Automated driving integrated control system

Abstract

To provide an automated driving integrated control system which, when an abnormality occurs in an automated driving vehicle entering an intersection, generates intersection escape routes corresponding to an abnormal state of the vehicle, and selects a route having a high collision-avoidance level, thereby allowing the vehicle to travel.SOLUTION: An automated driving integrated control system 1 includes multiple roadside monitoring devices 110 each including a sensor and a transmitter for transmitting information on a detected object, and an autonomous driving apparatus 300 which includes: a driving control device 320 including a reception unit, a vehicle abnormal state detection unit, an escape route generation unit for generating, for each of the roadside monitoring devices, an intersection escape route for escaping from an intersection, a selection unit for comparing the generated intersection escape routes in terms of collision-avoidance levels and making a selection, a travelable distance calculation unit for calculating an abnormal-time travelable distance, and a vehicle control unit for allowing a vehicle to travel along a high-level intersection escape route corresponding to the abnormal-time travelable distance when an abnormal state of the vehicle is detected; and a vehicle operation unit 317.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present application relates to an automatic driving comprehensive control device. [Background technology]

[0002] In recent years, there has been a demand for autonomous transport vehicles. The introduction of autonomous driving is expected to solve various social problems, such as the shortage of drivers in the logistics field, traffic congestion, and the last-mile problem.

[0003] Traditionally, vehicles operating within a specific area such as a factory have been transported by drivers using trucks or towing dollies. However, to improve the factory's operating rate, it is desirable to operate the in-factory transport system at all times. Methods such as increasing the number of vehicles that can be towed and expanding the transport time period require increased costs for equipment and drivers, which is a heavy burden for businesses. For this reason, automated driving is also desired for transport vehicles operating within a specific area.

[0004] SAE (Society of Automotive Engineers) defines the technological levels of autonomous driving for general vehicles traveling on public roads. Many manufacturers and social organizations have adopted these technological levels. Level 4 autonomous driving specifies the conditions for a vehicle to continue autonomous driving without requiring the driver to constantly monitor the vehicle.

[0005] The Ministry of Land, Infrastructure, Transport and Tourism's safety technology guidelines for autonomous vehicles mandate the MRM (Minimum Risk Maneuver) function for autonomous driving at level 4. MRM is a function that automatically and safely stops the vehicle when the autonomous driving device determines that it is difficult to continue autonomous driving, such as when the vehicle is outside the operational design domain or when a fault occurs in the autonomous vehicle.

[0006] If an abnormality occurs in an autonomous vehicle that has entered an intersection, there is a greater risk of collision than when the vehicle is traveling on a normal road. If the vehicle stops in the intersection, it may cause traffic obstruction and may result in a collision. Therefore, if an abnormality occurs in an autonomous vehicle that has entered an intersection, it is desirable to prioritize escaping the intersection. In addition, it is desirable to compare multiple escape routes based on collision avoidance levels and select an escape route with a high collision avoidance level when determining the route to escape the intersection.

[0007] A technology has been disclosed in which, when an autonomous vehicle is driving at an intersection, the vehicle's autonomous driving device and a roadside monitoring device share information, and when a high-risk situation such as a collision with or approach to an obstacle is predicted, the autonomous driving route of the vehicle is overridden (forcibly operated) from outside the vehicle to drive the vehicle along an appropriate route (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0008] [Patent Document 1] JP 2022-174519 A Summary of the Invention [Problem to be solved by the invention]

[0009] In the technology disclosed in Patent Document 1, information about the road surroundings is acquired by a recognition device installed in the vehicle and a monitoring device installed in an external monitoring system. The recognition device and monitoring device are assumed to be cameras that capture the surrounding conditions, radars that detect surrounding obstacles, etc. These two devices detect obstacles around the road. The automatic driving route is then verified to ensure that the vehicle does not collide with or approach obstacles. When an abnormality is detected by the abnormality detection unit that detects abnormalities in the vehicle, an automatic driving route is specified from outside the vehicle so as not to come into contact with obstacles around the road, and acceleration or deceleration instructions are given.

[0010] However, Patent Document 1 does not describe a process that prioritizes the escape of the vehicle from the intersection when an abnormality occurs in the autonomous vehicle that has entered the intersection. Furthermore, there is no mention of comparing and selecting multiple escape routes depending on the collision avoidance level. With the technology described in Patent Document 1, if an abnormality occurs in the vehicle within the intersection, it is possible that the vehicle will be externally overridden (forcibly operated) and will slow down and stop within the intersection. Stopping a vehicle within an intersection will obstruct the passage of other vehicles and increase the risk of collision with other vehicles.

[0011] The present application has been made to solve the above problems, and aims to provide an automated driving comprehensive control device that, when an abnormality occurs in an automated driving vehicle that has entered an intersection, generates an intersection escape route according to the abnormal state of the vehicle, selects a route with a high collision avoidance level, and drives the vehicle. [Means for solving the problem]

[0012] The automatic driving integrated control device according to the present application comprises: A sensor for detecting surrounding objects; and A plurality of roadside monitoring devices each having a transmitter for transmitting information on a viewing angle of a sensor and information on an object detected by the sensor; a receiving unit that receives the viewing angle information and the object information from the transmitter of the roadside monitoring device; a vehicle abnormality detection unit that detects an abnormality in the vehicle; an escape route generating unit that generates an intersection escape route for escaping the intersection when an abnormality occurs in a vehicle that has entered the intersection, for each roadside monitoring device and for each of a plurality of possible driving distances in the event of an abnormality, based on information received from the roadside monitoring device; a selection unit that compares the intersection escape routes generated by the escape route generation unit according to a collision avoidance level at which the intersection escape routes can avoid collision with an object while the vehicle is traveling, and selects the intersection escape route with the highest collision avoidance level as a high-level intersection escape route for each possible travel distance in the event of an abnormality; a travelable distance calculation unit that calculates an abnormality-state travelable distance according to the abnormal state when an abnormal state of the vehicle is detected by the vehicle abnormality detection unit; a vehicle control unit that outputs a command signal to cause the vehicle to travel along a high-level intersection escape route corresponding to the abnormality-state travelable distance calculated by the travelable distance calculation unit among the high-level intersection escape routes selected by the selection unit when an abnormality state of the vehicle that has entered the intersection is detected by the vehicle abnormality detection unit; and and an autonomous driving device having a vehicle operation unit that drives an actuator based on a command signal output by a vehicle control unit. Effect of the Invention

[0013] According to the automated driving integrated control device of the present application, when an abnormality occurs in an automated driving vehicle that has entered an intersection, an intersection escape route according to the abnormal state of the vehicle is generated, and a route with a high collision avoidance level is selected so that the vehicle can travel along the route. As a result, even if an abnormality occurs in an automated driving vehicle that has entered an intersection, it is possible to prioritize the escape of the vehicle from the intersection and prevent the vehicle from obstructing the passage of other vehicles and increasing the risk of collision with other vehicles. [Brief description of the drawings]

[0014] [Figure 1] 1 is a layout diagram of a roadside monitoring device according to a first embodiment. [Diagram 2] 1 is an overall configuration diagram of an automatic driving integrated control device according to a first embodiment. [Diagram 3] 1 is a configuration diagram of a data collection device according to a first embodiment. [Figure 4] FIG. 2 is a configuration diagram of an edge server according to the first embodiment. [Diagram 5] FIG. 1 is a configuration diagram of an operating device according to a first embodiment. [Figure 6] FIG. 2 is a hardware configuration diagram of a control device according to the first embodiment. [Figure 7] 4 is a diagram for explaining the relationship between an object and a collision avoidance level according to the first embodiment. FIG. [Figure 8] 4 is a diagram illustrating the relationship between a stopping position and a collision avoidance level addition point according to the first embodiment. FIG. [Figure 9] FIG. 4 is a first diagram illustrating the relationship between an abnormal state and a travelable distance according to the first embodiment. [Figure 10] FIG. 11 is a second diagram illustrating the relationship between an abnormal state and a travelable distance according to the first embodiment. [Figure 11] FIG. 2 is a diagram illustrating an intersection escape route according to the first embodiment. [Figure 12] 5 is a flowchart showing a process of the edge server according to the first embodiment. [Figure 13] 4 is a flowchart showing the process of the driving device according to the first embodiment. [Figure 14] 4 is a flowchart showing a process for calculating a travelable distance of the driving device according to the first embodiment. [Figure 15] FIG. 11 is an overall configuration diagram of an automatic driving integrated control device according to a second embodiment. [Figure 16] FIG. 11 is a configuration diagram of an edge server according to the second embodiment. [Figure 17] FIG. 11 is a configuration diagram of an operating device according to a second embodiment. [Figure 18] 13 is a first flowchart showing a process of an edge server according to the second embodiment. [Figure 19] 13 is a second flowchart showing the process of the edge server according to the second embodiment. [Figure 20] 10 is a first flowchart showing the process of the driving device according to the second embodiment. [Figure 21] 10 is a second flowchart showing the process of the driving device according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] 1. First embodiment <Location of roadside monitoring devices> FIG. 1 is a diagram showing the arrangement of roadside monitoring devices 110 and 120 at an intersection according to the first embodiment. The roadside monitoring devices are also called RSUs (Road Side Units). FIG. 1 shows an example of an intersection, depicting a vehicle 801 that has entered the intersection and is about to turn right, and a vehicle 811 that is approaching the intersection from the right of the vehicle 801. The roadside monitoring device 110 installed around the road has a sensor that monitors surrounding objects, and the range that the sensor can grasp is indicated by a viewing angle 501. Similarly, the range that the sensor that monitors surrounding objects of the roadside monitoring device 120 can grasp is indicated by a viewing angle 502. The roadside monitoring devices 110 and 120 can monitor not only the center of the intersection, but also the periphery of the intersection. The roadside monitoring devices 110 and 120 are arranged so that blind spots around the intersection are minimized.

[0016] In Fig. 1, roadside monitoring devices 110, 120 are placed diagonally so that their field of view widens toward the intersection, and identify the position, speed, moving direction, size, shape, type, etc. of pedestrians, four-wheeled vehicles, stationary obstacles, etc. within their field of view. Although two roadside monitoring devices 110, 120 are placed in Fig. 1, three or more roadside monitoring devices may be installed at an intersection to increase the monitoring area and ensure redundancy of the roadside monitoring devices.

[0017] One of the roads constituting the intersection in Fig. 1 is provided with a stopping area 510 where vehicles can be stopped. The stopping area refers to areas where vehicles can be stopped, such as stopping lanes, time-limited parking areas, and roadside strips, as well as stopping areas in front of bus stops and road shoulders where vehicles can be stopped in an emergency. No-parking roadside strips and pedestrian roadside strips as defined by the Road Traffic Act may be excluded.

[0018] <Configuration of the Autonomous Driving Integrated Control Device> Fig. 2 is an overall configuration diagram of the autonomous driving integrated control device 1 according to the first embodiment. A vehicle 801 to which the autonomous driving integrated control device 1 is applied is autonomously driven by a driving device 300 that communicates with an edge server 200 that receives information from roadside monitoring devices 110 and 120 outside the vehicle. As shown in Fig. 1, in the case of an intersection without a traffic light with a stop instruction, the vehicle stops at the stop line just before the intersection. If no abnormality is occurring with the vehicle, the driving device 300 determines whether it is possible to turn right at the intersection based on input information from the edge server 200, and starts to turn right.

[0019] The driving device 300 checks in advance that there will be no collision with other objects during a right turn, and has determined that there will be no collision with an oncoming vehicle 811 during a right turn. The determination of whether or not to perform a right turn is made when the vehicle 801 crosses a stop line (or a virtual intersection entrance boundary line if there is no stop line), and when the vehicle 801 crosses the stop line (or a virtual intersection entrance boundary line), the vehicle 801 automatically travels along a target route for passing through the intersection.

[0020] The autonomous driving comprehensive control device 1 is composed of a data collection device 100 having multiple roadside monitoring devices such as roadside monitoring devices 110, 120, an edge server 200, and a driving device 300 mounted on a vehicle 801. The vehicle 801 can automatically drive by grasping not only the external environment using an object sensor 303 equipped on the driving device 300 mounted on the vehicle 801 but also the environment around roads and intersections using sensors such as the roadside monitoring devices 110, 120. The edge server 200 and the driving device 300 constitute an autonomous driving device 901 with the data collection device 100 as an external device.

[0021] <Data collection devices and roadside monitoring devices> Fig. 3 is a configuration diagram of the data collection device 100 according to the first embodiment. The data collection device 100 is composed of a plurality of roadside monitoring devices including roadside monitoring devices 110 and 120. The plurality of roadside monitoring devices may be arranged in a cluster at the same location and may share the responsibility of monitoring in all directions. Alternatively, as shown in Fig. 1, the plurality of roadside monitoring devices may be arranged separately and may monitor an intersection from different directions.

[0022] The roadside monitoring device 110 will be described below, but the same can be said for the roadside monitoring device 120, etc. The roadside monitoring device 110 is equipped with a sensor that grasps the external environment within a predetermined viewing angle 501 and detects objects around intersections and roads. Possible objects to be detected include four-wheeled vehicles, two-wheeled vehicles, pedestrians, animals, other moving objects, fallen objects, signs, other stationary objects, etc. The roadside monitoring device 110 is equipped with an image sensor 112, a radio wave sensor 113, and an optical sensor 114 as sensors that grasp the external environment.

[0023] The image sensor 112, as typified by a surveillance camera, photographs an object and calculates the distance to the object from image data photographed within a certain viewing angle range. The size, moving direction, moving speed, attributes, etc. of the object can also be obtained from the image data. The image sensor 112 can be a visible light camera, an infrared camera, etc.

[0024] A millimeter wave radar (MMWR) using a frequency band of 24 to 79 GHz or the like can be used as the radio wave sensor 113. The radio wave sensor 113 can detect the position of an object and detect the moving speed of the object by the Doppler effect.

[0025] A laser radar, LiDAR (Light Detection and Ranging), etc. can be used as the optical sensor 114. By irradiating a laser beam within a certain field of view and detecting point cloud data obtained by reflection of the laser beam from an object, the position and shape of the object can be grasped.

[0026] The sensor information calculation unit 117 receives information from the image sensor 112, the radio wave sensor 113, and the optical sensor 114, which are sensors that grasp the external environment. The sensor information calculation unit 117 can use a sensor fusion technique that combines this information to detect objects. Combining multiple types of sensor information can remove noise information, enabling highly reliable distance measurement, speed detection, and attribute identification of objects. Furthermore, obstacles may be identified based on reinforcement learning such as deep learning.

[0027] All of the information from these sensors may be processed by the sensor information calculation unit 117, but information processing may be performed for each sensor, data acquired by the various sensors may be processed, and only information on the position, outer shape, speed, and attributes of the identified object may be transmitted to the sensor information calculation unit 117. In this way, the processing of the sensor information can be distributed and executed, so that the amount of information processed by a single sensor information calculation unit 117 can be reduced.

[0028] The roadside monitoring device 110 may use all of the image sensor 112, the radio wave sensor 113, and the optical sensor 114, or may use only some of the sensors. In addition, other sensors such as ultrasonic sensors may be used to grasp the external environment. In addition, the roadside monitoring device 110 may be combined with other sensors that provide information on traffic flow and weather conditions as sensor information.

[0029] The roadside monitoring device 110 also monitors abnormal states of various sensors by the sensor information calculation unit 117. When a failure occurs in the image sensor 112, the radio wave sensor 113, or the optical sensor 114, or when a power supply abnormality occurs, the sensor information calculation unit 117 detects this.

[0030] The sensor information calculation unit 117 transmits object information, such as the object's position, speed, outer shape, and attributes, and sensor viewing angle information, based on the information received from the various sensors, to the edge server 200 via the communication module 118. The sensor information calculation unit 117 also transmits information regarding an abnormal state of the sensor to the edge server 200 via the communication module 118.

[0031] <Edge Server> With the recent advancement of communication technology and processing technology represented by 5G and LTE (Long Term Evolution), mobile edge computing (MEC) and multi-access edge computing (MEC) are beginning to be used, which perform processing by installing a server near the outside of the terminal instead of processing information on the terminal alone. In the past, proposals for the operation of autonomous vehicles have been made to rely on information from sensors installed in the vehicle to grasp the surrounding environment and to receive support for processing using a large-scale remote server on the cloud. However, it is becoming possible to process large amounts of information using edge servers installed near the vehicle and share the information with the vehicle without delay. Note that the edge servers may be distributed to the terminals themselves.

[0032] 4 is a configuration diagram of edge server 200 according to embodiment 1. Edge server 200 communicates at high speed with data collection device 100 and roadside monitoring devices 110, 120 constituting data collection device 100 via communication module 201 to share information. Edge server 200 is a server provided near a terminal (in this case, driving device 300 mounted on vehicle 801) to achieve low latency, and may be provided at each intersection.

[0033] The communication module 201 of the edge server 200 receives object information and sensor viewing angle information as roadside monitoring information IRSU for each roadside monitoring device, and transmits it to the map generation unit 202. The communication module 201 also receives information on an abnormal state of the sensor for each roadside monitoring device, and transmits it as abnormality information FRSU, including information on communication abnormalities (such as communication interruption) with the roadside monitoring device, to the selection unit 206 and the communication module 207. The communication module 207 is a module for communicating with the communication module 302 of the driving device 300 mounted on the vehicle 801. The communication module 201 and the communication module 207 may be a single communication module.

[0034] The map generating unit 202 of the edge server 200 generates combined map information IOBM by superimposing the roadside monitoring information IRSU of all roadside monitoring devices and the map information 208. The map generating unit 202 also transmits the first combined map information IOBM1, which is generated by superimposing the roadside monitoring information by the roadside monitoring device 110 and the map information 208, to the first escape route generating unit 204. The first escape route generating unit 204 generates a plurality of routes for escaping the intersection for each possible driving distance in an abnormality based on the first combined map information IOBM1 and the position information of the vehicle 801. The first escape route generating unit 204 transmits the first escape route (RESC1) based on the roadside monitoring information by the roadside monitoring device 110 to the selecting unit 206. Here, the route for escaping the intersection is generated within the viewing angle 501 of the sensor of the roadside monitoring device 110, and the route that deviates from the viewing angle 501 of the sensor is excluded.

[0035] The map generating unit 202 transmits second combined map information IOBM2 obtained by superimposing the roadside monitoring information from the roadside monitoring device 120 and the map information 208 to the second escape route generating unit 205. The second escape route generating unit 205 generates a plurality of routes for escaping the intersection for each possible driving distance in an abnormality based on the second combined map information IOBM2 and the position information of the vehicle 801. The second escape route generating unit 205 transmits the second escape route RESC2 based on the roadside monitoring information from the roadside monitoring device 120 to the selecting unit 206. Here, the route for escaping the intersection is generated within the viewing angle 502 of the sensor of the roadside monitoring device 120, and the route deviating from the viewing angle 502 of the sensor is excluded.

[0036] The selection unit 206 of the edge server 200 compares the intersection escape routes generated by the first escape route generation unit 204 and the second escape route generation unit 205 for each roadside monitoring device and each possible driving distance in an abnormal state based on the collision avoidance level at which the intersection escape route can avoid collision with an object while the vehicle is traveling, and selects the intersection escape route with the highest collision avoidance level as the high-level intersection escape route RESCB. In Fig. 4, the intersection escape routes generated based on the roadside monitoring information from the two roadside monitoring devices 110 and 120 are compared and selected, but the number of intersection escape routes to be compared increases as the number of roadside monitoring devices increases.

[0037] Details of the abnormality-state driving distance and collision avoidance level will be explained later with reference to Figs. 7 to 10. In Fig. 7, the collision avoidance level is set in four stages, and in Fig. 8, the collision avoidance level points are set from +1 to +3. In Figs. 9 and 10, the abnormality-state driving distance is specified as three or four types of distance depending on the abnormal state of the vehicle. The edge server 200 generates intersection escape routes for multiple types of distance in advance. By generating, selecting, and storing intersection escape routes in advance, they can be used quickly when a vehicle abnormality occurs.

[0038] The high-level intersection escape route RESCB selected by the selection unit 206 for each of the multiple types of possible driving distances in an abnormality is transmitted to the driving device 300 via the communication module 207. At this time, abnormality information FRSU regarding each roadside monitoring device is also transmitted to the driving device 300 via the communication module 207.

[0039] The edge server 200 holds combined map information for each roadside monitoring device, in which roadside monitoring information received from the roadside monitoring device is superimposed on map information, and generates an escape route for each roadside monitoring device. Therefore, even if an abnormality occurs in an individual roadside monitoring device, the influence of the abnormal roadside monitoring device can be quickly eliminated. Then, among the escape routes based on the roadside monitoring information of roadside monitoring devices that are not abnormal, the intersection escape route RESCB with the highest collision avoidance level can be selected for each possible driving distance in the event of an abnormality.

[0040] In the above description, the edge server 200 stores combined map information for each roadside monitoring device, in which roadside monitoring information received from the roadside monitoring device is superimposed on map information. However, if there are many roadside monitoring devices, the edge server 200 may create groups of the roadside monitoring devices and create and store combined map information for each roadside monitoring device group.

[0041] The edge server 200 is not limited in information processing speed and capacity as the driving device 300, which is an in-vehicle device, is. The edge server 200 can generate, compare, and select a large number of intersection escape routes while processing a large amount of information based on roadside monitoring information received from multiple roadside monitoring devices. This makes it possible to select a more appropriate intersection escape route while reducing the burden on the driving device 300 mounted on the vehicle 801.

[0042] <Operating device> 5 is a configuration diagram of driving device 300 according to embodiment 1. Driving device 300 mounted on vehicle 801 is provided with vehicle position detection unit 301 that determines the position of the vehicle, communication module 302 that transmits and receives information to edge server 200, and object sensor 303 that detects the position, speed, and type of objects around the vehicle.

[0043] The driving control device 320 receives signals from the vehicle position detection unit 301, the communication module 302, and the object sensor 303, and drives the lateral direction operation unit 313, the longitudinal direction operation unit 314, and the longitudinal direction braking unit 315 via the vehicle control unit 312. This causes the vehicle 801 to travel automatically. The lateral direction operation unit 313, the longitudinal direction operation unit 314, and the longitudinal direction braking unit 315 are collectively referred to as the vehicle operation unit 317.

[0044] The vehicle position detection unit 301 can calculate the vehicle position using positioning information from a Global Navigation Satellite System (GNSS) that detects the vehicle position, a travel distance sensor that detects the number of rotations of the wheels of the vehicle 801, a gyro sensor that detects the acceleration, speed, angular acceleration, and angular velocity of the vehicle 801, and the like. Furthermore, the detection accuracy of the travel distance sensor is affected by tire air pressure. For this reason, the vehicle 801 may be equipped with a tire air sensor. The object sensor 303 may be any one or a combination of all of the sensors such as the image sensor 112, radio wave sensor 113, and optical sensor 114 equipped in the roadside monitoring device 110.

[0045] The driving control device 320 detects abnormalities in the object sensor 303, the vehicle position detection unit 301, etc., by the vehicle abnormal state detection unit 305. The driving control device 320 can also detect abnormalities in the communication module 302, and abnormal states of the roadside monitoring devices 110 and 120 by communicating with the edge server 200 via the communication module 302.

[0046] The driving control device 320 includes a map information integrating unit 304. The map information integrating unit 304 receives the vehicle position information IS output from the vehicle position detecting unit 301 and the object sensor information IO output from the object sensor 303. Furthermore, the map information integrating unit 304 receives the combined map information IOBM that the communication module 302 receives from the edge server 200, and high-level intersection escape routes RESCB for each of multiple types of possible driving distances in an abnormal situation.

[0047] The map information integrating unit 304 receives the above information, integrates the object information detected by the data collecting device 100 and the driving device 300, and updates the integrated map 316. The map information integrating unit 304 transmits the updated integrated map 316 to the driving route switching unit 307.

[0048] <Normal target route calculation unit> The normal-state target route calculation unit 308 of the driving route switching unit 307 calculates a normal-state target route based on the updated integrated map 316. Since the vehicle 801 was planned to turn right at the intersection in advance, a target route that does not collide with obstacles is generated based on the integrated map 316 and transmitted to the second selection unit 311.

[0049] When the normal target route calculated by the normal target route calculation unit 308 is confirmed to be appropriate, a second selection unit 311 selects the normal target route as the target route. Here, the target route may be expressed as a vector including any parameter among the vehicle position, vehicle speed, vehicle acceleration, vehicle jerk, etc. In order to travel along the target route, a target lateral operation amount, a target longitudinal operation amount, and a target longitudinal braking amount are calculated based on a motion model of the vehicle 801. Then, an operation command signal is output from the vehicle control unit 312 to the lateral operation unit 313, the longitudinal operation unit 314, and the longitudinal braking unit 315 of the vehicle operation unit 317.

[0050] Vehicle control unit 312 may specify a target steering angle, a target steering angular velocity, a target steering torque, etc. as the target lateral operation amount, and perform feedback control, feedforward control, or a combination of these controls to follow the target operation amount. Lateral operation unit 313 operates according to the target operation amount output from vehicle control unit 312. As lateral operation unit 313, an electric power steering device with a steering angle sensor and an electric power steering controller that controls the rotation of the electric power steering device are mounted on the vehicle. In this way, a configuration that performs follow-up control on the target steering angle, target angular velocity, or their derivative values ​​can be realized.

[0051] Vehicle control unit 312 may specify a target vehicle speed, a target acceleration or a target jerk, a target drive torque, etc. as the target longitudinal operation amount, and perform feedback control, feedforward control, or a combination of these controls to follow the target operation amount. A longitudinal operation unit 314 operates according to the target longitudinal operation amount output from vehicle control unit 312. As longitudinal operation unit 314, driving device 300 is provided with a configuration of a drive electric motor equipped with a rotation sensor and an inverter that controls the rotation of the drive electric motor. In this way, a configuration that performs control to follow the target longitudinal distance, the target vehicle speed, or a derivative value thereof can be realized.

[0052] The vehicle control unit 312 may specify the brake pressure as the target longitudinal braking operation amount, and perform feedback control, feedforward control, or a combination of these controls so as to follow the target operation amount. The longitudinal braking unit 315 operates according to the target longitudinal braking operation amount output from the vehicle control unit 312. As the longitudinal braking unit 315, a hydraulic brake and a control unit that controls the hydraulic brake pressure are mounted on the vehicle. In this way, a configuration can be realized in which the brake pressure is controlled so that the deceleration of the vehicle body follows the target deceleration.

[0053] <Vehicle-side escape route generation unit> The driving device 300 prioritizes escaping the intersection when an abnormal vehicle condition is detected, even if communication with the roadside monitoring devices 110, 120 or the edge server 200 is interrupted. For this purpose, the vehicle-side escape route generation unit 309 of the driving route switching unit 307 generates an intersection escape route based on the integrated map 316.

[0054] At this time, in order to prevent the load on the driving device from becoming too heavy, it is possible to generate only an escape route for the possible driving distance L1 (15 m) in the event of an abnormality of abnormality level 3. Since the driving distance is short, escape route calculation can be performed without delay, and an intersection escape route can be generated without communication with an edge server. In addition, a predetermined vehicle-side escape route generation distance may be set, and an intersection escape route may be generated for the vehicle-side escape route generation distance.

[0055] <High Level Escape Route Storage Unit> The driving route switching unit 307 stores the high-level intersection escape route RESCB transmitted from the edge server 200 in the high-level escape route storage unit 310. Then, a second selection unit 311 selects a route based on the vehicle abnormal state detected by the vehicle abnormal state detection unit 305 and the abnormality-condition travelable distance according to the vehicle abnormal state.

[0056] If there is no vehicle abnormality, the normal target route is selected. If the abnormality level is 3 and the abnormality travelable distance is L1 (15 m), the escape route generated by the vehicle-side escape route generation unit 309 is selected.

[0057] When the abnormality level is 1 to 2 and the remaining driving distance under abnormality is L3, L2.5, L2 (100 m, 50 m, 30 m), the high-level intersection escape route RESCB transmitted from the edge server 200 is selected. Since it is sufficient to adopt the high-level intersection escape route RESCB that has been generated and selected in advance by the edge server 200, the load on the driving device 300 does not become large. In addition, since the remaining driving distance under abnormality L3, L2.5, L2 (100 m, 50 m, 30 m) is relatively long, there are many intersection escape route generation patterns, and it is rational to have the edge server 200 generate and select the route, which also distributes the load.

[0058] The driving route calculated by the edge server 200 has a wider range than the intersection exit route calculated by the driving device 300. Therefore, the target stopping position can be set at a more appropriate point on the route to exit the intersection. In addition, the roadside monitoring device can monitor other vehicles traveling along the target route. This has the advantage that the vehicle 801 can be moved to a more appropriate point in a comprehensive view.

[0059] In the case of abnormality level 3 and abnormality-time travelable distance L1 (15 m), the high-level intersection escape route RESCB generated and selected in advance by the edge server 200 may be adopted. The escape route generated by the vehicle-side escape route generating unit 309 may be used only when communication with the roadside monitoring devices 110 and 120 or the edge server 200 becomes impossible. In addition, a vehicle-side escape route generating distance may be set in addition to the abnormality-time travelable distance L1, and when the abnormality-time travelable distance is equal to or less than the vehicle-side escape route generating distance, the escape route generated by the vehicle-side escape route generating unit 309 may be selected, and when the abnormality-time travelable distance is longer than the vehicle-side escape route generating distance, the high-level intersection escape route RESCB may be adopted. Only when the distance of the escape route is short, the vehicle-side escape route generating unit 309 of the travel route switching unit 307 generates the escape route, so that the load is not excessive. In addition, the range of the high-level intersection escape route RESCB generated and selected by the edge server 200 side can be set arbitrarily, so that the load on the edge server 200 side can also be adjusted.

[0060] <Control device hardware configuration> FIG. 6 is a hardware configuration diagram of the control device. The hardware configuration shown in FIG. 6 can be applied to the data collection device 100, the edge servers 200, 600, and the driving devices 300, 700. It can also be applied individually to the roadside monitoring devices 110, 120. Here, the case where it is applied to the driving device 300 will be described as a representative. In this embodiment, the driving device 300 is an electronic control device mounted on the vehicle 801 to make the vehicle 801 automatically drive. Each function of the driving device 300 is realized by a processing circuit provided in the driving device 300. Specifically, the driving device 300 includes, as processing circuits, an arithmetic processing device 90 (computer) such as a CPU (Central Processing Unit), a storage device 91 that exchanges data with the arithmetic processing device 90, an input circuit 92 that inputs an external signal to the arithmetic processing device 90, and an output circuit 93 that outputs a signal from the arithmetic processing device 90 to the outside. Each piece of hardware such as the arithmetic processing device 90, the storage device 91, the input circuit 92, and the output circuit 93 are connected to each other by a wired network such as a bus or a wireless network.

[0061] The arithmetic processing device 90 may be an application specific integrated circuit (ASIC), an integrated circuit (IC), a digital signal processor (DSP), a graphics processing unit (GPU), a field programmable gate array (FPGA), various logic circuits, various signal processing circuits, etc. Also, the arithmetic processing device 90 may be a plurality of devices of the same type or different types, and each process may be shared and executed. The storage device 91 may be a random access memory (RAM) configured to be able to read and write data from the arithmetic processing device 90, a read only memory (ROM) configured to be able to read data from the arithmetic processing device 90, etc. As the storage device 91, a non-volatile or volatile semiconductor memory such as a flash memory, a solid state drive (SSD), an EPROM, an EEPROM, etc., a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, etc. may be used. The input circuit 92 is connected to various sensors, switches, and communication lines, and includes an A / D converter, a communication circuit, and the like that inputs output signals and communication information of these sensors and switches to the arithmetic processing device 90. The output circuit 93 includes a drive circuit that outputs control signals from the arithmetic processing device 90, a communication circuit, and the like. The interfaces of the input circuit 92 and the output circuit 93 may be based on specifications such as CAN (Control Area Network) (registered trademark), Ethernet (registered trademark), USB (Universal Serial Bus) (registered trademark), DVI (Digital Visual Interface) (registered trademark), and HDMI (High-Definition Multimedia Interface) (registered trademark). In addition to the input circuit 92 and the output circuit 93, the arithmetic processing device 90 may be directly connected to a communication device 94 for communication.

[0062] Each function of the driving device 300 is realized by the calculation processing device 90 executing software (programs) stored in a storage device 91 such as a ROM, and cooperating with other hardware of the driving device 300 such as the storage device 91, the input circuit 92, and the output circuit 93. Setting data such as thresholds and judgment values ​​used by the driving device 300 is stored in the storage device 91 such as a ROM as part of the software (programs). Each function of the driving device 300 may be configured as a software module, or may be configured as a combination of software and hardware.

[0063] <Collision avoidance level> 7 is a diagram for explaining the relationship between objects and collision avoidance levels according to embodiment 1. The selection unit 206 of the edge server 200 calculates and evaluates the collision avoidance levels for the escape route RESC1 generated by the first escape route generation unit 204 and the escape route RESC2 generated by the second escape route generation unit 205. The selection unit 206 calculates the collision avoidance level for each escape route by calculating the collision risk with an object that will become an obstacle on the escape route.

[0064] If there are no obstacles and no objects that may collide with the vehicle 801, the score is 3 points, if there is only a stationary obstacle but no collision with the vehicle 801, the score is 2 points, if there is a moving obstacle but it will not collide with the vehicle 801 because it is moving in a different direction, etc., the score is 1 point, and if there is an obstacle that may cause a collision, or if there is an obstacle and the possibility of a collision is unknown, the score is 0 points. These scores are just an example, and the scores can be changed by adding, deleting, or changing conditions as necessary.

[0065] A collision avoidance level is calculated for each escape route. The collision avoidance route associated with the collision avoidance level is deactivated based on abnormality information of the roadside monitoring device. In other words, when an abnormal state of the roadside monitoring device is detected, the collision avoidance level of the escape route generated based on the roadside monitoring information IRSU obtained by the roadside monitoring device is set to 0 points.

[0066] Furthermore, the edge server 200 monitors the communication status between the roadside monitoring devices 110 and 120 and the edge server 200. If a delay longer than a predetermined value occurs in communication with a specific roadside monitoring device, the collision avoidance level of the escape route generated based on the roadside monitoring information obtained by that roadside monitoring device is set to 0 points.

[0067] Fig. 8 is a diagram for explaining the relationship between the stopping position and the collision avoidance level addition points according to the first embodiment. When the stopping positions that are the end points of the first escape route RESC1 and the second escape route RESC2 generated by the first escape route generating unit 204 and the second escape route generating unit 205 correspond to a possible stopping area, points are added to the collision avoidance level. The method of adding points can be determined arbitrarily by the designer, and for example, a bus stop stopping section, a stopping lane, and a time-limited parking section can be set to +3 points, a roadside strip can be set to +2 points, and a shoulder of a travel road can be set to +1 point. These points are just an example, and the points can be changed by adding, deleting, or changing conditions as necessary.

[0068] The selection unit 206 of the edge server 200 calculates the collision avoidance level of the escape route in this way, and evaluates it by adding the additional points. For each of multiple types of possible driving distances in an abnormal state, the escape route with the highest collision avoidance level score is selected as the high-level intersection escape route RESCB. Then, the edge server 200 transmits the high-level intersection escape route RESCB to the driving device 300.

[0069] In addition, locations where passengers can appropriately disembark, such as bus stop sections, parking lanes, time-limited parking sections, roadside strips, and road shoulders, which are candidate locations for the possible stopping areas, are recorded on the map. The additional points for the collision avoidance level for each candidate location for the possible stopping area may be recorded in the map information.

[0070] <Drivable distance in case of abnormality> Fig. 9 is a first diagram for explaining the relationship between an abnormal state and a driving distance of the vehicle according to the first embodiment. The driving control device 320 includes a driving distance calculation unit 306. The driving distance calculation unit 306 determines the abnormality level and the abnormality-state driving distance according to the abnormal state calculated by the vehicle abnormality detection unit 305. Fig. 9 shows how far the vehicle is allowed to travel when an abnormality occurs in the vehicle.

[0071] When there is no abnormality in the driving device 300, the abnormality level is set to 0, and the abnormality-state possible driving distance is set to L3 (first distance). A specific example of the abnormality-state possible driving distance L3 is exemplified as 100 m.

[0072] Next, consider the case of a minor abnormality that does not affect driving. This corresponds to the case where the temporary detection abnormality flag of one of the multiple sensors of the object sensor 303 and the vehicle position detection unit 301 is set to High. It is assumed that a temporary failure occurred, but the vehicle has now returned to normal. In this case, the abnormality level is defined as 1, and the possible driving distance under abnormality is defined as L3.

[0073] If one of the multiple sensors of the object sensor 303 and the vehicle position detection unit 301 is out of order, the abnormality level is set to 2 and the abnormality-state driving distance is set to L2 (second distance). A specific example of the abnormality-state driving distance L2 is 30 m. In this case, abnormalities in the weather sensor may be excluded. This is because abnormalities in the weather sensor often do not seriously affect the drivability of the autonomous vehicle.

[0074] When the object sensor 303 and the vehicle position detection unit 301 have multiple failures, the abnormality level is set to 3 and the abnormality-state driving distance is set to L1 (third distance). A specific example of the abnormality-state driving distance L1 is 15 m.

[0075] Fig. 10 is a second diagram for explaining the relationship between the abnormal state and the travelable distance according to the embodiment 1. Compared to Fig. 9, it differs in that items for abnormality level 1.5 and abnormality level 4 have been added. Only the different items will be explained.

[0076] Although the weather sensor abnormality is outside the operation design range, it is often possible to determine that the vehicle is still capable of traveling, so the abnormality level is set to 1.5, and the distance that can be traveled under abnormal conditions is set to L2.5 (fourth distance). As a specific example of the distance that can be traveled under abnormal conditions, L2.5 is given as 50 m.

[0077] Also, if it is difficult to move the vehicle 801 due to an abnormality in the drive motor or brakes, the abnormality level is set to 4, and the possible driving distance under abnormality is set to L0 (fifth distance). As a specific example of the possible driving distance under abnormality L0, 0 m is shown. In other words, the vehicle is not allowed to drive to escape the intersection and is immediately stopped.

[0078] As described above, the abnormal-state traveling distance is calculated by the traveling distance calculation unit 306 based on the vehicle abnormal state detected by the vehicle abnormal state detection unit 305. Then, the second selection unit 311 of the traveling route switching unit 307 selects an intersection escape route according to the abnormal-state traveling distance.

[0079] Here, the tire pressure sensor provided on the vehicle 801 and the abnormality-state drivable distance will be described. When the tire pressure is out of a predetermined standard value, the abnormality-state drivable distance may be changed and set according to the amount by which the tire pressure is out of the standard value. This is because a change in tire pressure changes the amount of vehicle movement per wheel rotation, which affects the calculation of the vehicle position.

[0080] Furthermore, the abnormality-state possible driving distance may be changed and set based on the abnormality of the vehicle position calculated using the vehicle position detection unit 301. The degree of abnormality of the vehicle position can be obtained by comparing the calculated vehicle position with the vehicle position obtained from the relative positions of the vehicle 801 and landmark topography, landmark buildings, objects, and signs detected by the object sensor 303. The abnormality-state possible driving distance is changed and set according to the degree of abnormality of the vehicle position.

[0081] When GNSS is used as the vehicle position detection unit 301, the abnormality drivable distance may be changed and set according to an abnormality state of the GNSS positioning. The positioning accuracy of the GNSS positioning is affected by deterioration of the accuracy of the receiving device provided in the vehicle 801, deterioration of the radio wave condition of the positioning satellite, deterioration of the accuracy of the clock of the positioning satellite, deviation of the orbit of the positioning satellite, etc. The driving control device 320 can detect an abnormality state of the GNSS positioning, and change and set the abnormality drivable distance according to the degree of the abnormality.

[0082] Also, the abnormality possible driving distance may be changed and set according to the abnormal state of the gyro sensor provided in the vehicle 801. For example, the abnormality possible driving distance may be changed and set according to the drift amount of the gyro sensor. This is significant because when the gyro sensor becomes abnormal, the error of the vehicle position calculated based on the output of the gyro sensor is accumulated according to the distance traveled by the vehicle.

[0083] The abnormality drivable distance may be changed and set according to the degree of the abnormality of the weather sensor equipped in the vehicle 801. Although an abnormality in the weather sensor often does not directly affect the detection of the vehicle's position, it is significant because it has a medium- to long-term effect.

[0084] In addition, the amount of delay in communication performed by the communication module 302 of the driving device 300 may be monitored, and the abnormality travelable distance may be changed and set based on the state of communication delay. This is significant because the occurrence of communication delays causes delays in object recognition based on external information.

[0085] <Selecting an escape route from an intersection> Fig. 11 is a diagram for explaining an intersection escape route according to the embodiment 1. Fig. 11 visually represents a possible driving distance in an abnormality. Fig. 11 shows a case where an abnormality occurs in the driving device 300 or the roadside monitoring device 120 after the vehicle 801 enters an intersection.

[0086] A drivable area 601, a drivable area 602, and a drivable area 603 are shown. These drivable areas correspond to abnormality-state drivable distances L1, L2, and L3, which are set according to the abnormality state of the vehicle. The drivable distance is the amount of travel permitted from when the driving device 300 detects an abnormality until the vehicle 801 is brought to an emergency stop. For example, the range is set by position information such as the maximum travel distance, latitude, and longitude. The drivable areas shown in FIG. 11 are defined based on the drivable distance. Specifically, when the abnormality-state drivable distance is set as L3>L2>L1, the areas of the respective areas are set as drivable area 603>drivable area 602>drivable area 601.

[0087] The abnormality-state drivable distance is calculated by the drivable distance calculation unit 306 based on the abnormal state of the vehicle detected by the vehicle abnormality state detection unit 305. If the abnormality level is 0 or 1, the abnormality-state drivable distance is set to L3 (the drivable area 603 is specified). This includes cases where there is no abnormality in the driving device 300 or where there is a minor abnormality that does not affect driving. This applies when one of the multiple object sensors 303 has a temporary detection abnormality flag set to High.

[0088] When the abnormality level becomes 2, the abnormality-state possible driving distance is set to L2 (the possible driving area 602 is specified). It is assumed that the abnormal state of one of the plurality of object sensors 303 continues.

[0089] When the abnormality level is 3, the abnormality-state travelable distance is set to L1 (the travelable area 601 is specified). The abnormal state is assumed to be the failure of multiple sensors. For example, when the driving device 300 detects an abnormality in the GNSS and yaw rate sensor, it can determine that travel is difficult within a range of the abnormality-state travelable distance L2 or more. When the abnormality-state travelable distance is L1, the top priority is given to escaping the intersection.

[0090] When the abnormality level is 4, the remaining driving distance under abnormality is set to L0. This corresponds to a case where it is difficult to move the vehicle 801 due to an abnormality in the drive motor or brakes. In this case only, it is physically impossible for the driving device 300 to move the vehicle 801. For this reason, the driving device 300 promptly stops the vehicle 801 at the spot where the abnormality is detected.

[0091] In this way, the vehicle 801, whose travelable distance during an abnormality is limited according to the abnormality level of the driving device 300, is limited in the range in which it can move from the point where the abnormality is detected to the designated stopping position. Within that range, the vehicle needs to stop at a position where the risk of colliding with an obstacle is low.

[0092] Therefore, the following describes switching to an escape route when the vehicle is abnormal, which is different from the normal target route, using the abnormality-state travelable distance output from the travelable distance calculation unit 306 as input information. This process is executed by the travel route switching unit 307.

[0093] When the remaining driving distance calculation unit 306 has set the remaining driving distance in an abnormal state to L3 (the remaining driving area 603 is specified), the end points of the escape route are the target stopping positions 853 and 856. In FIG. 11, the roadside monitoring device 120 is in an abnormal state, and it is necessary to select a target stopping position within the viewing angle 502 of the roadside monitoring device 110.

[0094] The target stopping position 856 is outside the viewing angle 502 and is therefore unsuitable. Note that a left turn is excluded from the candidates for the escape route of the vehicle 802. The area 511 is outside the viewing angle 502 of the roadside monitoring device 110 and therefore cannot be monitored, so it is excluded from the candidates.

[0095] Although an obstacle 520 exists along the escape route 505, it is determined that the obstacle is a stationary object that will not collide with, and the collision avoidance level is 2. And, since the target stopping position 853 exists in the stopping area 510, which is a stopping lane, the collision avoidance level is increased by +3.

[0096] Therefore, when the target stopping position 853 is set according to the escape route 505, the collision avoidance level is evaluated as 5, and is selected by the second selection unit 311. This means that a possible stopping area within the travel restriction range that can better avoid a collision can be selected, compared to stopping at a point after turning right.

[0097] When the abnormality-state driving distance is set to L2 (the driving area 602 is specified), the end points of the escape route are the target stopping positions 852 and 855. In FIG. 11, both of the target stopping positions 852 and 855 are within the viewing angle 502 of the roadside monitoring device 110.

[0098] Although an obstacle 520 exists on escape route 505 leading to target stopping position 852, it is determined that the obstacle 520 is a stationary object that will not collide with, and the collision avoidance level is 2. And since target stopping position 852 exists in stopping possible area 510, which is a low vehicle zone, the collision avoidance level is incremented by +3. Therefore, when target stopping position 852 is reached via escape route 505, the collision avoidance level is evaluated as 5.

[0099] Since escape route 504 leading to target stopping position 855 crosses the path of vehicle 811, there is a moving object that will not collide with, and the collision avoidance level becomes 1. If target stopping position 855 is the shoulder of the road, the additional point of the collision avoidance level becomes +1. Therefore, when target stopping position 855 is set by escape route 504, the collision avoidance level is evaluated as 2. In this case, escape route 505 with target stopping position 852 is selected by second selection unit 311. This means that a possible stopping area with a higher collision avoidance level within the abnormality travelable distance L2 can be selected, rather than stopping at a point after turning right.

[0100] When the abnormality-state driving distance is set to L1 (driving area 601 is specified), the end points of the escape route are target stopping positions 851 and 854. In Fig. 11, both target stopping positions 851 and 854 are within the viewing angle 502 of the roadside monitoring device 110. Since an obstacle 520 exists in the immediate vicinity of the target stopping position 851 and there is a possibility of a collision, the collision avoidance level is determined to be 0.

[0101] And since target stopping position 851 is in stopping possible area 510, which is a low vehicle zone, the collision avoidance level should be increased by +3. However, since obstacle 520 exists at target stopping position 851 and it is a location where stopping is physically impossible, the additional points are invalidated. Therefore, the collision avoidance level of the escape route to target stopping position 851 is evaluated as 0.

[0102] The escape route 504 leading to the target stopping position 854 crosses the path of the vehicle 811. Therefore, the collision avoidance level is 1. If the target stopping position 854 is the shoulder of the travelling road, the additional point of the collision avoidance level is +1. Therefore, if the target stopping position 854 is set based on the escape route 504, the collision avoidance level is evaluated as 2. In this case, the second selection unit 311 selects the escape route 504 as the target stopping position 854.

[0103] <Edge server processing> Fig. 12 is a flowchart showing the processing of the edge server 200 according to the first embodiment. The processing shown in Fig. 12 is executed by the arithmetic processing device of the edge server 200. This processing may be executed at predetermined time intervals (for example, every 5 ms). It may also be executed in response to an event such as receiving data by the communication module 302, instead of at predetermined time intervals.

[0104] 12 starts, and in step S101, data is received from a roadside monitoring device (abbreviated as RSU in the figure). Object information, information on the viewing angle of the sensor, and information on an abnormal state of the sensor are received via the communication module 201.

[0105] In step S102, combined map information is generated for each roadside monitoring device by superimposing the roadside monitoring information received from the roadside monitoring device on the map information. Specifically, the map generating unit 202 of the edge server 200 transmits first combined map information IOBM1 by superimposing the roadside monitoring information from the roadside monitoring device 110 on the map information 208 to the first escape route generating unit 204. Then, the map generating unit 202 transmits second combined map information IOBM2 by superimposing the roadside monitoring information from the roadside monitoring device 120 on the map information 208 to the second escape route generating unit 205.

[0106] In step S103, an intersection escape route is generated for each possible driving distance in an abnormal state based on the combined map information generated for each roadside monitoring device. In step S104, a collision avoidance level is calculated for each intersection escape route. At this time, the collision avoidance level addition point is also added.

[0107] In step S105, the collision avoidance levels are compared to evaluate the intersection escape routes. In step S106, the intersection escape route with the highest collision avoidance level is selected as the high-level intersection escape route RESCB. The high-level intersection escape route RESCB is selected by the selection unit 206 for each possible driving distance in an abnormal situation.

[0108] In step S107, the selected high-level intersection escape route RESCB is transmitted to the driving device. Specifically, the high-level intersection escape route RESCB selected by the selection unit 206 for each possible driving distance in an abnormality is transmitted to the driving device 300 via the communication module 207. At this time, the abnormality information FRSU regarding each roadside monitoring device is also transmitted to the driving device 300 via the communication module 207. Then, the process ends.

[0109] <Operating device processing> Fig. 13 is a flowchart showing the processing of driving device 300 according to embodiment 1. The processing shown in Fig. 13 is executed by the arithmetic processing device of driving device 300. This processing may be executed at predetermined time intervals (for example, every 5 ms). Instead of at predetermined time intervals, the processing may be executed in response to an event such as receiving data via communication module 302 or vehicle 801 traveling a predetermined distance.

[0110] 13 starts, and in step S201, the driving device 300 receives data from the edge server 200. Specifically, the communication module 302 receives the combined map information IOBM, the high-level intersection escape route RESCB, and the like from the edge server 200.

[0111] In step S202, the driving device 300 acquires vehicle position information. Specifically, the vehicle position detection unit 301 calculates the vehicle position using positioning information from a Global Navigation Satellite System (GNSS) that detects the vehicle position, a travel distance sensor that detects the number of rotations of the wheels of the vehicle 801, a gyro sensor that detects the acceleration, speed, angular acceleration, and angular velocity of the vehicle 801, and the like.

[0112] In step S203, the driving device 300 acquires object sensor information. Specifically, information is acquired from the object sensor 303 that detects the position, speed, and type of an object around the vehicle.

[0113] In step S204, the integrated map 316 is updated. The map information integrating unit 304 receives the vehicle position information IS output from the vehicle position detecting unit 301 and the object sensor information IO output from the object sensor 303. Furthermore, the map information integrating unit 304 receives the combined map information IOBM and the high-level intersection escape route RESCB received by the communication module 302 from the edge server 200. The map information integrating unit 304 receives these pieces of information as input, integrates the object information detected by the data collecting device 100 and the driving device 300, and updates the integrated map 316.

[0114] In step S205, the driving device 300 detects an abnormal vehicle state. The abnormal vehicle state detection unit 305 detects abnormalities in the object sensor 303, the vehicle position detection unit 301, and the like.

[0115] In step S300, driving device 300 calculates the remaining driving distance using remaining driving distance calculation unit 306. The remaining driving distance is calculated according to the detected abnormal vehicle state. Details of the process in step S300 are described in FIG.

[0116] In step S206, it is determined whether or not an abnormality has been detected by the vehicle abnormal state detection unit 305. If an abnormality has been detected (determination is YES), the process proceeds to step S207. If no abnormality has been detected (determination is NO), the process proceeds to step S212, where the normal-state target route calculation unit 308 calculates the normal-state target route. Then, the process proceeds to step S210.

[0117] In step S207, it is determined whether the vehicle abnormality level is 3. If it is not 3 (determination is NO), proceed to step S208. If the vehicle abnormality level is 3 (determination is YES), proceed to step S213 and generate a vehicle-side intersection escape route. The vehicle-side escape route generation unit 309 generates the intersection escape route based on the integrated map 316 in order to prioritize escaping the intersection when a vehicle abnormality state of abnormality level 3 is detected. Then, proceed to step S210.

[0118] In step S208, it is determined whether the vehicle's abnormality level is greater than 3. If it is greater than 3 (determination is YES), proceed to step S209, issue an instruction to stop the vehicle, and proceed to step S210. This is because the vehicle's abnormality level is 4, so it will be difficult to escape the intersection and the vehicle should stop immediately. In step S208, if the vehicle's abnormality level is not greater than 3 (determination is NO) (the vehicle's abnormality level is 1 or 2), proceed to step S214, and read out the high-level intersection escape route RESCB corresponding to the remaining driving distance in the abnormal state. Then, proceed to step S210.

[0119] In step S210, the route that was generated or read immediately before is set as the travel route. Then, in step S211, vehicle control unit 312 executes vehicle travel control. Then, the process ends.

[0120] <Drivable distance calculation process> Fig. 14 is a flowchart showing the process of calculating the remaining driving distance by the remaining driving distance calculation unit 606 of the driving device 300 according to the first embodiment based on Fig. 10. The process in Fig. 14 illustrates details of step S300 in Fig. 13.

[0121] In step S301, the abnormal state of the driving device 300 detected by the vehicle abnormal state detection unit 305 is read. In step S302, it is determined whether the abnormality level is 1 or less. If the abnormality level is 1 or less (determination is YES), the process proceeds to step S303, where the remaining driving distance is set to L3, and the process ends.

[0122] In step S302, if the abnormality level is not 1 or less (determination is NO), the process proceeds to step S304. In step S304, it is determined whether the abnormality level is 1.5 or less. If the abnormality level is 1.5 or less (determination is YES), the process proceeds to step S305, the remaining driving distance is set to L2.5, and the process ends.

[0123] In step S304, if the abnormality level is not 1.5 or less (determination is NO), the process proceeds to step S306. In step S306, it is determined whether the abnormality level is 2 or less. If the abnormality level is 2 or less (determination is YES), the process proceeds to step S307, where the remaining driving distance is set to L2, and the process ends.

[0124] In step S306, if the abnormality level is not 2 or less (determination is NO), proceed to step S308. In step S308, it is determined whether the abnormality level is 3 or less. If the abnormality level is 3 or less (determination is YES), proceed to step S309, set the remaining driving distance to L1, and end the process. In step S308, if the abnormality level is not 3 or less (determination is NO), proceed to step S310. In step S310, set the remaining driving distance to L0, and end the process.

[0125] 2. Second embodiment <Configuration of the Autonomous Driving Integrated Control Device> Fig. 15 is an overall configuration diagram of an autonomous driving integrated control device 2 according to embodiment 2. Fig. 16 is a configuration diagram of an edge server 600 according to embodiment 2. Fig. 17 is a configuration diagram of an operation device 700 according to embodiment 2.

[0126] The data collection device 100 according to the second embodiment is the same as that according to the first embodiment, and therefore a description thereof will be omitted. As shown in Figs. 15 to 17, in the driving device 700 according to the second embodiment, the functions of the driving range calculation unit 606 and the vehicle control unit 612 have been transferred to the edge server 600. This is what makes it different from the driving device 300 according to the first embodiment. However, the functions performed by the entire autonomous driving integrated control device 2 according to the second embodiment correspond to the functions performed by the entire autonomous driving integrated control device 1 according to the first embodiment. The edge server 600 and the driving device 700 constitute an autonomous driving device 902 in which the data collection device 100 is an external device.

[0127] 17, driving device 700 according to the second embodiment includes a vehicle position detection unit 701 that detects vehicle position information, an object sensor 703 that detects objects around the vehicle, and a vehicle abnormality detection unit 705 that detects an abnormal state of the vehicle. Vehicle position information IS, object information IO, and vehicle abnormality state information FV are transmitted to edge server 600 via communication module 702.

[0128] High-speed communication is performed between the data collection device 100, the communication module 702 of the driving device 700, and the communication modules 621 and 626 of the edge server 600. The communication module 702 also receives vehicle control information transmitted from the edge server 600. The vehicle control information operates a lateral operation unit 713, a longitudinal operation unit 714, and a longitudinal braking unit 715, causing the vehicle 802 to travel automatically. The lateral operation unit 713, the longitudinal operation unit 714, and the longitudinal braking unit 715 are collectively referred to as a vehicle operation unit 717.

[0129] The edge server 600 is equipped with a communication module 621, which transmits the vehicle position information IS and object sensor information IO from the driving device 700, and the roadside monitoring information IRSU and abnormality information FRSU from the roadside monitoring devices 110 and 120 to the map information integrator 604. In addition, the edge server 600 transmits vehicle abnormality state information FV to a vehicle abnormality state reflector 605, and a driving distance calculator 606 calculates the driving distance.

[0130] The map information integrating unit 604 is provided in the driving device 300 in the first embodiment, but is provided in the edge server 600 in the second embodiment. If there is no abnormality in the vehicle 802 or the roadside monitoring devices 110, 120, etc., the integrated map 616 is updated by superimposing on the map the vehicle position information IS, the roadside monitoring information IRSU measured by the roadside monitoring devices, and the object sensor information IO detected by the vehicle.

[0131] <Edge server processing> Fig. 18 is a first flowchart showing the processing of the edge server 600 according to embodiment 2. Fig. 19 is a second flowchart showing the processing of the edge server.

[0132] 18 is executed by the arithmetic processing unit of the edge server 600. This process may be executed at predetermined time intervals (for example, every 5 ms). It may also be executed in response to an event such as receiving data by the communication module 621, instead of at predetermined time intervals.

[0133] The process in Fig. 18 differs from the process in Fig. 12 according to the first embodiment only in that step S107 in Fig. 12 is deleted. In the first embodiment, high-level escape route data and the like are transmitted to the driving device and the subsequent processing is left to the driving device. In the second embodiment, the edge server 600 itself performs the subsequent processing, so the transmission process is deleted.

[0134] The process shown in Fig. 19 is executed by the arithmetic processing unit of the edge server 600. This process may be executed at predetermined time intervals (for example, every 5 ms). It may also be executed in response to an event such as receiving data by the communication module 621, rather than at predetermined time intervals. The process in Fig. 19 may also be executed following the process in Fig. 18.

[0135] The process in FIG. 19 corresponds to the process in FIG. 13 that was performed by the driving device 300 in the first embodiment. The difference is that steps S201, S205, and S211 in FIG. 13 are changed to steps S221, S225, and S231 in FIG. 19. Only the changed parts will be explained. For the other steps, the execution subject will be changed from the driving device 300 to the edge server 600. The details of the process of step S300 in FIG. 19 are also applied to the process in FIG. 14, with the execution subject changed from the driving device 300 to the edge server 600.

[0136] 19, data is received from the driving device 700. This is because the edge server 600 obtains the vehicle position information IS, the object sensor information IO, and the vehicle abnormality state information FV collected by the driving device 700.

[0137] In step S225 of Fig. 19, the vehicle abnormal state is reflected. Since the vehicle abnormal state information FV has already been received from the driving device 700, there is no need to newly detect the vehicle abnormal state. Referencing already received data is described as reflecting the vehicle abnormal state.

[0138] In step S231 in Fig. 19, vehicle driving data is sent to the driving device 700. In Fig. 13, it was described that vehicle driving control is executed by the driving device 300. In Fig. 19, the edge server 600 creates vehicle driving data and requests the driving device 700 to execute it.

[0139] <Operating device processing> Figure 20 is a first flowchart showing the processing of driving device 700 according to embodiment 2. The processing shown in Figure 20 is executed by the arithmetic processing unit of driving device 700. This processing may be executed at predetermined time intervals (for example, every 5 ms). Instead of at predetermined time intervals, the processing may be executed in response to an event such as the completion of each of the following processes: detection of the vehicle position by vehicle position detection unit 701, detection of objects around the vehicle by object sensor 703, and detection of an abnormal vehicle state by vehicle abnormal state detection unit 705.

[0140] The process of FIG. 20 starts, and in step S202, the driving device 700 acquires vehicle position information. In step S203, the driving device 700 acquires object sensor information. In step S241, the driving device 700 detects an abnormal vehicle state. In step S242, the driving device 700 transmits the vehicle position information, object information, and vehicle abnormal state information to the edge server 600 via the communication module 702. Then, the process ends.

[0141] Fig. 21 is a second flowchart showing the processing of driving device 700 according to embodiment 2. The processing shown in Fig. 21 is executed by the arithmetic processing unit of driving device 700. This processing may be executed at predetermined time intervals (for example, every 5 ms). Instead of at predetermined time intervals, it may be executed in response to an event such as receiving data by communication module 702.

[0142] 21 starts, and in step S251, the communication module 702 receives vehicle driving data from the edge server 600. Then, in step S211, the communication module 702 transmits an operation signal to the vehicle operation unit 717 based on the vehicle control information. That is, the operation signal is transmitted to the lateral direction operation unit 713, the longitudinal direction operation unit 714, and the longitudinal direction braking unit 715, causing the vehicle 802 to automatically drive. Then, the process ends.

[0143] In the second embodiment, the functions of the driving device 700 mounted on the vehicle 802 are transferred to the edge server 600, so that the driving device 700 can be made smaller and lighter. This can reduce the cost of the driving device 700 and save labor in changing specifications. Specifically, the capacity of the CPU and memory mounted on the driving device 700 can be reduced. Also, while the capacity of the driving device 700 mounted on the vehicle 802 is limited, the edge server 600 installed outside the vehicle 802 can share the processing contents with multiple hardware devices, and the capacity is not severely limited. This is therefore advantageous for executing a large amount of processing at high speed.

[0144] Although various exemplary embodiments and examples are described in this application, various features, aspects, and functions described in one or more embodiments are not limited to the application of a specific embodiment, but can be applied to the embodiments alone or in various combinations. Therefore, countless modifications not illustrated are expected within the scope of the technology disclosed in this specification. For example, the following are included: when at least one component is modified, added, or omitted, and further, when at least one component is extracted and combined with a component of another embodiment.

[0145] Various aspects of the present disclosure are summarized below as appendices.

[0146] (Appendix 1) A sensor for detecting surrounding objects; and a plurality of roadside monitoring devices each having a transmitter for transmitting information on a viewing angle of the sensor and information on the object detected by the sensor; a receiving unit that receives the viewing angle information and the object information from the transmitter of the roadside monitoring device; a vehicle abnormality detection unit that detects an abnormality in the vehicle; an escape route generating unit that generates an intersection escape route for escaping the intersection when an abnormality occurs in the vehicle that has entered the intersection, for each of the roadside monitoring devices and for each of a plurality of possible travel distances in the event of an abnormality, based on information received from the roadside monitoring devices; a selection unit that compares the intersection escape routes generated by the escape route generation unit according to a collision avoidance level at which the intersection escape routes can avoid a collision with the object while the vehicle is traveling, and selects the intersection escape route having the highest collision avoidance level as a high-level intersection escape route for each of the abnormality-time travelable distances; a travelable distance calculation unit that calculates an abnormality-state travelable distance according to the abnormal state when an abnormal state of the vehicle is detected by the vehicle abnormality detection unit; a vehicle control unit that outputs a command signal to cause the vehicle to travel along the high-level intersection escape route selected by the selection unit, the high-level intersection escape route corresponding to the abnormality-state travelable distance calculated by the travelable distance calculation unit, when an abnormality state of the vehicle that has entered the intersection is detected by the vehicle abnormality detection unit; and and an autonomous driving device having a vehicle operation unit that drives an actuator based on a command signal output by the vehicle control unit. (Appendix 2) The autonomous driving integrated control device of claim 1, wherein the selection unit of the autonomous driving device sets the collision avoidance level to 3 when no object is detected on the intersection escape route, to 2 when a stationary object is present on the intersection escape route but will not collide, to 1 when a moving object is present on the intersection escape route but will not collide, and to 0 when an object is present on the intersection escape route and there is a collision or the possibility of a collision is unknown. (Appendix 3) The automatic driving integrated control device of Appendix 2, wherein the selection unit of the autonomous driving device adds +3 to the collision avoidance level if the stopping point that is the end point of the intersection escape route is a bus stop stopping section, a stopping lane, or a time-limited parking section, +2 if the stopping point of the intersection escape route is a shoulder strip, and +1 if the stopping point of the intersection escape route is a shoulder. (Appendix 4) The autonomous driving device has a plurality of on-board sensors mounted on the vehicle to detect surrounding objects, The autonomous driving integrated control device according to any one of Appendices 1 to 3, wherein the drivable distance calculation unit of the autonomous driving device sets, as the drivable distance in an abnormal state, a predetermined first distance when one of the on-board sensors experiences a temporary failure, sets a predetermined second distance shorter than the first distance when one of the on-board sensors experiences a continuous failure, and sets a predetermined third distance shorter than the second distance when multiple on-board sensors experience continuous failures. (Appendix 5) The autonomous driving device has a plurality of on-board sensors mounted on the vehicle to detect surrounding objects, and a weather sensor mounted on the vehicle to detect surrounding weather conditions, The autonomous driving integrated control device according to any one of Appendices 1 to 3, wherein the drivable distance calculation unit of the autonomous driving device sets, as the drivable distance in an abnormal state, a predetermined first distance when one of the on-board sensors experiences a temporary failure, a predetermined fourth distance shorter than the first distance when the weather sensor experiences a failure, a predetermined second distance shorter than the fourth distance when one of the on-board sensors experiences a continuous failure, a predetermined third distance shorter than the second distance when multiple on-board sensors experience a continuous failure, and a predetermined fifth distance shorter than the third distance when a failure occurs in the vehicle's drive device or braking device. (Appendix 6) The autonomous driving integrated control device according to claim 5, wherein the driving range calculation unit of the autonomous driving device sets the fifth distance to 0 as the abnormality-state driving range. (Appendix 7) The autonomous driving device has a plurality of on-board sensors mounted on the vehicle to detect surrounding objects, The vehicle control unit of the autonomous driving device, a vehicle-side escape route generation unit that generates an intersection escape route for escaping and stopping the intersection when an abnormality occurs in the vehicle that has entered the intersection, for a predetermined vehicle-side escape route generation distance based on the information received from the roadside monitoring device and the information detected by the in-vehicle sensor, and sets the generated intersection escape route as a vehicle-side intersection escape route; a second selection unit that, when an abnormal state of the vehicle that has entered the intersection is detected by the vehicle abnormal state detection unit, selects the high-level intersection escape route corresponding to the abnormal state travelable distance if the abnormal state travelable distance calculated by the travelable distance calculation unit is longer than the vehicle-side escape route generation distance, and selects the vehicle-side intersection escape route generated by the vehicle-side escape route generation unit if the abnormal state travelable distance calculated by the travelable distance calculation unit is equal to or shorter than the vehicle-side escape route generation distance, The vehicle control unit outputs a command signal to cause the vehicle to travel along the high-level intersection escape route or the vehicle-side intersection escape route selected by the second selection unit. (Appendix 8) The vehicle control unit of the autonomous driving device, a vehicle-side escape route generating unit that generates an intersection escape route for escaping and stopping the vehicle from the intersection when an abnormality occurs in the vehicle that has entered the intersection, for the third distance based on the information received from the roadside monitoring device and the information detected by the in-vehicle sensor, and sets the generated intersection escape route as a vehicle-side intersection escape route; a second selection unit that, when an abnormal state of the vehicle that has entered the intersection is detected by the vehicle abnormal state detection unit, selects the high-level intersection escape route corresponding to the abnormal state travelable distance if the abnormal state travelable distance calculated by the travelable distance calculation unit is the first distance or the second distance, and selects the vehicle-side intersection escape route generated by the vehicle-side escape route generation unit if the abnormal state travelable distance calculated by the travelable distance calculation unit is the third distance, The vehicle control unit outputs a command signal to cause the vehicle to travel along the high-level intersection escape route or the vehicle-side intersection escape route selected by the second selection unit. (Appendix 9) The autonomous driving comprehensive control device according to any one of appendix 4 to 8, wherein the on-board sensor of the autonomous driving device includes at least one of a GNSS, a gyro sensor, or a tire pressure sensor. (Appendix 10) The vehicle abnormality state detection unit of the autonomous driving device detects an abnormal state of the vehicle by including a communication delay amount, The autonomous driving device is configured to calculate the abnormality-state driving distance according to an amount of delay in communication. (Appendix 11) The vehicle abnormality state detection unit of the autonomous driving device detects an abnormal state of the vehicle and includes a position abnormality of the vehicle, The autonomous driving integrated control device according to any one of appendices 1 to 10, wherein the driving range calculation unit of the autonomous driving device calculates the abnormality-state driving range according to the amount of abnormality in the position of the vehicle. (Appendix 12) The vehicle abnormality state detection unit of the autonomous driving device detects an abnormality in the positioning of a GNSS mounted on the vehicle as an abnormality of the vehicle, The autonomous driving integrated control device according to any one of claims 1 to 11, wherein the driving range calculation unit of the autonomous driving device calculates the abnormal driving range in response to an abnormal state of the GNSS positioning. (Appendix 13) The vehicle abnormality state detection unit of the autonomous driving device detects an abnormality of the vehicle, the abnormality state being detected by the vehicle abnormality sensor mounted on the vehicle, The autonomous driving integrated control device according to any one of appendixes 1 to 12, wherein the driving range calculation unit of the autonomous driving device calculates the abnormal driving range in response to an abnormal state of the gyro sensor. (Appendix 14) The vehicle abnormality detection unit of the autonomous driving device detects an abnormality of the vehicle, the abnormality being detected by a weather sensor mounted on the vehicle, The autonomous driving device is configured to calculate the abnormality-state drivable distance in accordance with an abnormality state of the weather sensor. (Appendix 15) The vehicle abnormality state detection unit of the autonomous driving device detects an abnormality of the vehicle, the abnormality being an abnormality of a tire pressure sensor mounted on the vehicle, The autonomous driving device is configured to calculate the abnormality-state drivable distance in accordance with an abnormality state of the tire pressure sensor. (Appendix 16) The automatic driving integrated control device according to any one of appendices 1 to 15, wherein the selection unit of the autonomous driving device compares the collision avoidance levels of the intersection escape routes generated based on information received from the roadside monitoring device in which a communication delay has occurred with a collision avoidance level of 0, and selects the intersection escape route with the highest collision avoidance level as a high-level intersection escape route for each possible driving distance in the event of an abnormality. (Appendix 17) An autonomous driving integrated control device according to any one of appendixes 1 to 16, wherein the escape route generation unit of the autonomous driving device generates an intersection escape route whose driving route falls within the field of view of the sensor based on the field of view angle information received from the roadside monitoring device. (Appendix 18) The autonomous driving device includes an edge server connected to the vehicle via communication, and the vehicle abnormality detection unit, the driving range calculation unit, the vehicle control unit, and the vehicle operation unit are mounted on the vehicle. (Appendix 19) An autonomous driving integrated control device as described in any one of Appendices 1 to 17, wherein the receiving unit, the escape route generating unit, the selection unit, the driving distance calculating unit, and the vehicle control unit of the autonomous driving device are provided in an edge server connected via communication with the vehicle, and the vehicle abnormality state detection unit and the vehicle operation unit are mounted on the vehicle. [Explanation of symbols]

[0147] 1, 2 Autonomous driving integrated control device, 100 Data collection device, 110, 120 Roadside monitoring device, 112 Image sensor, 113 Radio wave sensor, 114 Optical sensor, 118, 201, 207, 302, 621, 626, 702 Communication module, 200, 600 Edge server, 204 First escape route generation unit, 205 Second escape route generation unit, 206 Selection unit, 300, 700 Driving device, 301, 701 Vehicle position detection unit, 303, 703 Object sensor, 305, 705 Vehicle abnormal state detection unit, 306, 606 Driving distance calculation unit, 309 Vehicle side escape route generation unit, 311 Second selection unit, 312, 612 Vehicle control unit, 317, 717 Vehicle operation unit, 801, 802, 811 Vehicle, 851, 852, 853, 854, 855, 856 Target stopping position, 901, 902 Autonomous driving device

Claims

1. A sensor for detecting surrounding objects; and a plurality of roadside monitoring devices each having a transmitter for transmitting information on a viewing angle of the sensor and information on the object detected by the sensor; a receiving unit that receives the viewing angle information and the object information from the transmitter of the roadside monitoring device; a vehicle abnormality detection unit that detects an abnormality in the vehicle; an escape route generating unit that generates an intersection escape route for escaping the intersection when an abnormality occurs in the vehicle that has entered the intersection, for each of the roadside monitoring devices and for each of a plurality of possible travel distances in the event of an abnormality, based on information received from the roadside monitoring devices; a selection unit that compares the intersection escape routes generated by the escape route generation unit according to a collision avoidance level at which the intersection escape routes can avoid a collision with the object while the vehicle is traveling, and selects the intersection escape route having the highest collision avoidance level as a high-level intersection escape route for each of the abnormality-time travelable distances; a travelable distance calculation unit that calculates an abnormality-state travelable distance according to the abnormal state when an abnormal state of the vehicle is detected by the vehicle abnormality detection unit; a vehicle control unit that outputs a command signal to cause the vehicle to travel along the high-level intersection escape route selected by the selection unit, the high-level intersection escape route corresponding to the abnormality-state travelable distance calculated by the travelable distance calculation unit, when an abnormality state of the vehicle that has entered the intersection is detected by the vehicle abnormality detection unit; and and an autonomous driving device having a vehicle operation unit that drives an actuator based on a command signal output by the vehicle control unit.

2. The automatic driving integrated control device of claim 1, wherein the selection unit of the autonomous driving device sets the collision avoidance level to 3 when no object is detected on the intersection escape route, to 2 when a stationary object is present on the intersection escape route but will not collide, to 1 when a moving object is present on the intersection escape route but will not collide, and to 0 when an object is present on the intersection escape route and there is a collision or the possibility of a collision is unknown.

3. 3. The automatic driving integrated control device according to claim 2, wherein the selection unit of the autonomous driving device adds +3 to the collision avoidance level if the stopping point that is the end point of the intersection escape route is a bus stop stopping section, a stopping lane, or a time-limited parking section, +2 if the stopping point of the intersection escape route is a shoulder strip, and +1 if the stopping point of the intersection escape route is a shoulder.

4. The autonomous driving device has a plurality of on-board sensors mounted on the vehicle to detect surrounding objects, The autonomous driving integrated control device of claim 1, wherein the remaining distance calculation unit of the autonomous driving device sets a predetermined first distance as the remaining distance in an abnormal state when one of the on-board sensors experiences a temporary failure, sets a predetermined second distance shorter than the first distance when one of the on-board sensors experiences a continuous failure, and sets a predetermined third distance shorter than the second distance when multiple on-board sensors experience continuous failures.

5. The autonomous driving device includes a plurality of on-board sensors mounted on the vehicle to detect surrounding objects, and a weather sensor mounted on the vehicle to detect surrounding weather conditions, The autonomous driving integrated control device of claim 1, wherein the remaining distance calculation unit of the autonomous driving device sets, as the remaining distance in an abnormal state, a predetermined first distance when one of the on-board sensors experiences a temporary failure, a predetermined fourth distance shorter than the first distance when the weather sensor experiences a failure, a predetermined second distance shorter than the fourth distance when one of the on-board sensors experiences a continuous failure, a predetermined third distance shorter than the second distance when multiple on-board sensors experience a continuous failure, and a predetermined fifth distance shorter than the third distance when a failure occurs in the vehicle's drive device or braking device.

6. The autonomous driving integrated control device according to claim 5 , wherein the remaining driving distance calculation unit of the autonomous driving device sets the fifth distance to 0 as the remaining driving distance in an abnormal state.

7. The autonomous driving device has a plurality of on-board sensors mounted on the vehicle to detect surrounding objects, The vehicle control unit of the autonomous driving device, a vehicle-side escape route generation unit that generates an intersection escape route for escaping and stopping the intersection when an abnormality occurs in the vehicle that has entered the intersection, for a predetermined vehicle-side escape route generation distance based on the information received from the roadside monitoring device and the information detected by the in-vehicle sensor, and sets the generated intersection escape route as a vehicle-side intersection escape route; a second selection unit that, when an abnormal state of the vehicle that has entered the intersection is detected by the vehicle abnormal state detection unit, selects the high-level intersection escape route corresponding to the abnormal state travelable distance if the abnormal state travelable distance calculated by the travelable distance calculation unit is longer than the vehicle-side escape route generation distance, and selects the vehicle-side intersection escape route generated by the vehicle-side escape route generation unit if the abnormal state travelable distance calculated by the travelable distance calculation unit is equal to or shorter than the vehicle-side escape route generation distance, The automatic driving integrated control device according to claim 1, wherein the vehicle control unit outputs a command signal to cause the vehicle to travel along the high-level intersection escape route or the vehicle-side intersection escape route selected by the second selection unit.

8. The vehicle control unit of the autonomous driving device, a vehicle-side escape route generating unit that generates an intersection escape route for escaping and stopping the vehicle from the intersection when an abnormality occurs in the vehicle that has entered the intersection, for the third distance based on the information received from the roadside monitoring device and the information detected by the in-vehicle sensor, and sets the generated intersection escape route as a vehicle-side intersection escape route; a second selection unit that, when an abnormal state of the vehicle that has entered the intersection is detected by the vehicle abnormal state detection unit, selects the high-level intersection escape route corresponding to the abnormal state travelable distance if the abnormal state travelable distance calculated by the travelable distance calculation unit is the first distance or the second distance, and selects the vehicle-side intersection escape route generated by the vehicle-side escape route generation unit if the abnormal state travelable distance calculated by the travelable distance calculation unit is the third distance, The automatic driving integrated control device according to claim 4, wherein the vehicle control unit outputs a command signal to cause the vehicle to travel along the high-level intersection escape route or the vehicle-side intersection escape route selected by the second selection unit.

9. The autonomous driving comprehensive control device according to claim 4 , wherein the on-board sensor of the autonomous driving device includes at least one of a GNSS, a gyro sensor, or a tire pressure sensor.

10. The vehicle abnormality state detection unit of the autonomous driving device detects an abnormal state of the vehicle by including a communication delay amount, The autonomous driving integrated control device according to claim 1 , wherein the remaining driving distance calculation unit of the autonomous driving device calculates the remaining driving distance in an abnormal state according to an amount of delay in the communication.

11. The vehicle abnormality state detection unit of the autonomous driving device detects an abnormal state of the vehicle and includes a position abnormality of the vehicle, The autonomous driving integrated control device according to claim 1 , wherein the remaining driving distance calculation unit of the autonomous driving device calculates the remaining driving distance in an abnormal state according to an amount of abnormal position of the vehicle.

12. The vehicle abnormality state detection unit of the autonomous driving device detects an abnormality in the positioning of a GNSS mounted on the vehicle as an abnormality of the vehicle, The autonomous driving integrated control device according to claim 1 , wherein the remaining distance calculation unit of the autonomous driving device calculates the remaining distance in an abnormal state according to an abnormal state of the GNSS positioning.

13. The vehicle abnormality state detection unit of the autonomous driving device detects an abnormality of the vehicle, the abnormality state being an abnormality of a gyro sensor mounted on the vehicle, The autonomous driving integrated control device according to claim 1 , wherein the remaining distance calculation unit of the autonomous driving device calculates the remaining distance under abnormal conditions in accordance with the abnormal state of the gyro sensor.

14. The vehicle abnormality detection unit of the autonomous driving device detects an abnormality of the vehicle, the abnormality being detected by a weather sensor mounted on the vehicle, The autonomous driving integrated control device according to claim 1 , wherein the remaining distance calculation unit of the autonomous driving device calculates the remaining distance under abnormal conditions in accordance with an abnormal state of the weather sensor.

15. The vehicle abnormality state detection unit of the autonomous driving device detects an abnormality of the vehicle, the abnormality being an abnormality of a tire pressure sensor mounted on the vehicle, The autonomous driving integrated control device according to claim 1 , wherein the remaining driving distance calculation unit of the autonomous driving device calculates the remaining driving distance under abnormal conditions in accordance with an abnormal state of the tire pressure sensor.

16. The automatic driving comprehensive control device of claim 1, wherein the selection unit of the autonomous driving device compares the collision avoidance levels of the intersection escape routes generated based on information received from the roadside monitoring device in which a communication delay has occurred with a collision avoidance level of 0, and selects the intersection escape route with the highest collision avoidance level as a high-level intersection escape route for each possible driving distance in the event of an abnormality.

17. The autonomous driving integrated control device according to claim 1, wherein the escape route generation unit of the autonomous driving device generates an intersection escape route whose driving route falls within the field of view of the sensor based on the field of view angle information received from the roadside monitoring device.

18. The autonomous driving device according to any one of claims 1 to 17, wherein the receiving unit, the escape route generating unit, and the selecting unit of the autonomous driving device are provided in an edge server connected to the vehicle via communication, and the vehicle abnormality detection unit, the driving range calculation unit, the vehicle control unit, and the vehicle operation unit are mounted on the vehicle.

19. An autonomous driving comprehensive control device as described in any one of claims 1 to 17, wherein the receiving unit, the escape route generating unit, the selection unit, the driving range calculating unit, and the vehicle control unit of the autonomous driving device are provided in an edge server connected to the vehicle via communication, and the vehicle abnormality state detection unit and the vehicle operation unit are mounted on the vehicle.

Images