Roadway identification device and roadway identification method

Abstract

To enable a runway to be identified with good accuracy, even when using a peripheral monitoring sensor and map information to identify the runway of a vehicle.SOLUTION: The present invention comprises: a sensor acquisition unit 104 for acquiring sensing results from a peripheral monitoring sensor 14 that is used in a vehicle; a map acquisition unit 101 for acquiring map information; a runway identification unit 108 which is capable of identifying the runway of the vehicle, using both of the sensing results acquired by the sensor acquisition unit 104 and the map information acquired by the map acquisition unit 101; and a prescribed environment identification unit 107 for identifying a prescribed environment where, with the sensing results out of the sensing results and the map information, the accuracy of identifying a runway is estimated to decrease. The runway identification unit 108 switches to runway identification using only the map information out of the sensing results and the map information, on the basis of the fact that the prescribed environment is identified by the prescribed environment identification unit 107.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to a lane identification device and a lane identification method. [Background technology]

[0002] There is a known technology for supporting vehicle driving by recognizing the driving environment of the vehicle using a perimeter monitoring sensor such as a camera. However, it may be difficult to identify the vehicle's driving path using only the perimeter monitoring sensor because the dividing lines on the road are unclear or do not exist.

[0003] In response to this, Patent Document 1 discloses a technique for determining a local route for maintaining a position in a lane by using both a driving image captured by a camera and map information. In the technique disclosed in Patent Document 1, the constant term of a polynomial of a fitting curve expressing a local route is estimated from the driving image, and the linear and quadratic terms are estimated from the map information. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2019-139762 Summary of the Invention [Problem to be solved by the invention]

[0005] In Patent Document 1, when the accuracy of either the driving image or the map information is low, the accuracy of the estimated fitting curve is also low. Therefore, even if the surroundings monitoring sensor and the map information are used, the accuracy of identifying the vehicle's route is low.

[0006] One object of this disclosure is to provide a vehicle control device and a vehicle control method that enable more accurate identification of the vehicle's path, even when surrounding monitoring sensors and map information are used to identify the vehicle's path. [Means for solving the problem]

[0007] The above object is achieved by the combination of features recited in the independent claims, and the subclaims define further advantageous embodiments of the disclosure. The reference characters in parentheses in the claims indicate a correspondence with specific means described in the following embodiment as one aspect, and do not limit the technical scope of the present disclosure.

[0008] In order to achieve the above object, the path identification device of the present disclosure includes a sensor acquisition unit (104) that acquires sensing results from a perimeter monitoring sensor (14) used in the vehicle, a map acquisition unit (101) that acquires map information, a path identification unit (108, 108a) that can identify the path of the vehicle using both the sensing results acquired by the sensor acquisition unit and the map information acquired by the map acquisition unit, and a predetermined environment identification unit (107, 107a) that identifies a predetermined environment in which the sensing results out of the sensing results and the map information are estimated to reduce the accuracy of path identification, and the path identification unit switches to path identification using only the map information out of the sensing results and the map information based on the identification of the predetermined environment by the predetermined environment identification unit. the predetermined environment specification unit specifies, as the predetermined environment, a predetermined road structure which is at least any one of an intersection, a branch point, a curved road with a certain degree of curvature or more, and an uphill road with a certain gradient or more; and the running path specification unit switches from specifying a running path using only the sensing results out of the sensing results and the map information, via specifying a running path using both the sensing results and the map information, to specifying a running path using only the map information out of the sensing results and the map information, as the distance from the vehicle to the predetermined road structure becomes shorter based on the predetermined road structure specified by the predetermined environment specification unit, and switches from specifying a running path using only the map information out of the sensing results and the map information, via specifying a running path using both the sensing results and the map information, to specifying a running path using only the sensing results out of the sensing results and the map information, as the distance from the vehicle to the predetermined road structure becomes longer .

[0009] In order to achieve the above object, the path identification method of the present disclosure includes a sensor acquisition step of acquiring sensing results from a perimeter monitoring sensor (14) used in a vehicle, which are executed by at least one processor, a map acquisition step of acquiring map information, a path identification step capable of identifying a path of the vehicle using both the sensing results acquired in the sensor acquisition step and the map information acquired in the map acquisition step, and a predetermined environment identification step of identifying a predetermined environment in which it is estimated that the sensing results out of the sensing results and the map information will reduce the accuracy of identifying the path, and in the path identification step, based on the identification of the predetermined environment in the predetermined environment identification step, switching to identifying the path using only the map information out of the sensing results and the map information. In the predetermined environment specifying step, a predetermined road structure is specified as the predetermined environment, which is at least any one of an intersection, a branch point, a curved road with a certain degree of curvature or more, and an uphill road with a certain gradient or more. In the running path specifying step, as the distance from the vehicle to the predetermined road structure becomes shorter based on the predetermined road structure specified in the predetermined environment specifying step, the method switches from specifying a running path using only the sensing results out of the sensing results and the map information to specifying a running path using both the sensing results and the map information to specifying a running path using only the map information out of the sensing results and the map information, while as the distance from the vehicle to the predetermined road structure becomes longer, the method switches from specifying a running path using only the map information out of the sensing results and the map information to specifying a running path using both the sensing results and the map information to specifying a running path using only the sensing results out of the sensing results and the map information. .

[0010] According to the above configuration, in a specified environment where it is estimated that the accuracy of identifying the vehicle's path is reduced using the sensing results of the perimeter monitoring sensor used in the vehicle and the map information, it is possible to switch to identifying the path using only the map information of the sensing results and the map information. Therefore, even in a place where it is estimated that the accuracy of identifying the vehicle's path is reduced using the sensing results, it is possible to identify the path with high accuracy using the map information. On the other hand, in a case other than such a specified environment, it is possible to identify the path with high accuracy using the sensing results. As a result, even when the perimeter monitoring sensor and the map information are used to identify the vehicle's path, it is possible to identify the path with high accuracy. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram showing an example of a schematic configuration of a vehicle system 1. FIG. [Diagram 2] FIG. 2 is a diagram illustrating an example of a schematic configuration of an autonomous driving ECU 10. [Diagram 3] FIG. 13 is a diagram showing an example of dividing a region into areas based on intersections; [Figure 4] FIG. 4 is a diagram showing an example of switching of lane specification by area shown in FIG. 3. [Diagram 5] 10 is a flowchart showing an example of a flow of a road identification-related process in the autonomous driving ECU 10. [Figure 6] FIG. 2 is a diagram illustrating an example of a schematic configuration of an autonomous driving ECU 10a. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] A number of embodiments for disclosure will be described with reference to the drawings. For convenience of description, in a number of embodiments, parts having the same functions as parts shown in the drawings used in the previous description may be given the same reference numerals and their description may be omitted. For parts given the same reference numerals, the description in other embodiments may be referred to.

[0013] (Embodiment 1) <Overall configuration of vehicle system 1> Hereinafter, a first embodiment of the present disclosure will be described with reference to the drawings. A vehicle system 1 shown in FIG. 1 can be used in a vehicle. As shown in FIG. 1, the vehicle system 1 includes an automatic driving ECU 10, a communication module 11, a GNSS (Global Navigation Satellite System) receiver 12, a vehicle state sensor 13, a surroundings monitoring sensor 14, and a vehicle control ECU 15. For example, the automatic driving ECU 10, the communication module 11, the GNSS receiver 12, the vehicle state sensor 13, the surroundings monitoring sensor 14, and the vehicle control ECU 15 may be configured to be connected to an in-vehicle LAN (see the LAN in FIG. 1). A vehicle using the vehicle system 1 is not necessarily limited to an automobile, but the following description will be given taking an example of use in an automobile.

[0014] The communication module 11 transmits and receives information to and from a server external to the vehicle via wireless communication. In other words, it performs wide-area communication. The server external to the vehicle may be, for example, a cloud server. The communication module 11 receives map information, traffic congestion information, and the like from the server via wide-area communication.

[0015] The map information may be high-precision map data. The high-precision map data is map data with higher precision than map data used for route guidance in the navigation function. The map information may also include map data used for route guidance. The high-precision map data includes information that can be used for automated driving, such as three-dimensional shape information of roads, information on the number of lanes, and information indicating the permitted traveling direction for each lane. In addition, the high-precision map data may include node point information indicating the positions of both ends of road markings such as dividing lines. The map data used for route guidance includes link data, node data, and the like. The link data includes a unique number (link ID) that identifies a link, a link length that indicates the length of the link, a link direction, link shape information, node coordinates of the start and end of the link, and road attribute data. The road attributes include the road name, road type, road width, number of lanes, the presence or absence of right-turn / left-turn dedicated lanes and the number of dedicated lanes, and speed limit value. Node data consists of various data such as node ID, which is a unique number assigned to each node on the map, node coordinates, node name, node type, connecting link ID, which describes the link ID of the link connecting to the node, center coordinates of intersections, etc. Intersections and branching points correspond to nodes.

[0016] The GNSS receiver 12 receives positioning signals from a plurality of positioning satellites. The vehicle state sensor 13 is a group of sensors for detecting various states of the vehicle. The vehicle state sensor 13 includes a vehicle speed sensor, an acceleration sensor, and the like. The vehicle speed sensor detects the speed of the vehicle. The acceleration sensor detects the acceleration of the vehicle. The acceleration sensor may be, for example, a three-axis acceleration sensor. The vehicle state sensor 13 outputs the detection result to an in-vehicle LAN. The detection result detected by the vehicle state sensor 13 may be configured to be output to the in-vehicle LAN via an ECU mounted in the vehicle.

[0017] The perimeter monitoring sensor 14 monitors the environment around the vehicle. The perimeter monitoring sensor 14 can also be called an autonomous sensor. As an example, the perimeter monitoring sensor 14 detects obstacles around the vehicle, such as moving objects such as pedestrians and other vehicles, and stationary objects such as objects fallen on the road. In addition, the perimeter monitoring sensor 14 detects road markings such as dividing lines around the vehicle. The dividing lines include lane center lines, lane boundary lines, and roadway outer lines. The perimeter monitoring sensor 14 is, for example, a perimeter monitoring camera that captures a predetermined range around the vehicle, a millimeter wave radar that transmits a search wave to a predetermined range around the vehicle, a sonar, a LIDAR, or other sensor. The predetermined range may be a range that includes at least the front of the vehicle. The perimeter monitoring camera sequentially outputs the captured images to the automatic driving ECU 10 as sensing information. The sonar, millimeter wave radar, LIDAR, or other sensor that transmits a search wave sequentially outputs a scan result based on a received signal obtained when receiving a reflected wave reflected by an obstacle to the automatic driving ECU 10 as sensing information. The sensing information detected by the periphery monitoring sensor 14 may be configured to be output to the automatic driving ECU 10 without passing through an in-vehicle LAN.

[0018] The vehicle control ECU 15 is an electronic control device that performs driving control of the host vehicle. Examples of driving control include acceleration / deceleration control and / or steering control. The vehicle control ECU 15 includes a steering ECU that performs steering control, a power unit control ECU that performs acceleration / deceleration control, and a brake ECU. The vehicle control ECU 15 performs driving control by outputting control signals to each driving control device such as an electronically controlled throttle, a brake actuator, and an EPS (Electric Power Steering) motor mounted on the host vehicle. The vehicle control ECU 15 controls the driving of the host vehicle based on instructions from the autonomous driving ECU 10.

[0019] The autonomous driving ECU 10 is mainly composed of a computer equipped with a processor, volatile memory, non-volatile memory, I / O, and a bus connecting these. The autonomous driving ECU 10 executes processes related to autonomous driving by executing a control program stored in the non-volatile memory. This autonomous driving ECU 10 corresponds to a road identification device. It performs autonomous driving by controlling the vehicle control ECU 15. There can be multiple levels of autonomous driving stages (hereinafter, automation levels), as defined by, for example, the SAE. The automation levels are divided into LV0 to 5, for example, as follows:

[0020] LV0 is the level at which the driver performs all driving tasks without the system intervening. The driving task may be rephrased as a dynamic driving task. The driving task may be, for example, steering, acceleration / deceleration, and surrounding monitoring. LV0 corresponds to so-called manual driving. LV1 is the level at which the system assists with either steering or acceleration / deceleration. LV1 corresponds to so-called driving assistance. LV2 is the level at which the system assists with both steering and acceleration / deceleration. LV2 corresponds to so-called partial driving automation. Note that LV1-2 are also considered to be part of autonomous driving.

[0021] Level 3 autonomous driving is a level where the system can perform all driving tasks under certain conditions, with the driver taking over driving operations in emergencies. LV3 corresponds to so-called conditional driving automation. Level 4 autonomous driving is a level where the system can perform all driving tasks except under certain circumstances such as unmanageable roads and extreme environments. LV4 corresponds to so-called high driving automation. Level 5 autonomous driving is a level where the system can perform all driving tasks in any environment. LV5 corresponds to so-called full driving automation.

[0022] The autonomous driving performed by the autonomous driving ECU 10 may be autonomous driving of LV1 or higher. The autonomous driving performed by the autonomous driving ECU 10 may be driving assistance of LV1 or autonomous driving of LV2 or higher.

[0023] <Outline of Autonomous Driving ECU 10> Next, a schematic configuration of the vehicle's route identification in the autonomous driving ECU 10 will be described with reference to FIG. 2. The route of the vehicle is a planned route along which the vehicle will travel in autonomous driving including driving assistance. The route may be an area with a width or a trajectory. For example, the route may be a short-term planned travel trajectory for realizing travel along a planned route to a set destination. In this embodiment, for convenience, a description of the configuration of the autonomous driving ECU 10 other than the route identification of the vehicle will be omitted. As shown in FIG. 2, the autonomous driving ECU 10 includes a map acquisition unit 101, a vehicle information acquisition unit 102, a position identification unit 103, a sensor acquisition unit 104, a map route generation unit 105, an autonomous route generation unit 106, a predetermined environment identification unit 107, and a route identification unit 108 as functional blocks for route identification. In addition, the execution of the processing of each functional block of the autonomous driving ECU 10 by a computer corresponds to the execution of a route identification method. Some or all of the functions executed by the autonomous driving ECU 10 may be configured in hardware using one or more ICs, etc. Some or all of the functional blocks included in the autonomous driving ECU 10 may be realized by a combination of software execution by a processor and hardware components.

[0024] The map acquisition unit 101 acquires map information. The processing in the map acquisition unit 101 corresponds to a map acquisition step. The map acquisition unit 101 acquires map information from a server external to the vehicle via the communication module 11. The map acquisition unit 101 may be configured to acquire map information limited to the vicinity of the vehicle position by sending the vehicle position identified by the position identification unit 103 described later to the server. The map acquisition unit 101 may be configured to acquire map information stored in a non-volatile memory used in the vehicle.

[0025] The vehicle information acquisition unit 102 acquires the detection result detected by the vehicle state sensor 13. The vehicle information acquisition unit 102 acquires the vehicle speed detected by the vehicle speed sensor. The vehicle information acquisition unit 102 acquires the acceleration detected by the acceleration sensor.

[0026] The position identifying unit 103 sequentially identifies the vehicle position of the vehicle (hereinafter, the vehicle position). The position identifying unit 103 identifies the vehicle position by combining the positioning signal received by the GNSS receiver 12 with the detection results of the vehicle speed and acceleration acquired by the vehicle information acquiring unit 102. The vehicle position may be expressed, for example, by latitude and longitude coordinates.

[0027] The sensor acquisition unit 104 acquires the sensing result of the perimeter monitoring sensor 14. That is, it acquires sensing information obtained by the perimeter monitoring sensor 14. This processing by the sensor acquisition unit 104 corresponds to a sensor acquisition step.

[0028] The map / path generating unit 105 generates a path for the vehicle by using the map information acquired by the map acquiring unit 101 and the vehicle position identified by the position identifying unit 103. As an example, the map / path generating unit 105 may generate a planned driving trajectory for the vehicle on a high-precision map.

[0029] The autonomous path generation unit 106 generates a path for the host vehicle using the sensing results acquired by the sensor acquisition unit 104. As an example, the autonomous path generation unit 106 recognizes the positions, shapes, and movement states of objects around the host vehicle using the sensing results acquired by the sensor acquisition unit 104. Next, the autonomous path generation unit 106 generates a virtual space that reproduces an actual driving environment from the recognition results. Then, the autonomous path generation unit 106 generates a planned driving trajectory of the host vehicle in this virtual space.

[0030] The predetermined environment specifying unit 107 specifies the predetermined environment. The predetermined environment is a road structure (hereinafter, a predetermined road structure) that is estimated to reduce the accuracy of identifying the vehicle's path based only on the sensing results of the surrounding monitoring sensor 14 and the sensing results of the map information. The processing in the predetermined environment specifying unit 107 corresponds to a predetermined environment specifying step. The predetermined environment specifying unit 107 may specify, for example, a predetermined road structure ahead of the vehicle's path. The predetermined road structure that the predetermined environment specifying unit 107 specifies may be an intersection, a branch point, a curved road with a certain degree of curvature or more (hereinafter, a target curved road), an uphill slope road with a certain degree of gradient or more (hereinafter, a target slope road), etc. The "certain" referred to here may be a value that is estimated to reduce the accuracy of identifying the vehicle's path based only on the sensing results. The degree of curvature of a curved road may be expressed by either a curvature or a curvature radius. An intersection corresponds to a predetermined road structure because an area without a dividing line exists. A branch point corresponds to a predetermined road structure because an area without a dividing line may exist. The branching point also includes a branching point where the number of lanes increases. The target curved road and the target sloped road correspond to the predetermined road structure because the dividing lines may be difficult to detect by the perimeter monitoring sensor 14. The predetermined environment specification unit 107 may specify the predetermined road structure ahead of the vehicle's path by using the map information acquired by the map acquisition unit 101 and the vehicle's position specified by the position specification unit 103.

[0031] The road identification unit 108 identifies the road of the vehicle. The process in the road identification unit 108 corresponds to a road identification step. The road identification unit 108 can use both the sensing result acquired by the sensor acquisition unit 104 and the map information acquired by the map acquisition unit 101 to identify the road. The road identification unit 108 switches to identifying the road using only the map information of the sensing result and the map information (hereinafter, map road identification) based on the specified road structure identified by the specified environment identification unit 107. According to this, even if the specified road structure is estimated to reduce the identification accuracy of the road of the vehicle in the sensing result, it is possible to identify the road with high accuracy using the map information. On the other hand, if the road structure is not the specified road structure, it is possible to identify the road with high accuracy using the sensing result. As a result, even if the perimeter monitoring sensor 14 and map information are used to identify the road of the vehicle, it is possible to identify the road with high accuracy.

[0032] In addition, the identification of a road using only the sensing result out of the sensing result and the map information is hereinafter referred to as autonomous road identification. In addition, the identification of a road using both the sensing result and the map information is hereinafter referred to as fusion road identification. The road identification unit 108 may switch from fusion road identification to map road identification based on the specified road structure identified by the specified environment identification unit 107. The road identification unit 108 may switch to map road identification from autonomous road identification via fusion road identification. The road identification unit 108 may switch to map road identification without going from autonomous road identification to fusion road identification.

[0033] It is preferable that the road identification unit 108 switches from autonomous road identification to map road identification based on the specified road structure identified by the specified environment identification unit 107. This is because, if the road structure is not the specified one, it is more likely that the road can be identified more accurately by using the sensing result rather than the map information. According to this, if the road structure is not the specified one, it is possible to identify the road more accurately by using only the sensing result out of the sensing result and the map information.

[0034] It is preferable that the road identification unit 108 switches from autonomous road identification to map road identification via fusion road identification based on the specified road structure identified by the specified environment identification unit 107. This makes it possible to suppress abrupt changes in the identified road compared to switching to map road identification without going through fusion road identification.

[0035] It is preferable that the road identification unit 108 switches from autonomous road identification to map road identification via fusion road identification as the distance from the vehicle to the predetermined road structure becomes shorter based on the predetermined road structure identified by the predetermined environment identification unit 107. It is also preferable that the road identification unit 108 switches from map road identification to autonomous road identification via fusion road identification as the distance from the vehicle to the predetermined road structure becomes longer. This makes it possible to maintain a higher level of road identification accuracy while suppressing steep changes in the road until the vehicle approaches, passes through, and leaves the predetermined road structure.

[0036] Here, an example of switching of lane specification according to the distance from the vehicle to a predetermined road structure will be described with reference to Figs. 3 and 4. Fig. 3 is a diagram showing an example of area division based on an intersection. HV in Fig. 3 indicates the vehicle. In the example of Fig. 3, an example is shown in which the predetermined road structure is an intersection. The distance from the vehicle to the intersection may be the distance from the vehicle position to the center coordinate of the intersection, or the distance from the vehicle position to the node coordinate of the intersection. In the following, the distance from the vehicle to the intersection (hereinafter, intersection distance) is set to X. In addition, in the case of the distance from the vehicle to a branch point, for example, the distance from the vehicle position to the branch point coordinate may be used. In the case of the distance from the vehicle to a target curved road, the distance may be used to the nearest end coordinate of a link or segment of the target curved road. In the case of the distance from the vehicle to a target sloped road, the distance may be used to the nearest end coordinate of a link or segment of the target sloped road.

[0037] As shown in FIG. 3, R1 is the area before the intersection that is more than the set distance D from the intersection. R2 is the area before the intersection that is less than the set distance D from the intersection. R3 is the area inside the intersection. R4 is the area after the intersection that is less than the set distance D from the intersection. R5 is the area after the intersection that is more than the set distance D from the intersection. The set distance D can be set to any distance.

[0038] Next, FIG. 4 is a diagram showing an example of switching of lane identification by area shown in FIG. 3. In FIG. 4, the sensing result is represented by "autonomous" and the map information is represented by "map". In FIG. 4, the sensing result and the map information used to identify the lane are indicated by a circle. As shown in FIG. 4, when the vehicle is located in the area R1, the lane identification unit 108 identifies the lane using only the sensing result of the sensing result and the map information. When the vehicle is located in the area R2, the lane identification unit 108 identifies the lane using both the sensing result and the map information. When the vehicle is located in the area R3, the lane identification unit 108 identifies the lane using only the map information of the sensing result and the map information. When the vehicle is located in the area R4, the lane identification unit 108 identifies the lane using both the sensing result and the map information. When the vehicle is located in the area R5, the road identification unit 108 identifies the road using only the sensing result out of the sensing result and the map information.

[0039] Furthermore, the switching of lane identification according to state transition is as follows. The lane identification unit 108 performs autonomous lane identification while the host vehicle position is R1. When the host vehicle position transitions from R1 to R2, the lane identification unit 108 switches from autonomous lane identification to fusion lane identification. When the host vehicle position transitions from R2 to R3, the lane identification unit 108 switches from fusion lane identification to map lane identification. When the host vehicle position transitions from R2 to R3, the lane identification unit 108 performs map lane identification while the host vehicle position is R3. When the host vehicle position transitions from R3 to R4, the lane identification unit 108 switches from map lane identification to fusion lane identification. When the host vehicle position transitions from R3 to R4, the lane identification unit 108 performs fusion lane identification while the host vehicle position is R4. The road identification unit 108 also switches from map road identification to fusion road identification when the vehicle position transitions from R3 to R2. The road identification unit 108 switches from fusion road identification to autonomous road identification when the vehicle position transitions from R4 to R5. The road identification unit 108 performs autonomous road identification while the vehicle position is R5.

[0040] When performing autonomous road specification, the road specification unit 108 may specify the road generated by the autonomous road generation unit 106 as the road of the vehicle. When performing map road specification, the road specification unit 108 may specify the road generated by the map road generation unit 105 as the road of the vehicle. When performing fusion road specification, the road specification unit 108 may specify the road obtained by integrating the road generated by the autonomous road generation unit 106 and the road generated by the map road generation unit 105 as the road of the vehicle. This integration may be performed by using the average value (hereinafter, the integrated average value) of the parameters of the road generated by the autonomous road generation unit 106 and the parameters of the road generated by the map road generation unit 105. The integrated average value may be calculated by arithmetic mean or weighted average. When using a fitting curve expressed by a two-variable polynomial as disclosed in Patent Document 1 as the road, it is sufficient to use the integrated average value for all of the linear term, quadratic term, and constant term. In addition to using the fitting curve, a configuration may be used in which a set of center points of passable areas ahead of the road is identified as the road. When performing fusion road identification in this configuration, it is sufficient to perform integration using the integrated average value of the center points of the road generated by the autonomous road generation unit 106 and the center points of the road generated by the map road generation unit 105.

[0041] When performing fusion road identification, it is preferable that the road identification unit 108 increases the ratio of map information used to identify the road as the distance from the vehicle to the specified road structure becomes shorter. This makes it possible to make the change in road smoother within the section where fusion road identification is performed. As an example, the road identification unit 108 may perform weighted integration shown in the formula: fusion road identification parameter = (X / D) x autonomous road identification parameter + (1-(X / D)) x map road identification parameter. X is the intersection distance mentioned above. D is the set distance mentioned above.

[0042] The road identified by the road identification unit 108 is used for automatic driving by the automatic driving ECU 10. For example, it is used for driving control during automatic driving along the road. Note that the road identified by the road identification unit 108 may be used for driving assistance other than driving control. For example, it may be used for driving assistance that provides audio and display guidance for driving along the road.

[0043] <Route identification-related processing in autonomous driving ECU 10> Here, an example of the flow of processing related to the identification of the vehicle's road in the automatic driving ECU 10 (hereinafter, road identification related processing) will be described using the flowchart of FIG. 5. The flowchart of FIG. 5 may be configured to be started when, for example, a switch (hereinafter, a power switch) for starting the internal combustion engine or motor generator of the vehicle is turned on. The flowchart of FIG. 5 may also be configured to be started when a driving assistance function using the road identified by the road identification unit 108 is turned on. In addition, the flowchart of FIG. 5 may also be configured to be started when the automation level of the vehicle reaches a specific level or higher, which is LV1 or higher. The specific level may be set arbitrarily. In FIG. 5, an example in which the predetermined road structure is an intersection will be described.

[0044] First, in step S1, the autonomous path generation unit 106 generates a path for the host vehicle using the sensing results acquired by the sensor acquisition unit 104. In step S2, the map path generation unit 105 generates a path for the host vehicle using the map information acquired by the map acquisition unit 101 and the host vehicle position identified by the position identification unit 103. The process of S1 and the process of S2 may be performed in reverse order, or may be performed in parallel.

[0045] In step S3, the predetermined environment determination unit 107 uses the map information acquired by the map acquisition unit 101 and the vehicle position identified by the position identification unit 103 to identify an intersection ahead on the path of the vehicle. In step S4, if the intersection distance X, which is the distance from the vehicle position identified by the position identification unit 103 to the intersection identified in S3, is equal to or greater than the set distance D (YES in S4), the process proceeds to step S5. On the other hand, if the intersection distance X is less than the set distance D (NO in S4), the process proceeds to step S6. The intersection distance X may be calculated, for example, by the predetermined environment determination unit 107 from the map information acquired by the map acquisition unit 101 and the vehicle position identified by the position identification unit 103.

[0046] In step S5, the road identification unit 108 performs autonomous road identification, and the process proceeds to step S10. In step S6, if it is determined that the vehicle is in an intersection (YES in S6), the process proceeds to step S9. On the other hand, if it is determined that the vehicle is not in an intersection (NO in S6), the process proceeds to step S7. Whether the vehicle is in an intersection may be determined, for example, by the predetermined environment identification unit 107 based on the intersection distance X. When the coordinates of the intersection center are used as the coordinates of the intersection, whether the vehicle is in an intersection may be determined, for example, based on whether the intersection distance X is less than the distance from the intersection center to the intersection entrance / exit. When the node coordinates of the intersection are used as the coordinates of the intersection, whether the vehicle is in an intersection may be determined, for example, based on whether the intersection distance X is 0.

[0047] In step S7, the lane identification unit 108 performs weighting integration of the parameters of the lane generated in S1 and the parameters of the lane generated in S2 according to the intersection distance X. In detail, weighting integration shown in the formula of fusion lane identification parameters = (X / D) × autonomous lane identification parameters + (1-(X / D)) × map lane identification parameters may be performed.

[0048] In step S8, the lane identification unit 108 performs fusion lane identification using the parameters that have been weighted and integrated in S7, and the process proceeds to step S10. Meanwhile, in step S9, the lane identification unit 108 performs map lane identification, and the process proceeds to step S10.

[0049] In step S10, if it is time to end the road identification-related process (YES in S10), the road identification-related process is terminated. On the other hand, if it is not time to end the road identification-related process (NO in S10), the process returns to S1 and is repeated. Examples of timing to end the road identification-related process include when the power switch of the vehicle is turned off, when the automation level of the vehicle becomes LV0, etc.

[0050] According to the configuration of the first embodiment, the route is identified by switching between the sensing result of the perimeter monitoring sensor 14 and map information based on the specified road structure. This makes it possible to identify the route using more accurate information. In addition, since it becomes possible to identify the route with high accuracy, the stability and continuity of the driving assistance function such as automatic driving are improved.

[0051] (Embodiment 2) In the first embodiment, the predetermined environment identified by the predetermined environment identification unit 107 is a predetermined road structure, but this is not necessarily limited to this. For example, the following configuration of the second embodiment may be used. An example of the configuration of the second embodiment will be described below with reference to the drawings. The vehicle system 1 of the second embodiment is similar to the vehicle system 1 of the first embodiment, except that the vehicle system 1 of the second embodiment includes an autonomous driving ECU 10a instead of the autonomous driving ECU 10.

[0052] <Outline of the configuration of the autonomous driving ECU 10a> Here, a schematic configuration of the autonomous driving ECU 10a for identifying the vehicle's route will be described with reference to FIG. 6. As shown in FIG. 6, the autonomous driving ECU 10a includes a map acquisition unit 101, a vehicle information acquisition unit 102, a position identification unit 103, a sensor acquisition unit 104, a map route generation unit 105, an autonomous route generation unit 106, a predetermined environment identification unit 107a, a route identification unit 108a, a traffic jam acquisition unit 109, and a dynamic environment determination unit 110 as functional blocks for identifying the route. The autonomous driving ECU 10a is similar to the autonomous driving ECU 10 of the first embodiment, except that the autonomous driving ECU 10a includes the predetermined environment identification unit 107a and the route identification unit 108a instead of the predetermined environment identification unit 107 and the route identification unit 108, and includes the traffic jam acquisition unit 109 and the dynamic environment determination unit 110. This autonomous driving ECU 10a also corresponds to a route identification device. Furthermore, the execution of the processing of each functional block of the autonomous driving ECU 10a by the computer corresponds to the execution of the road identification method.

[0053] The congestion acquisition unit 109 acquires congestion information. The congestion acquisition unit 109 acquires the congestion information from a server external to the vehicle via the communication module 11. The congestion information is information indicating the degree of congestion on a road. The degree of congestion may be a two-level level indicating whether or not there is congestion, or may be a level of three or more levels. For example, the congestion information may be information indicating the degree of congestion for each link.

[0054] The dynamic environment determination unit 110 interprets the situation indicated by the sensing result acquired by the sensor acquisition unit 104. The situation interpreted by the dynamic environment determination unit 110 is the degree of proximity between the host vehicle and other vehicles, the degree of obstruction of the lane markings by other vehicles, etc. The degree of proximity between the host vehicle and other vehicles may be, for example, the distance between the host vehicle and a vehicle ahead of the host vehicle, etc. The degree of obstruction of the lane markings by other vehicles may be, for example, the length of the lane markings that can be recognized from the sensing result without being obstructed by other vehicles.

[0055] The predetermined environment determination unit 107a determines a situation in which the sensing range of the perimeter monitoring sensor 14 is estimated to be blocked by a vehicle ahead of the vehicle by a specified range or more (hereinafter, the predetermined situation). The processing in the predetermined environment determination unit 107a corresponds to the predetermined environment determination processing. The predetermined range referred to here is a range in which the accuracy of identifying the road in the sensing result is estimated to decrease, and may be set arbitrarily. The predetermined environment determination unit 107a is similar to the predetermined environment determination unit 107 in the first embodiment, except that it determines the predetermined situation instead of the specified road structure.

[0056] The predetermined environment identification unit 107a may identify a traffic jam as the predetermined situation. This is because a traffic jam narrows the distance between the vehicle and the vehicle ahead, making it easier for the vehicle ahead to block the lane markings. The predetermined environment identification unit 107a may identify a predictable traffic jam (hereinafter, predictable traffic jam) as the predetermined situation from the traffic jam information acquired by the traffic jam acquisition unit 109. The predetermined environment identification unit 107a may also identify a certain amount of blocking of the lane markings by the vehicle ahead (hereinafter, blocking situation) or a traffic jam that is difficult to predict in advance (hereinafter, difficult to predict traffic jam) as the predetermined situation from the situation interpreted by the dynamic environment determination unit 110.

[0057] The lane identification unit 108a switches to map lane identification based on the specified situation identified by the specified environment identification unit 107a. The processing in the lane identification unit 108a also corresponds to lane identification processing. According to this, even in a specified situation where the sensing result is estimated to reduce the identification accuracy of the lane of the vehicle, it is possible to accurately identify the lane using the map information. On the other hand, if the specified situation is not present, it is possible to accurately identify the lane using the sensing result. As a result, even when the perimeter monitoring sensor 14 and map information are used to identify the lane of the vehicle, it is possible to identify the lane with higher accuracy. The lane identification unit 108a is similar to the lane identification unit 108 of the first embodiment, except that it switches to map lane identification based on the identification of a specified situation instead of the identification of a specified road structure.

[0058] It is preferable that the lane identification unit 108a switches from autonomous lane identification to map lane identification via fusion lane identification as the distance from the vehicle to the section corresponding to the predictable congestion becomes shorter based on the predictable congestion as the predetermined situation identified by the predetermined environment identification unit 107a. It is also preferable that the lane identification unit 108a switches from autonomous lane identification via fusion lane identification as the distance from the vehicle to the section corresponding to the predictable congestion becomes longer. This makes it possible to maintain a higher level of lane identification accuracy while suppressing abrupt changes in the lane until the vehicle approaches, passes through, and leaves the congestion section.

[0059] The switching of lane specification according to the distance from the vehicle to the section of predictable congestion may be performed in the same manner as the switching of lane specification according to the distance from the vehicle to the specified road structure, as described in the first embodiment. The distance from the vehicle to the section of predictable congestion may be the distance from the vehicle position to the nearest end coordinates of the link corresponding to the predictable congestion. If the section of predictable congestion spans multiple links, the coordinates of the start and end points of the section consisting of the multiple links as a single group may be treated as the end coordinates.

[0060] In addition, the roadway identification unit 108a may switch from autonomous roadway identification to map roadway identification via fusion roadway identification as the distance traveled increases after the predetermined environment identification unit 107a identifies the obstruction situation and the difficult-to-predict congestion. In addition, the roadway identification unit 108a may switch from autonomous roadway identification to map roadway identification via fusion roadway identification as the distance traveled increases after the predetermined environment identification unit 107a no longer identifies the obstruction situation and the difficult-to-predict congestion. This makes it possible to maintain a high level of roadway identification accuracy while suppressing abrupt changes in the roadway until the vehicle enters, passes through, and leaves the section corresponding to the obstruction situation and the difficult-to-predict congestion.

[0061] According to the configuration of the second embodiment, the route is identified by switching between the sensing result of the perimeter monitoring sensor 14 and map information based on the identification of a predetermined situation. Therefore, it is possible to identify the route using more accurate information. In addition, since it is possible to identify the route with high accuracy, the stability and continuity of the driving assistance function such as automatic driving are improved.

[0062] (Embodiment 3) In the above embodiment, the function of the road identification device is performed by the autonomous driving ECU 10, 10a, but this is not necessarily limited to the above. For example, an electronic control device other than the autonomous driving ECU 10, 10a may perform the function of the road identification device.

[0063] The present disclosure is not limited to the above-described embodiment, and various modifications are possible within the scope of the claims. The embodiments obtained by appropriately combining the technical means disclosed in the different embodiments are also included in the technical scope of the present disclosure. The control unit and the method described in the present disclosure may be realized by a dedicated computer constituting a processor programmed to execute one or more functions embodied by a computer program. Alternatively, the device and the method described in the present disclosure may be realized by a dedicated hardware logic circuit. Alternatively, the device and the method described in the present disclosure may be realized by one or more dedicated computers configured by a combination of a processor that executes a computer program and one or more hardware logic circuits. The computer program may be stored in a computer-readable non-transient tangible recording medium as instructions executed by the computer. [Explanation of symbols]

[0064] 1 Vehicle system, 10, 10a Autonomous driving ECU (Route identification device), 14 Surrounding monitoring sensor, 101 Map acquisition unit, 104 Sensor acquisition unit, 107, 107a Predetermined environment identification unit, 108, 108a Route identification unit

Claims

1. A sensor acquisition unit (104) that acquires a sensing result from a surroundings monitoring sensor (14) used in a vehicle; A map acquisition unit (101) for acquiring map information; A route identification unit (108, 108a) capable of identifying a route of the vehicle by using both the sensing result acquired by the sensor acquisition unit and the map information acquired by the map acquisition unit; a predetermined environment specification unit (107, 107a) that specifies a predetermined environment in which it is estimated that the accuracy of specifying the road is reduced in the sensing result out of the sensing result and the map information; The road identification unit switches to identification of the road using only the map information of the sensing result and the map information based on the fact that the predetermined environment has been identified by the predetermined environment identification unit, The predetermined environment specifying unit specifies, as the predetermined environment, a predetermined road structure that is at least one of an intersection, a branch point, a curved road with a certain degree of curvature or more, and an uphill road with a certain degree of gradient or more; The path identification unit switches from identifying the path using only the sensing results of the sensing results and the map information, via identifying the path using both the sensing results and the map information, to identifying the path using only the map information of the sensing results and the map information, as the distance from the vehicle to the specified road structure becomes shorter based on the specified road structure identified by the specified environment identification unit, while switching from identifying the path using only the map information of the sensing results and the map information, via identifying the path using both the sensing results and the map information, to identifying the path using only the sensing results of the sensing results and the map information, as the distance from the vehicle to the specified road structure becomes longer.

2. The lane identification device according to claim 1, The path identification unit is a path identification device that switches from identifying the path using only the sensing results of the sensing results and the map information to identifying the path using only the map information of the sensing results and the map information based on the specified environment being identified by the specified environment identification unit.

3. The lane identification device according to claim 2, The path identification unit, based on the identification of the specified environment by the specified environment identification unit, switches from identifying the path using only the sensing results of the sensing results and the map information, to identifying the path using both the sensing results and the map information, to identifying the path using only the map information of the sensing results and the map information.

4. The lane identification device according to any one of claims 1 to 3, The road identification unit is a road identification device that, when identifying the road using both the sensing results and the map information, increases the proportion of the map information used in identifying the road as the distance from the vehicle to the specified road structure becomes shorter.

5. Executed by at least one processor, A sensor acquisition process for acquiring a sensing result from a surroundings monitoring sensor (14) used in the vehicle; a map acquisition step of acquiring map information; a route identification process capable of identifying a route of the vehicle by using both the sensing result acquired in the sensor acquisition process and the map information acquired in the map acquisition process; a predetermined environment specifying step of specifying a predetermined environment in which it is estimated that the accuracy of specifying the road is reduced in the sensing result out of the sensing result and the map information, In the path identification step, based on the fact that the predetermined environment has been identified in the predetermined environment identification step, a step of switching to identification of the path using only the map information out of the sensing result and the map information is performed; In the predetermined environment specification step, a predetermined road structure is specified as the predetermined environment, which is at least one of an intersection, a branch point, a curved road with a certain degree of curvature or more, and an uphill road with a certain degree of gradient or more; In the path identification process, based on the specified road structure identified in the specified environment identification process, as the distance from the vehicle to the specified road structure becomes shorter, the path identification process switches from using only the sensing results of the sensing results and the map information to using both the sensing results and the map information, and then to using only the map information of the sensing results and the map information, while as the distance from the vehicle to the specified road structure becomes longer, the path identification process switches from using only the map information of the sensing results and the map information to using both the sensing results and the map information, and then to using only the sensing results of the sensing results and the map information,

Images