Vehicle driving control device

Abstract

A vehicle driving control device is provided, which can continue automatic driving with excellent driving stability without hindering other vehicles even when the left and right dividing lines dividing the lane in which the vehicle is traveling cannot be identified. When traveling on a snowy road where the dividing lines cannot be identified, the driving lane of the vehicle M is estimated based on the vehicle position received by a GNSS receiver (14) and a high-precision road map (S1), the lateral position x_i of surrounding vehicles based on the vehicle M is calculated (S2), the driving lanes of the surrounding vehicles are estimated based on the lateral position x_i and the high-precision road map (S3), the offset amount i_offset of the surrounding vehicles is statistically calculated (S15), and when it is determined that the surrounding vehicles are offset toward the road shoulder (S16, S18), the offset amount d_offset for offsetting the vehicle M toward the road shoulder is calculated (S7), and the target travel path for automatic driving is reset (S9, S10).

Description

Technical Field

[0001] The present invention relates to a vehicle travel control device capable of continuing automatic driving even when left and right dividing lines dividing a lane in which the vehicle is traveling are covered with snow and cannot be recognized. Background Art

[0002] The vehicle's onboard driving assistance unit maps the vehicle's location onto a high-precision road map (dynamic map) based on location information received from positioning satellites, including GPS satellites and GNSS (Global Navigation Satellite System) satellites. Furthermore, if the passenger (primarily the driver) sets a destination on the high-precision road map, the driving assistance unit constructs a driving route connecting the vehicle's location and the destination.

[0003] The target path for the vehicle to follow the driving route is then set several kilometers away from the vehicle. The high-precision road map contains road information necessary for autonomous driving. This road information includes information such as the number of lanes (two or three lanes), road width, and curvature of curves. The driving assistance unit sets the target path for the vehicle to stay in the center of the selected lane based on the road information in the high-precision road map.

[0004] Therefore, as disclosed in Patent Document 1 (Japanese Patent Application Laid-Open No. 2016-181015), for example, it is possible to guide the vehicle along a target travel path constructed using high-precision road map information and vehicle position information estimated based on position information received from positioning satellites. As a result, even when the vehicle's lane is covered with snow, making it impossible to directly identify the left and right lane dividing lines using sensors such as cameras, driving assistance control can continue.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2016-181015

[0008] Patent Document 2: Japanese Patent Application Laid-Open No. 2018-41194 Summary of the Invention

[0009] Technical issues

[0010] However, when driving on a drivable road with accumulated snow after snow removal (hereinafter referred to as a "snowy road"), drivers tend to drive closer to the shoulder of the road, as they need to be careful of vehicles approaching from the opposite lane if there is no central divider, or they need to be wary of vehicles approaching in the same direction to prevent being overtaken by approaching vehicles. Furthermore, even if there are actually three lanes, if snow accumulates on the shoulder, the driver may use up most of the road width of one lane, driving as if it were a two-lane road.

[0011] The target path set based on a high-precision road map is usually set in the center of the vehicle's lane on a snow-free road. Therefore, when driving on snowy roads, the driver's vehicle and surrounding vehicles may choose different routes, potentially hindering the movement of other vehicles.

[0012] As a countermeasure, for example, as disclosed in Patent Document 2 (Japanese Patent Gazette No. 2018-41194), it is also possible to consider finding the driving trajectory of the center of the vehicle width direction of the preceding vehicle, and generating a target travel path for the vehicle based on the driving trajectory, so that the vehicle travels along the target travel path.

[0013] However, if the target travel path of the vehicle is set based on the driving trajectory of the preceding vehicle, the vehicle will actually travel along the route preferred by the driver of the preceding vehicle. If the preceding vehicle travels differently from surrounding vehicles, the travel of surrounding vehicles will be hindered.

[0014] In view of the above situation, the present invention aims to provide a vehicle driving control device that does not hinder the driving of other vehicles and can continue driving control with excellent driving stability even when the left and right dividing lines dividing the lane in which the vehicle is traveling cannot be recognized.

[0015] Technical Solution

[0016] The present invention provides a vehicle driving control device, comprising: a surrounding environment information acquisition unit, which is mounted on the vehicle and acquires the surrounding environment information of the vehicle; a map information storage unit, which stores road map information; a vehicle position estimating unit, which estimates the vehicle position of the vehicle; a road surface information acquisition unit, which acquires information about the road surface on which the vehicle is traveling; and a target travel path setting unit, which sets the target travel path of the vehicle during automatic driving based on the vehicle position information estimated by the vehicle position estimating unit and the input destination information, with reference to the road map information stored in the map information storage unit. The vehicle driving control device also comprises: a surrounding vehicle lateral position information estimating unit, which, when the road surface acquired by the road surface information acquiring unit is a snow-covered road surface and the left and right lanes dividing the lane on which the vehicle is traveling cannot be visually recognized. In the case of a dividing line, the lateral position information of a surrounding vehicle traveling on the road on which the host vehicle is traveling is estimated relative to the driving lane on the road map information stored in the map information storage unit; an offset determination unit determines whether the surrounding vehicle is offset relative to the driving lane based on the lateral position information of the surrounding vehicle estimated by the surrounding vehicle lateral position information estimation unit; and a target travel path correction unit sets an offset amount for offsetting the host vehicle toward the dividing line on the shoulder side of the lane on which the host vehicle is traveling on the road map information stored in the map information storage unit, when the offset determination unit determines that the surrounding vehicle is offset. The lateral position of the target travel path set by the target travel path setting unit is corrected using the offset amount to set a new target travel path.

[0017] Technical Effects

[0018] According to the present invention, when the host vehicle is traveling on a snowy road surface where the left and right dividing lines dividing the lanes cannot be visually identified, the lateral position information of surrounding vehicles traveling on the road where the host vehicle is traveling relative to the driving lane on the road map information is estimated, and based on the lateral position information, it is determined whether the surrounding vehicles are offset from the driving lane. If it is determined that the surrounding vehicles are offset, an offset amount is set to offset the host vehicle toward the dividing line on the shoulder side of the lane where the host vehicle is traveling on the road map information, and the lateral position of the target travel path is corrected using the offset amount, and a new target travel path is set. Therefore, even in a situation where the left and right dividing lines dividing the lane where the host vehicle is traveling cannot be identified, the travel of other vehicles is not hindered, and driving control with excellent driving stability can continue. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a brief diagram of the entire automatic driving assistance system.

[0020] Figure 2This is a functional block diagram of a driving assistance unit installed in a vehicle.

[0021] Figure 3 is a flowchart showing a target travel path correction routine.

[0022] Figure 4 1 is a flowchart showing a surrounding vehicle statistical deviation estimation subroutine.

[0023] Figure 5 is a flowchart showing a drivable area verification subroutine.

[0024] Figure 6 It is an explanatory diagram showing a traveling state when traveling on a road without snow using traveling control.

[0025] Figure 7 This is an explanatory diagram showing a driving state when driving on a snowy road where the left and right dividing lines cannot be recognized due to driving control.

[0026] Figure 8 This is an explanatory diagram showing a driving state in which a snow wall protrudes toward the dividing line on the road shoulder side.

[0027] Figure 9 This is an explanatory diagram of a vehicle traveling on a two-lane snow-covered road together with surrounding vehicles.

[0028] Figure 10 This is an explanatory diagram of a state in which a three-lane snowy road is driven as a two-lane road.

[0029] Figure 11 It is a distribution diagram showing the trend of the travel routes taken by traveling vehicles.

[0030] Figure 12 It is an explanatory diagram showing the offset amount of a vehicle when traveling on a snowy road.

[0031] Figure 13 This is a flowchart showing a characteristic portion of a routine for setting the lateral position of the vehicle based on another method.

[0032] Figure 14 This is a flowchart showing a characteristic portion of a routine for setting the lateral position of the vehicle based on another method.

[0033] Explanation of symbols

[0034] 1: Cloud Server

[0035] 2: Traffic Information Center

[0036] 3: Base Station

[0037] 4: Weather Information Center

[0038] 5. Internet

[0039] 6: Driving control device

[0040] 11: Positioning unit

[0041] 12: Map positioning calculation unit

[0042] 12a: Vehicle position estimation calculation unit

[0043] 12b: Driving route / target travel path setting calculation unit

[0044] 13: Road information transceiver

[0045] 14: GNSS receiver

[0046] 15: Autonomous driving sensor

[0047] 16: Route information input unit

[0048] 17: High-precision road map database

[0049] 21: Camera unit

[0050] 21a: Main camera

[0051] 21b: Secondary camera

[0052] 21c: Image Processing Unit (IPU)

[0053] 21d: Forward driving environment recognition unit

[0054] 22: Peripheral monitoring unit

[0055] 22a: Surrounding environment recognition sensor

[0056] 22b: Surrounding environment recognition department

[0057] 31: Steering control unit

[0058] 32: Brake control unit

[0059] 33: Acceleration and deceleration control unit

[0060] 34: Notification device

[0061] A: Offset processing unit

[0062] F: Following vehicle

[0063] Fr: adjacent following vehicle

[0064] F_off: offset determination flag

[0065] M: This vehicle

[0066] P: Vehicle ahead

[0067] Pr: adjacent preceding vehicle

[0068] TYl, TYs: tire

[0069] W: width of snowy road

[0070] d_sl: offset determination threshold

[0071] d_offset, f_offset: offset

[0072] i_offset: statistical offset

[0073] n: number of lanes

[0074] wd: surplus value

[0075] x_i: horizontal position

[0076] x_j: lateral position deviation

[0077] x_jsl: deviation judgment value DETAILED DESCRIPTION

[0078] Hereinafter, one embodiment of the present invention will be described with reference to the drawings. Figure 1 The automated driving assistance system shown includes a cloud server 1 serving as an external information aggregation device, various traffic information centers 2, a base station 3, and a weather information center 4, all of which are connected via the Internet 5. Furthermore, a driving control device 6 mounted on the vehicle M, which receives cloud information from the cloud server 1 via the base station 3, is also included in the automated driving assistance system.

[0079] Furthermore, each traffic information center 2, located within the jurisdiction of private and public institutions, collects and aggregates constantly changing traffic information (e.g., the number of vehicles traveling in each section of the country divided according to a predetermined method) and environmental information from probe information transmitted by probe vehicles (not shown), and transmits this information to the cloud server 1 as traffic information. This probe information includes the probe vehicle's vehicle ID, vehicle information (model, width, etc.), transmission history, and the probe vehicle's intended travel route. The transmission history includes the transmission time (year, month, day, time), the location at the time of transmission (latitude, longitude), the vehicle's speed, and the direction of travel.

[0080] Furthermore, the public institution's traffic information center 2 aggregates snow removal information (such as snow removal areas and times) for each region during snowfall and transmits it to the cloud server 1. Meanwhile, the weather information center 4, located within the jurisdiction of private and public institutions, sequentially aggregates weather information for each region and the current snowfall amount (cm / h) and transmits it to the cloud server 1.

[0081] Based on traffic information and snow removal information transmitted from traffic information centers 2 and weather information transmitted from weather information center 4, cloud server 1 aggregates traffic information, snow removal information, weather information, and road surface information (dry road surface, snow-covered road surface) for each section from the current time to a predetermined time. This information is then stored as cloud information for each pre-defined region (section) on a global dynamic map maintained by cloud server 1, and is continuously updated.

[0082] The global dynamic map described above has a four-layer structure, with the bottommost static information layer as the foundation, upon which additional map information required to support autonomous driving is layered. The static information layer is high-precision three-dimensional map information. It is the bottommost basic information layer, storing minimally changing static information such as road shape (curvature, etc.), lane information (number of lanes, lane width, etc.), three-dimensional structures (guardrails, shoulder walls, etc.), and permanent regulatory information.

[0083] The additional map information overlaid on this static information layer is divided into three layers: from the bottom up, a quasi-static information layer, a quasi-dynamic information layer, and a dynamic information layer. These layers are divided based on the degree of change (variability) on the time axis. Cloud information such as traffic information, weather information (e.g., snowfall information), snow removal information, and road surface information (dry roads, snow-covered roads) are the most variable and require real-time updates, so they are stored in the dynamic information layer. It should be noted that this global dynamic map is the road map required for autonomous vehicles to operate autonomously.

[0084] The cloud server 1 transmits information required for autonomous driving to the autonomous driving vehicle via the base station 3. In this embodiment, the autonomous driving vehicle is described as an example of a vehicle (host vehicle) M in which the driver (in manual driving) is riding.

[0085] The vehicle M is equipped with a driving control device 6 for autonomous driving within an automated driving zone (e.g., a highway) without driver input. The driving control device 6 includes a positioning unit 11 and an automated driving control unit 26. The positioning unit 11 is connected to a road information transceiver 13 and a GNSS (Global Navigation Satellite System) receiver 14. The positioning unit 11 estimates the vehicle's position (latitude and longitude) based on positioning signals received by the GNSS receiver 14 from multiple positioning satellites.

[0086] Furthermore, the positioning unit 11 accesses the cloud server 1 from the road information transceiver 13 via the base station 3 and the internet 5 to obtain various information required for autonomous driving, as well as map information stored on a global dynamic map. Based on the map information received from the road information transceiver 13, the positioning unit 11 then performs map matching on the vehicle's position on the map and constructs a driving route connecting the input destination and the vehicle's position. Furthermore, the target path for autonomous driving is set several kilometers ahead of the vehicle M on the driving route constructed by the positioning unit 11.

[0087] like Figure 2 As shown, the positioning unit 11 of the driving control device 6 mounted on the host vehicle M includes a map positioning calculation unit 12 and a high-precision road map database 17 serving as a map information storage unit. The map positioning calculation unit 12, the forward driving environment recognition unit 21d and the surrounding environment recognition unit 22b (described later), and the automatic driving control unit 26 are composed of a well-known microcomputer including a CPU, RAM, ROM, and a non-volatile storage unit, and its peripheral devices. The ROM stores programs executed by the CPU and fixed data such as data tables.

[0088] The input side of the map positioning calculation unit 12 is connected to the aforementioned road information transceiver 13 and GNSS receiver 14, as well as to an autonomous driving sensor 15 and a route information input unit 16. The autonomous driving sensor 15 enables autonomous driving in environments such as tunnels, where the sensitivity of receiving information from GNSS satellites is low and positioning signals cannot be effectively received. It is composed of a vehicle speed sensor, a yaw rate sensor, and a longitudinal acceleration sensor.

[0089] The route information input unit 16 is a terminal device operated by a passenger (primarily the driver) and is capable of inputting a series of information required for setting a driving route in the map positioning operation unit 12, such as the destination and via points. Specifically, the route information input unit 16 is an input unit of a car navigation system (e.g., a touch panel on a monitor), a mobile terminal such as a smartphone, a personal computer, etc., and is connected to the map positioning operation unit 12 via a wired or wireless connection.

[0090] When the passenger operates the route information input unit 16 to input destination and via point information (facility name, address, telephone number, etc.), the input information is read by the map positioning calculation unit 12. If a destination or via point is input, the map positioning calculation unit 12 sets the position coordinates (latitude and longitude) of the destination or via point.

[0091] The map positioning calculation unit 12 includes: a vehicle position estimation calculation unit 12a as a vehicle position estimation unit for estimating the vehicle position information, and a driving route / target driving route setting calculation unit 12b for setting a driving route from the vehicle position to the destination (and the via point) and a target driving route for automatically driving the vehicle M in an automatic driving area (for example, a highway).

[0092] Furthermore, the high-precision road map database 17 is a large-capacity storage medium such as an HDD, storing known high-precision road map information (local dynamic map). This high-precision road map information has the same layered structure as the global dynamic map stored in the cloud server 1, sharing the underlying static information layer. Furthermore, overlaid on this bottom-level static information layer is additional map information required to support autonomous driving. This additional map information acquires the surrounding information required for the vehicle M to autonomously drive along a set driving route from the global dynamic map and is sequentially updated.

[0093] The host vehicle position estimation calculation unit 12a obtains the current position coordinates (latitude and longitude) of the host vehicle M based on the positioning signal received by the GNSS receiver 14. It then performs map matching on the high-precision road map information to estimate the vehicle's position on the road map (current position). Furthermore, in environments where the GNSS receiver 14's sensitivity decreases, preventing it from receiving valid positioning signals from positioning satellites, such as when traveling in a tunnel, the host vehicle position estimation calculation unit 12a switches to autonomous navigation and calculates the vehicle's travel distance and orientation based on the autonomous driving sensors 15 (such as vehicle speed detected by a speed sensor, yaw rate (yaw angular velocity) detected by a yaw rate sensor, and longitudinal acceleration detected by a longitudinal acceleration sensor) to perform local positioning.

[0094] The driving route / target route setting calculation unit 12b refers to the high-precision road map information stored in the high-precision road map database 17 based on the position information (latitude and longitude) of the vehicle position estimated by the vehicle position estimation calculation unit 12a and the position information (latitude and longitude) of the input destination (and via points). The driving route / target route setting calculation unit 12b constructs a driving route connecting the vehicle position and the destination (or, if via points have been set, the destination via the via points) on the high-precision road map information based on pre-set route conditions (such as a recommended route and a fastest route).

[0095] Next, a target route for autonomous driving of the host vehicle M is set several kilometers ahead of the host vehicle M. Items set as the target route include the lane in which the host vehicle M is to travel (for example, which lane the host vehicle is to travel in if there are three lanes), the lane change to overtake the preceding vehicle, and the time to initiate the lane change. Note that in this embodiment, the autonomous driving section is assumed to be a highway, and the target route is set to the lane on the shoulder side of the road. Therefore, the driving route / target route setting calculation unit 12b functions as the target route setting unit of the present invention.

[0096] Furthermore, the travel control device 6 includes a camera unit 21 that recognizes the travel environment in front of the host vehicle M and a surrounding monitoring unit 22 that monitors the travel environment around the host vehicle M.

[0097] The camera unit 21 includes an onboard camera (stereo camera) consisting of a main camera 21a and a sub-camera 21b fixed to the upper center of the front interior of the vehicle M and arranged in bilaterally symmetrical positions across the center in the vehicle width direction; an image processing unit (IPU) 21c; and a forward driving environment recognition unit 21d. The camera unit 21 captures reference image data using the main camera 21a and captures comparison image data using the sub-camera 21b.

[0098] The IPU 21c then processes these two image data sets in a predetermined manner. The forward driving environment recognition unit 21d reads the reference image data and comparison image data processed by the IPU 21c, identifies the same object in the two images based on their parallax, and calculates its distance data (the distance from the vehicle M to the object) using the principle of triangulation. This forward driving environment information, which serves as forward surrounding environment information, includes information such as the left and right dividing lines that separate the lane in which the vehicle M is traveling, preceding vehicles traveling in lanes ahead of or adjacent to the vehicle M, and the boundary between the snow wall formed by snow removal on the roadside and the snow-covered road surface.

[0099] Meanwhile, the surrounding monitoring unit 22 includes a surrounding environment recognition sensor 22a comprised of an ultrasonic sensor, millimeter-wave radar, light detection and ranging (LIDAR), a camera, or a combination thereof, and a surrounding environment recognition unit 22b that recognizes information about moving objects around the host vehicle M, i.e., surrounding environment information, based on signals from the surrounding environment recognition sensor 22a. The surrounding environment recognition sensor 22a detects moving objects (such as parallel vehicles, following vehicles, and following vehicles traveling in adjacent lanes) around the host vehicle M. It should be noted that these two units 21 and 22 constitute the surrounding environment information acquisition unit of the present invention.

[0100] The input side of the automatic driving control unit 26 is connected to the forward driving environment recognition unit 21d of the camera unit 21 and the surrounding environment recognition unit 22b of the surrounding monitoring unit 22. Furthermore, the automatic driving control unit 26 is connected to the map positioning calculation unit 12 via an in-vehicle communication line (e.g., CAN (Controller Area Network)) for bidirectional communication.

[0101] On the other hand, connected to the output side of the automatic driving control unit 26 are a steering control unit 31 for controlling the direction of travel of the host vehicle M, a braking control unit 32 for decelerating the host vehicle M through forced braking, an acceleration / deceleration control unit 33 for controlling the speed of the host vehicle M, and a notification device 34 such as a monitor and a speaker. When a target travel path is set by the map positioning calculation unit 12, the automatic driving control unit 26 controls the steering control unit 31, the braking control unit 32, and the acceleration / deceleration control unit 33 in a predetermined manner, and automatically drives the host vehicle M along the target travel path based on the positioning signal indicating the host vehicle's position received by the GNSS receiver 14.

[0102] However, in Figure 7 、 Figure 8 and Figure 12 The snow-covered road shown in the figure is piled up on the road shoulder when the snow is removed during snow removal, forming a snow wall. As a result, the snow-covered road surface (the road surface formed by snow removal) between the snow walls becomes a drivable area (usable space). Figure 12 As shown, the two ends of the width of the snow-covered road surface (snow-covered road surface width) W_free, which is the road width of the drivable area, are the boundaries between the snow-covered road surface and the snow wall. The boundary is based on the front driving environment information recognized by the front driving environment recognition unit 21d of the camera unit 21 and is detected according to brightness difference, etc.

[0103] The road width W_free in the drivable area is narrowed by the snow wall. Furthermore, the driver, driving the vehicle themselves, cannot visually identify the dividing line because it is covered by snow. Therefore, the snow wall serves as a guide for determining the vehicle's lateral position. In such circumstances, the driver tends to drive closer to the roadside to ensure driving stability and avoid contact with vehicles attempting to overtake them.

[0104] In addition, for example, Figure 10As shown in the figure, even if the road is actually three-lane, the road width W_free in the drivable area may become 2.5 lanes due to the snow wall protruding toward the driving lane. In such cases, the driver of a vehicle driving on a snowy road cannot visually recognize the dividing lines and tends to ensure a wider width for one lane, thus driving in the drivable area as two lanes.

[0105] Since the target travel path generated on the high-precision road map does not correspond to the road width W_free narrowed by snow accumulation, if the vehicle M is driven along the target travel path inconsistent with the actual travel path of other vehicles, it is possible to come into contact with the snow wall or hinder the travel of other vehicles.

[0106] Therefore, when the vehicle M is traveling on a snowy road, the automatic driving control unit 26 confirms the behavior of other vehicles and the position of the snow wall, and corrects and updates the target travel path set by the driving route / target travel path setting operation unit 12b of the map positioning operation unit 12 as planned in a manner so as not to hinder the driving of other vehicles when traveling on the snowy road and to continue automatic driving in a stable state.

[0107] Specifically, the update of the target travel path performed by the automatic driving control unit 26 is performed according to Figure 3 The target travel path correction routine shown is used to handle this.

[0108] In this routine, first, in step S1, road surface information indicating the road surface condition is acquired. Then, the routine proceeds to step S2, where the road surface information is used to determine whether the road surface is covered with snow. If so, the routine proceeds to step S3. Otherwise, if the road surface is clear of snow, the routine exits.

[0109] Regarding the road surface condition, for example, the road surface is determined to be snowy based on whether the left and right lane dividing lines separating the lanes in which the vehicle M is traveling are recognized based on the forward driving environment information detected by the forward driving environment recognition unit 21d of the camera unit 21, and if the lane dividing lines cannot be recognized based on the road surface brightness and the brightness is high. Alternatively, cloud information for the section in which the vehicle M is about to travel, stored in the cloud server 1, can be obtained to determine whether the road surface is snowy. It should be noted that the processing in step S1 corresponds to the road surface information acquisition unit of the present invention.

[0110] If the routine is exited directly from step S2, the automatic driving control unit 26 automatically drives the host vehicle M along the target travel path set by the travel route / target travel path setting calculation unit 12b of the map positioning calculation unit 12. Figure 6As shown, based on the forward driving environment information recognized by the forward driving environment recognition unit 21d of the camera unit 21, the left and right dividing lines are detected, and the lateral positional deviation between the center of the width (lane width) W_road between the left and right dividing lines and the center of the vehicle width W_car of the host vehicle M is calculated. Feedback correction is then performed on the target travel path set on the high-precision road map so that this lateral positional deviation is zero.

[0111] It should be noted that in this embodiment, the vehicle width W_car of a vehicle smaller than an ordinary vehicle with a relatively narrow vehicle width is set to the value obtained by adding the tire width to the outer side of the tire (winter tire) TYs, that is, the wheelbase. In addition, the vehicle width W_car of a large vehicle with a relatively wide vehicle width is set to the value obtained by subtracting the tire width from the inner side of the tire (winter tire) TYl, that is, the wheelbase. Figures 6 to 10 In FIG, for convenience, the vehicle M is shown as an ordinary vehicle.

[0112] Then, in step S3, the system acquires surrounding vehicle information, which indicates whether there are any other vehicles traveling nearby, based on the current location coordinates of the host vehicle M. This surrounding vehicle information is acquired based on the forward driving environment information recognized by the forward driving environment recognition unit 21d of the camera unit 21 and the surrounding environment information recognized by the surrounding environment recognition unit 22b of the surrounding monitoring unit 22. Alternatively, the information may be acquired through inter-vehicle communication between the host vehicle M and other vehicles.

[0113] In this case, if Figure 9 As shown, the surrounding vehicles include a preceding vehicle P traveling ahead of the vehicle M, a following vehicle F following behind the vehicle M, an adjacent preceding vehicle Pr traveling ahead in the adjacent lane on the opposite side of the road shoulder, and an adjacent following vehicle Fr traveling behind the adjacent lane on the opposite side of the road shoulder. Furthermore, although not shown, parallel vehicles traveling in adjacent lanes are also included in the surrounding vehicles.

[0114] Then, the routine proceeds to step S4. If no surrounding vehicles are detected, that is, if there is not even one vehicle traveling around the host vehicle, the routine is exited because it is not necessary to shift the host vehicle M toward the road shoulder. Otherwise, if surrounding vehicles are detected, the routine proceeds to step S5.

[0115] If the routine is exited directly from step S4, since the vehicle M is traveling on a snowy road, it is unable to obtain information about the dividing line based on the forward driving environment information recognized by the forward driving environment recognition unit 21d of the camera unit. Therefore, autonomous driving becomes radio navigation based on positioning signals from the GNSS receiver 14 and a high-precision road map. In this case, for example, even if the autonomous driving control unit 26 shifts the target path set by the driving route / target path setting calculation unit 12b of the map positioning calculation unit 12 toward the center of the road width W_free calculated based on the forward driving environment information recognized by the forward driving environment recognition unit 21d, this will not hinder the travel of other vehicles.

[0116] Furthermore, if the process proceeds to step S5, the lateral position information of each surrounding vehicle, that is, the offset of the center of the vehicle width direction relative to the center of the lane on the high-precision road map, is detected. The offset is detected according to Figure 4 The surrounding vehicle statistical deviation estimation subroutine shown is performed.

[0117] In this subroutine, first, in step S11 , the current position (latitude, longitude) of the host vehicle M is checked on the high-precision road map stored in the high-precision road map database 17 of the positioning unit 11 , and the lane in which the host vehicle M is traveling is estimated.

[0118] Next, the process proceeds to step S12, where the position coordinates (latitude and longitude) of each surrounding vehicle are detected with respect to the vehicle M. The position coordinates of the surrounding vehicles are obtained by marking the positions of each surrounding vehicle on a high-precision road map with respect to the position coordinates of the vehicle M. Then, based on the position coordinates of the vehicle M and the position coordinates of each surrounding vehicle, the lateral position x_i of the center in the vehicle width direction of each surrounding vehicle P, F, Pr, and Fr is detected with respect to the center in the vehicle width direction of the vehicle M (refer to Figure 9 ).

[0119] Then, in step S13, the lanes on the high-precision road map in which each surrounding vehicle is traveling are estimated based on the lateral positions x_i of each surrounding vehicle relative to the host vehicle M. The processing in steps S12 and S13 corresponds to the surrounding vehicle lateral position information estimating unit of the present invention.

[0120] Next, the process proceeds to step S14, where the offset f_offset of the adjacent subsequent vehicle Fr traveling in the lane opposite the shoulder of the lane currently being traveled by the host vehicle M, relative to the center of the lane, is compared with a deviation determination threshold d_sl. This deviation determination threshold d_sl is a threshold used to determine whether the driver intentionally deviated from the lane, taking into account the lateral deviation width from the center of the lane during normal travel. This value is determined in advance through experiments and is set on the side of the adjacent lane closer to the shoulder.

[0121] Furthermore, if the offset amount f_offset of the adjacent subsequent vehicle Fr is greater than the offset determination threshold d_sl and is closer to the driving lane of the host vehicle M, it is determined that the adjacent subsequent vehicle Fr is offset toward the driving lane of the host vehicle M, and the process branches to step S17. Furthermore, if the offset amount f_offset is closer to the lane center than the offset determination threshold d_sl, it is determined that the adjacent subsequent vehicle Fr is not offset, and the process proceeds to step S15.

[0122] Once the process proceeds to step S15, the offsets of the vehicle widthwise centers of the preceding vehicle P, the following vehicle F, and the adjacent preceding vehicle Pr, etc., relative to the center of the lane on the high-precision road map in which the preceding vehicle P and the following vehicle F are traveling are calculated in a time series manner. A statistical offset (offset trend) i_offset is estimated for each lane. By statistically calculating the offset i_offset, high detection accuracy can be achieved.

[0123] At this time, if Figure 11 As shown by the dashed line in the middle, on a dry road, the driver drives roughly in the center of the lane while recognizing the left and right dividing lines that separate the vehicle's lanes. Therefore, the mode of the statistical offset is roughly aligned with the center of the lane (i.e., a value close to 0). On the other hand, as shown by the solid line in the figure, on a snowy road where the dividing lines are not discernible, each vehicle tends to drive closer to the shoulder, as described above. Consequently, the mode of the statistical offset is biased toward the shoulder.

[0124] Next, the process proceeds to step S16, where the statistical offset i_offset of the preceding vehicle P and the following vehicle F traveling in the lane of the host vehicle M is compared with an offset determination threshold d_sl set toward the road shoulder relative to the lane center. If the statistical offset i_offset is greater than the offset determination threshold d_sl and toward the road shoulder, it is determined that the majority of the preceding vehicle P and the following vehicle F are offset toward the road shoulder, and the process branches to step S17.

[0125] If the statistical offset i_offset is closer to the lane center than the offset determination threshold d_sl, it is determined that most of the preceding vehicle P and the following vehicle F are not offset, and the process proceeds to step S18. Note that the processing in steps S14 and S16 corresponds to the offset determination unit of the present invention.

[0126] If step S17 is entered, the offset determination flag F_off is set (F_off←1), and the process enters Figure 3 In addition, if the process enters step S18, the offset determination flag F_off is cleared (F_off←0), and the process enters step S6. Figure 3 Step S6.

[0127] However, for example, if a subsequent vehicle Fr, traveling in a lane opposite the road shoulder, approaches the vehicle M while offset from the lane, and gradually overtakes it, the lateral positions of the subsequent vehicle Fr and the vehicle M become extremely close, causing a sense of unease for the driver. Therefore, in step S14, if offsetting of the subsequent vehicle Fr is detected, the process immediately branches to step S17, setting the offset determination flag F_off (F_off←1). By prioritizing the determination in step S14 over the determination of the statistical offset amount i_offset in step S16, driving in accordance with the driver's intent is possible.

[0128] Then, if you enter Figure 3 In step S6, the value of the deviation determination flag F_off is checked. If F_off = 1, the routine proceeds to step S7. Otherwise, if F_off = 0, the routine is exited. If the routine is exited directly from step S6, the automatic driving control unit 26 automatically drives the vehicle along the target route set by the driving route / target route setting calculation unit 12b of the map positioning calculation unit 12.

[0129] When the process proceeds to step S7, the offset d_offset, or lateral position correction value of the host vehicle M toward the roadside shoulder, is calculated. As described above, when the driver is driving on a snowy road where the left and right lane dividing lines cannot be visually identified, the vehicle tends to move closer to the roadside shoulder. In this embodiment, after checking the deviation trends of surrounding vehicles in step S5, and determining in step S6 that there is deviation (F_off = 1), the offset d_offset of the host vehicle M from the center of the lane on the high-precision road map information toward the roadside shoulder is calculated independently of the deviations of surrounding vehicles (particularly the preceding vehicle P and the following vehicle F).

[0130] In this case, if Figure 12 As shown, from the perspective of driving stability, the vehicle width W_car of ordinary vehicles with a narrower vehicle width is set to the value obtained by adding the tire width to the outer side of the tire TYs, that is, the wheelbase. In addition, the vehicle width W_car of large vehicles with a wider vehicle width is set to the value obtained by subtracting the tire width from the inner side between the tires TYl, that is, the wheelbase, and thus is set for each vehicle model.

[0131] That is, the offset d_offset of the vehicle width W_car of the ordinary vehicle is d_offset←(W_road−W_car) / 2+wd.

[0132] Furthermore, the offset d_offset of the vehicle width W_car of a large vehicle is d_offset←(W_road - W_car) / 2-wd. Here, wd is a margin to prevent the tires from running over the dividing line, and in this embodiment, it is set to approximately 0 to 0.3 m. Incidentally, if the vehicle M is traveling in a two-lane passing lane or a three-lane second lane or passing lane, if the vehicle overtakes into an adjacent lane, it will hinder the movement of vehicles in the adjacent lane. Therefore, the vehicle width W_car of a large vehicle is also set to the value obtained by adding the tire width to the wheelbase, similar to that of a standard vehicle. The offset d_offset is calculated as d_offset←(W_road - W_car) / 2-wd. In this case, the margin wd is set to approximately 0.3 m to prevent the tires from running over the dividing line.

[0133] Thereafter, the process proceeds to step S8 to verify the travelable area of ​​the host vehicle M. The process in step S8 corresponds to the travelable area verification unit of the present invention.

[0134] In step S8, it is verified whether the vehicle M can actually automatically drive along the dividing line on the shoulder side. The verification of the drivable area is carried out according to Figure 5 The drivable area verification subroutine shown is used for processing.

[0135] In this subroutine, first, the driving space of the road shoulder is detected in step S21. The driving space of the road shoulder is determined by, for example, a high-precision road map and the position of the vehicle to determine the dividing line on the road shoulder side of the lane currently being traveled (in Figure 7 、 Figure 8 At the same time, the boundary between the snow-covered road surface and the wall is detected based on the front running environment information recognized by the front running environment recognition unit 21d of the camera unit 21.

[0136] Next, the distance (lateral distance) between the determined dividing line and the detected boundary is calculated and set as the driving space. This lateral distance is calculated, for example, by subtracting the distance between the vehicle's position marked on the high-precision road map and the detected dividing line on the shoulder side from the lateral position to the boundary calculated using the vehicle's position as a reference.

[0137] Then, in step S22, if the lateral position to the boundary minus the distance between the vehicle position and the dividing line>the margin width, then Figure 7 As shown, it is determined that a running space wider than the margin width is secured on the road shoulder, and the process proceeds to step S23.

[0138] In addition, when the lateral position to the boundary - the distance between the vehicle position and the dividing line ≤ the margin width, for example, Figure 8 As shown, the boundary is in a state where it protrudes from the dividing line on the shoulder side (the left dividing line in the figure) toward the road side, or the boundary is in a state where it approaches the dividing line from the shoulder side.

[0139] In this state, it is difficult to make the tires of the host vehicle M run along the dividing line on the road shoulder side, so the process branches to step S26. For example, when a three-dimensional object such as a stopped work vehicle is recognized on the road shoulder based on the forward driving environment information recognized by the forward driving environment recognition unit 21d of the camera unit 21, the side surface of the three-dimensional object on the dividing line side is detected as the boundary.

[0140] Then, the process proceeds to step S23 to detect the road width of the drivable area (available space), that is, the snowy road width W_free. The snowy road width W_free is the distance between the left and right boundaries recognized by the front driving environment recognition unit 21 d of the camera unit 21 .

[0141] Next, the process proceeds to step S24, where the width of the snow-covered road surface W_free is divided by the number of lanes n of the road currently being traveled on to calculate the lane width W_free / n of each lane, and the process proceeds to step S25, where, based on the statistical offset in each driving lane calculated in the above-mentioned step S14, the process investigates whether the surrounding vehicles are traveling roughly along the center (W_free / (n·2)) of each lane width W_free / n.

[0142] Typically, when a driver manually controls a vehicle on a snowy road where the dividing lines cannot be visually identified, they identify the width of the snowy road surface between the boundaries, W_free, and divide this width by the number of lanes to determine the approximate lane width W_free / n that the vehicle should travel in. In most cases, the driver drives the vehicle with a target of approximately the center of this lane width, W_free / (n·2), of W_free / n.

[0143] So, for example, Figure 7 As shown, when the road has two lanes and the dividing line is covered with snow and cannot be identified, the driver divides the snow road surface width W_free by the number of lanes to identify the lane width W_free / 2 of the current lane and drives with the center (W_free / (2·2)) as the target.

[0144] On the other hand, Figure 10 As shown, even if a road has three lanes, only about 2.5 lanes of snow are cleared. In this case, the lane width W_free / 3 is narrowed for each lane. Even if the vehicle attempts to travel in the center of this lane width, W_free / (3·2), surrounding vehicles (in the figure, the preceding vehicle P and the adjacent preceding vehicle Pr) will divide the snowy road width W_free into two equal halves and travel in the approximate center, W_free / (2·2), ensuring that one lane is wider.

[0145] As a result, the host vehicle M is different from surrounding vehicles and travels extremely close to the road shoulder. However, since the surrounding vehicles travel with the lane width of each lane being ensured to be wide, the driver (and passengers) of the host vehicle M may feel uncomfortable.

[0146] Therefore, in step S25, the absolute value |x_j| of the lateral positional deviation x_j between the center (W_free / (n·2) of the lane width W_free / n) of each surrounding vehicle (P, Pr, F, Fr in the figure) estimated in step S13 and the lateral center of the surrounding vehicle (P, Pr, F, Fr) is calculated and compared with a preset deviation determination value x_jsl. If |x_j| ≤ x_jsl, it is determined that the surrounding vehicle is traveling within the lane width W_free / n obtained by dividing the snowy road surface, and the process proceeds to step S9.

[0147] If |x_j|>x_jsl, the surrounding vehicle is determined to have deviated from the lane width W_free / n obtained by dividing the snowy road surface, and the routine branches to step S26. If the routine proceeds from step S22 or step S25 to step S26, the automatic driving is canceled and the routine ends.

[0148] The automatic driving control unit 26 then activates the notification device 34 to notify the driver that automatic driving has ended, requesting the driver to hold the steering wheel, and after a predetermined period of time, the system switches from automatic driving to driving assistance control. Driving assistance control uses information from the camera unit 21 and the surrounding monitoring unit 22 to assist the driver through well-known functions such as following distance control (ACC: Adaptive Cruise Control), rear vehicle detection warning, and rear side collision avoidance assistance.

[0149] When the automatic driving is released by branching from step S25 to step S26, the driver can drive the vehicle along the virtual lane in which the preceding vehicle or the following vehicle is traveling, thereby enabling the vehicle to drive in accordance with the surrounding vehicles without causing discomfort to the driver.

[0150] On the other hand, if the process proceeds from step S25 Figure 3 In step S9, the offset d_offset of the host vehicle M calculated in step S7 is used to correct the lateral position of the target path set by the driving route / target path setting calculation unit 12b of the map positioning calculation unit 12. A new target path is set in step S10, and the routine exits. It should be noted that the processing in steps S9 and S10 corresponds to the target path correction unit of the present invention.

[0151] Then, the automatic driving control unit 26 notifies the driver of the snowy road driving condition by the driving notification device 34, and then moves the host vehicle M toward the road shoulder by the offset amount d_offset, and performs automatic driving along the target travel path. Figure 12 As shown, when the host vehicle M is a normal vehicle, the outer sides of the tires run along the inner side of the dividing line on the shoulder side.

[0152] Furthermore, if the vehicle M is a large vehicle, the inner side of the tire runs along the outer side of the dividing line on the shoulder side. If the tires roll along the dividing line, the grip is reduced and slip is likely to occur. However, in this embodiment, ordinary vehicles run along the inner side of the dividing line, and large vehicles run along the outer side of the dividing line, so the grip is not reduced more than necessary.

[0153] Thus, in the present embodiment, first, when traveling along the target travel path of the lane set on the shoulder side based on the position of the vehicle estimated by the positioning signal from the GNSS receiver 14 and the high-precision road map, on a snowy road surface where the left and right dividing lines dividing the lane in which the vehicle M is traveling cannot be identified, the lateral positions x_i of the vehicle M and surrounding vehicles (the preceding vehicle P, the adjacent preceding vehicle Pr, the following vehicle F, the adjacent following vehicle Fr, etc.) are detected, and the lanes in which the surrounding vehicles are traveling are determined with reference to the high-precision road map.

[0154] Next, a statistical deviation amount is calculated based on the center of the lane in which each surrounding vehicle is traveling, and the system checks whether each surrounding vehicle is traveling in a state of deviation from the center toward the roadside. If it is determined that each surrounding vehicle is traveling in a state of deviation toward the roadside, the host vehicle M traveling in the lane on the roadside is caused to travel in a state of deviation toward the roadside. This allows the autonomous driving to continue without hindering the travel of surrounding vehicles.

[0155] In addition, the offset of the vehicle M toward the shoulder side is independent of the lateral position of the preceding vehicle and / or the following vehicle. If it is an ordinary vehicle, the outer side of the tire is set on the inner side of the dividing line on the shoulder side. In addition, if it is a large vehicle, the inner side of the tire is set on the outer side of the dividing line on the shoulder side. Therefore, stable driving performance can be obtained.

[0156] Furthermore, when an adjacent subsequent vehicle Fr traveling in an adjacent lane approaches in a state of being offset toward the lane in which the vehicle M is traveling, the target travel path of the vehicle M is immediately offset and corrected toward the dividing line side of the road shoulder. Therefore, when the vehicle M is overtaken by the adjacent subsequent vehicle Fr, the driver will not feel uneasy.

[0157] Furthermore, when the host vehicle M is offset, it is verified whether snow removal has ensured a clear driving space on the shoulder side of the road, thereby enabling safer driving without causing discomfort to the driver. Furthermore, the road width, which is the drivable area (usable space) on a snowy road, is determined by snow removal. When the road width is narrow and the host vehicle M is traveling on a snowy road with fewer lanes than usual, the automated driving mode is disengaged and switched to driving assistance control. This allows the driver to steer along the virtual lanes of the preceding and / or following vehicles, achieving good driving performance.

[0158] [Other methods]

[0159] However, you can also apply Figure 13 The offset processing unit A shown above is used to replace the above Figure 3 The target travel path correction routine shown is the offset processing section A (steps S3 to S5).

[0160] That is, in this method, if Figure 3 In step S2 of the target travel path correction routine shown, if it is determined that the road surface is snow-covered and step S31 is entered, cloud information for a predetermined section set on the target travel path of the present vehicle M is obtained from the cloud server 1, and lateral position information, i.e., offset data, of other vehicles traveling in the same section from the relatively recent past (about 30 to 60 [min]) to the present with respect to the center of the travel lane on the high-precision road map is obtained. The cloud server 1 obtains and aggregates the offset data of each vehicle from detection information from detection vehicles and passing histories of traveling vehicles from vehicle-to-vehicle communication. It should be noted that the processing in this step S31 corresponds to the external information acquisition unit of the present invention.

[0161] Then, step S32 is entered, and the offset data of each lane of the aggregated other vehicles is aggregated, and the statistical offset amount i_offset of each travel lane is obtained in the same steps as in step S14 of the surrounding vehicle statistical offset determination subroutine shown above. Figure 4

[0162] Next, in step S33, the statistical offset amount i_offset of each travel lane is compared with the offset determination threshold d_sl. Then, when i_offset < d_sl, it is determined that there is no offset, the offset determination flag F_off is cleared (F_off ← 0) in step S34, and step S8 is entered. On the other hand, when i_offset ≥ d_sl, it is determined that there is an offset, and the process branches to step S35, the offset determination flag F_off is set (F_off ← 1), and step S8 is entered. It should be noted that the processing in this step S33 corresponds to the offset determination unit of the present invention.

[0163] According to this method, since the statistical offset amount is obtained based on cloud information, it is possible to easily estimate the driving conditions of surrounding vehicles in the target travel path ahead before entering an automatic driving section such as a highway. As a result, it is possible to perform offset correction of the target travel path and perform automatic driving immediately after entering the automatic driving section, and it is possible to automatically drive the present vehicle M without causing obstacles to the driving of surrounding vehicles (the preceding vehicle P, the adjacent preceding vehicle Pr, the following vehicle F, the adjacent following vehicle Fr, etc.).

[0164] [Another method]

[0165] In addition, in a snowfall area, the road surface in winter (winter road surface) is always in a snow-covered state. In the case of a snow-covered road surface formed by snow removal, it can be presumed that local drivers accustomed to driving on snow roads drive the vehicle in a state of offsetting towards the road shoulder without investigating the actual driving conditions.

[0166] Therefore, when the present vehicle M is traveling in a snowfall area, as Figure 14 As shown in FIG. 1 , the offset processing unit A can also be performed more simply. That is, in this manner, if Figure 3 In the target path correction routine shown, if a snowy road surface is determined in step S2 and the process proceeds to step S41, cloud information for a predetermined section of the target path of the host vehicle M is read from the cloud server 1. Winter road surface information is then acquired from this cloud information as lateral position information of surrounding vehicles. The processing in step S41 corresponds to the surrounding vehicle lateral position information estimating unit of the present invention.

[0167] Next, the process proceeds to step S42, where it checks whether the target route to be taken is a snowy road. If the road is clear, the process proceeds to step S43, where the deviation determination flag F_off is cleared (F_off←0), and the process proceeds to step S8. On the other hand, if the road is snowy, it is estimated that the surrounding vehicles have deviated toward the road shoulder, and the process branches to step S44, where the deviation determination flag F_off is set (F_off←1), and the process proceeds to step S8. It should be noted that the processing in step S42 corresponds to the deviation determination unit of the present invention.

[0168] According to this method, only whether the target travel path that the vehicle M is traveling on from now on is a snowy road is investigated. If it is a snowy road, it is estimated that the surrounding vehicles are deviating from the road. Therefore, the driving condition of the target travel path ahead can be estimated more simply before entering an automatic driving area such as a highway.

[0169] It should be noted that the present invention is not limited to the above-mentioned embodiment. For example, information on snowy roads and / or driving conditions of surrounding vehicles may be acquired through road-to-vehicle communication.

Claims

1. A vehicle driving control device, characterized in that: have: A surrounding environment information acquisition unit, mounted on the vehicle, and configured to acquire surrounding environment information of the vehicle; A map information storage unit for storing road map information; a vehicle position estimating unit for estimating the vehicle position of the vehicle; a road surface information acquiring unit for acquiring information about the road surface on which the vehicle is traveling; as well as a target route setting unit that sets a target route for the vehicle during automatic driving by referring to the road map information stored in the map information storage unit based on the vehicle position information estimated by the vehicle position estimating unit and the input destination information; The vehicle driving control device further comprises: a surrounding vehicle lateral position information estimating unit for estimating, based on the lateral position information of the surrounding vehicles detected with respect to the own vehicle as a reference, the lanes of surrounding vehicles traveling on the road on which the own vehicle is traveling, from the road map information stored in the map information storage unit, when the road surface acquired by the road surface information acquiring unit is a snow-covered road surface and left and right dividing lines dividing the lane on which the own vehicle is traveling are not visually recognizable, wherein the surrounding vehicles include a preceding vehicle traveling in front of the own vehicle, a following vehicle following behind the own vehicle, an adjacent preceding vehicle traveling in front of an adjacent lane on the opposite side of the road shoulder, and an adjacent following vehicle traveling behind an adjacent lane on the opposite side of the road shoulder; a deviation determination unit that determines, based on the lateral position information of the surrounding vehicle estimated by the surrounding vehicle lateral position information estimation unit, an amount of deviation of the center of the surrounding vehicle in the vehicle width direction relative to the center of the lane in which the surrounding vehicle is traveling, and determines, based on the deviation amount of the surrounding vehicle, whether the surrounding vehicle is deviated toward the road shoulder relative to the driving lane; as well as a target travel path correction unit that, when the deviation determination unit determines that the surrounding vehicle is deviating toward the road shoulder, sets an offset amount for causing the host vehicle to deviate toward a dividing line on the road shoulder side of the lane in which the host vehicle is traveling, as shown in the road map information stored in the map information storage unit, and uses the offset amount to correct the lateral position of the target travel path set by the target travel path setting unit toward the road shoulder side, thereby setting a new target travel path. The surrounding vehicle lateral position information estimating unit acquires the winter road conditions of the section where the host vehicle is traveling from an external information aggregation device, and uses the winter road conditions as the lateral position information of the surrounding vehicle. When the winter road surface is a snow-covered road surface, the deviation determination unit determines that the surrounding vehicle has deviated from the driving lane.

2. The vehicle travel control device according to claim 1, wherein: The surrounding vehicle lateral position information estimating unit estimates the lateral position information of the surrounding vehicle based on the surrounding environment information acquired by the surrounding environment information acquiring unit.

3. The vehicle travel control device according to claim 1, wherein: The surrounding vehicle lateral position information estimating unit acquires the lateral position information of the surrounding vehicles in a section in which the host vehicle is traveling from the recent past to the present from an external information collecting device.

4. The vehicle travel control device according to any one of claims 1 to 3, characterized in that: The offset amount set by the target travel path correction unit is set so that the tire of the host vehicle approaches the dividing line on the road shoulder side.

5. The vehicle travel control device according to any one of claims 1 to 3, characterized in that: The vehicle travel control device further includes a travelable region verification unit that checks whether the host vehicle can be traveled while being offset to the road shoulder side. The drivable area verification unit cancels the automatic driving when it is determined that no drivable space is ensured on the road shoulder based on the surrounding environment information acquired by the surrounding environment information acquisition unit.

6. The vehicle travel control device according to any one of claims 1 to 3, characterized in that: The vehicle travel control device further includes a travelable region verification unit that checks whether the host vehicle can be traveled while being offset to the road shoulder side. The drivable area verification unit detects the width of the snowy road surface based on the surrounding environment information acquired by the surrounding environment information acquisition unit, divides the width of the snowy road surface by the number of lanes on the road map information stored in the map information storage unit to set an imaginary lane, calculates the lateral position deviation of the surrounding vehicles relative to the imaginary lane, and releases the automatic driving when the lateral position deviation is greater than a predetermined judgment value.

7. The vehicle travel control device according to any one of claims 1 to 3, characterized in that: When the offset determination unit determines that an adjacent subsequent vehicle traveling in a lane adjacent to the side opposite to the shoulder side relative to the lane in which the host vehicle is traveling is approaching the lane in which the host vehicle is traveling, the target travel path correction unit sets an offset amount for immediately offsetting the host vehicle toward a dividing line on the shoulder side of the lane in which the host vehicle is traveling on the road map information stored in the map information storage unit.

Images