Road recognition device

Abstract

The road recognition device includes a detection unit that detects a road marking of a road on which a host vehicle is traveling, a road marking discrimination unit that discriminates a category of the road marking corresponding to a color and a shape of the road marking detected by the detection unit, and a road determination unit that determines whether or not the road on which the host vehicle is traveling is a two-way road on which two-way traffic is performed, based on road marking information including information on the category of the road marking discriminated by the road marking discrimination unit.

Description

TECHNICAL FIELD

[0001] The present application relates to a road recognition device that recognizes a category of a road on which a vehicle is traveling. BACKGROUND

[0002] As such a device, a device that determines whether a road on which a vehicle is traveling is a bidirectional road has been known. For example, in Patent Literature 1, a device is described that determines whether a road on which a vehicle is traveling is a bidirectional road, based on map information including current position information of the vehicle and a category of a road, which is acquired from a navigation device.

[0003] However, if determination is made based on information from a navigation device as in the device described in Patent Literature 1, it is difficult to determine whether it is a bidirectional road with high precision.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT LITERATURE

[0006] Patent Literature 1: Japanese Patent Application Publication No. JP 2013-190962 A SUMMARY

[0007] A road recognition device according to an aspect of the present application includes a detection unit that detects a road marking of a road on which a host vehicle is traveling, a road marking discrimination unit that discriminates a category of the road marking corresponding to a color and a shape of the road marking detected by the detection unit, and a road determination unit that determines whether the road on which the host vehicle is traveling is a bidirectional road on which bidirectional traffic is performed, based on road marking information including the category of the road marking discriminated by the road marking discrimination unit. BRIEF DESCRIPTION OF DRAWINGS

[0008] Objects, features, and advantages of the present application will be further clarified by the following description of embodiments with reference to the accompanying drawings.

[0009] Figure 1 is a block diagram that schematically shows an overall configuration of a vehicle control system of an autonomous vehicle that has a road recognition device according to an embodiment of the present application.

[0010] Figure 2 is a diagram that shows an example of a driving scenario assumed by a road recognition device according to an embodiment of the present application.

[0011] Figure 3A is a diagram that shows an example of a road marking between a host lane and an adjacent lane, which is different from Figure 2

[0012] Figure 3B Figure 2 ​​FIG. 1 is a diagram showing another example of a road marking between a subject lane and an adjacent lane.

[0013] Figure 4 FIG. 2 is a block diagram showing a main part configuration of a road recognition device according to an embodiment of the present application.

[0014] Figure 5 FIG. 3 is a diagram showing an example of a road observed from a camera provided to the road recognition device according to the embodiment of the present application.

[0015] Figure 6 FIG. 4 is a flowchart showing an example of processing performed by a controller of the road recognition device according to the embodiment of the present application. Figure 4

[0016] Figure 7 FIG. 5 is a diagram showing an example of a change in a determination result of the road recognition device according to the embodiment of the present application.

[0017] Figure 8A FIG. 6 is a diagram showing an example of a right-side passing road.

[0018] Figure 8B FIG. 7 is a diagram showing another example of a right-side passing road. DETAILED DESCRIPTION

[0019] Hereinafter, the embodiment of the present application will be described with reference to the drawings. Figures 1-8B A road recognition device according to an embodiment of the present application is a device that recognizes a category of a road, and can be applied to both a vehicle having an automatic driving function, i.e., an automatic driving vehicle, and a vehicle not having the automatic driving function, i.e., a manual driving vehicle. Hereinafter, an example in which the road recognition device is applied to the automatic driving vehicle will be described. Note that the vehicle to which the road recognition device according to the embodiment is applied can be referred to as a subject vehicle in distinction from other vehicles.

[0020] The subject vehicle can be any one of an engine vehicle having an internal combustion engine (engine) as a travel drive source, an electric vehicle having a travel motor as a travel drive source, and a hybrid vehicle having an engine and a travel motor as travel drive sources.

[0021] The subject vehicle (automatic driving vehicle) can travel not only in an automatic driving mode in which a driver does not perform a driving operation, but also in a manual driving mode in which the driver performs the driving operation. For example, the subject vehicle can start to travel in the manual driving mode by operating a steering wheel (opening hand control) from a state of traveling in the automatic driving mode without operating the steering wheel (closing hand control). Alternatively, the subject vehicle can start to travel in a manner of opening hand control by reducing an automatic driving level by one or more levels according to an instruction from a vehicle control system from a state of traveling in a manner of closing hand control at a predetermined automatic driving level.

[0022] ​First, a general configuration of automatic driving will be described. Figure 1 is a block diagram showing a general configuration of a vehicle control system 100 having a road recognition device of an embodiment of the present application. As shown in Figure 1 , the vehicle control system 100 mainly has a controller 10 and an external sensor group 1, an internal sensor group 2, an input / output device 3, a positioning unit 4, a map database 5, a navigation device 6, a communication unit 7, an actuator AC for running, which are communicably connected to the controller 10 respectively through a CAN communication line or the like.

[0023] The external sensor group 1 is a general term of a plurality of sensors (external sensors) that detect the surrounding information of the host vehicle, that is, external conditions. For example, the external sensor group 1 includes a laser radar that detects the position (distance from the host vehicle, direction) of an object in the vicinity of the host vehicle by irradiating laser light to detect reflected light, a radar that detects the position of an object in the vicinity of the host vehicle by irradiating electromagnetic waves to detect reflected waves, and a camera that has a CCD (Charge Coupled Device), CMOS (Complementary Metal Oxide Semiconductor), or the like, to take an image of the vicinity of the host vehicle, and the like. The laser radar and the radar can detect an object within the imaging area of the camera.

[0024] The internal sensor group 2 is a general term of a plurality of sensors (internal sensors) that detect the running state of the host vehicle. For example, the internal sensor group 2 includes a vehicle speed sensor that detects the vehicle speed of the host vehicle, an acceleration sensor that detects the acceleration in the front-rear direction and the left-right direction of the host vehicle, a rotation speed sensor that detects the rotation speed of a running drive source, and the like. A sensor that detects the driving operation of the driver in a manual driving mode, such as the operation of an accelerator pedal, the operation of a brake pedal, the operation of a steering wheel, and the like, is also included in the internal sensor group 2.

[0025] The input / output device 3 is a general term of a device that inputs an instruction from the driver and outputs information to the driver. For example, the input / output device 3 includes various switches that allow the driver to input various instructions by operating an operation member, a microphone that allows the driver to input an instruction by voice, a display that provides information to the driver by displaying an image, a speaker that provides information to the driver by voice, and the like.

[0026] The positioning unit (GNSS unit) 4 has a positioning sensor that receives a positioning signal transmitted from a positioning satellite. The positioning sensor can also be included in the internal sensor group 2. The positioning satellite is a GPS satellite, a quasi-zenith satellite, or the like, artificial satellite. The positioning unit 4 determines the current position (latitude, longitude, altitude) of the host vehicle using the positioning information received by the positioning sensor.

[0027] The map database 5 is a device that stores general map information used in the navigation device 6, for example, constituted by a hard disk, a semiconductor element. The map information includes: position information of roads, information of road shapes (curvatures, etc.), position information of intersections, branch points. Note that the map information stored in the map database 5 is different from the high-precision map information stored in the storage section 12 of the controller 10.

[0028] The navigation device 6 is a device that searches for a target route on a road to a destination input by the driver and performs guidance along the target route. The input of the destination and the guidance along the target route are performed by the input / output device 3. The target route is calculated based on the current position of the host vehicle determined by the positioning unit 4 and the map information stored in the map database 5. It is also possible to determine the current position of the host vehicle using the detection values of the external sensor group 1, and it is also possible to calculate the target route based on the current position and the high-precision map information stored in the storage section 12.

[0029] The communication unit 7 communicates with various servers not shown using a network including a wireless communication network typified by the Internet, a mobile phone network, etc., and acquires map information, travel record information, traffic information, etc. from the servers periodically or at arbitrary timing. The network includes not only a public wireless communication network but also a closed communication network such as a wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc. set for each prescribed management area. The acquired map information is output to the map database 5, the storage section 12, and the map information is updated.

[0030] The actuator AC is a travel actuator for controlling the travel of the host vehicle. In the case where the travel drive source is an engine, the actuator AC includes a throttle actuator that adjusts the opening degree of a throttle valve of the engine. In the case where the travel drive source is a travel motor, the actuator AC includes the travel motor. A brake actuator that operates a brake device of the host vehicle and a steering actuator that drives a steering device are also included in the actuator AC.

[0031] The controller 10 is constituted by an electronic control unit (ECU). More specifically, the controller 10 is constituted by a computer including an arithmetic section 11 having a CPU (microprocessor), etc., a storage section 12 such as a ROM (read only memory), a RAM (random access memory), an I / O (input / output) interface, and other peripheral circuits not shown. Note that a plurality of ECUs having different functions such as an engine control ECU, a travel motor control ECU, a brake device ECU, etc. can be provided separately, but for the sake of convenience, the controller 10 is shown as a collection of these ECUs in the following description. Figure 1

[0032] ​High-precision road map information is stored in the storage section 12. The road map information includes: position information of a road, information of a road shape (curvature, etc.), information of a road slope, position information of an intersection, a branch, information of a number of lanes, a width of a lane, and position information of each lane (a central position of a lane, information of a boundary line of a lane position), position information of a landmark (a signal, a sign, a building, etc.) as a marker on a map, information of a road surface profile such as a road surface bump, etc. The map information stored in the storage section 12 includes: map information acquired from the outside of the host vehicle by the communication unit 7 and map information made by the host vehicle itself using a detection value of the external sensor group 1 or a detection value of the external sensor group 1 and the internal sensor group 2.

[0033] The arithmetic section 11 has, as a functional structure, a host vehicle position recognition section 13, an outside recognition section 14, a behavior plan generation section 15, and a travel control section 16.

[0034] The host vehicle position recognition section 13 recognizes a position of the host vehicle on a map (host vehicle position) from the position information of the host vehicle acquired by the positioning unit 4 and the map information of the map database 5. The host vehicle position can also be recognized using the map information stored in the storage section 12 and the surrounding information of the host vehicle detected by the external sensor group 1, whereby the host vehicle position can be recognized with high precision. Note that when the host vehicle position can be measured by a sensor provided outside, such as on or beside a road, the host vehicle position can also be recognized by communicating with the sensor via the communication unit 7.

[0035] The outside recognition section 14 recognizes an outside situation around the host vehicle from a signal from the external sensor group 1 such as a laser radar, a radar, a camera, etc. For example, the outside recognition section 14 recognizes a position, a speed, an acceleration of a surrounding vehicle (a front vehicle, a rear vehicle) traveling around the host vehicle, a position of a surrounding vehicle parked or standing around the host vehicle, and a position, a state of another object, etc. The another object includes: a sign, a signal, a marker such as a road marking or a stop line of a road, a building, a guardrail, a utility pole, a billboard, a pedestrian, a bicycle, etc. The state of the another object includes: a color (red, green, yellow) of a signal, a moving speed, a direction of a pedestrian, a bicycle, etc.

[0036] The action plan generation section 15 generates a travel trajectory (target trajectory) of the host vehicle from the current time point to a prescribed time, for example, based on the target path calculated by the navigation device 6, the map information stored in the storage section 12, the host vehicle position recognized by the host vehicle position recognition section 13, and the external situation recognized by the external recognition section 14. When there are a plurality of trajectories as candidates for the target trajectory on the target path, the action plan generation section 15 selects the best trajectory that complies with the law and satisfies criteria such as efficient and safe travel, from among them, and sets the selected trajectory as the target trajectory. Then, the action plan generation section 15 generates an action plan corresponding to the generated target trajectory. The action plan generation section 15 generates various action plans corresponding to, for example, passing travel for overtaking a preceding vehicle, lane change travel for changing a travel lane, following travel for following a preceding vehicle, lane keeping travel for keeping a travel lane without deviating therefrom, deceleration travel, or acceleration travel. The action plan generation section 15 first determines a travel mode when generating the target trajectory, and generates the target trajectory based on the travel mode.

[0037] In the automatic driving mode, the travel control section 16 controls each actuator AC so that the host vehicle travels along the target trajectory generated by the action plan generation section 15. More specifically, the travel control section 16 calculates a required driving force for obtaining the target acceleration per unit time calculated by the action plan generation section 15, taking into account the travel resistance determined by the road slope and the like in the automatic driving mode. Then, for example, feedback control is performed on the actuators AC so that the actual acceleration detected by the internal sensor group 2 becomes the target acceleration. That is, the actuators AC are controlled so that the host vehicle travels at the target vehicle speed and the target acceleration.

[0038] Note that, in the manual driving mode, the travel control section 16 controls each actuator AC based on a travel instruction (steering operation and the like) from the driver obtained by the internal sensor group 2. Even in the manual driving mode, the host vehicle is able to automatically travel in a prescribed driving condition without relying on the driving operation of the driver in a state where the front monitoring obligation by the driver is generated, not in a completely manual driving state. For example, in a same lane in an expressway and the like, the host vehicle is able to travel in a closed hand control mode in which the driver releases the steering wheel, that is, automatically drive in the same lane, by the travel control section 16 controlling the actuators AC (for example, a steering actuator) based on the inter-vehicle distance with a preceding vehicle and the like. Note that, in a state where the automatic driving level is lowered to a prescribed level, the same lane automatic driving is also able to be performed in the automatic driving mode.

[0039] With the above configuration as a premise, the characteristic structure of the road recognition device of the embodiment of the present application will be described. The road recognition device of the embodiment is a device that recognizes whether a road on which the host vehicle is traveling is a two-way road on which two-way traffic is performed, as a category of the road. There are sometimes sections on which two-way traffic is performed without a central separation on an expressway or a general road. In this case, when the host vehicle continues to perform a prescribed driving state (for example, automatic driving in the same lane) while traveling in a state in which hand control is closed, it is difficult to perform a quick evasive action in a case where an oncoming vehicle approaches the host vehicle so that the host vehicle needs to give way to the oncoming vehicle. Therefore, it is particularly important for a vehicle that can travel in a state in which hand control is closed to recognize whether a road on which the vehicle is traveling is a two-way road.

[0040] Figure 2 is a drawing that shows an example of a driving scenario assumed by the road recognition device of the embodiment, and shows a one-lane two-way road. In Figure 2 , the host vehicle 101 is traveling in the host lane (first lane) LN1 in the arrow A1 direction, and the other vehicle 102 is traveling in the adjacent lane (second lane) LN2 adjacent to the host lane LN1 in the arrow A2 direction opposite to the arrow A1 direction. The host lane LN1 is defined by a pair of road lines L1, L2 on the left and right, and the adjacent lane LN2 is defined by a pair of road lines L3, L4 on the left and right. The road lines L1, L4 on the road end side are white (W) solid lines. The road lines L2, L3 on the central side of the road are yellow (Y) solid lines that show a no-passing line for passing. The road line L5 on the road center inside the road lines L2, L3 is a white (W) broken line. Note that the yellow road line can also be orange.

[0041] The road line between the host lane LN1 and the adjacent lane LN2, that is, the boundary between the host lane LN1 and the adjacent lane LN2 is not limited to Figure 2 the example shown in the drawing. There are various forms. Figure 3A , 3B is a drawing that shows this example. In Figure 3A , the boundary line between the host lane LN1 and the adjacent lane LN2 is shown by a single road line L2 (center line). The road line L2 is indicated by a white (W) broken line. In Figure 3B , the boundary line between the host lane LN1 and the adjacent lane LN2 is shown by a pair of road lines L2, L3 on the left and right. The road line L2 on the host lane side is indicated by a white (W) broken line, and the road line L3 on the adjacent lane side is indicated by a yellow (Y) solid line.

[0042] In the embodiment, a road that satisfies a prescribed road line condition is treated as a two-way road. As Figure 2As shown, the road marking condition includes the road marking L2 of the center line side (the adjacent lane side) of the road markings L1, L2 of the subject lane LN1 being yellow. Therefore, Figure 3A , 3B The road shown does not satisfy the road marking condition, and thus is not treated as a two-way traffic road. Note that illustration is omitted, but the two-way traffic road includes a case where the boundary line between the subject lane and the adjacent lane is indicated by a single yellow solid line, a case where it is indicated by a single yellow solid line and a single white solid line or a single white dashed line in parallel, and the like Figure 2 The mode of the road marking other than the example shown.

[0043] Figure 4 is a block diagram showing the main part configuration of the road recognition device 50 of the embodiment of the present application. As shown in Figure 4 , the road recognition device 50 mainly has the camera la, the controller 10, the actuator AC, and the notification section 3a.

[0044] The camera la is a monocular camera capable of color recognition of an object, having an imaging element (image sensor) such as a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), and the like, and constitutes a part of the external sensor group 1 of Figure 1 . The camera la can also be a stereo camera. The camera la is installed at a prescribed position of the front portion of the host vehicle 101, for example, and continuously captures the front space of the host vehicle 101 to acquire an image (camera image) of an object. The object includes other vehicles 102 and road markings on the road (for example, the road markings L1 to L5 of Figure 2 . Note that the object can also be detected by a laser radar or the like instead of or in addition to the camera la.

[0045] The notification section 3a is a member that prompts the driver to perform a prescribed driving operation, and constitutes a part of the input / output device 3 of Figure 1 . Specifically, the notification section 3a notifies the driver who closes the manual control of opening the manual control in a case where the host vehicle 101 is in the same lane automatic driving. The notification section 3a is constituted by a monitor provided in front of the driver's seat, a speaker in the vehicle cabin, and the like.

[0046] Figure 4 The controller 10 of Figure 1 functions as a functional structure assumed by the arithmetic section 11, and has a road marking discrimination section 141 and a road determination section 142 in addition to the travel control section 16. The road marking discrimination section 141 and the road determination section 142 are used to discriminate the category of the road marking and the category of the road, respectively, which are included in the outside recognition section 14 of Figure 1 , for example.

[0047] The road marking recognition unit 141 identifies multiple road markings on the road based on camera images acquired by camera 1a, and classifies each road marking. Specifically, it classifies whether the road marking is white or yellow, and whether it is a solid or dashed line. This classified information is used as road marking information.

[0048] The road determination unit 142 determines whether the road in motion is a two-way road based on road marking information, including information about the type of road markings, identified by the road marking identification unit 141, and information about other vehicles detected by the camera 1a. Specifically, it determines whether the road in motion meets the prescribed road marking conditions; if the road marking conditions are met, it is determined to be a two-way road. It should be noted that the following assumes a two-way road with one lane on each side, and determines whether it is a two-way road.

[0049] like Figure 2 As shown, the road marking conditions include specifying that the road marking L2 on the right side of the center line of lane LN1 is a yellow road marking L2 (referred to as the R road marking). Additionally, the road marking conditions include specifying that the road marking L1 on the opposite side (left side) of the center line of lane LN1 is a solid white road marking L1 (referred to as the L road marking). It should be noted that a two-way traffic road cannot be determined if the R road marking L2 is not identified.

[0050] When identifying the type of road markings based on camera images, if road marking L1 is blurred, it may not be recognized as a solid white line. To address this, the road marking condition also includes the absence of sufficient area outside (to the left) of road marking L1. Figure 5 This diagram illustrates an example of a road as seen from camera 1a when the road marking L1 is blurred and the white solid line is not clearly visible. (See diagram 1.) Figure 5 As shown, a sidewall 103 is erected outside the road marking L1 along the direction of road extension to demarcate the road. The base end of the sidewall 103, i.e. the boundary line L10 of the road edge, extends approximately parallel to the road marking L2.

[0051] At this time, the road determination unit 142 sequentially calculates the distance D from the R road line L2 to the boundary line L10 along the direction of travel, that is, the distance D from the feature point Pa on the R road line L2 to the feature point Pb on the boundary line L10, as identified by the camera image. Then, by determining whether the distance D is less than a predetermined value D1, it is determined that there is not enough area to the left of the L road line L1, that is, one lane on one side. The predetermined value D1 is set to the width of two lanes (for example, 2 × 3.5m). Thus, even if the L road line L1 is not identified as a solid white line, the road determination unit 142 can still accurately determine whether it is a two-way road.

[0052] Even if the distance D is set to be equal to or greater than the prescribed value D1 (D ≥ D1), there is not necessarily a lane on the left side of the L road line L1. Therefore, the presence of a road line on the outer side (left side) of the L road line L1 (referred to as an LL road line) is also included in the road line condition. In the state where D ≥ D1, when an LL road line is present, the road determination portion 142 determines that it is not a two-way traffic road (non-two-way traffic road).

[0053] In order to improve the determination accuracy of two-way traffic, in the present embodiment, the road line condition includes a case where a prescribed condition is recognized on the right side of the R road line L2 from the camera image. Specifically, as shown in FIG. 6, the presence of a white road line (referred to as an RR road line) L5 on the outer side (right side) of the R road line L2 from the camera image is included in the road line condition. Figure 2

[0054] In the case where the RR road line L5 is recognized, the road determination portion 142 also determines, from the camera image, whether or not there is an oncoming vehicle (other vehicle 102) that is traveling on the adjacent lane LN2 and approaching the host vehicle 101 from the front. Then, the determination result level (reliability of the determination result) of the two-way traffic road is set to different values when the oncoming vehicle is recognized and when the oncoming vehicle is not recognized. That is, when the oncoming vehicle is present, the likelihood that it is a two-way traffic road is high, and therefore the determination result level is set to be high (two-way traffic (high)), and when the oncoming vehicle is not present, the determination result level is set to be lower than when the oncoming vehicle is present (two-way traffic (low)).

[0055] On a two-way traffic road, there is a case where the RR road line L5 is not detected from the camera image due to the absence of the RR road line L5 and the like. In this case, the road determination portion 142 determines whether or not it is a two-way traffic road depending on the presence or absence of an oncoming vehicle. Specifically, when the oncoming vehicle is recognized, the likelihood that it is a two-way traffic road is high, and therefore the road determination portion 142 determines that it is a two-way traffic road (two-way traffic (high)).

[0056] On the other hand, when the oncoming vehicle is not recognized, the road determination portion 142 determines that it is a non-two-way traffic road (not a two-way traffic road). However, even in the case where the oncoming vehicle is not recognized, there is a possibility that it is a two-way traffic road. Therefore, in the case where the oncoming vehicle is not recognized, the road determination portion 142 sets the determination result level of the non-two-way traffic road to be low (non-two-way traffic (low)). Note that, as described above, in the case where D ≥ D1 and an LL road line is present, the road determination portion 142 sets the determination result level of the non-two-way traffic road to be high (non-two-way traffic (high)).

[0057] ​In the same lane automatic driving, when the road determination section 142 determines that it is a two-way traffic road (two-way traffic (high) or two-way traffic (low)), the travel control section 16 controls the notification section 3a to output a voice, display that prompts the driver to hold the steering wheel. The travel control section 16 also controls the actuator AC to suspend the same lane automatic driving on the two-way traffic road. In this case, the manner of notification (the content of notification) and the manner of actuator control (for example, the timing of suspension of the same lane automatic driving) can be made different from each other depending on the two-way traffic (high) and the two-way traffic (low). Note that, in the case where the road determination section 142 cannot determine that the R road line L2 is yellow due to the R road line L2 being blurred or the like, the travel control section 16 also outputs a control signal to the actuator AC to prompt the driver to turn on the manual control and suspend the same lane automatic driving.

[0058] On the other hand, in the same lane automatic driving, when the road determination section 142 determines that it is a non-two-way traffic road (non-two-way traffic (high) or non-two-way traffic (low)), the travel control section 16 controls the actuator AC to continue the same lane automatic driving. In addition, after the driver is prompted to turn on the manual control upon determining that it is a two-way traffic road, when the road determination section 142 determines that it is a non-two-way traffic road, the travel control section 16 outputs a control signal to the notification section 3a to notify the driver that the manual control can be turned off.

[0059] Figure 6 is a flowchart showing an example of processing performed by the controller of Figure 4 . Figure 6 The processing mainly regarding the determination of a two-way traffic road is repeatedly performed at a prescribed cycle, for example, starting in the same lane automatic driving.

[0060] As shown in Figure 6 , first, in S1 (S: processing step), an image signal from the camera 1a is read in. Next, in S2, the class of a road line (R road line, L road line, RR road line, LL road line, or the like) included in the camera image is recognized. That is, it is determined whether the color of the road line is white or yellow, whether the shape of the road line is a solid line or a dashed line, and the class of the road line is recognized. Next, in S3, it is determined whether the R road line L2 on the right side of the own lane LN1 is yellow. When S3 is negative (S3: No), it proceeds to S4, and when it is affirmative (S3: Yes), it proceeds to S5. In S4, the determination of the non-two-way traffic is not possible and the processing ends.

[0061] In S5, determine whether the L road line L1 to the left of lane LN1 is a solid white line. If S5 is affirmative (S5: Yes), proceed to S6, and determine whether the RR road line L5 to the right of R road line L2 is white. If S6 is affirmative (S6: Yes), proceed to S7, and determine whether there are oncoming vehicles. If S7 is affirmative (S7: Yes), proceed to S8; otherwise, proceed to S9. In S8, the road is determined to be a two-way road (two-way traffic (high)) based on the higher determination result level, and the process ends. In S9, the road is determined to be a two-way road (two-way traffic (low)) based on the lower determination result level, and the process ends.

[0062] In S6, if the RR road marking L5 is determined to be non-white, S6 is negative (S6: No) and proceeds to S10. In S10, the same as S7, it is determined whether there are oncoming vehicles. If S10 is positive (S10: Yes), proceed to S8; if it is negative (S10: No), proceed to S11. In S11, the road is determined to be a non-two-way road (non-two-way traffic (low)) based on the lower determination result level, and the process ends.

[0063] In S5, if it is determined that road marking L1 is not a solid white line, S5 is negative (S5: No) and proceeds to S12. In S12, it is determined whether there is sufficient space in the adjacent lane to the left of lane LN1. That is, as follows... Figure 5 As shown, the distance D from road line L2 (R) to the road boundary line L10 is calculated based on the camera image, and it is determined whether the distance D is greater than or equal to the specified value D1. If S12 is affirmative (S12: Yes), proceed to S13; if it is negative (S12: No), proceed to S6. In S13, it is determined whether there is another road line (LL road line) to the left of road line L1 (L). If S13 is affirmative (S13: Yes), proceed to S14; if it is negative (S13: No), proceed to S6. In S14, the road is determined to be a non-two-way road (non-two-way traffic (high)) based on the higher determination result level, and the process ends.

[0064] The operation of the road recognition device 50 in this embodiment will be explained in more detail. For example... Figure 2As shown, the right-hand (opposite lane side) R road line L2 (first road line) of lane LN1 is yellow (e.g., a solid yellow line), and the left-hand L road line L1 (second road line) is a solid white line. The right-hand RR road line L5 of R road line L2 is white (e.g., a dashed white line). When an oncoming vehicle (other vehicle 102) is detected in the adjacent lane LN2, all the conditions for multiple road lines are met, and therefore it is determined to be a two-way traffic road (two-way traffic (high)) (S8). Thus, not only are the categories of road lines L1 and L2 of lane LN1 defined, but the category of RR road line L5 and the presence or absence of oncoming vehicles are also considered, so it is possible to more accurately determine whether it is a two-way traffic road. When the vehicle 101 is automatically driving in the same lane and it is determined to be a two-way traffic road, the notification unit 3a instructs the driver to activate manual control, that is, the monitor and speaker remind the driver to activate manual control, thereby stopping the automatic driving in the same lane. Therefore, it is possible to instruct manual control to be activated and automatic driving in the same lane to be stopped at the appropriate time, thereby improving the driving safety of autonomous vehicles.

[0065] Even for two-way roads, there are cases where the road marking condition 102 (the presence of oncoming vehicles) is not met. In this case, if the road marking condition R (road marking L2) is yellow and L (road marking L1) is a solid white line, it is determined to be a two-way road (two-way traffic (low)) (S9). Therefore, even if all the necessary road marking conditions are not met, a road is still determined to be a two-way road when some of the required road marking conditions are met, thus enabling the determination of two-way roads corresponding to various road structures. In this case, the determination result level is set lower than when all road marking conditions are met, thus allowing for appropriate determination of two-way roads.

[0066] Sometimes, due to blurring of road line L1, the solid white line of road line L1 may not be recognized in the camera image. In this case, the boundary line L10 from road line L2 to the left end of the road (…) Figure 5 When the distance D is less than the specified value D1 (S12→S6), there is no adjacent lane to the left of lane LN1, and it is determined to be a two-way road (S8, S9). Furthermore, even if the distance D is greater than the specified value D1, if no L1 road marking is identified to the left of lane L1 (S13→S6), there is no adjacent lane to the left of lane LN1, and it is determined to be a two-way road (S8, S9). Therefore, even when lane L1 is not identified, it is possible to accurately determine whether a road is a two-way road.

[0067] Note that, above, when it is determined that the road marking condition is established, it is determined that it is a two-way road, but sometimes the camera 1a erroneously detects a road marking (for example, the R road marking L2) due to backlight or the like, in which case the determination of two-way traffic can be erroneous. Therefore, in a state in which it is determined that it is not a two-way road, when the road marking condition is established for a prescribed time or more, the road determination unit 142 can also determine that it is a two-way road. Also, in a state in which it is determined that it is a two-way road, when the road marking condition is not established for a prescribed time or more, the road determination unit 142 can also determine that it is not a two-way road. That is, a prescribed state for a prescribed time can also be added to the determination condition for a two-way road.

[0068] Figure 7 is a graph showing an example of a change in the determination result of a two-way road made by the road recognition device 50 constituted taking this into account over time. In Figure 7 an example of a change in the determination result after it is determined by the road determination unit 142 that it is a two-way road (Embodiment 1) and a comparative example thereof (Comparative Example 1), and an example of a change in the determination result after it is determined that it is not a two-way road (Embodiment 2) and a comparative example thereof (Comparative Example 2). Note that, in the graph, the case in which it is determined that it is a two-way road is shown by hatching.

[0069] As shown in Figure 7 , in Embodiment 1, even if the road marking condition is established at time point tl, it is not immediately determined by the road determination unit 142 that it is a two-way road, and at time point t2, when a prescribed time Δtl (for example, 2 seconds) has passed after the road marking condition is established, it is determined that it is a two-way road. Thereafter, at time point t3, when the road marking condition is not established, in Comparative Example 1, it is immediately determined that it is not a two-way road. In contrast, in Embodiment 1, even if the road marking condition is not established, it is not immediately determined that it is not a two-way road, and at time point t4, which is after a state in which the road marking condition is not established has continued for a prescribed time Δt2 (for example, 20 seconds), it is determined by the road determination unit 142 that it is not a two-way road. Therefore, as shown in Embodiment 1 of Figure 7 , as long as the duration for which the road marking condition is not established is less than the prescribed time Δt2, the determination that it is a two-way road is maintained.

[0070] Thus, for example, when the host vehicle 101 is traveling in a state of backlight or the like, it is possible to prevent a yellow road marking from being erroneously detected as a white road marking and erroneously determined to be not a two-way road. That is, the likelihood of erroneous detection due to backlight continuing for a prescribed time Δt2 or more is low, and therefore, by using the fact that the road marking condition is not established for a prescribed time Δt2 or more as a determination condition, the road determination unit 142 can accurately determine a change from a two-way road to a non-two-way road.

[0071] In Comparative Example 2, in a state where it is determined that the road on which the host vehicle 101 is traveling is a non-two-way road, at the time point t5, when it is determined that the road marking condition is satisfied, it is immediately determined that the road is a two-way road. In contrast, in Example 2, even if it is determined by the road determination unit 142 that the road marking condition is satisfied, it is not immediately determined that the road is a two-way road, and when the state in which the road marking condition is satisfied continues until the time point t6, which is after the prescribed time Δtl, it is determined that the road is a two-way road. Thus, as shown in Example 2, as long as the duration for which the road marking condition is satisfied is less than the prescribed time Δtl, the determination that the road is a non-two-way road is maintained. Figure 7

[0072] Thus, in the case where the road marking condition is temporarily satisfied due to a false detection by the camera la, it is possible to prevent the road determination unit 142 from erroneously determining that the road is a two-way road, and it is possible to accurately determine a change from a non-two-way road to a two-way road. Here, the prescribed time Δt2 is set to be longer than the prescribed time Δtl, and thus, compared to the case where it is determined that the road is a two-way road, it is less likely to be determined that the road is a non-two-way road. Thus, it is possible to well prevent the same-lane automatic driving and the like on a two-way road from being assumed to be a non-two-way road, and the safety is high.

[0073] By adopting the present embodiment, the following effects can be achieved.

[0074] (1) The road recognition device 50 includes the camera la that detects a road marking of a road on which the host vehicle 101 is traveling, the road marking discrimination unit 141 that discriminates a category of the road marking corresponding to the color and shape of the road marking detected by the camera la, and the road determination unit 142 that determines whether the road on which the host vehicle 101 is traveling is a two-way road on which two-way traffic is performed (S2) on the basis of road marking information including the category of the road marking discriminated by the road marking discrimination unit 141. In a two-way road, the color and shape of the road marking around the host lane LNl are in a prescribed manner. Thus, by determining whether it is a two-way road on the basis of the color and shape of the road marking detected by the camera la, accurate determination can be performed. Figure 4 (2) The camera la is configured to also be able to detect the other vehicle 102 that is an oncoming vehicle traveling on an oncoming lane (adjacent lane) LN2. The road determination unit 142 also determines whether the road on which the host vehicle 101 is traveling is a two-way road (S2) on the basis of whether the oncoming vehicle 102 is detected by the camera la. In this way, by taking into account whether the oncoming vehicle 102 is actually detected, it is possible to more accurately determine whether it is a two-way road.

[0075] (3) The road determination unit 142 determines whether the road on which the host vehicle 101 is traveling is a two-way road (S2) on the basis of whether the other vehicle 102 is detected by the camera la and whether the road marking condition is satisfied. Thus, by taking into account whether the other vehicle 102 is actually detected and whether the road marking condition is satisfied, it is possible to more accurately determine whether it is a two-way road. Figure 6

[0076] ​​(3) The road determination portion 142 determines whether the road on which the host vehicle 101 is traveling is a bidirectional road based on information on the types of the pair of road markings L1, L2 on the left and right of the host lane LN1 on which the host vehicle 101 is traveling. Figure 6 That is, the R road marking L2 is yellow, and the L road marking L1 is a white solid line is included in the road marking condition, and it is determined whether it is a bidirectional road based on whether the road marking condition is satisfied. Thus, the color and shape of the road marking are taken into consideration, and it is possible to determine whether it is a bidirectional road with good accuracy.

[0077] (4) The road determination portion 142 determines whether the road on which the host vehicle 101 is traveling is a bidirectional road based on the length (distance D) in the vehicle width direction of the outer side region of the host lane LN1. Figure 6 That is, the distance D is less than a predetermined value D1 is included in the road marking condition, and it is determined whether it is a bidirectional road based on whether the road marking condition is satisfied. Thus, in the case where the L road marking L1 is unclear, or even in the case where it is not determined that the L road marking L1 is a white solid line, it is possible to determine whether it is a bidirectional road with good accuracy.

[0078] (5) The road determination portion 142 determines whether the road on which the host vehicle 101 is traveling is a bidirectional road based on information on the road markings on the outer sides of the road markings L1, L2 of the host lane LN1, such as the LL road marking and the RR road marking L5. Figure 6 Thus, it is possible to more accurately determine whether the road on which the host vehicle 101 is traveling is a bidirectional road.

[0079] (6) The road determination portion 142 determines whether the road marking condition is satisfied based on the road marking information, and determines that the road on which the host vehicle 101 is traveling is a bidirectional road when it is determined that the road marking condition is satisfied. On the other hand, after it is determined that it is a bidirectional road, when it is determined that the road marking condition is not satisfied for a predetermined time (first predetermined time) At2 or more, it is determined that the road on which the host vehicle 101 is traveling is not a bidirectional road. Figure 7 Thus, even in the case where the camera 1a erroneously detects the types of the road markings due to backlighting or the like, it is possible to accurately perform determination of a bidirectional road.

[0080] (7) The road determination portion 142 determines that the road on which the host vehicle 101 is traveling is a bidirectional road when it is determined that the road marking condition is satisfied for a predetermined time (second predetermined time) At1 that is shorter than the predetermined time At2 after it is determined that the road on which the host vehicle 101 is traveling is not a bidirectional road. Figure 7 Thus, it is possible to accurately and quickly determine that it is a bidirectional road.

[0081] The above-described embodiments can be modified in various ways. Several modifications will be described below. In the above-described embodiments, the road markings and the oncoming vehicle in the road being traveled on are detected by the camera 1a, but the road markings and the oncoming vehicle can also be detected by different detection units. In the above-described embodiments, the R road marking L2 being yellow (a prescribed first category) and the L road marking L1 being a white solid line (a prescribed second category) are included in the road marking condition, and it is determined whether the road being traveled on is a two-way road. However, in a two-way road, the categories of the pair of road markings L1, L2 on the left and right of the subject lane LN1 are sometimes different depending on the country. Therefore, it is also possible to determine whether the road being traveled on is a two-way road by taking into account the standards of each country for two-way roads. Thus, the manner in which the two-way road determination is made by the road determination unit based on the information on the categories of the pair of road markings is not limited to that described above.

[0082] In the above-described embodiments, an example of a left-side traffic road is shown, but the present application can also be applied to a right-side traffic road. Figure 8A 、 8B is an example of a right-side traffic non-two-way road, and is an example of a road in the United States. In Figure 8A , there are yellow road markings L11, L12 on the left side of the subject lane LN1. Thus, it is also possible for the road determination unit to determine whether the road being traveled on is a non-two-way road by including the L road markings and the LL road markings being yellow in the road marking condition. In Figure 8B , there are yellow road markings L11 on the left side of the subject lane LN1, and white dashed road markings L13 on the right side. In the above-described embodiments, it is determined whether the road is a two-way road with a single lane, but it is also possible to determine whether the road is a two-way road with multiple lanes in the same manner as described above.

[0083] In the above-described embodiments, the road determination unit 142 determines whether the road being traveled on is a two-way road based on the length (distance D) in the vehicle width direction of the outer side region of the subject lane LN1 and the information on the categories of the road markings on the outer sides of the pair of road markings L1, L2 on the left and right. However, the configuration of the road determination unit can be any form as long as it determines whether the road being traveled on is a two-way road that allows two-way traffic based on road marking information including information on the categories of the road markings identified by the road marking identification unit. In the above-described embodiments, the road determination unit 142 determines that the road being traveled on is not a two-way road when it is determined that the prescribed road condition does not hold for a prescribed time Δt2 or more after determining that it is a two-way road, but it is also possible to omit this condition of the prescribed time.

[0084] In the above-described embodiments, the example in which the road recognition device 50 is applied to the automated vehicle is described, but the present application can also be applied to a manual driving vehicle having a driving assistance function.

[0085] One or more of the above-described embodiments and modified examples can be combined arbitrarily, and each modified example can be combined with each other.

[0086] With the present application, it is possible to accurately determine whether a road on which a vehicle is traveling is a two-way road.

[0087] The present application has been described above in connection with preferred embodiments, but it will be understood by those skilled in the art that various modifications and changes can be made without departing from the scope of the disclosure of the following claims.

Claims

1. A road recognition device, characterized in that, have: Inspection unit (1a) inspects the road markings on the road in which the vehicle (101) is traveling; The road marking identification unit (141) identifies the category of road markings corresponding to the color and shape of the road markings detected by the detection unit (1a); and The road determination unit (142) determines whether the road in motion is a two-way road for two-way traffic based on road marking information, including information on the type of road markings identified by the road marking identification unit (141). The road determination unit (142) determines whether the prescribed road conditions are met based on the road marking information. When the prescribed road conditions are met, the road in motion is determined to be a two-way road. On the other hand, if the prescribed road conditions are not met for a specified period of time (Δt2) or more after the road is determined to be a two-way road, the road in motion is determined not to be a two-way road. The specified time (Δt2) is the first specified time. After determining that the road in motion is not a two-way road, the road determination unit (142) determines that the road in motion is a two-way road when the determination that the specified road conditions are continuously shorter than the first specified time (Δt2) for more than a second specified time (Δt1) is established.

2. The road recognition device according to claim 1, characterized in that, The detection unit (1a) is configured to also detect oncoming vehicles (102). The road determination unit (142) further determines whether the road in motion is a two-way road based on whether the detection unit (1a) has detected an oncoming vehicle (102).

3. The road recognition device according to claim 1 or 2, characterized in that, The road determination unit (142) determines whether the road in motion is a two-way road based on the information of the left and right pairs of road markings (L1, L2) of the lane (LN1) in which the vehicle (101) is traveling.

4. The road recognition device according to claim 3, characterized in that, The road determination unit (142) further determines whether the road in motion is a two-way road based on the length (D) of the outer region of the lane (LN1) in the vehicle width direction.

5. The road recognition device according to claim 4, characterized in that, The lane (LN1) is defined as having a pair of road markings on the left and right sides, consisting of a first road marking (L2) on the opposite side of the lane and a second road marking (L1) on the opposite side of the lane. The road determination unit (142). When the first road marking (L2) is classified as the first category and the second road marking (L1) is classified as the second category, regardless of the length (D) of the outer area of ​​the lane in the vehicle width direction, the road in motion is determined to be a two-way road. When the category of the first road marking (L2) is the first category as specified, and the category of the second road marking (L1) is not the second category as specified, the road in motion is determined to be a two-way road as long as the length (D) of the outer area of ​​the lane in the vehicle width direction is less than the specified value (D1).

6. The road recognition device according to claim 3, characterized in that, The road determination unit (142) also determines whether the road in motion is a two-way road based on the information of the type of the road markings on the outer side of the pair of left and right road markings (L1, L2).

7. The road recognition device according to claim 1 or 2, characterized in that, It also includes a driving control unit (16), which controls a steering actuator to enable automatic driving in the same lane without driver intervention. When the vehicle is driving automatically in the same lane, if the road determination unit (142) determines that the road is a two-way road, the driving control unit (16) controls the steering actuator to stop the automatic driving in the same lane.

8. The road recognition device according to claim 7, characterized in that, It also includes a notification unit (3a), which notifies the driver of information. When the vehicle is driving automatically in the same lane, if the road determination unit (142) determines that the road is a two-way road, the driving control unit (16) controls the notification unit (3a) to notify and remind the driver to hold the steering wheel.

Images