a detection result acquisition unit 12, a boundary line identification unit 13, and a walking recognition unit 14 by reading the control program stored in the ROM 20, expanding it in the RAM 30, and executing it.

[0019] The turning identification unit 11 identifies whether the vehicle V1 is performing a turning operation. Specifically, it acquires the detection result of the yaw rate sensor 300, and if the detected yaw rate is equal to or greater than a predetermined threshold, it identifies that a turning operation is being performed, and if it is less than the predetermined threshold, it identifies that no turning operation is being performed.

[0020] The detection result acquisition unit 12 acquires the detection results of the sensor 200, that is, the detection results of the front sensor 210 and the peripheral sensor 220. Specifically, it acquires information indicating the shape and position of an obstacle or a white line detected by the sensor 200. The detection result acquisition unit 12 stores the acquired detection results in the ROM 20. In addition, the detection result acquisition unit 12 updates the coordinates of a structure such as an obstacle detected in the past among the detection results stored in the ROM 20 according to the movement of the vehicle V1.

[0021] The boundary line specifying unit 13 specifies the left and right boundary lines of the planned driving lane using the detection results acquired by the detection result acquisition unit 12. When the planned driving lane is included in a general road, the above-mentioned "boundary line" means the lane line (white line) of the road. Also, for example, when the planned driving lane is included in the lane C1 in the parking lot P1 shown in FIG. 2, the left and right boundary lines OBL and IBL mean lines connecting portions that will face the vehicle V1 traveling in the two parallel parking rows L11 and L12. The "parking row" means a row in which at least one of the parked vehicles and the parking spaces are arranged continuously in a common direction for a predetermined number of consecutive threshold values or more. The "parking space" means a frame for indicating a partitioned area for a vehicle to park. Specifically, rectangles, straight lines, etc. drawn by white lines or yellow lines on the road surface of the parking lot correspond to this. In the example of FIG. 2, the parking row L11 means a row consisting of seven parking spaces PA11 to PA17 and four parked vehicles PV11 to PV14. In the parking row L11, the four parked vehicles PV11 to PV14 are all parked so as to be within the parking spaces. Therefore, the boundary line OBL corresponds to a line connecting the straight line portions that face the vehicle traveling in the lane C1 among the rectangular lines in each of the seven parking spaces PA11 to PA17. Also, the parking row L12 means a row consisting of seven parking spaces PA21 to PA27 and six parked vehicles PV21 to PV26. The six parked vehicles PV21 to PV26 are all parked so as to be within the parking spaces. Therefore, the boundary line IBL corresponds to a line connecting the straight line portions that face the vehicle traveling in the lane C1 among the rectangular lines in each of the seven parking spaces PA21 to PA27.

[0022] In the example of FIG. 2, the vehicle V1 is traveling in a direction orthogonal to the two parking rows L11 and L12 and is about to turn right to enter the road C1. At this time, although a part of the boundary line OBL is included in the detection range Sa1 of the front sensor 210, the boundary line IBL is not included in the detection range Sa1. Therefore, when starting a right turn from the state of FIG. 2, a portion corresponding to a relatively long distance within the detection range Sa1 is detected for the boundary line OBL located outside the turning direction, while for the boundary line IBL inside the turning direction, only a portion corresponding to a relatively short distance within the detection range Sa1 is detected. However, according to the road recognition device 100 of the present embodiment, by executing the road recognition process described later, the boundary line IBL can be detected from the beginning of the turning operation for a right turn, and the planned travel road can be accurately recognized. Hereinafter, the boundary line OBL outside the turning direction will also be referred to as the "outer boundary line OBL". Also, the boundary line IBL inside the turning direction will also be referred to as the "inner boundary line IBL".

[0023] Note that the detection result acquisition unit 12 acquires the detection result of the sensor 200 and detects a parking row using the acquired detection result. Specifically, the detection result acquisition unit 12 performs image analysis processing such as edge extraction, pattern matching, and feature point extraction on the captured image obtained by the imaging camera included in the sensor 200 to detect a parking frame. Whether or not the section composed of the detected white line or the like is a parking frame can be determined, for example, by comparing the dimensions of the section composed of the white line or the like with the dimensions of the parking lot frame in a general parking lot. If the difference is within a predetermined dimension range, it is determined to be a parking frame, and if the difference exceeds the predetermined dimension range, it may be determined not to be a parking frame. Similarly for parked vehicles, the detection result acquisition unit 12 uses the detection result of the sensor 200 to detect that the obstacle is a stationary object and that it is a vehicle (parked vehicle). Then, the detection result acquisition unit 12 determines whether at least one of the detected parked vehicle and the parking frame is lined up continuously in a common direction for a predetermined number of consecutive threshold values or more. If it is determined that they are lined up, at least one of the detected parked vehicle and the parking frame is detected as a parking row. In this embodiment, the number of consecutive threshold values is "3". Note that the number of consecutive threshold values is not limited to 3 and may be any number recognizable as a row.

[0024] The driving path recognition unit 14 shown in FIG. 1 recognizes the planned driving path using the boundary lines specified by the boundary line specifying unit 13. Specifically, the coordinates at the center in the width direction of the left and right boundary lines and located at predetermined distances such as 0.5 m along the length direction of the boundary line are specified, and a set of such coordinates is specified as the planned driving path.

[0025] A2. Driving Path Recognition Process: The driving path recognition process shown in FIG. 3 is a process for recognizing the planned driving path of the vehicle V1 and is executed when the power of the driving path recognition device 100 is turned on.

[0026] In step S101, the detection result acquisition unit 12 acquires the detection results of the front sensor 210 and the peripheral sensors 220. Hereinafter, the term "step" may be omitted and simply described as "S". In S102, the boundary line identification unit 13 identifies the left and right boundary lines using the acquired detection results of the front sensor 210 and the peripheral sensors 220. In S102, the left and right boundary lines are respectively identified only within the ranges detected by both sensors 210 and 220.

[0027] In S105, the turning identification unit 11 determines whether the vehicle V1 is performing a turning operation.

[0028] When it is determined that the turning operation is not being performed (S105: NO), the boundary line identification unit 13 identifies the left and right boundary lines using the acquired detection results of the front sensor 210 and the peripheral sensors 220 (S145). Specifically, the boundary line identification unit 13 compares the two left and right boundary lines obtained using the detection results of both sensors 210 and 220 identified in S102. For the overlapping portion of the two left and right boundary lines when viewed in the width direction of the planned driving path, the boundary line identified using the detection result of the peripheral sensor 220 is prioritized. Also, when only one of the two left and right boundary lines exists when viewed in the width direction of the planned driving path, such a boundary line is prioritized. Further, when neither of the two left and right boundary lines exists when viewed in the width direction of the planned driving path, in other words, when there is a gap, in this embodiment, the boundary line obtained by extending and estimating the boundary line obtained using the detection result of the front sensor 210 is compared with the boundary line obtained by extending and estimating the boundary line obtained using the detection result of the peripheral sensor 220, and the more inwardly located boundary line is prioritized. By connecting the thus prioritized boundary lines, the left and right boundary lines are identified. Note that when there is such a gap, one of the boundary lines obtained by extending and estimating the boundary line obtained using the detection result of the front sensor 210 and the boundary line obtained by extending and estimating the boundary line obtained using the detection result of the peripheral sensor 220 may be prioritized as predetermined.

[0029] In S145, when the left and right boundary lines are specified, the lane recognition unit 14 recognizes the planned travel lane of the vehicle V1 using the specified left and right boundary lines (S135). After the completion of S135, the process returns to S105.

[0030] In the above-mentioned S105, when it is determined that the vehicle V1 is performing a turning operation (S105: YES), the boundary line specifying unit 13 specifies the outer boundary line using at least a part of the outer boundary line OBL (hereinafter, also referred to as the "outer detected part") included in the detection result of the front sensor 210 (S120). Note that the outer boundary line specified by the boundary line specifying unit 13 is also referred to as the "outer boundary line BLL".

[0031] FIG. 4 shows the detection ranges Sa1 and Sa10 immediately after starting a turning operation in the right direction from the state of FIG. 2. In FIG. 4, for the sake of illustration, the two parking rows L11 and L12 are omitted.

[0032] In the state of FIG. 4, the detection result of the front sensor 210, that is, the detection range Sa1 includes the outer detected part OP1 which is a part of the front side of the outer boundary line OBL. The "front side" means the side closer to the position where the vehicle V1 starts the turning operation in the direction along the lane C1. The side farther from the position where the vehicle V1 starts the turning operation in the direction along the lane C1 is also referred to as the "rear side". Note that a part of the front side of the outer detected part OP1 is not included in the detection range Sa1 in the state of FIG. 4. However, in a state (not shown) that is earlier in time than the state of FIG. 4, such a part of the front side is included in the detection range Sa1 and is stored in the ROM20 as a detection result. Therefore, in the state of FIG. 4, the part obtained by combining all the parts of the boundary line OBL that are included in the detection range Sa1 in the state of FIG. 4 from the front end of the front side is specified as the outer boundary line BLL. Note that in the state of FIG. 4, the detection range Sa10 includes a part of the inner boundary line IBL (hereinafter, also referred to as the "inner detected part") which is the inner detected part IP1.

[0033] In S130 shown in FIG. 3, the boundary line specifying unit 13 specifies the inner boundary line BLR by extending and estimating a part of the inner boundary line (inner detection part) included in the detection result of the peripheral sensor 220 using the outer detection part OP1. In other words, in S130, the boundary line specifying unit 13 estimates the boundary on the inner side of the turn from the information on the outer side of the turn. Specifically, as shown in FIG. 4, the boundary line specifying unit 13 extends the inner detection part IP1 included in the detection range Sa10 so as to be parallel to the outer detection part OP1. As a result, the extended part IPv extending from the end on the back side of the inner detection part IP1 is extrapolated, and the inner boundary line BLR is specified. Note that "extension" has the same meaning as "extrapolation". The position along the traveling direction of the vehicle V1 at the back side end of the extended part IPv (hereinafter, also simply referred to as the "traveling direction") is substantially equal to the position along the traveling direction of the back side end of the outer detection part OP1 (outer boundary line BLL).

[0034] After the completion of S130 shown in FIG. 3, the above-described S135 is executed, and the planned travel path is recognized. In the example of FIG. 4, the outer boundary line BLL is only a part on the front side of the actual outer boundary line OBL. Similarly, the inner boundary line BLR is only a part on the front side of the actual inner boundary line IBL. Therefore, in S135, the planned travel path is not recognized over the entire actual travel path C1, but the planned travel path is recognized for a part on the front side in the travel path C1.

[0035] According to the travel path recognition device 100 of the first embodiment described above, among the left and right boundary lines of the planned travel path, the inner boundary line BLR, which is the boundary line on the inner side of the turning direction, is specified by extending and estimating a part of the inner boundary line IBL (inner detection part IP1) included in the detection result using the outer detection part OP1. Therefore, the inner boundary line BLR can be accurately specified from the beginning of the turning operation, and the planned travel path can be recognized.

[0036] In addition, by extending and estimating a part of the inner boundary line IBL (inner detection part IP1) included in the detection result of the peripheral sensor 220 using the outer detection part OP1 included in the detection result of the front sensor 210, the inner boundary line BLR is specified. In other words, the boundary line specifying unit 13 estimates the boundary on the inner side of the turn from the information on the outer side of the turn. Therefore, even when only a part of the inner boundary line (inner detection part IP1) can be detected by the peripheral sensor 220 with a narrow detection range, the detection result of the front sensor 210 can be used to estimate (complement) the inner boundary line BLR in a wide range with high accuracy.

[0037] In addition, the inner boundary line BLR is specified by extending and estimating a part of the inner boundary line IBL (inner detection part IP1) so as to be parallel to the outer detection part OP1, so that the inner boundary line BLR can be specified with high accuracy.

[0038] B. Second Embodiment: The walking recognition device 100a according to the second embodiment shown in FIG. 5 is different from the walking recognition device 100 according to the first embodiment shown in FIG. 1 in that the CPU 10 also functions as the correlation determination unit 15. Since other configurations in the walking recognition device 100a according to the second embodiment are the same as those in the walking recognition device 100, the same reference numerals are given to the same configurations, and detailed descriptions thereof are omitted.

[0039] The correlation determination unit 15 determines whether a part of the inner boundary line IBL (inner detection part IP1) included in the detection result of the peripheral sensor 220 and the outer detection part OP1 included in the detection result of the front sensor 210 correlate with each other as the left and right boundary lines of the same walking path. Specifically, in the present embodiment, the correlation determination unit 15 determines that they "correlate with each other" when all of the following conditions (i) to (iii) are satisfied, and determines that they "do not correlate with each other" when any one of them is not satisfied. (i) The average of the distances (so-called road widths) between the inner detection part IP1 and the outer detection part OP1 is within a predetermined range. (ii) The angle formed by the inner detection part IP1 and the outer detection part OP1 is equal to or less than a predetermined threshold angle. (iii) The difference in thickness between the inner detection part IP1 and the outer detection part OP1 is equal to or less than a predetermined threshold value. Note that the determination conditions are not limited to the above (i) to (iii), and any conditions that can be recognized as having a correlation as the left and right boundary lines of the runway may be adopted.

[0040] The runway recognition process of the second embodiment shown in FIG. 6 is different from the runway recognition process of the first embodiment shown in FIG. 3 only in that S115 is executed between S105 and S120. Since the other procedures in the runway recognition process of the second embodiment are the same as those in the runway recognition process of the first embodiment, the same reference numerals are assigned to the same procedures, and detailed descriptions thereof are omitted.

[0041] As shown in FIG. 6, when it is determined in S105 that a turning operation is being performed (S105: YES), the correlation determination unit 15 determines whether the inner detection part IP1 and the outer detection part OP1 correlate with each other as the left and right boundary lines of the same runway (S115). The correlation determination unit 15 specifies the success or failure of the above conditions (i) to (iii) and makes a determination based on the result.

[0042] When it is determined that they correlate with each other (S115: YES), the above-described S120 to S135 are executed. Therefore, the inner detection part IP1 is estimated to extend using the outer detection part OP1, and the inner boundary line BLR is specified.

[0043] On the other hand, when it is determined that they do not correlate with each other (S115: NO), S120 and S130 are omitted and the process proceeds to S135. Therefore, in this case, the estimation of the extension of the inner detection part IP1 using the outer detection part OP1 is not executed.

[0044] According to the travel path recognition device 100a of the second embodiment described above, the same effects as those of the travel path recognition device 100 of the first embodiment can be achieved. In addition, it is determined whether a part (inner detection part IP1) of the inner boundary line IBL included in the detection result of the peripheral sensor 220 and an outer detection part OP1 included in the detection result of the front sensor 210 are correlated with each other as the left and right boundary lines of the same travel path. When it is determined that they are correlated with each other, the inner boundary line BLR is specified by stretching and estimating the inner detection part IP1 so as to be parallel to the outer detection part OP1. Therefore, it is possible to suppress the wasteful execution of stretching and estimation in the case where they are not correlated with each other, and the processing load on the travel path recognition device 100a (CPU 10) can be reduced.

[0045] C. Third Embodiment: The travel path recognition device 100 of the third embodiment has the same configuration as the travel path recognition device 100 of the first embodiment. Therefore, the same reference numerals are given to the same components, and detailed descriptions thereof are omitted.

[0046] The travel path recognition process of the third embodiment shown in FIGS. 7 and 8 is different from the travel path recognition process of the first embodiment shown in FIG. 3 in that S122, S124, S126, and S128 are additionally executed. Since the other procedures in the travel path recognition process of the third embodiment are the same as those in the travel path recognition process of the first embodiment, the same reference numerals are given to the same procedures, and detailed descriptions thereof are omitted.

[0047] As shown in FIG. 7, when S120 is executed and the outer boundary line BLL is specified, the boundary line specifying unit 13 determines whether a part of the inner boundary line IBL is included in the detection result of the front sensor 210 (S122). Note that a part of the inner boundary line IBL included in the detection result of the front sensor 210 is also referred to as the "second inner detection part". In addition, a part of the inner boundary line IBL included in the detection result of the peripheral sensor 220 is also referred to as the "first inner detection part".

[0048] FIG. 9 shows detection ranges Sa1 and Sa10 in a state where the turning operation further to the right is continued from the state of FIG. 4. In FIG. 9, as in FIG. 4, for the sake of illustration, the two parking rows L11 and L12 are omitted.

[0049] In the state of FIG. 9, the detection result of the front sensor 210, that is, the outer detection part OP1a included in the detection range Sa1, includes a more inner part on the outer boundary line OBL compared to the state of FIG. 3. Further, the detection range Sa1 also includes a part behind the inner boundary line IBL (the second inner detection part IP1b). At this time, the detection range Sa10 includes the first inner detection part IP1a that extends further behind the first inner detection part IP1 shown in FIG. 4.

[0050] When it is determined that the second inner detection part is not included in the detection result of the front sensor 210 (S122: NO), the above-described S130 and S135 are executed. Therefore, in this case, similar to the first embodiment, the inner boundary line BLR is specified by extending and estimating the inner detection part IP1 (the first inner detection part IP1).

[0051] On the other hand, when it is determined that the second inner part is included in the detection result of the front sensor 210 (S122: YES), as shown in FIG. 8, the first inner detection part is extended and estimated so as to be parallel to the outer detection part to specify the first candidate boundary line (S124). The "candidate boundary line" means a boundary line that is a candidate for the inner boundary line. S124 and S130 differ from each other in whether to specify the result of extending and estimating the first inner detection part as the "inner boundary line" or the "first candidate boundary line", and the other processes are the same.

[0052] In the example of FIG. 10, the first inner detection part IP1a is extended and estimated so as to be parallel to the outer detection part OP1a, and the first candidate boundary line CL1 is specified. In FIG. 10, for the sake of illustration, the left half of the running path C1 is omitted.

[0053] As shown in FIG. 8, in S126, the boundary line specifying unit 13 extends and estimates the second inner detection part so as to be parallel to the outer detection part OP1a, and specifies a second candidate boundary line. In the example of FIG. 10, the second inner detection part IP1b is extended and estimated, and the second candidate boundary line CL2 is specified. As shown in FIG. 10, in the second candidate boundary line CL2, the position along the traveling direction of the front end portion of the extending portion IPVb is substantially equal to the position along the traveling direction of the front end portion of the first inner detection part IP1a. Further, in the first candidate boundary line CL1 described above, the position along the traveling direction of the rear end portion of the extending portion IPVa is substantially equal to the position along the traveling direction of the rear end portion of the second inner detection part IP1b.

[0054] As shown in FIG. 8, the boundary line specifying unit 13 specifies, as the inner boundary line, the boundary line closer to the center in the width direction of the planned driving lane among the first candidate boundary line and the second candidate boundary line (S128). The width direction is parallel to the X direction.

[0055] In the example of FIG. 10, among the first candidate boundary line CL1 and the second candidate boundary line CL2, the first candidate boundary line CL1 is closer to the center CR in the width direction of the lane C1. Therefore, in this example, the first candidate boundary line CL1 is specified as the inner boundary line BLR.

[0056] Both the detection result of the front sensor 210 and the detection result of the surrounding sensor 220 include detection errors. Therefore, as shown in FIG. 10, the first candidate boundary line CL1 and the second candidate boundary line CL2 may not match. In this case, in the third embodiment, by specifying the candidate boundary line closer to the center CR in the width direction of the lane C1 as the inner boundary line, it is possible to suppress misrecognizing a part of the parking row as being within the lane and recognizing an incorrect lane.

[0057] According to the travel path recognition device 100 of the third embodiment described above, the same effects as those of the travel path recognition device 100 of the first embodiment are achieved. In addition, by extending and estimating the first inner detection portion IP1a so as to be parallel to the outer detection portion OP1a, the first candidate boundary line CL1, which is a candidate for the inner boundary line, is specified. By extending and estimating the second inner detection portion IP1b so as to be parallel to the outer detection portion OP1a, the second candidate boundary line CL2, which is a candidate for the inner boundary line, is specified. Among the first candidate boundary line CL1 and the second candidate boundary line CL2, the boundary line closer to the center in the width direction of the travel path C1 is specified as the inner boundary line BLR, so that a travel path with a smaller width can be specified as the planned travel path. Therefore, when the vehicle V1 travels on the specified planned travel path, it is possible to prevent the vehicle V1 from deviating from the travel path and traveling.

[0058] D. Fourth Embodiment: The travel path recognition device 100 of the fourth embodiment has the same configuration as the travel path recognition device 100 of the first embodiment. Therefore, the same reference numerals are given to the same components, and detailed descriptions thereof are omitted.

[0059] The travel path recognition process of the fourth embodiment shown in FIG. 11 differs from the travel path recognition process of the third embodiment shown in FIGS. 7 and 8 in that S124 is omitted, S126a is executed instead of S126, and S129 is executed instead of S128. Other procedures in the travel path recognition process of the fourth embodiment are the same as those in the travel path recognition process of the third embodiment. Therefore, the same reference numerals are given to the same procedures, and detailed descriptions thereof are omitted. Note that the process shown in FIG. 7 is exactly the same in the fourth embodiment, so the illustration is omitted in the fourth embodiment.

[0060] When it is determined in S122 that the detection result of the front sensor 210 includes the second inner detection part (S122: YES), as shown in FIG. 11, the boundary line specifying part 13 extends and estimates the second inner detection part so as to be parallel to the outer detection part, and specifies a candidate boundary line (S126a). In the fourth embodiment, there is only one candidate boundary line. Therefore, S126a is different from S126 only in that it is specified as a mere "candidate boundary line" instead of the "second candidate boundary line".

[0061] The boundary line specifying part 13 specifies, as the inner boundary line, the set of the other part (hereinafter also referred to as the "overlapping part") of the candidate boundary lines that does not overlap with the first inner detection part in the width direction of the lane with respect to the first inner detection part, and the first inner detection part (S129).

[0062] In FIG. 12, the inner boundary line BLR specified from the detection results of the front sensor 210 and the peripheral sensor 220 is shown in the same state as FIG. 9 of the third embodiment. In FIG. 12, similar to FIG. 10, for the sake of illustration, the left half of the lane C1 is omitted.

[0063] As shown in FIG. 12, the set of the other part CL3 of the candidate boundary line CL2 that does not overlap with the first inner detection part IP1a in the width direction (X direction) of the lane C1 and the first inner detection part IP1a is specified as the inner boundary line BLR. Note that the "other part CL3" corresponds to the combined part of the second inner detection part IP1b and the part of the extended part IPVb excluding the overlapping part OVP.

[0064] As shown in FIG. 11, after the completion of S129, the above-described S135 is executed, and the planned driving lane is recognized.

[0065] According to the running recognition device 100 of the fourth embodiment described above, the same effects as those of the running recognition device 100 of the third embodiment can be achieved. In addition, the set of the other part CL3 of the candidate boundary line CL2 excluding the overlapping part OVP that overlaps in the width direction (X direction) of the running C1 with respect to the first inner detection part IP1a and the first inner detection part IP1a is specified as the inner boundary line BLR. Therefore, the actually detected first inner detection part IP1a is specified as the inner boundary line, and for the part that does not overlap with the first inner detection part IP1a in the width direction, the candidate boundary line CL2 is specified as the inner boundary line. Thus, the inner boundary line BLR can be specified in a wide range while suppressing a decrease in the specification accuracy.

[0066] E. Other Embodiments: (E1) In each of the embodiments, the sensor 200 includes the front sensor 210 and the peripheral sensor 220, but it may be configured to include only the front sensor 210. Even in such a configuration, as shown in FIG. 9, when the second inner detection part IP1b is detected within the detection range Sa1 of the front sensor 210, the boundary line obtained by extending and estimating the second inner detection part IP1b so as to be parallel to the outer detection part OP1a, that is, the second candidate boundary line CL2, may be specified as the inner boundary line BLR. Even in such a configuration, when a part of the inner boundary line IBL can be detected within the detection range Sa1 of the front sensor 210 immediately after the start of the turning operation, the inner boundary line BLR can be detected with high accuracy and the planned running path can be recognized.

[0067] (E2) In each of the embodiments, when extending and estimating the inner detection parts IP1 and IP1a, they are extended and estimated so as to be parallel to the outer detection parts OP1 and OP1a, but the present disclosure is not limited to this. They may be extended and estimated so that the angle between them and the outer detection parts OP1 and OP1a is greater than 0° and less than or equal to a predetermined angle.

[0068] (E3) In each embodiment, the turning determination unit 11 determines whether the vehicle V1 is performing a turning operation by using the detection result of the yaw rate sensor 300. However, the present disclosure is not limited thereto. For example, it may be determined by using the detection result of a steering angle sensor. Further, for example, in a configuration in which the vehicle V1 is equipped with map information and can specify its own vehicle position by GPS or the like, when it is specified that the current position of the vehicle V1 is at an intersection or within a curve, it may be determined that the vehicle V1 is performing a turning operation.

[0069] (E4) The travel recognition devices 100 and 100a and the methods thereof described in the present disclosure may be realized by a dedicated computer provided by configuring a processor and a memory programmed to execute one or more functions embodied by a computer program. Alternatively, the travel recognition devices 100 and 100a and the methods thereof described in the present disclosure may be realized by a dedicated computer provided by configuring a processor with one or more dedicated hardware logic circuits. Or, the travel recognition devices 100 and 100a and the methods thereof described in the present disclosure may be realized by one or more dedicated computers configured by a combination of a processor and a memory programmed to execute one or more functions and a processor configured by one or more hardware logic circuits. Further, the computer program may be stored in a computer-readable non-transitory tangible recording medium as instructions to be executed by a computer.

[0070] The present disclosure is not limited to the above-described embodiments, and can be realized in various configurations without departing from the gist thereof. For example, the technical features in each embodiment corresponding to the technical features in the forms described in the summary of the invention can be appropriately replaced or combined in order to solve some or all of the above-described problems, or to achieve some or all of the above-described effects. Further, if the technical feature is not described as essential in this specification, it can be appropriately deleted. The present disclosure can be realized, for example, in the form of a control method for controlling a lane departure prevention function, a computer program for realizing such a method, a non-transitory recording medium recording such a computer program, and the like.

Explanation of Signs

[0071] 11…Turn identification section, 12…Detection result acquisition section, 13…Boundary line identification section, 100…Lane recognition device, 100a…Lane recognition device, 200…Sensor, 210…Front sensor, 220…Peripheral sensor, BLL…Outer boundary line, BLR…Inner boundary line, C1…Lane (planned driving lane), IBL…Inner boundary line, IP1…First inner detection portion, IP1a…First inner detection portion, IP1b…Second inner detection portion, OBL…Outer boundary line, OP1…Outer detection portion, OP1a…Outer detection portion, V1…Vehicle

Claims

1. A travel route recognition device that recognizes a planned travel route of a vehicle, comprising: a turning identification unit (11) that identifies whether or not the vehicle (V1) is performing a turning operation; a detection result acquisition unit (12) that acquires a detection result of a sensor (200) mounted on the vehicle and detecting the environment around the vehicle; a boundary line identification unit (13) that identifies left and right boundary lines (BLL, BLR) of the planned travel route (C1) using the detection result; wherein when it is identified that the vehicle is performing a turning operation, the boundary line identification unit (i) identifies an outer boundary line (BLL), which is the boundary line on the outer side of the turning direction among the left and right boundary lines, using at least a part (OP1, OP1a) of the outer detection part of the outer boundary line included in the detection result; and (ii) identifies an inner boundary line (BLR), which is the boundary line on the inner side of the turning direction among the left and right boundary lines, by extending and estimating a part (IP1, IP1a) of the inner boundary line included in the detection result using the outer detection part. Travel route recognition device.

2. The travel route recognition device according to claim 1, wherein the sensor includes a front sensor (210) and a peripheral sensor (220), the front sensor detecting at least the front environment of the vehicle and being capable of detecting a distant obstacle with higher accuracy than the peripheral sensor, and the peripheral sensor detecting at least the side or rear environment of the vehicle and being capable of detecting an obstacle in a wider angle range than the front sensor; when it is identified that the vehicle is performing a turning operation, the boundary line identification unit identifies the outer boundary line using the outer detection part included in the detection result of the front sensor; and identifies the inner boundary line by extending and estimating a part of the inner boundary line included in the detection result of the peripheral sensor using the outer detection part included in the detection result of the front sensor.

3. The travel route recognition device according to claim 2, wherein the boundary line identification unit identifies the inner boundary line by extending and estimating a part of the inner boundary line to be parallel to the outer detection part.

4. The travel route recognition device according to claim 3, A correlation determination unit (15) that determines whether or not a part of the inner boundary line included in the detection result of the peripheral sensor and the outer detection part included in the detection result of the front sensor correlate with each other as the left and right boundary lines of the same lane, is further provided. The boundary line specifying unit is a lane recognition device that, when determined by the correlation determination unit to correlate with each other, specifies the inner boundary line by extending and estimating a part of the inner boundary line so as to be parallel to the outer detection part. **Claim 5** In the lane recognition device according to any one of claims 1 to 4, the sensor includes a front sensor (210) and a peripheral sensor (220), the front sensor that detects at least the front environment of the vehicle and can detect a far obstacle with higher accuracy than the peripheral sensor, and the peripheral sensor that detects at least the side or rear environment of the vehicle and can detect an obstacle in a wider angle range than the front sensor. The boundary line specifying unit is a case where it is specified that the vehicle is performing a turning operation, and the detection result of the peripheral sensor includes a first inner detection part (IIP1a) that is a part of the inner boundary line, and the detection result of the front sensor includes both the outer detection part (OP1a) and a second inner detection part (IP1b) that is a part of the inner boundary line. (iii) By extending and estimating the first inner detection part so as to be parallel to the outer detection part, a first candidate boundary line (CL1) that is a candidate for the inner boundary line is specified. (iv) By extending and estimating the second inner detection part so as to be parallel to the outer detection part, a second candidate boundary line (CL2) that is a candidate for the inner boundary line is specified. (v) Among the first candidate boundary line and the second candidate boundary line, a boundary line closer to the center (CR) in the width direction of the planned travel lane is specified as the inner boundary line. The lane recognition device. **Claim 6** In the lane recognition device according to any one of claims 1 to 4, the sensor includes a front sensor (210) and a peripheral sensor (220), the front sensor that detects at least the front environment of the vehicle and can detect a far obstacle with higher accuracy than the peripheral sensor, and the peripheral sensor that detects at least the side or rear environment of the vehicle and can detect an obstacle in a wider angle range than the front sensor. When the boundary line specifying unit specifies that the vehicle is performing a turning operation, and the detection result of the peripheral sensor includes a first inner detection portion (IP1a) that is a part of the inner boundary line, and the detection result of the front sensor includes both the outer detection portion (OP1a) and a second inner detection portion (IP1b) that is a part of the inner boundary line, (vi) specifying a candidate boundary line (CL2) that is a candidate for the inner boundary line by extending and estimating the second inner detection portion so as to be parallel to the outer detection portion; (vii) a travel path recognition device that specifies, as the inner boundary line (BLR), a set of the other portion (CL3) of the candidate boundary line excluding a portion (OVP) that overlaps with the first inner detection portion in the width direction of the planned travel path with respect to the first inner detection portion.

7. A travel path recognition method for recognizing a planned travel path of a vehicle, a step of specifying, by a travel path recognition device (100, 100a), whether or not the vehicle (V1) is performing a turning operation; a step of acquiring, by the travel path recognition device, a detection result of a sensor (200) mounted on the vehicle and detecting an environment around the vehicle; a step of specifying, by the travel path recognition device, left and right boundary lines (BLL, BLR) of the planned travel path using the detection result; comprising: the step of specifying the left and right boundary lines includes, when it is specified that the vehicle is performing a turning operation, (i) a step of specifying, as an outer boundary line (BLL) that is an outer boundary line in the turning direction among the left and right boundary lines, using at least a part of the outer boundary line, i.e., an outer detection portion (OP1, OP1a), included in the detection result; (ii) a step of specifying, as an inner boundary line (BLR) that is an inner boundary line in the turning direction among the left and right boundary lines, by extending and estimating a part (IP1, IP1a) of the inner boundary line included in the detection result using the outer detection portion; A travel path recognition method including the above.

8. A computer program for recognizing a planned travel path of a vehicle, a function of specifying whether or not the vehicle (200) is performing a turning operation; a function of acquiring a detection result of a sensor (200) mounted on the vehicle and detecting an environment around the vehicle; a function of specifying left and right boundary lines (BLL, BLR) of the planned travel path; to be realized by a computer, The function of specifying the left and right boundary lines, when it is specified that the vehicle is performing a turning operation, (i) a function of specifying an outer boundary line (BLL), which is the boundary line on the outer side of the turning direction among the left and right boundary lines, using an outer detection part (OP1, OP1a) which is at least a part of the outer boundary line included in the detection result; (ii) a function of specifying an inner boundary line (BLR), which is the boundary line on the inner side of the turning direction among the left and right boundary lines, by extending and estimating a part (IP1, IP1a) of the inner boundary line included in the detection result using the outer detection part; A computer program including the above.

Images