Vehicle capable of autonomous driving

Abstract

[Problem] To enhance the safety of the passage of a vehicle capable of autonomous driving through an intersection. [Solution] This vehicle capable of autonomous driving includes a sensor capable of detecting a space around the vehicle, a map acquisition device that acquires map data, and a travel control device capable of executing travel by autonomous driving of the vehicle. When the vehicle is traveling towards an intersection with an intersecting road or a merging road, the travel control device determines whether or not the field of view of the sensor for the intersecting road or the merging road in the map data is ensured.

Description

Autonomous vehicles

[0001] This application primarily discloses a vehicle capable of autonomous driving.

[0002] Patent Document 1 discloses determining whether or not there is an intersection with poor visibility in the direction of travel of the vehicle based on images captured by a camera. Patent Document 1 also discloses stopping repeatedly at predetermined speeds after moving a predetermined distance at an intersection with poor visibility. Patent Document 2 discloses determining the edge position of a building located near the intersection as the camera's field of view boundary, and determining the stopping position of the vehicle at a position that overlaps with the field of view boundary. Patent Document 3 discloses detecting obstacles that form a blind spot based on images captured by a camera while the vehicle is stopped at a stop line, and measuring the distance from the front end of the vehicle to the position in front of the obstacle. Patent Document 3 also discloses moving the vehicle forward at a low speed until the distance between the front end of the vehicle and the position in front of the obstacle is less than or equal to a predetermined distance, and then stopping. Thus, conventional vehicle driving control is based on the premise that a certain field of view is obtained by a camera, and that safety confirmation regarding entry into an intersecting road can be performed based on the detection by that camera.

[0003] Japanese Patent Publication No. 2019-067295, Japanese Patent Publication No. 2005-189936, Japanese Patent Publication No. 2019-095938

[0004] However, the viewing angle of a camera that images the traveling direction of a vehicle in a vehicle generally ranges from about 170 degrees on the left and right as a whole around the traveling direction. Even if the viewing angle of the camera is widened further, the distortion at the peripheral part of the captured image increases, and the accuracy of the recognition processing of other vehicles tends to decrease. For this reason, in the conventional vehicle running control, there is a certain limitation on the viewing field of the camera, and it is difficult to assert that high safety can be ensured for the running environment such as an intersection road when the host vehicle enters an intersection with an intersection road or the like. For example, when the host vehicle road on which the vehicle travels and the intersection road intersect from a state where they are substantially parallel, the camera of the vehicle cannot detect other vehicles on the intersection road coming from the substantially rear side of the host vehicle. Also, even when the intersection angle between the host vehicle road and the intersection road is not small, if there is a high shielding object between the host vehicle road and the intersection road or the like, the camera of the vehicle cannot detect other vehicles on the intersection road on the back side of the shielding object. These non-detections can occur even when the host vehicle is in a state of entering the intersection. That is, on the premise that the running safety can be confirmed by the viewing field of the camera provided in the vehicle as in Patent Documents 1 to 3, it is difficult to assert that high safety can be ensured for the running control for the host vehicle to enter the intersection road.

[0005] Thus, for a vehicle capable of autonomous driving, it is required to enhance the safety regarding passing through an intersection.

[0006] An autonomous driving vehicle according to an embodiment of the present invention includes a sensor capable of detecting the surrounding space of the vehicle including at least the road in the traveling direction of the vehicle, a map acquisition device that acquires map data including the road on which the vehicle travels, and a travel control device capable of executing travel by autonomous driving of the vehicle. The travel control device executes a viewing field determination process for determining whether or not the viewing field of the sensor regarding the intersection road or the merging road in the map data is ensured when traveling toward an intersection with an intersection road or a merging road.

[0007] In one embodiment of the present invention, the driving control device determines whether the sensor's field of view of the intersecting road or merging road is secured when the vehicle, which is the vehicle itself, is traveling toward an intersection with an intersecting road or merging road. In particular, the driving control device determines whether the sensor's field of view of the intersecting road or merging road is secured in the map data. As a result, the autonomously driven vehicle can drive in a way that allows it to safely enter the intersection while the sensor's field of view of the intersecting road or merging road is secured. The autonomously driven vehicle can ensure a high level of safety when it enters an intersection between its own road and an intersecting road or merging road.

[0008] Figure 1 is an explanatory diagram showing a vehicle traveling on its own road towards an intersection with a merging road. Figure 2 is an explanatory diagram showing the vehicle from Figure 1 traveling on its own road towards an intersection with an intersecting road. Figure 3 is an explanatory diagram showing the main configuration of an example of a control system for an autonomously drivable vehicle according to an embodiment of the present invention. Figure 4 is an explanatory diagram showing an example of the detection range and field of view of the LiDAR in Figure 3. Figure 5 is an explanatory diagram showing an example of the detection range and field of view of a multi-front camera that can be used as the camera in Figure 3. Figure 6 is an example of the main flowchart of the autonomous driving control by the driving control device in Figure 3. Figure 7 is an example of a flowchart showing an example of the detailed processing flow of the intersection entry control in Figure 6. Figure 8 is an explanatory diagram showing the first, second, and third stopping positions of the vehicle. Figure 9 is an explanatory diagram showing an example of a combination of the position of the road edge on the vehicle's side of a single-lane merging road and the positions of multiple detected road edge points. Figure 10 is an explanatory diagram showing the relationship between the road surface of the merging road in Figure 9 and the LiDAR point cloud data. Figure 11 is an explanatory diagram of predetermined conditions for determining the degree of agreement between the positions of multiple detected road endpoints and the positions of the road shoulders in high-precision map data. Figure 12 is an explanatory diagram of the first state in which the vehicle, the owner, has started to travel toward the intersection in Figure 9 under autonomous driving conditions. Figure 13 is an explanatory diagram of the second state in which the vehicle, the owner, is closer to the intersection than in Figure 12 under autonomous driving conditions. Figure 14 is an explanatory diagram of the third state in which the vehicle, the owner, is closer to the intersection than in Figure 13 under autonomous driving conditions.

[0009] Hereinafter, embodiments of the present invention will be described with reference to the drawings in the following order. In the embodiments, first, an overview of the embodiments will be described. Then, two specific examples of the vehicle's driving state, an example of the configuration and control of an autonomously drivable vehicle, a basic explanation of the method for confirming the sensor field of view of intersecting or merging roads, an example of multiple detection road endpoints, an example of predetermined conditions for determining the degree of matching, and an example of judgment according to driving will be described in order.

[0010] (Overview) Vehicles operating autonomously may travel towards intersections with intersecting or merging roads. In this case, the vehicle's driving control system confirms safety based on detections from sensors installed on the vehicle and executes driving control to enter the intersection. However, sensors installed on the vehicle have limitations in their detection range. Moreover, if there are obstacles such as guardrails, buildings, or walls on either side of the road the vehicle is traveling on, the sensors installed on the vehicle can only detect the surrounding driving environment in a narrower range than the detection range that the sensors can detect due to their field of view. In this case, even if safety checks are performed based on detections from the sensors installed on the vehicle, it is difficult to ensure a high level of safety for the vehicle to enter the intersection. Therefore, the driving control system of this embodiment determines whether the sensor's field of view for the intersecting or merging road in the map data is secured when the vehicle is traveling towards an intersection with an intersecting or merging road. In this field of view determination process, the driving control system extracts, for example, multiple detected road endpoints on the vehicle side for the intersecting or merging road that are included in the sensor's detection range from the sensor's detection information. These multiple detected road endpoints on the vehicle's side include detected road endpoints located away from the intersection in the direction of the intersecting or merging road. Here, the driving control device may extract locations on the road surface of the intersecting or merging road included in the sensor's detection information where a step of height greater than or equal to a predetermined value is detected, as the locations of the detected road endpoints on the vehicle's side. Next, the driving control device maps the extracted multiple detected road endpoints to the road surface of the intersecting or merging road based on map data, at the locations detected by the sensor. Next, the driving control device determines the degree of agreement between the location of the road edge on the vehicle's side of the intersecting or merging road in the map data and the locations of the multiple detected road endpoints based on the sensor's detection. If a predetermined degree of agreement is obtained, the driving control device determines that the sensor's field of view for the intersecting or merging road is secured. If the sensor's field of view is secured after performing this sensor field of view confirmation process, the driving control device further performs a safety confirmation process for entering the intersection based on the sensor's detection.Furthermore, the driving control system executes driving control to enter the intersection by autonomous driving when it determines that safety can be confirmed through safety confirmation processing based on sensor detection. As a result, a vehicle driving by autonomous driving can enter an intersection by autonomous driving when the driving environment ensures that the sensors installed on the vehicle have a field of view of the intersecting or merging road. A vehicle driving by autonomous driving can drive in a way that allows it to enter the intersection safely after safety confirmation has been made under the appropriate field of view of the sensors. The driving control system may switch the vehicle's operation from autonomous driving to manual driving if it cannot confirm that the field of view of the sensors is ensured through field of view confirmation processing, or if it does not determine that safety can be confirmed through safety confirmation processing based on sensor detection.

[0011] (First specific example of driving conditions) Figure 1 is an explanatory diagram of a state in which vehicle 1 is traveling on its own road 31 toward the intersection with the merging road 32. In Figure 1, the vehicle 1, which is capable of autonomous driving, is traveling on its own road 31 toward the intersection with the merging road 32. Here, the merging road 32 may be a road to which the vehicle's own road 31 is connected, or a road connected to the vehicle's own road 31. Also, there is no traffic light at the intersection of the vehicle's own road 31 and the merging road 32. A stop line is drawn on the road surface of the vehicle's own road 31 at position P1 before the intersection with the merging road 32. Vehicle 1 is required by law and other regulations to stop at the position P1 of the stop line.

[0012] Incidentally, if vehicle 1, which is stopped at stop line position P1, is being operated manually, the driver of vehicle 1 visually checks the driving environment of merging road 32 before entering the intersection with merging road 32. If the safety of merging road 32 can be confirmed, the driver drives vehicle 1 from stop line position P1 and enters the intersection with merging road 32. In this case, if the safety of merging road 32 cannot be sufficiently visually confirmed due to obstructions such as guardrails 34, the driver may advance vehicle 1 to the vicinity of boundary line position P2 between merging road 32 and the vehicle's own road 31, stop vehicle 1 again at boundary line position P2, and visually confirm the safety of merging road 32 once more. Through this two-stage stopping and checking, the driver can more reliably confirm that the driving environment of merging road 32 is safe for the vehicle to enter the intersection.

[0013] (Second specific example of driving conditions) Figure 2 is an explanatory diagram of the state in which vehicle 1 from Figure 1 is traveling on its own road 41 toward the intersection with intersecting road 42. In Figure 2, the vehicle 1, which is capable of autonomous driving, is traveling on its own road 41 toward the intersection with intersecting road 42. Here, intersecting road 42 is simply the road that intersects with the vehicle's own road 41. In Figure 2, the vehicle's own road 41 and the intersecting road 42 intersect in a cross shape. There are no traffic lights at the intersection of the vehicle's own road 41 and the intersecting road 42. A stop line is drawn on the surface of the vehicle's own road 41 at position P1, just before the intersection with the intersecting road 42. Vehicle 1 is required by law and other regulations to stop at the position P1 of the stop line.

[0014] Incidentally, if vehicle 1, which is stopped at stop line position P1, is being operated manually, the driver of vehicle 1 visually checks the driving environment of intersecting road 42 before entering the intersection with intersecting road 42. If the safety of intersecting road 42 can be confirmed, the driver drives vehicle 1 from stop line position P1 and enters the intersection with intersecting road 42. In this case, if the safety of intersecting road 42 cannot be sufficiently visually confirmed due to the obstruction shown in Figure 2, the driver may advance vehicle 1 to the vicinity of the boundary line position P2 between intersecting road 42 and the vehicle's own road 41, stop vehicle 1 again at boundary line position P2, and visually confirm the safety of intersecting road 42 once more. By performing a two-stage stop, the driver can more reliably confirm that the driving environment of intersecting road 42 is safe for the vehicle to enter the intersection.

[0015] In this embodiment, the terms "merging road 32" and "intersecting road 42" are used with the following definitions: When the vehicle road 41 continues on the opposite side of the road that intersects it, that road is called an intersecting road 42. When the vehicle road 41 does not continue on the opposite side of the road that intersects it, that is, when the vehicle road 41 terminates at the road that intersects it, that road is called a merging road 32.

[0016] (Example of Configuration of an Autonomous Driving Vehicle) First, an example of the configuration of an autonomous driving vehicle 1 will be described. The autonomous driving vehicle 1 described here is a vehicle 1 that can switch between autonomous driving and manual driving. Figure 3 is an explanatory diagram of the main configuration of an example of the control system 10 of an autonomous driving vehicle 1 according to an embodiment of the present invention. The control system 10 of the autonomous driving vehicle 1 in Figure 3 includes a LiDAR (Light Detection and Ranging) 12, a camera 13, a map storage device 15, a communication device 16, and a driving control device 11 to which these are connected, as sensors provided on the vehicle 1. In addition, a GNSS (Global Navigation Satellite System) receiver 14, a drive device 17, a braking device 18, a steering device 19, and an operating device 20 are further connected to the driving control device 11. The driving control device 11 and other devices of the control system 10 may be connected by a vehicle network. Vehicle networks include, for example, CAN (Controller Area Network) and LIN (Local Interconnect Network). Alternatively, general communication networks such as IEEE (Institute of Electrical and Electronics Engineers) 802.3 may be used for the vehicle network. Each device shown in Figure 2 may be connected to the vehicle network individually, or it may be divided into multiple devices and connected to the vehicle network. Furthermore, multiple devices shown in Figure 2 may be integrated into a single device and connected to the vehicle network.

[0017] The LiDAR 12 is installed facing forward at the front of the vehicle body of the vehicle 1, for example, as shown in Figure 1. The LiDAR 12 scans the front of the vehicle 1 with a laser and receives reflected light. Based on the distribution of reflected light and the timing of light reception, the LiDAR 12 generates point cloud data as information about the space in front of the vehicle 1, including the direction of travel. The LiDAR 12 is a sensor capable of detecting the surrounding space of the vehicle 1, including at least the road in the direction of travel of the vehicle 1.

[0018] Camera 13 is mounted facing forward, for example, at the roof height of vehicle 1. Camera 13 captures images of the space in front of vehicle 1, including the direction of travel. Camera 13 is a sensor capable of detecting the surrounding space of vehicle 1, including at least the road in the direction of travel, using the captured images. Camera 13 captures images of the detection range with a predetermined field of view.

[0019] The GNSS receiver 14 receives radio waves from a GNSS satellite and generates information about the location of the vehicle 1 on which the GNSS receiver 14 is installed.

[0020] The map storage device 15 stores high-precision map data 21. The map storage device 15 can acquire high-precision map data 21, including the roads on which the vehicle 1 travels. The map storage device 15 may be provided as part of the driving control device 11.

[0021] The high-precision map data 21 includes information about the shape and width of each lane, such as multiple lanes, shoulders, and sidewalks on a road. The direction of each lane is recorded by the coordinate information of multiple nodes and the link information connecting the multiple nodes. In the high-precision map data 21, intersections are represented by nodes that are commonly included in the link information of multiple roads. Each node has position information in a coordinate system. The node information and link information may include information such as stopping positions based on regulations before the intersection, the number of lanes at the intersection, and the width of each lane. The node and link information at the intersection may be provided for each lane. The number of lanes at the intersection may include lanes passing through the intersection and dedicated lanes for turning right or left at the intersection. The link information may include information about the width of the lane. In this case, the positions of the road edges on both sides of the intersecting road 42 or merging road 32 can be calculated from the number of lanes constituting the road and the width of each lane. The area between the two road edges becomes the road surface of the intersecting road 42 or merging road 32. Here, the road edge on the side of the vehicle refers to the road edge on the side of the intersecting road 42 or merging road 32 where the vehicle is located.

[0022] The communication device 16 establishes a wireless communication path with an external base station or the like, and receives information on high-precision map data 21 and information to be added to the high-precision map data 21 from an external map server or the like. The communication device 16 can acquire high-precision map data 21 that includes the roads on which the vehicle 1 travels.

[0023] The drive unit 17 includes an internal combustion engine, a motor, a battery, a drive transmission mechanism, etc. The drive unit 17 rotates the wheels of the vehicle 1 using the driving force generated by the internal combustion engine and the motor.

[0024] The braking system 18 includes brakes, a regenerative braking mechanism, and the like. The braking system 18 slows down or stops the rotation of the wheels of the vehicle 1 by braking or regeneration.

[0025] The steering device 19 has a steering mechanism. The steering device 19 changes the direction of the wheels of the vehicle 1 by the steering mechanism.

[0026] The operating device 20 includes a steering wheel, shift lever, accelerator pedal, brake pedal, and the like for the driver to operate the vehicle.

[0027] The driving control device 11 controls the movement of the vehicle 1. The driving control device 11 can switch between manual driving and automatic driving. When controlling the movement of the vehicle 1 in manual driving mode, the driving control device 11 outputs control values ​​corresponding to the driver's operation on the operating device 20 to the drive unit 17, braking unit 18, and steering unit 19. In this case, the driving control device 11 may generate and output control values ​​to assist the driver's operation. When controlling the movement of the vehicle 1 in automatic driving mode, the driving control device 11 acquires the vehicle's position information from the GNSS receiver 14 and acquires high-precision map data 21 of the area around the vehicle from the map storage device 15 or communication device 16. The driving control device 11 also outputs control values ​​to the drive unit 17, braking unit 18, and steering unit 19 for moving along the roads included in the high-precision map data 21 of the area around the vehicle according to the route to the destination in automatic driving mode. Furthermore, the driving control device 11 acquires detection information from the vehicle's sensors, such as LiDAR 12 and camera 13, and controls the vehicle's movement to avoid interference with surrounding vehicles. Through this automated driving, the vehicle 1 can, for example, enter an intersection from its own road 31 without interfering with other vehicles in Figure 1, turn at the intersection, and travel on the merging road 32. Alternatively, the vehicle 1 can enter an intersection from its own road 41 without interfering with other vehicles 45 in Figure 2, and travel straight or turn at the intersection to travel on one of the lanes of the intersecting road 42. The intersecting road 42 in Figure 2 has two lanes, 43 and 44. The driving control device 11 is a control device installed in the vehicle 1. As shown in Figure 3, the vehicle 1 is equipped with multiple devices. In this case, the driving control device 11 may be configured as a single device as shown in Figure 3, or it may be composed of multiple devices that can cooperate.

[0028] (Detection limits of sensors installed on the vehicle) Figure 4 is an explanatory diagram illustrating an example of the detection range and field of view of the LiDAR 12 in Figure 3.

[0029] LiDAR 12 can scan the area in front of the extension of the front edge of vehicle 1, shown by the dashed line in Figure 4, and detect roads and other vehicles that are within the detectable range in front of vehicle 1. In this case, the field of view of LiDAR 12 is approximately 180 degrees. The field of view of LiDAR 12 can be adjusted by changing the structure of LiDAR 12, etc. Here, the field of view of LiDAR 12 is simply the angle of view when the horizontal width of the detection range of LiDAR 12 is viewed from the viewpoint of LiDAR 12. In Figure 1, if there is no guardrail 34, the road 31 has good visibility. In this case, LiDAR 12 can detect roads and other vehicles that are within the detection range in front of the extension of the front edge of vehicle 1, shown by the dashed line in Figure 1, using the detection range in Figure 4. In contrast, the road 41 in Figure 2 is an environment with poor visibility, surrounded on the left and right by obstructions 46 and 47. In this case, LiDAR 12 can detect roads and other vehicles that are located within a detection range between the left and right obstructions 46 and 47, which is narrower than the detection range shown in Figure 4.

[0030] Figure 5 is an explanatory diagram illustrating an example of the detection range and field of view of a multi-front camera that can be used as camera 13 in Figure 3.

[0031] The multi-front camera in Figure 5 has a camera 13 positioned to the left relative to the front direction of the vehicle 1, and a camera 13 positioned to the right relative to the forward direction. By having multiple cameras 13 with different orientations in this way, the multi-front camera in Figure 5 can obtain a field of view of 180 degrees or more. In contrast, the field of view of a monocular camera used in a vehicle 1 to image the direction of travel of the vehicle 1 is generally about 170 degrees in total to the left and right, centered on the direction of travel. The multi-front camera can achieve a wider field of view than a monocular camera. Here, the field of view of camera 13 can be defined as the angle of view when the horizontal width of the detection range of camera 13 is viewed from the viewpoint position of camera 13. In the example in Figure 1, in an environment with good visibility where there is no guardrail 34, the multi-front camera in Figure 5 and the camera 13 in Figure 3 can detect roads and other vehicles that are present in the detection range corresponding to the field of view of each camera, according to the detection range in Figure 5. In contrast, in an environment with poor visibility where the left and right sides are surrounded by obstructions 46 and 47, as in the example in Figure 2, the multi-front camera in Figure 5 and the camera 13 in Figure 3 can detect roads and other vehicles that are located within a detection range between the left and right obstructions, which is narrower than the detection range in Figure 5.

[0032] Incidentally, in a vehicle 1 equipped with sensors such as LiDAR 12 and camera 13, the driving control device 11 will perform driving control for the vehicle to enter an intersection with an intersecting road 42 or merging road 32, as shown in the examples in Figures 1 and 2, based on the detection of the sensors. In this case, it is desirable that the driving control device 11 confirms the safety of the intersecting road 42 or merging road 32 based on the sensor detection results and controls the vehicle's driving so that it enters the intersection only when the safety of the intersecting road 42 or merging road 32 is ensured.

[0033] However, as illustrated in Figures 4 and 5, sensors such as the LiDAR 12 and camera 13 installed on the vehicle 1 can generally only detect within a certain range. Therefore, in the driving environment shown in Figure 1, the sensors on the vehicle 1 cannot detect other vehicles 33 that are behind the vehicle, which are outside the sensor's detection range. In particular, as shown in Figure 1, when the road 41 on which the vehicle 1 is traveling and the merging road 32 intersect from a nearly parallel state, the sensors on the vehicle 1 cannot detect other vehicles 33 on the merging road 32 approaching from approximately the rear of the vehicle. Furthermore, in the driving environment shown in Figure 2, the sensors on the vehicle 1 cannot detect other vehicles 45 that are behind the obstruction 47. Even when the intersection angle between the road 41 and the intersecting road 42 is not small, as shown in Figure 2, if there is a tall obstruction between the road 41 and the intersecting road 42, the sensors on the vehicle 1 cannot detect other vehicles 45 on the intersecting road 42 that are behind the obstruction. These failures to detect can also occur when the vehicle is entering the intersection. As a result, the driving control device 11 cannot determine the possibility of interference with other vehicles that have not been detected by sensors installed on the vehicle 1, and therefore cannot confirm that safety is ensured in the driving environment of the intersecting road 42 or merging road 32.

[0034] In other words, even if the sensors installed on vehicle 1 can secure a wide field of view, or even if safety checks can be performed regarding entry into intersecting roads 42 and merging roads 32 based on sensor detection, it does not necessarily mean that a high level of safety can be ensured when vehicle 1 enters the intersection. Thus, the driving control of autonomously driven vehicle 1 requires enhanced safety when passing through intersections.

[0035] (Example of control in a vehicle) Next, an example of control in vehicle 1 will be described. Figure 6 is an example of the main flowchart of automatic driving control by the driving control device 11 in Figure 3. The driving control device 11 repeatedly executes the main flowchart in Figure 6. Note that Figure 6 shows only the part of the main flowchart that pertains to the case where vehicle 1, which is the vehicle itself, is traveling towards an intersection.

[0036] In step ST1, the driving control device 11 acquires the position information of its own vehicle, vehicle 1. The driving control device 11 may acquire the latest position information of its own vehicle from the GNSS receiver 14.

[0037] In step ST2, the driving control device 11 identifies the vehicle's position on the high-precision map data 21. The driving control device 11 acquires the high-precision map data 21 surrounding the vehicle's latest position from the map storage device 15 or communication device 16, displays it on the road surface, and maps the vehicle's latest position onto the displayed road surface. This allows the driving control device 11 to confirm the current position of the vehicle and the surrounding static driving environment, including the roads.

[0038] In step ST3, the driving control device 11 determines whether the vehicle is traveling on the road 41 towards the intersection. If the vehicle's driving environment is such that it is traveling towards the intersection as shown in Figures 1 and 2, the driving control device 11 determines that the vehicle is traveling on the road 41 towards the intersection. In this case, the driving control device 11 proceeds to step ST3. Otherwise, the driving control device 11 determines that the vehicle is not traveling on the road 41 towards the intersection and terminates this control. At this time, the driving control device 11 may perform driving control (not shown), for example, driving control to maintain the current driving position and continue traveling on the road 41.

[0039] In step ST4, the driving control device 11 performs intersection entry control. When the vehicle is traveling towards an intersection, the driving control device 11 needs to confirm the safety of the intersecting road 42 and merging road 32 that intersect with the vehicle's road 41 at the intersection. The driving control device 11 also needs to confirm that the driving environment of the intersecting road 42 and merging road 32 can be confirmed to be safe by sensors installed on the vehicle. Then, if these confirmations have been made, the driving control device 11 needs to control the vehicle's movement to enter the intersection. In the intersection entry control of step ST4, the driving control device 11 performs these processes as shown in Figure 8, which will be described later. After that, the driving control device 11 terminates this control.

[0040] Figure 7 is an example flowchart showing a detailed processing flow of the intersection entry control in Figure 6. In Figure 7, the driving control device 11 repeatedly performs setting processes, including field of view judgment processing and safety confirmation processing, at each stage of approaching the intersection. If it is determined at any stage that a field of view is secured and the safety of the intersecting road 42 or merging road 32 is confirmed, the driving control device 11 executes driving control to allow the vehicle to enter the intersection by automatic driving in step ST27. Here, the field of view judgment processing is the process of determining whether or not the sensor's field of view of the intersecting road 42 or merging road 32 is secured. The safety confirmation processing is the process of confirming the safety of the vehicle's entry into the intersection of the vehicle's road 41 and the intersecting road 42 or merging road 32, based on the sensor's detection. Note that here we will explain the case where there is no preceding vehicle in front of the vehicle. If there is a preceding vehicle in front of the vehicle, the driving control device 11 controls the vehicle's driving so that the vehicle and the preceding vehicle do not interfere with each other.

[0041] In step ST11, the driving control device 11 performs automatic driving control to stop the vehicle, which is traveling on the road 41 toward the intersection, at the first stopping position. Here, the first stopping position may be the position P1 of the stop line in Figures 1 and 2. The driving control device 11 may obtain information on the position P1 of the stop line from high-precision map data 21 or based on images captured by the camera 13. The high-precision map data 21 may include, for example, information on the first stopping position based on regulations for the road 41 connected to the intersection, as link and node information.

[0042] In step ST12, the driving control device 11 determines whether the sensor's field of view for the intersecting road 42 or merging road 32 is secured while the vehicle is stopped at the first stopping position. The sensor may be, for example, a LiDAR 12. Here, the driving control device 11 determines whether the sensor can detect the road edge on the vehicle's side of the intersecting road 42 or merging road 32 in the high-precision map data 21 while the vehicle is stopped at the first stopping position. The reference position of the road edge on the vehicle's side of the intersecting road 42 or merging road 32 can be obtained by calculation from the information on the number of road lanes and lane width included in the high-precision map data 21. The method and specific examples of determining whether the sensor's field of view for the intersecting road 42 or merging road 32 is secured based on the detection of the road edge on the vehicle's side of the intersecting road 42 or merging road 32 will be explained in detail in Figures 9 to 14. If the road edge on the vehicle's side of the intersecting road 42 or merging road 32 is detected and the sensor's field of view is secured, the driving control device 11 proceeds to step ST13. The driving control device 11 proceeds to step ST13 if the sensor's field of view is secured while the vehicle is stopped at the first stopping position. However, if the sensor's field of view of the intersecting road 42 or merging road 32 is not secured, the driving control device 11 proceeds to step ST14. The driving control device 11 proceeds to step ST14 in order to proceed to the next stage if the sensor's field of view is not secured while the vehicle is stopped at the first stopping position.

[0043] In step ST13, the driving control device 11 determines whether it is safe for the vehicle to travel when it starts moving from a state where it is stopped at the first stopping position and enters the intersection of its own road 41 and the intersecting road 42 or merging road 32. Here, the driving control device 11 determines the possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32 when the vehicle, which is stopped at the first stopping position, enters the intersection. Other vehicles traveling on the intersecting road 42 or merging road 32 are detected by sensors that have a field of view. The driving control device 11 calculates, for example, the timing at which the other vehicles detected by the sensors arrive at the intersection and the timing at which the vehicle arrives at the intersection, based on the distance, speed, etc., of each vehicle 1. Then, if the timing at which the vehicle is at the intersection and the timing at which the other vehicles are at the intersection overlap, the driving control device 11 may determine that there is a possibility of interference between the vehicle and the other vehicles at the intersection. Conversely, if there is a significant gap between the timing of the vehicle's presence at the intersection and the timing of other vehicles' presence at the intersection, the driving control device 11 may determine that there is no possibility of interference between the vehicle and other vehicles at the intersection. Specifically, to determine the possibility of interference between the vehicle and other vehicles, the driving control device 11 may use an ST chart that maps the vehicle and other vehicles. An ST chart is a two-axis chart in which one axis represents the position on the road and the other axis represents the elapsed time from the present moment. The ST chart can map the current position of the vehicle and other vehicles, as well as their future positional changes according to their respective speeds. By using the ST chart to determine the possibility of interference, the driving control device 11 can determine the possibility of interference between the vehicle and other vehicles more accurately. If there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is safe for the vehicle to travel and proceeds to step ST27 to execute driving control to enter the intersection. The driving control device 11 proceeds to step ST27 if it can confirm that the sensor's field of view is secured while the vehicle is stopped at the first stopping position, and that there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32.On the other hand, when there is a possibility of interference with other vehicles traveling on the intersecting road 42 or the merging road 32, the driving control device 11 determines that the driving of the host vehicle is not safe, and advances the process to step ST14 in order to proceed to the next stage. Here, when there are a plurality of other vehicles traveling on the intersecting road 42 or the merging road 32, the driving control device 11 determines interference with all of the plurality of other vehicles, and when there is no possibility of interference with all of the plurality of other vehicles, the process may proceed to step ST27. On the other hand, when there is a possibility of interference with even one of the plurality of other vehicles, the driving control device 11 may advance the process to step ST14.

[0044] In step ST14, the driving control device 11 executes driving control of automatic driving to slowly drive the vehicle 1 stopped at the first stop position on the host vehicle road 41 to the second stop line on the host vehicle road 41. Here, the second stop position may be the boundary line position P2 between the intersecting road 42 or the merging road 32 in FIGS. 1 and 2 and the host vehicle road 41. The driving control device 11 may obtain the information of the boundary line position P2 by calculating based on the high-precision map data 21 or the like, or by extracting it from the captured image of the camera 13. The high-precision map data 21 includes information on links and nodes, and width information of a traffic lane such as a lane for calculating the road width of the intersecting road 42 or the merging road 32. The driving control device 11 can calculate the closest boundary line position P2 that does not intrude into the intersecting road 42 or the merging road 32 by applying the road width of the intersecting road 42 or the merging road 32 to the connection state between the host vehicle road 41 and the intersecting road 42 or the merging road 32. The driving control device 11 starts driving from the stopped first stop position and starts driving control of automatic driving by slow travel to stop at the second stop line.

[0045] In step ST15, the driving control device 11 determines whether the sensor's field of view for the intersecting road 42 or merging road 32 is secured while the vehicle is moving from the first stopping position to the second stopping position. The driving control device 11 determines whether the sensor can detect the vehicle's side of the road edge of the intersecting road 42 or merging road 32 in the high-precision map data 21 while the vehicle is moving from the first stopping position to the second stopping position. If the vehicle's side of the road edge for the intersecting road 42 or merging road 32 is detected and the sensor's field of view is secured, the driving control device 11 proceeds to step ST16. The driving control device 11 proceeds to step ST16 if the sensor's field of view is secured while the vehicle is moving from the first stopping position to the second stopping position. On the other hand, if the sensor's field of view for the intersecting road 42 or merging road 32 is not secured, the driving control device 11 proceeds to step ST17. If the driving control device 11 determines that the sensor's field of view is not secured while the vehicle is moving from the first stopping position to the second stopping position, it proceeds to step ST17 in order to move to the next stage of processing.

[0046] In step ST16, the driving control device 11 determines whether it is safe for the vehicle to travel if it is moving from the first stopping position to the second stopping position and then entering the intersection of its own road 41 and the intersecting road 42 or merging road 32. The driving control device 11 determines the possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32 if the vehicle enters the intersection with the current driving control. Other vehicles traveling on the intersecting road 42 or merging road 32 are detected by sensors that have a field of view. If there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is safe for the vehicle to travel and proceeds to step ST27 to execute driving control to enter the intersection while continuing to travel with the current driving control. The driving control device 11 proceeds to step ST27 if it can confirm that the sensor's field of view is secured during movement from the first stopping position to the second stopping position, and that there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32. On the other hand, if there is a possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is not safe for its own vehicle to travel and proceeds to step ST17 in order to proceed to the next stage. Here, the judgment conditions when there are multiple other vehicles traveling on the intersecting road 42 or merging road 32 may be the same as in step ST13.

[0047] In step ST17, the driving control device 11 performs automatic driving control to stop the vehicle, which is traveling along the road 41 toward the second stopping position, at the second stopping position.

[0048] In step ST18, the travel control device 11 determines whether the field of view of the sensor for the intersecting road 42 or the merging road 32 is secured while the host vehicle is stopped at the second stop position. The travel control device 11 determines whether the sensor can detect the road edge on the host vehicle side of the intersecting road 42 or the merging road 32 in the high-precision map data 21 while stopped at the second stop position. When the road edge on the host vehicle side of the intersecting road 42 or the merging road 32 is detected and the field of view of the sensor is secured, the travel control device 11 advances the process to step ST19. The travel control device 11 advances the process to step ST19 when the field of view of the sensor is secured while stopped at the second stop position. On the other hand, when the field of view of the sensor for the intersecting road 42 or the merging road 32 is not secured, the travel control device 11 advances the process to step ST20. The travel control device 11 advances the process to step ST20 to proceed to the next stage when the field of view of the sensor is not secured while stopped at the second stop position.

[0049] In step ST19, the driving control device 11 determines whether it is safe for the vehicle to travel when it starts moving from a state where it is stopped at the second stopping position and enters the intersection of its own road 41 and the intersecting road 42 or merging road 32. The driving control device 11 determines the possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32 when the vehicle, stopped at the second stopping position, enters the intersection. Other vehicles traveling on the intersecting road 42 or merging road 32 are detected by sensors whose field of view is ensured. If there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is safe for the vehicle to travel and proceeds to step ST27 to execute driving control to enter the intersection. The driving control device 11 proceeds to step ST27 when it can confirm that the field of view of the sensors is ensured while the vehicle is stopped at the second stopping position and that there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32. In contrast, if there is a possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is not safe for its own vehicle to travel and proceeds to step ST20 in order to move to the next stage of processing. Here, the judgment conditions when there are multiple other vehicles traveling on the intersecting road 42 or merging road 32 may be the same as in step ST13.

[0050] In step ST20, the driving control device 11 performs automatic driving control to move the vehicle 1, which is stopped at the second stopping position on the road 41, slowly to a position where it will move out onto the intersecting road 42 or merging road 32. Here, the position where the vehicle will move out may be a position that is slightly ahead of the boundary line position P2 in Figures 1 and 2 towards the intersection. The driving control device 11 may obtain the position where the vehicle will move out by calculating it based on high-precision map data 21 or by calculating it from images captured by the camera 13. The driving control device 11 starts driving from the second stopping position where the vehicle is stopped and starts automatic driving control by moving slowly to stop at the position where the vehicle will move out.

[0051] In step ST21, the driving control device 11 determines whether the sensor's field of view for the intersecting road 42 or merging road 32 is secured while the vehicle is moving from the second stopping position to the front-end position. The driving control device 11 determines whether the sensor can detect the vehicle's side of the road edge of the intersecting road 42 or merging road 32 in the high-precision map data 21 while the vehicle is moving from the second stopping position to the front-end position. If the vehicle's side of the road edge for the intersecting road 42 or merging road 32 is detected and the sensor's field of view is secured, the driving control device 11 proceeds to step ST22. The driving control device 11 proceeds to step ST22 if the sensor's field of view is secured while the vehicle is moving from the second stopping position to the front-end position. On the other hand, if the sensor's field of view for the intersecting road 42 or merging road 32 is not secured, the driving control device 11 proceeds to step ST23. If the driving control device 11 determines that the sensor's field of view is not secured while the vehicle is moving from the second stopping position to the front-end position, it proceeds to step ST23 in order to move to the next stage of processing.

[0052] In step ST22, the driving control device 11 determines whether it is safe for the vehicle to proceed if it is moving from the second stopping position to the front-end position and then entering the intersection of its own road 41 and the intersecting road 42 or merging road 32. The driving control device 11 determines the possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32 if the vehicle enters the intersection with the current driving control. Other vehicles traveling on the intersecting road 42 or merging road 32 are detected by sensors that have a field of view. If there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is safe for the vehicle to proceed and proceeds to step ST27 to execute driving control to enter the intersection while continuing to drive with the current driving control. The driving control device 11 proceeds to step ST27 if it can confirm that the sensor's field of view is secured while moving from the second stopping position to the front-end position, and that there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32. On the other hand, if there is a possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is not safe for its own vehicle to travel and proceeds to step ST23 in order to proceed to the next stage. Here, the judgment conditions when there are multiple other vehicles traveling on the intersecting road 42 or merging road 32 may be the same as in step ST13.

[0053] In step ST23, the driving control device 11 performs automatic driving control to stop the vehicle, which is traveling toward the front-end position, at the front-end position.

[0054] In step ST24, the driving control device 11 determines whether the sensor's field of view for the intersecting road 42 or merging road 32 is secured while the vehicle is stopped at the front-end position. The driving control device 11 determines whether the sensor can detect the vehicle's side of the road edge of the intersecting road 42 or merging road 32 in the high-precision map data 21 while the vehicle is stopped at the front-end position. If the vehicle's side of the road edge for the intersecting road 42 or merging road 32 is detected and the sensor's field of view is secured, the driving control device 11 proceeds to step ST25. The driving control device 11 proceeds to step ST25 if the sensor's field of view is secured while the vehicle is stopped at the front-end position. On the other hand, if the sensor's field of view for the intersecting road 42 or merging road 32 is not secured, the driving control device 11 proceeds to step ST26. The driving control device 11 proceeds to step ST26 in order to switch the vehicle's driving from automatic driving to manual driving when the sensor's field of view is not secured while the vehicle is stopped at the front-end position.

[0055] In step ST25, the driving control device 11 determines whether it is safe for the vehicle to drive when it starts moving from a state where it is stopped in the forward position and enters the intersection of its own road 41 and the intersecting road 42 or merging road 32. The driving control device 11 determines the possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32 when the vehicle, stopped in the forward position, enters the intersection. Other vehicles traveling on the intersecting road 42 or merging road 32 are detected by sensors whose field of view is ensured. If there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is safe for the vehicle to drive and proceeds to step ST27 to execute driving control to enter the intersection. The driving control device 11 proceeds to step ST27 when it can confirm that the field of view of the sensors is ensured while the vehicle is stopped in the forward position and that there is no possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32. In contrast, if there is a possibility of interference with other vehicles traveling on the intersecting road 42 or merging road 32, the driving control device 11 determines that it is not safe for the vehicle to travel and proceeds to step ST26 in order to switch the vehicle's driving from automatic driving to manual driving. Here, the determination conditions when there are multiple other vehicles traveling on the intersecting road 42 or merging road 32 may be the same as in step ST13.

[0056] Step ST26 is a process that is executed when it cannot be determined that it is possible to safely enter and drive into the intersection using autonomous driving. The driving control device 11 executes this process when the field of view of the sensors for the intersecting road 42 or merging road 32 is not secured, or when it cannot be determined that the driving environment of the intersecting road 42 or merging road 32 is safe. The driving control device 11 executes a process to switch from autonomous driving to manual driving in order to hand over the driving of the vehicle from autonomous driving to manual driving when the vehicle is stopped in the position where the front of the vehicle is sticking out. The process of switching from autonomous driving to manual driving includes, for example, disabling the autonomous driving of the vehicle and notifying the driver of the vehicle of the switch. The driving control device 11, for example, disabling the autonomous driving mode of the vehicle and setting it to manual driving mode. The driving control device 11 requests the driver inside the vehicle to take over the driving. When the driver starts operating the control device 20, the driving control device 11 outputs control values ​​to the drive unit 17, brake unit 18, and steering unit 19 based on the driver's operation. As a result, vehicle 1 can enter the intersection according to the driver's input and continue driving manually. Subsequently, the driving control device 11 terminates this control and the main flowchart in Figure 6. This cancels the automatic driving for vehicle 1.

[0057] Step ST27 is a process that is executed when it is possible to safely enter and drive into the intersection using autonomous driving. The driving control device 11 executes this process when the field of view of the sensors for the intersecting road 42 or merging road 32 is secured, and the driving environment of the intersecting road 42 or merging road 32 is determined to be safe based on the detection of the sensors whose field of view is secured. The driving control device 11 executes driving control to enter and drive into the intersection of the intersecting road 42 or merging road 32 using autonomous driving, from the current driving state or stopped state of the vehicle. For example, when crossing the intersecting road 42 in Figure 2 and continuing to drive on the vehicle's own road 41, the driving control device 11 executes driving control to pass through the intersection using autonomous driving and drive towards the vehicle's own road 41 beyond it. When turning right or left onto the intersecting road 42 in Figure 2 and driving on the intersecting road 42, the driving control device 11 executes driving control to enter the intersection using autonomous driving and then turn right or left onto the intersecting road 42. When the vehicle turns left onto the merging road 32 in Figure 1 and travels along the merging road 32, the driving control device 11 executes driving control to enter the intersection in automatic driving mode and then turn left onto the merging road 32. Based on the route taken to the destination in automatic driving mode, the driving control device 11 obtains the vehicle's immediate route and outputs control values ​​to the drive unit 17, braking unit 18, and steering unit 19. As a result, the vehicle 1 can enter the intersection from its own road 41 in automatic driving mode, pass through the intersection, and continue driving toward its destination. After that, the driving control device 11 terminates this control and returns the process to the main flowchart in Figure 6.

[0058] Figure 8 is an explanatory diagram of the first, second, and third stopping positions of vehicle 1, which is the vehicle itself. In Figure 8, the vehicle road 41 on which the autonomously driven vehicle 1 travels is connected to an intersecting road 42. The intersecting road 42 in Figure 8 is a single-lane road. In this case, the high-precision map data 21 includes node and link information, lane width information, node and link position information, and stop line position information for vehicle road 41. Here, the road width in Figure 1 is the same as the lane width. The positions of the nodes and links are set to the center of the lane.

[0059] In this case, the driving control device 11 can obtain a first stopping position on the vehicle's road 41 based on the information about the position of the stop line on the vehicle's road 41 in the high-precision map data 21. The driving control device 11 also calculates the position of the road edge on the vehicle's side of the intersecting road 42 from the position information of the nodes and links of the vehicle's road 41 and the intersecting road 42 in the high-precision map data 21, and the lane width information of the intersecting road 42. The driving control device 11 can obtain the boundary line position P2 between the intersecting road 42 and the vehicle's road 41 as a second stopping position through calculations based on the information in the high-precision map data 21. If the road width is W and the remaining distance from the vehicle to the center of the road or lane is L, the remaining distance D to the second stopping position can be calculated using the following equation 1.

[0060] D = L-W / 2...Formula 1

[0061] The travel control device 11 calculates the third stop position by adding a predetermined forward-extending distance to the second stop position. The forward-extending distance may be, for example, several tens of centimeters.

[0062] (Basic explanation of the method for confirming that the sensor field of view of the intersecting or merging road is secured) Next, an example of a method for confirming whether the sensor field of view of the intersecting road 42 or merging road 32 is secured by the driving control device 11 will be described. In this embodiment, the driving control device 11 first scans the intersecting road 42 multiple times along the road width direction. As a result, the driving control device 11 extracts multiple detected road endpoints on the vehicle side of the intersecting road 42 from the point cloud data of the LiDAR 12, which is a sensor installed on the vehicle 1, as shown in Figure 9. Next, the driving control device 11 maps the extracted multiple detected road endpoints on the vehicle side to the road surface based on high-precision map data 21. Then, the driving control device 11 determines the degree of agreement between the road edge position of the intersecting road 42 in the high-precision map data 21 and the positions of the mapped multiple detected road endpoints on the vehicle side. If the degree of agreement is high enough to satisfy a predetermined condition, the driving control device 11 determines that the sensor field of view of the intersecting road 42 or merging road 32 is secured. If the degree of matching is not high enough to satisfy the predetermined conditions, the driving control device 11 determines that the sensor's field of view for the intersecting road 42 or merging road 32 is not secured. This allows the driving control device 11 to reliably determine whether or not the sensor's field of view for the intersecting road 42 or merging road 32 is secured.

[0063] Furthermore, if there are obstructions around the vehicle's road 41, the road surface detection range detected by the sensor on the vehicle 1 will be narrower than the detection range detectable by the sensor's field of view. In this case, the driving control device 11 may not be able to properly extract multiple detection road endpoints on the vehicle's side of the intersecting road 42 from the detection results of the sensor's road surface detection range. For example, in the scanning process to extract detection road endpoints located far from the intersection in the direction of the intersecting road 42 or merging road 32, the driving control device 11 may not be able to extract detection road endpoints that satisfy the extraction conditions even if it performs scanning up to the boundary of the sensor's road surface detection range. Also, if the sensor's road surface detection range is narrow, the driving control device 11 may not be able to perform scanning up to the boundary of the sensor's road surface detection range. In these cases, the driving control device 11 may extract the position on the boundary line of the sensor's road surface detection range as the position of the detection road endpoint on the vehicle's side.

[0064] (Example of multiple detected road endpoints) Figure 9 is an explanatory diagram of an example of a combination of the position of the road edge on the vehicle's side of a single-lane merging road 62 and the positions of multiple detected road endpoints. In Figure 9, the vehicle's road 61 is connected to a single-lane merging road 62. The single-lane merging road 62 has curbs 66 on both the left and right sides of the lane. The curbs 66 distinguish the road surface of the lane on which the vehicle 1 is traveling from the area outside the lane. In this case, the high-precision map data 21 includes information on the nodes and links of the merging road 62, including the position information of the center of the lane of the merging road 62 in the direction of the road, and information on the width of the road surface between the left and right curbs 66. The vehicle 1's driving control device 11 can calculate and obtain the positions of the left and right shoulders of the merging road 62 based on the information in the high-precision map data 21.

[0065] In Figure 9, vehicle 1 is, for example, at the second stopping position or the front-end position. In this case, the LiDAR 12 sensor installed on vehicle 1 can detect the road surface of the merging road 62 in the hatched area in Figure 9, which is in front of the vehicle. The point cloud data 70 detected by LiDAR 12 may include point cloud data 70 of the height of the road surface of the merging road 62 and point cloud data 70 of the height of the curb 66. In Figure 9, only the point cloud data 70 on the curb 66 is shown. Point cloud data 70, not shown, is also mapped to the road surface of the merging road 62, etc.

[0066] The driving control device 11 determines whether the sensor's field of view is secured for the single-lane merging road 62. To this end, the driving control device 11 first maps the LiDAR 12 point cloud data 70 onto the road surface, including the area around the intersection as shown in Figure 9, based on the information from the high-precision map data 21. At this time, the driving control device 11 maps the LiDAR 12 point cloud data 70 onto the road surface with the vehicle's own position as the reference.

[0067] Next, the driving control device 11 scans the road surface of the merging road 62 in the generated mapping data and extracts the position of the shoulder of the merging road 62. For example, as shown by multiple arrows in Figure 9, the driving control device 11 scans the road surface of the merging road 62 multiple times, shifting its position in the direction of the road from the center of the merging road 62 in the vehicle width direction toward the shoulder. With each scan, the driving control device 11 extracts the position of the point cloud data 70 that first detects a step height of a predetermined value or more as the position of the detected road endpoint 71 on the vehicle side. The driving control device 11 also scans the road surface of the merging road 62 at positions away from the intersection of the vehicle road 61 and the merging road 62 in the direction of the road of the merging road 62. As a result, the driving control device 11 extracts the positions of multiple detected road endpoints 71 on the vehicle side that indicate the position of the shoulder of the merging road 62.

[0068] Figure 10 is an explanatory diagram illustrating the relationship between the road surface 91 of the merging road 62 shown in Figure 9 and the point cloud data 70 of LiDAR 12. Figure 10 shows a cross-section of the merging road 62 in the road surface width direction. The point cloud data 70 of LiDAR 12 is mapped onto the road surface 91 of the merging road 62 and onto the curb 66.

[0069] The driving control device 11 scans the road surface of the merging road 62, for example, as shown by the arrow in Figure 10, from the center of the merging road 62 in the vehicle width direction toward the shoulder. The center of the merging road 62 in the vehicle width direction is obtained from the node and link information of the high-precision map data 21. The driving control device 11 extracts the position of the point cloud data 70 in which a step H of a height greater than or equal to a predetermined value is first detected by scanning the road surface of the merging road 62 as the position of the detected road endpoint 71 on the vehicle side. As a result, the driving control device 11 can extract the position of the curb 66 detected by the LiDAR 12 as the position of the detected road endpoint 71 on the vehicle side indicating the position of the shoulder of the merging road 62. Here, the reference HT for a step of a height greater than or equal to a predetermined value may be based on the height of the center of the road surface from which the scan begins, as shown in Figure 9, on the height of the point cloud data 70 that was operated immediately before, or on the average of the heights of multiple point cloud data 70 that were operated immediately before. By repeatedly performing this road surface operation on the merging road 62, the driving control device 11 extracts multiple detected road endpoints 71 on the vehicle side that indicate the position of the shoulder of the merging road 62, including not only detected road endpoints 71 located near the intersection of the vehicle's road 61 and the merging road 62, but also detected road endpoints 71 located at positions further away from the intersection in the direction of the merging road 62.

[0070] As shown by the dashed line in Figure 10, sensors such as the LiDAR 12 installed on the vehicle 1 may not be able to partially detect the road surface near the curb 66 due to the shadow of the curb 66 that forms a step H of a predetermined height or higher. In this case, the driving control device 11 may add auxiliary lines along the scanning direction, as shown by the dashed line in Figure 10, based on the multiple point cloud data 70 of the detected road surface, and add temporary point cloud data 70 on top of the auxiliary lines. Also, on the merging road 62, other vehicles 69 may be moving, as shown in Figure 9. In this case, the point cloud data 70 of the LiDAR 12 will include point cloud data 70 due to the reflection position of the other vehicles 69. In this case, the driving control device 11 may perform calculation processing, for example, to remove the difference between multiple sets of point cloud data 70 acquired with a time difference, and map the remaining point cloud data 70, excluding the point cloud data 70 from the moving other vehicles 69, to the road surface. Missing road surface and point cloud data 70 may be supplemented. This allows the driving control device 11 to scan the road surface while suppressing the influence of other vehicles 69, and to extract multiple detected road endpoints 71 on the vehicle's side that indicate the position of the shoulder of the merging road 62.

[0071] Next, the driving control device 11 maps the multiple detected road endpoints 71 on the vehicle's side, which indicate the position of the shoulder of the extracted merging road 62, to the road surface of the intersecting road 42 or merging road 62 based on the high-precision map data 21, at the position where each is detected by the sensor. After this mapping, the road surface will be in the state shown in Figure 9. The driving control device 11 then determines the degree of agreement between the position of the vehicle's side road edge 26 of the merging road 62 in the high-precision map data 21 and the positions of the multiple detected road endpoints 71. If the position of the vehicle's side road edge 26 in the high-precision map data 21 and the positions of the multiple detected road endpoints 71 match in a way that satisfies predetermined conditions, the driving control device 11 determines that the sensor's field of view for the intersecting road 42 or merging road 62 is secured. Conversely, if it cannot be determined that the match satisfies the predetermined conditions, the driving control device 11 determines that the sensor's field of view for the intersecting road 42 or merging road 62 is not secured. Here, the predetermined conditions may be, for example, conditions for determining that the position of the road edge 26 on the vehicle's side in the high-precision map data 21 and the positions of the multiple detected road edge points 71 are close enough that they are not misaligned with each other. For example, a threshold value can be used to determine the proximity of the distance between the position of the road edge 26 and the position of the detected road edge points 71.

[0072] (An example of predetermined conditions for determining the degree of match) Figure 11 is an explanatory diagram of predetermined conditions for determining the degree of match between the positions of multiple detected road endpoints 71 and the positions of the road shoulders 82 of the intersecting road 81 in the high-precision map data 21. Figure 11 shows the road surface portion to the right of Y0 in the high-precision map data 21 for the intersecting road 81 that intersects with the vehicle road 61. Here, Y0 refers to the line segment in the longitudinal direction at the center of the vehicle width direction of the vehicle 1.

[0073] The multiple detected road endpoints 71 are mapped in Figure 11 so as to roughly follow the road shoulder 82 on the vehicle's side of the road surface of the intersecting road 81 in the high-precision map data 21. However, the further away from Y0 in the intersection is in the direction of the intersecting road 81, the further the mapping position of the detected road endpoints 71 is from the position of the road shoulder 82 of the intersecting road 81. In this case, the driving control device 11 determines the degree of agreement between the multiple detected road endpoints 71 and the position of the road shoulder 82 of the intersecting road 81 in the high-precision map data 21, starting from the detected road endpoint 71 on the Y0 side. The degree of agreement can be determined, for example, by the following equation 2. Here, W1 is the road width of the intersecting road 81 in the high-precision map data 21 at the scanning position of the detected road endpoint 71. W2 is the road width detected by the sensor at the scanning position of the detected road endpoint 71. The driving control device 11 may extract not only the detected road endpoint 71 on the vehicle's side but also the detected road endpoint on the opposite side, indicated by the curb 65 in Figure 9, for the intersecting road 81 within the sensor's detection range, and use the distance between them as the road width detected by the sensor. The curb 65 in Figure 9 is located on the opposite side of the road shoulder for the intersecting road 81 in the high-precision map data 21.

[0074] W2 ≥ 0.7 × W1 ... Equation 2

[0075] If all of the multiple detected road endpoints 71 on the vehicle side within a predetermined distance L1 from Y0 satisfy the predetermined condition for the degree of match, the driving control device 11 determines that the position of the road edge 26 on the vehicle side in the high-precision map data 21 matches the positions of the multiple detected road endpoints 71 in a way that satisfies the predetermined condition. In this case, the driving control device 11 determines that the sensor's field of view for the intersecting road 81 is secured. On the other hand, if a predetermined number or more of the multiple detected road endpoints 71 on the vehicle side within a predetermined distance L1 from Y0 do not satisfy the predetermined condition for the degree of match, the driving control device 11 does not determine that the position of the road edge 26 on the vehicle side in the high-precision map data 21 matches the positions of the multiple detected road endpoints 71 in a way that satisfies the predetermined condition. In this case, the driving control device 11 determines that the sensor's field of view for the intersecting road 81 is not secured.

[0076] Furthermore, the degree of agreement between the positions of the multiple detected road endpoints 71 and the positions of the road shoulders 82 of the intersecting roads 81 in the high-precision map data 21 may be determined by criteria other than those described above. For example, the driving control device 11 may determine the degree of agreement based on whether the cumulative value of the positional error between the positions of the multiple detected road endpoints 71 within a predetermined distance L1 from Y0 and the positions of the road shoulders 82 of the intersecting roads 81 in the high-precision map data 21 is less than or equal to a predetermined value.

[0077] (Examples of decisions based on driving) Next, based on Figures 12 to 14, specific examples of decisions regarding securing the sensor's field of view for the aforementioned intersecting road 81 or merging road 62 at each stage of the vehicle's approach to the intersection will be explained.

[0078] Figure 12 is an explanatory diagram of the first state in which vehicle 1, the vehicle itself, begins to travel towards the intersection in Figure 9 under autonomous driving. In Figure 12, the road 61 on which the vehicle is traveling is connected to a two-lane merging road 62 63 and 64. There is a curb 66 on the side of the road on the vehicle's side of the merging road 62. Obstructions such as buildings, walls, and guardrails exist on both the left and right sides of the vehicle road 61. The sensors of vehicle 1 traveling on the vehicle road 61 towards the intersection have their field of view obstructed by the obstructions on the left and right, and cannot detect other vehicles 69 traveling on the merging road 62. The sensor installed on vehicle 1 in Figure 12 is a LiDAR 12. The LiDAR 12 detects the road surface detection range indicated by the single point difference in the figure. The road surface detection range in Figure 12 is narrower than the range detectable by the field of view of the LiDAR 12. When the vehicle is stopped at the first stopping position as shown in Figure 12, the driving control device 11 determines, for example, in step ST12 of Figure 7, whether the sensor's field of view for the merging road 62 is secured. In Figure 12, the vehicle 1 is positioned between the left and right obstructions 68, 68, away from the intersection. In this case, the LiDAR 12's field of view is obstructed by the left and right obstructions 68, 68, and it can only detect the intersection portion of the road surface of the merging road 62 in front of the vehicle. The road surface detection range of the LiDAR 12 is limited to the intersection portion in front of the vehicle.

[0079] In this case, the driving control device 11 maps multiple point cloud data 70, which are detection information from the LiDAR 12, onto the road surface in Figure 12 based on the high-precision map data 21. The driving control device 11 then performs the process of scanning the road surface of the intersecting or merging road multiple times, shifted in the direction of the intersecting or merging road. For each scan, the driving control device 11 extracts the position where it first detects a step height of a predetermined value or higher as the position of the detected road endpoint 71 on the vehicle side for each scan. In this way, the driving control device 11 extracts multiple detected road endpoints 71 on the vehicle side of the intersecting road 81 from the LiDAR 12 point cloud data 70. Ideally, the multiple detected road endpoints 71 on the vehicle side are aligned so as to overlap with the road edge 26 on the vehicle side of the road surface based on the high-precision map data 21, as shown in Figure 9.

[0080] However, the road surface detection range of the LiDAR 12 in Figure 12 does not include the curb 66 on the shoulder of the merging road 62 at a position far from the intersection in the direction of the merging road 62. Therefore, even if the LiDAR 12 scans up to the boundary line 72 of its road surface detection range, the driving control device 11 cannot extract a detected road endpoint 71 that generates a step of a predetermined height at a position far from the intersection in the direction of the merging road 62. In this case, when it is not possible to detect a step of a height greater than the predetermined height, the driving control device 11 extracts the scanning position on the boundary line 72 of the LiDAR 12's road surface detection range, shown by the dashed line in Figure 12, as the position of the detected road endpoint 71 on the vehicle side instead. Furthermore, if scanning of the road surface itself is not possible within the road surface detection range of the LiDAR 12, the driving control device 11 extracts a scanning position on the boundary line 72 of the road surface detection range of the LiDAR 12, shown by the dashed line in Figure 12, as the position of the detected road endpoint 71 on the vehicle side. Here, the driving control device 11 may use the intersection point of the line segment along each scanning direction and the boundary line 72 of the road surface detection range of the LiDAR 12 as the scanning position on the boundary line 72 of the road surface detection range of the LiDAR 12.

[0081] Next, the driving control device 11 maps the extracted multiple detected road endpoints 71 on the vehicle's side to the road surface using high-precision map data 21. As a result, the driving control device 11 can obtain information that maps the multiple detected road endpoints 71 to the road surface of the merging road 62 in the high-precision map data 21, as shown in Figure 12. The driving control device 11 then determines the degree of agreement between the position of the road edge 26 of the intersecting road 81 in the high-precision map data 21 and the positions of the mapped multiple detected road endpoints 71 on the vehicle's side. In the case of Figure 12, the position of the road edge 26 of the intersecting road 81 in the high-precision map data 21 and the positions of the mapped multiple detected road endpoints 71 on the vehicle's side do not match, satisfying the predetermined conditions in Figure 11. In this case, the driving control device 11 determines that the sensor's field of view for the merging road 62 is not secured. For example, in step ST12 of Figure 7, the driving control device 11 determines that the sensor's field of view for the merging road 62 is not secured, and in steps ST14 and ST17, it executes driving control to move to the second stopping position and stop.

[0082] Figure 13 is an explanatory diagram of the second state in which the vehicle 1, which is the vehicle itself, is closer to the intersection than in Figure 12 due to autonomous driving. When the vehicle is in the state of Figure 13, where it is stopped at the second stopping position, the driving control device 11 determines, for example, in step ST18 of Figure 7, whether or not the sensor's field of view for the merging road 62 is secured. In Figure 13, the vehicle 1 is positioned between the left and right obstructions 68, 68. In this case, the LiDAR 12's field of view is obstructed by the left and right obstructions 68, 68, and it can only detect the intersection portion in front of the vehicle on the road surface of the merging road 62. However, since the vehicle 1 is closer to the intersection than in Figure 12, the field of view of the LiDAR 12 is wider than in Figure 12.

[0083] In this case, the driving control device 11 maps multiple point cloud data 70, which are detection information from the LiDAR 12, onto the road surface of the high-precision map data 21. The driving control device 11 then performs the process of scanning the road surface of the intersecting or merging road multiple times, shifted in the direction of the intersecting or merging road. For each scan, the driving control device 11 extracts the position where it first detects a step of a height greater than a predetermined value as the position of the detected road endpoint 71 on the vehicle side for each scan. In this way, the driving control device 11 extracts multiple detected road endpoints 71 on the vehicle side of the intersecting road 81 from the point cloud data 70 of the LiDAR 12. Ideally, the multiple detected road endpoints 71 on the vehicle side are aligned so as to overlap with the road edge 26 on the vehicle side of the road surface based on the high-precision map data 21, as shown in Figure 9. However, the road surface detection range of the LiDAR 12 in Figure 13 does not include the curb 66 on the shoulder of the merging road 62 at a position far from the intersection in the direction of the merging road 62. Therefore, even if the driving control device 11 scans up to the boundary line 72 of the road surface detection range of the LiDAR 12, it cannot extract a detected road endpoint 71 that generates a step of a predetermined height at a position far from the intersection in the direction of the merging road 62. In this case, when it is not possible to detect a step of a height greater than the predetermined height, the driving control device 11 extracts the scanning position on the boundary line 72 of the road surface detection range of the LiDAR 12, shown by the dashed line in Figure 13, as the position of the detected road endpoint 71 on the vehicle side instead. Furthermore, if the LiDAR 12 is unable to scan the road surface within its road surface detection range, the driving control device 11 extracts a scanning position on the boundary line 72 of the LiDAR 12's road surface detection range as the position of the detected road endpoint 71 on the vehicle side.

[0084] Next, the driving control device 11 maps the extracted multiple detected road endpoints 71 on the vehicle's side to the road surface using high-precision map data 21. As a result, the driving control device 11 can obtain information that maps the multiple detected road endpoints 71 to the road surface of the merging road 62 in the high-precision map data 21, as shown in Figure 13. The driving control device 11 then determines the degree of agreement between the position of the road edge 26 of the intersecting road 81 in the high-precision map data 21 and the positions of the mapped multiple detected road endpoints 71 on the vehicle's side. In the case of Figure 13, the position of the road edge 26 of the intersecting road 81 in the high-precision map data 21 and the positions of the mapped multiple detected road endpoints 71 on the vehicle's side do not match, satisfying the predetermined conditions in Figure 11. In this case, the driving control device 11 determines that the sensor's field of view for the merging road 62 is not secured. For example, in step ST18 of Figure 7, the driving control device 11 determines that the sensor's field of view for the merging road 62 is not secured, and in steps ST20 and ST23, it executes driving control to move to the front-end position and stop.

[0085] Figure 14 is an explanatory diagram of a third state in which the vehicle 1, which is the owner vehicle, is closer to the intersection than in Figure 13 due to autonomous driving. When the vehicle is in the state of Figure 14, where the owner vehicle is stopped at the front-end position, the driving control device 11 determines, for example, in step ST24 of Figure 7, whether the sensor's field of view for the merging road 62 is secured. In Figure 14, the vehicle 1, which is the owner vehicle, is not in a position between the left and right obstructions 68, 68. In this case, the LiDAR 12's field of view is not obstructed by the left and right obstructions 68, 68. The field of view of the LiDAR 12 is wider than in the case of Figure 13. The LiDAR 12 can detect the entire road surface of the merging road 62 with a road surface detection range that has a width equivalent to the detection range corresponding to the field of view of the LiDAR 12.

[0086] In this case, the driving control device 11 maps multiple point cloud data 70, which are detection information from the LiDAR 12, onto the road surface of the high-precision map data 21. The driving control device 11 then performs the process of scanning the road surface of the intersecting or merging road multiple times, shifted in the direction of the intersecting or merging road. For each scan, the driving control device 11 extracts the position where it first detects a step of a height greater than a predetermined value as the position of the detected road endpoint 71 on the vehicle side for each scan. In this way, the driving control device 11 extracts multiple detected road endpoints 71 on the vehicle side of the intersecting road 81 from the point cloud data 70 of the LiDAR 12. In this case, the multiple detected road endpoints 71 on the vehicle side may be arranged so as to overlap with the road edge 26 on the vehicle side of the road surface based on the high-precision map data 21, as shown in Figure 14. In the case of the road surface detection range of the LiDAR 12, the driving control device 11 can scan up to the curb 68 at the boundary of the intersecting road 81 included in the road surface detection range of the LiDAR 12 in each scan, unlike in the cases of Figures 12 and 13. In each scan, the driving control device 11 can extract the detected road endpoint 71 on the vehicle side that creates a step of a predetermined height due to the curb 66 of the merging road 62 from the point cloud data 70 of the LiDAR 12. The driving control device 11 can also extract the detected road endpoint 71 on the vehicle side based on the curb 66 of the merging road 62 even at a position far from the intersection in the direction of the road of the merging road 62.

[0087] Next, the driving control device 11 maps the extracted multiple detected road endpoints 71 on the vehicle's side to the road surface of the high-precision map data 21. As a result, the driving control device 11 can obtain information that maps the multiple detected road endpoints 71 to the road surface of the merging road 62 in the high-precision map data 21, as shown in Figure 14. The driving control device 11 then determines the degree of agreement between the position of the road edge 26 of the intersecting road 81 in the high-precision map data 21 and the positions of the multiple detected road endpoints 71 on the vehicle's side that have been mapped. In the case of Figure 14, the position of the road edge 26 of the intersecting road 81 in the high-precision map data 21 and the positions of the multiple detected road endpoints 71 on the vehicle's side that have been mapped match in a way that satisfies the predetermined conditions in Figure 11. In this case, the driving control device 11 determines that the sensor's field of view for the merging road 62 is secured. For example, in step ST24 of Figure 7, the driving control device 11 determines that the sensor's field of view for the merging road 62 is secured, and further, if the driving environment of the merging road 62 is safe, in step ST27, it executes driving control to start driving from the front position by automatic driving and enter the intersection.

[0088] As described above, in this embodiment, when the vehicle 1 is traveling towards an intersection with an intersecting or merging road, the driving control device 11 determines whether the sensor's field of view for the intersecting or merging road in the high-precision map data 21 is secured. By performing this field of view determination process, the driving control device 11 enables the vehicle to travel towards the intersection while ensuring that the sensor's field of view for the intersecting or merging road is secured. Conversely, if the sensor's field of view for the intersecting or merging road is not secured, the driving control device 11 can adjust the vehicle's travel toward the intersection. As a result, the vehicle 1 can travel in a manner that allows it to enter the intersection while ensuring the sensor's field of view for the intersecting or merging road is secured and the safety of the intersecting or merging road can be confirmed. The vehicle 1 can ensure a high level of safety when entering an intersection with an intersecting or merging road.

[0089] In this embodiment, in the field of view determination process, the driving control device 11 first extracts a plurality of detected road endpoints 71 on the vehicle side for intersecting or merging roads included in the sensor's detection range from the sensor's detection information. These plurality of detected road endpoints 71 on the vehicle side include detected road endpoints 71 located away from the intersection in the direction of the intersecting or merging road. The driving control device 11 maps the extracted plurality of detected road endpoints 71 to the road surface of the intersecting or merging road based on high-precision map data 21 at the positions detected by the sensor. The driving control device 11 then determines the degree of agreement between the position of the road edge on the vehicle side of the intersecting or merging road in the high-precision map data 21 and the positions of the plurality of detected road endpoints 71 based on the sensor's detection. If these match to satisfy predetermined conditions, the driving control device 11 determines that the sensor's field of view for the intersecting or merging road is secured; otherwise, it determines that it is not secured. As a result, the driving control device 11 can determine that the sensor's field of view is secured for the intersecting or merging road if the sensor's field of view can detect the road edge on the vehicle's side of the intersecting or merging road to a degree that meets predetermined conditions. If the field of view is not of such a degree, the driving control device 11 can determine that the sensor's field of view is not secured.

[0090] In this embodiment, the driving control device 11 extracts the location where a step of a certain height or greater is detected on the road surface of an intersecting or merging road, as included in the sensor's detection information, as the location of the vehicle's detection road endpoint 71. This allows the driving control device 11 to extract the location of a step on the road shoulder or other part of the road surface of an intersecting or merging road as the location of the vehicle's detection road endpoint 71 for the intersecting or merging road. For example, the driving control device 11 can extract the location where a step caused by a curb 66 on the road shoulder of an intersecting or merging road is detected as the location of the vehicle's detection road endpoint 71 for the intersecting or merging road. Furthermore, if the driving control device 11 cannot detect a step of a certain height or greater even after scanning up to the boundary line of the sensor's road surface detection range, or if scanning itself is not possible, it extracts the scanning position on the boundary line 72 of the sensor's road surface detection range as the location of the vehicle's detection road endpoint 71. As a result, even if the driving control device 11 cannot detect a step of a predetermined height or higher within the sensor's road surface detection range, it can extract the scanning position on the boundary line 72 of the sensor's road surface detection range as the position of the detected road endpoint 71 on the vehicle side.

[0091] In this embodiment, if the field of view of the sensor is secured, the driving control device 11 further performs a safety confirmation process for entering an intersection based on the detection of the sensor with a secured field of view. In this embodiment, the driving control device 11 determines, based on the detection of the sensor with a secured field of view, that there is no possibility of interference with a moving object moving on an intersecting or merging road. Then, if the driving control device 11 determines that safety can be confirmed through the safety confirmation process based on the detection of the sensor with a secured field of view, it performs driving control to enter the intersection by automatic driving. In this way, the driving control device 11 does not simply determine whether safety can be confirmed based on the information of the sensor's detection range, but further confirms that the field of view of the sensor for the intersecting or merging road is secured. The driving control device 11 is able to perform a control that can ensure a high level of safety when the vehicle enters an intersection with an intersecting or merging road. In particular, if the field of view of the sensor is not secured by the field of view confirmation process, or if safety cannot be confirmed by the safety confirmation process based on the detection of the sensor with a secured field of view, the driving control device 11 performs a process to switch the operation of the vehicle 1 from automatic driving to manual driving. This allows the driving control device 11 to prevent the vehicle from forcibly entering the intersection through automatic driving if confirmation from both parties cannot be obtained. In this case, the vehicle 1 can safely enter the intersection through the driver's own confirmation and manual driving.

[0092] In this embodiment, the driving control device 11 can repeatedly execute a set of field of view judgment processing and safety confirmation processing in each of the following cases: first case when the vehicle 1 is stopped at a first stopping position based on the regulations of the road on which it is traveling; second case when the vehicle 1 is stopped at a second stopping position just before entering the traffic lane of an intersecting or merging road from its own road; third case when the vehicle 1 is stopped at a third stopping position with only the front end of the vehicle 1 entering the traffic lane of an intersecting or merging road; fourth case when the vehicle is moving from the first stopping position to the second stopping position; and fifth case when the vehicle is moving from the second stopping position to the third stopping position. As a result, the vehicle 1 can, for example, safely enter the intersection without stopping again after stopping at the first stopping position, or stop at the second or third stopping position to check its field of view and safety before entering the intersection. Furthermore, if the field of view and safety cannot be confirmed at the third stopping position, the vehicle 1 can safely enter the intersection through the driver's own confirmation and manual driving, without forcing its way into the intersection through automatic driving.

[0093] The embodiments described above are examples of preferred embodiments of the present invention, but the present invention is not limited thereto, and various modifications or changes are possible without departing from the spirit of the invention.

[0094] (Modification) In the embodiment described above, the driving control device 11 can repeatedly execute the set process of field of view judgment processing and safety confirmation processing in all of the first to fifth cases described above. Alternatively, for example, the driving control device 11 may execute the set process of field of view judgment processing and safety confirmation processing in only some of the first to fifth cases described above, or in only one of them. Furthermore, the driving control device 11 may stop the vehicle at at least one stopping position, rather than all of the stopping positions from the first to the third stopping position. Here, if the driving control device 11 does not determine that the field of view of the intersecting road or merging road is secured and that it is safe to enter the intersecting road or merging road, it may switch the vehicle's operation from automatic driving to manual driving. For example, the driving control device 11 may repeatedly execute the set process in multiple cases selected from the first to the fifth cases. In this case, the control of the driving control device 11 will repeatedly execute the set process according to the stage of the vehicle 1's operation. The driving control device 11 will execute the setting process multiple times. In each stage of the setting process that is repeatedly executed, if the driving control device 11 determines that visibility of the intersecting or merging road is secured and that it is safe to enter the intersecting or merging road, it may execute driving control to enter the intersection by automatic driving. In this case, the driving control device 11 may interrupt the execution of the setting process in subsequent stages. On the other hand, if the driving control device 11 does not determine that visibility of the intersecting or merging road is secured and that it is safe to enter the intersecting or merging road in each stage of the setting process that is repeatedly executed, the driving control device 11 will execute the setting process for the next stage. In the final setting process that is executed after a predetermined number of stages, if the driving control device 11 does not determine that visibility of the intersecting or merging road is secured and that it is safe to enter the intersecting or merging road, it may switch the driving of vehicle 1 from automatic driving to manual driving.Figure 7 shows an example in which the set processing of the field of view judgment process and the safety confirmation process is repeatedly executed for each stage as the vehicle 1 progresses through the first, fourth, second, fifth, and third stages described above. In this example, the driving control device 11 will repeat the set processing of the field of view judgment process and the safety confirmation process up to five times. Furthermore, the driving control device 11 may repeat the set processing of the field of view judgment process and the safety confirmation process multiple times, rather than just once, at each stage. For example, in the fourth stage, when the vehicle moves from the stopping position of the first case to the stopping position of the second case, the driving control device 11 may repeatedly execute the set processing of the field of view judgment process and the safety confirmation process during the period until the vehicle 1 reaches the stopping position of the second case. Furthermore, in the fifth case, when the vehicle 1 moves from the stopping position in the second case to the stopping position in the third case, the driving control device 11 may repeatedly perform a set of processing for field of view judgment and safety confirmation during the period until the vehicle 1 reaches the stopping position in the third case.

[0095] In the embodiments described above, a road having only a carriageway is used as the intersecting or merging road. In reality, some roads have sidewalks or shoulders. In this case, the driving control device 11 may perform a set of visibility judgment processing and safety confirmation processing for each of the multiple lanes of the intersecting or merging road. Here, lanes refer to the lanes, sidewalks, and shoulders of the intersecting or merging road. The driving control device 11 may stop the vehicle at the second and third stopping positions for each of the multiple lanes of the intersecting or merging road and perform a set of visibility judgment processing and safety confirmation processing at each stopping position. As a result, the autonomously driven vehicle 1 can perform visibility judgment processing and safety confirmation processing not only for other vehicles traveling in the lanes of the intersecting or merging road, but also for pedestrians moving on the sidewalks, etc., of the intersecting or merging road, and is expected to ensure a high level of safety when entering each lane.

[0096] The above-described embodiment mainly describes the case in which a LiDAR 12 is used as the sensor provided on the vehicle 1. In addition to the LiDAR 12, the vehicle 1 may also be equipped with other ADAS (Advanced Driver Assistance System) sensors, such as a camera 13. The driving control device 11 may perform the same control as described above for the LiDAR 12 based on the detection of the camera 13, or based on the detection of both the camera 13 and the LiDAR 12. For example, the driving control device 11 can extract image components such as curbs 66 on the shoulder of intersecting or merging roads by analyzing the image components of the image captured by the camera 13. Based on the position and color components of the extracted curbs 66 in the captured image, the driving control device 11 can obtain the position on the road surface of the detected road endpoint 71 based on the curbs 66. The color of the concrete curb 66 and the color of the asphalt road surface are generally different.

[0097] In the embodiment described above, the driving control device 11, in the field of view determination process, extracts the position of a curb 66 that detects a step of a predetermined height or higher on the road surface of an intersecting or merging road included in the sensor's detection information, as the position of the detected road endpoint 71 on the vehicle side. In addition, for example, the driving control device 11 may extract the position of a lane boundary line drawn on the road surface of an intersecting or merging road, or the position of a guardrail, guard pole, planting, etc. that separates the roadway and sidewalk of an intersecting or merging road, as the position of the detected road endpoint 71 on the vehicle side.

[0098] 1...Vehicle, 10...Control system, 11...Driving control device, 12...LiDAR (sensor), 13...Camera (sensor), 14...GNSS receiver, 15...Map storage device (map acquisition device), 16...Communication device (map acquisition device), 17...Drive system, 18...Braking system, 19...Steering system, 20...Operating device, 21...High-precision map data, 26...Roadside, 31...Own road, 32...Merging road, 33...Other vehicles, 34...Guardrail (obstacle), 41...Own road, 42...Intersecting road, 45...Other vehicles, 46...Obstacle, 47...Obstacle, 61...Own road, 62...Merging road, 66...Curb, 68...Obstacle, 69...Other vehicles, 70...Point cloud data, 71...Detected roadside point, 72...Boundary of road surface detection range, 81...Intersecting road, 82...Shoulder, 91...Road surface

Claims

1. An autonomous vehicle comprising: a sensor capable of detecting the space surrounding the vehicle, including at least the road in the direction of travel of the vehicle; a map acquisition device that acquires map data including the road on which the vehicle is traveling; and a driving control device capable of performing autonomous driving of the vehicle, wherein the driving control device performs a field of view determination process to determine whether the sensor has a field of view of the intersecting road or merging road in the map data when the vehicle is traveling toward an intersection with an intersecting road or merging road.

2. The vehicle capable of autonomous driving according to claim 1, wherein the driving control device, in the field of view determination process, extracts a plurality of detected road endpoints on the vehicle side for intersecting roads or merging roads included in the detection range of the sensor from the detection information of the sensor, and the plurality of detected road endpoints on the vehicle side include a detected road endpoint located away from the intersection in the direction of the intersecting road or merging road, and determines whether the field of view of the sensor for the intersecting road or merging road in the map data is secured based on the degree of agreement between the positions of the plurality of detected road endpoints extracted based on the detection of the sensor and the positions of the road edges on the vehicle side of the intersecting road or merging road in the map data.

3. The vehicle capable of autonomous driving according to claim 2, wherein the driving control device, in the field of view determination process, performs a process of scanning the road surface of the intersecting road or merging road included in the detection information of the sensor multiple times, shifting in the direction of the intersecting road or merging road, and extracts a position in which a step of height greater than or equal to a predetermined value is detected each time, as the position of the detected road endpoint on the vehicle side, and if a scanning process to extract a detected road endpoint at a position far from the intersection in the direction of the intersecting road or merging road cannot be performed, or if a step of height greater than or equal to a predetermined value cannot be detected even after the scanning process, extracts a position on the boundary line of the road surface detection range of the sensor as the position of the detected road endpoint on the vehicle side.

4. The vehicle capable of autonomous driving according to any one of claims 1 to 3, wherein the driving control device, when it determines in the field of view determination process that the field of view of the sensor is secured, performs a safety confirmation process for when the vehicle enters the intersection based on the detection of the sensor that has a secured field of view, and when it determines that it is safe based on the safety confirmation process based on the detection of the sensor that has a secured field of view, performs driving control for entering the intersection by autonomous driving.

5. The vehicle capable of autonomous driving according to claim 4, wherein the driving control device switches the operation of the vehicle from autonomous driving to manual driving if it does not determine in the safety confirmation process based on the detection of the sensor in which the field of view is secured that the vehicle is safe.

6. The vehicle capable of autonomous driving according to claim 4, wherein the driving control device performs a set processing of the field of view judgment processing and the safety confirmation processing in at least one of the following cases: first case when the vehicle is stopped at a first stopping position based on the regulations of the road on which the vehicle is traveling; second case when the vehicle is stopped at a second stopping position immediately before entering an intersecting or merging road; third case when the vehicle is stopped at a third stopping position after the front end of the vehicle has entered an intersecting or merging road; fourth case when the vehicle is moving from the first stopping position to the second stopping position; and fifth case when the vehicle is moving from the second stopping position to the third stopping position.

7. The vehicle capable of automatic driving according to claim 6, wherein the driving control device is capable of repeatedly executing the set process in a plurality of cases selected from the first to fifth cases, and in the repeatedly executed set process, if it is determined that a field of view of the intersecting road or merging road is secured and that it is safe to enter the intersecting road or merging road, it executes driving control to enter the intersection by automatic driving; if it is not determined that a field of view of the intersecting road or merging road is secured and that it is safe to enter the intersecting road or merging road in the repeatedly executed set process, it executes the next set process; and in the last set process, if it is not determined that a field of view of the intersecting road or merging road is secured and that it is safe to enter the intersecting road or merging road, it switches the operation of the vehicle from automatic driving to manual driving.

Images