JP7254988B1 - Traffic management system, traffic management device, control method for traffic management device, and control program for traffic management device - Google Patents

Abstract

An object of the present invention is to provide highly accurate driving assistance in an automatic driving vehicle. [Solution] An operation management system for an autonomous vehicle that travels by referring to recommended speed data related to recommended speeds set for each road collects actual operation data collected by the autonomous vehicle in operation from the autonomous vehicle. The acquisition unit that acquires, the incident data related to the incident that occurred during the operation of the automated driving vehicle, and the external state data related to the external state of the automated driving vehicle at the time of the incident, which are included in the operation performance data, A storage unit that stores learning data associated with each road on which an accident has occurred, and a prediction model that predicts the occurrence of an incident on the road in operation is generated from the data outside the vehicle, using the learning data. and an updating unit for updating the recommended speed set for the road in operation in the recommended speed data according to the result of prediction by the prediction model. [Selection drawing] Fig. 5

Description

The present invention relates to an operation management system, an operation management device, a control method for the operation management device, and a control program for the operation management device.

2. Description of the Related Art Conventionally, there has been known a driving assistance device that performs driving assistance according to the surrounding conditions of a vehicle (for example, Patent Document 1). The driving support device described in Patent Document 1 corrects a target speed set based on the speed limit and road width of the road on which the vehicle is traveling according to the surrounding conditions of the vehicle, and calculates the corrected target speed and the vehicle speed. Driving assistance is executed by informing the driver that the vehicle needs to be decelerated.

JP 2020-060893 A

BACKGROUND ART In recent years, so-called self-driving vehicles that run without the need for driving operations such as steering and braking by the driver are spreading. Autonomous vehicles also require speed setting according to the road on which the vehicle travels and the surrounding conditions. It is difficult to apply to automatic driving vehicles that do not operate. In addition, for example, when a bus that carries passengers is realized with an autonomous vehicle, the safety and comfort of passengers will be ensured while considering the impact of the safety of passers-by and the traveling speed of the autonomous vehicle on traffic. , there has been a demand for more accurate driving assistance.

According to one embodiment of the present invention, an operation management system that manages the operation of a plurality of automated driving vehicles that run with reference to recommended speed data related to recommended speeds set for each road, and an automated driving vehicle in operation Acquisition unit that acquires the operation performance data collected by the automated driving vehicle from the automated driving vehicle, and the incident data related to the incident that occurred during the operation of the automated driving vehicle included in the operation performance data, and the automated driving vehicle at the time of the incident Using a storage unit that stores learning data that associates the outside state data related to the outside state of the vehicle with each road on which the incident occurred, and the learning data, from the outside state data of the automated driving vehicle that is in operation. A generation unit that generates a prediction model that predicts the occurrence of an incident on a road by machine learning, and an update that updates the recommended speed set for the road in operation in the recommended speed data according to the prediction result of the prediction model. and a part.

In the operation management system according to one embodiment of the present invention, the generation unit updates the prediction model using data with incidents and data outside the vehicle, which are included in actual operation data acquired from the currently operating automated driving vehicle, as learning data. you can

In the operation management system according to one embodiment of the present invention, the storage unit further stores, as learning data, vehicle exterior state data in the case where no incident occurs during operation of the automated driving vehicle in association with each road, and the generation unit may generate a predictive model that predicts the occurrence of an incident on the road on which the vehicle is traveling, using learning data including cases where no incident occurs.

In the operation management system according to one embodiment of the present invention, the generation unit uses, as learning data, outside-vehicle state data when no incident occurs, which is included in operation record data acquired from an automatically-operated vehicle that is currently in operation, to a prediction model can be updated.

In the operation management system according to one embodiment of the present invention, the update unit increases the recommended speed set for the road in operation when the result of the prediction by the prediction model is that no incident will occur on the road in operation. can be updated as

In the operation management system according to one embodiment of the present invention, when an incident is predicted to occur on the road in service as a result of prediction by the prediction model, the update unit reduces the recommended speed set for the road in service. can be updated to allow

In the operation management system according to one embodiment of the present invention, the acquisition unit further acquires in-vehicle state data related to the state inside the vehicle of the automated driving vehicle, which is included in the actual operation data, and the update unit performs prediction by the prediction model. In-vehicle condition data and a first coefficient preset according to the condition of passengers in the vehicle may be used to update the recommended speed accordingly.

In the operation management system according to one embodiment of the present invention, the updating unit adds a second coefficient set according to the number of occurrences of incidents over a predetermined period of time in updating the recommended speed according to the result of prediction by the prediction model. may be used.

In the operation management system according to one embodiment of the present invention, the recommended speed in the recommended speed data is set for each road section that divides the road according to a predetermined rule, and the generation unit is set for each road section on which the automatic driving vehicle travels. , the prediction model may be updated each time the road is driven.

In the operation management system according to one embodiment of the present invention, the automatically driven vehicle is an on-demand bus that travels by selecting a route between stops designated by the user from a plurality of sectioned roads, and the acquisition unit Operation record data may be obtained for each section road included in the route each time the vehicle travels.

An operation management system according to one embodiment of the present invention relates to an on-demand bus, and includes a reception unit that receives reservation data from a user regarding designation of a departure stop and departure time, and an arrival stop and arrival time, and based on the reservation data, A setting unit that sets a travel route between stops specified by the user, and a travel time required to travel the travel route set based on the reservation data is calculated from the recommended speed data set for the sectioned road included in the travel route. From the calculation unit, the first travel route and the first travel time set based on the first reservation data by the first user, and the second travel route and the second travel time based on the second reservation data by the second user, A determination unit that determines whether or not the first user and the second user can board the one on-demand bus may be further provided.

According to one embodiment of the present invention, an operation management device according to an operation management system that manages the operation of a plurality of automatically driven vehicles traveling with reference to recommended speed data related to recommended speeds set for each road, Acquisition unit that acquires the operation performance data collected by the automated driving vehicle from the automated driving vehicle, the incident data related to the incident that occurred during the operation of the automated driving vehicle, and the occurrence of the incident included in the operation performance data A storage unit that stores learning data that associates external state data related to the external state of the automated driving vehicle at the time with each road on which the incident occurred, and the external state of the automated driving vehicle that is in operation using the learning data. Based on the data, the generation unit generates a prediction model that predicts the occurrence of an incident on the road in operation, and the recommended speed set for the road in operation is updated in the recommended speed data according to the prediction result of the prediction model. and an updating unit.

According to an embodiment of the present invention, a control method for an operation management device according to an operation management system that manages operation of a plurality of automatically driven vehicles traveling with reference to recommended speed data related to recommended speeds set for each road, A step in which the operation management device acquires operation performance data collected from the automated driving vehicle in operation from the automated driving vehicle, and incident data related to an incident that occurred during operation of the automated driving vehicle, included in the operation performance data. and vehicle exterior state data relating to the exterior state of the automated driving vehicle at the time of the occurrence of the incident, a step of storing learning data associated with each road on which the incident occurred; A step of generating a prediction model for predicting the occurrence of an incident on a road on which the vehicle is in operation from the vehicle exterior state data; and updating the velocity.

According to one embodiment of the present invention, a control program for an operation management device according to an operation management system that manages operation of a plurality of automatically driven vehicles that run with reference to recommended speed data related to recommended speeds set for each road, The operation management device has a function to acquire the operation performance data collected from the automated driving vehicle in operation from the automated driving vehicle, and the incident data related to the incident that occurred during the operation of the automated driving vehicle, which is included in the operation performance data. and external state data related to the external state of the automated driving vehicle at the time of the incident, for each road on which the incident occurred. A function to generate a prediction model that predicts the occurrence of an incident on the road in operation from the vehicle's external condition data, and a recommendation set for the road in operation in the recommended speed data according to the prediction result of the prediction model. To achieve the ability to update speed.

According to the present invention, an operation management system capable of executing more accurate driving support that ensures the safety and comfort of passengers while considering the impact of the safety of passers-by and the traveling speed of an autonomous vehicle on traffic. can be provided.

1 is a schematic diagram of an operation management system configuration according to an embodiment of the present invention; FIG. FIG. 4 is a schematic diagram of a dynamic map for explaining setting of recommended speeds according to an embodiment of the present invention; FIG. It is an example of a recommended speed table in the operation management system according to one embodiment of the present invention. 1 is an example of a functional block diagram of an automatic driving vehicle according to an embodiment of the present invention; FIG. 1 is an example of a functional block diagram of a server (traffic management device) and a user terminal (communication terminal) according to an embodiment of the present invention; FIG. BRIEF DESCRIPTION OF THE DRAWINGS It is the schematic explaining the operation management system which concerns on one Embodiment of this invention. (a), (b) is a figure which shows an example of the external state data collected by the automatic driving vehicle in the operation management system which concerns on one Embodiment of this invention. It is an example of an incident data table in the operation management system which concerns on one Embodiment of this invention. It is a schematic diagram explaining the update of the recommended speed in the operation management system which concerns on one Embodiment of this invention. It is an example of a learning data table in the operation management system according to one embodiment of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS It is the schematic explaining the operation management system which concerns on one Embodiment of this invention. It is an example of a first coefficient that is set according to the state of passengers in the vehicle. It is an example of a second coefficient that is set according to the number of occurrences of incidents over a predetermined period of time in the past. 4 is a flow chart showing an operation example of a server according to one embodiment of the present invention; 1 is a schematic diagram illustrating an on-demand bus in an operation management system according to one embodiment of the present invention; FIG. 4 is a flowchart showing on-demand bus reservation processing in the operation management system according to one embodiment of the present invention. (a), (b) is a figure which shows an example of the display screen of a user terminal in the operation management system which concerns on one Embodiment of this invention.

An embodiment of the invention (also referred to as the present invention) according to the present disclosure will be described below with reference to the drawings. The drawings are only examples, and the present invention is not limited to those shown in the drawings. For example, the illustrated server (operation management device), user terminal (communication terminal), storage device, number of vehicles (automatic driving vehicle), data sets (tables), flowcharts, display screens, identifiers, etc. are examples, The present invention is not limited to these.

<System configuration>
FIG. 1 is a schematic diagram of an operation management system configuration according to one embodiment of the present invention. The operation management system 600 is a system that executes operation management of a shared bus in a service that realizes a shared bus that travels with one or more passengers by using an automatically driven vehicle. That is, the operation management system 600 is a system that remotely monitors and manages the operation of a plurality of autonomous vehicles 200 (200A to 200C) that are shared buses. Here, "automatic driving" may mean driving in which control of the engine, brakes, and steering is automatically performed without the driver's driving operation. Note that the automated driving vehicle 200 may be a vehicle of level 3 or higher in which the subject of driving is not the driver but the system side, but the driver may get in and perform an auxiliary driving operation. Also, in FIG. 1, only three buses are shown as the automatically driven vehicles, but more automatically driven vehicles managed by the operation management system 600 may exist. Further, the operation management system 600 according to one embodiment of the present invention is a system that provides a so-called on-demand bus service in which a shared bus runs on a route set according to a reservation made by a user who is a passenger. good.

The operation management system 600 includes at least automatic driving vehicles 200A to 200C, a server (operation management device) 100, communication terminals (user terminals) 300 (300A, 300B), and a storage device 400. The server 100 is an information processing device on the operation management side of the automatically driven vehicle, and is connected to the automatically driven vehicles 200A to 200C and the communication terminals 300A and 300B via the network 500. The server 100 can also execute various processes related to an on-demand bus reservation service (hereinafter simply referred to as “reservation service”) realized by the operation management system 600 . Although only one server 100 is shown in FIG. 1, the present invention is not limited to this, and a plurality of servers may exist. Moreover, the server 100 is a distributed server system that cooperates by communicating via a network, and may include an edge server, or may be a so-called cloud server. That is, the server 100 is not limited to physical servers, and may include virtual servers. Also, the function or processing of each functional unit of the server 100 may be realized by machine learning or AI (Artificial Intelligence) within a feasible range. Note that, hereinafter, the automatically driven vehicles 200A to 200C and the user terminals 300A and 300B may be simply referred to as the vehicle 200 and the user terminal 300 when there is no particular need to distinguish them.

The operation management system 600 may further include an operation management center (not shown) that remotely monitors and manages the vehicle 200 . Then, the server 100 may display the moving image inside and outside the vehicle acquired from each vehicle 200 on the display device installed in the operation control center where the monitor waits.

Network 500 includes wireless networks and wired networks. Specifically, for example, the network 500 includes a wireless LAN (WLAN), a wide area network (WAN), ISDNs (integrated service digital networks), wireless LANs, CDMA (code division multiple access), LTE (long term evolution), LTE-Advanced, fourth generation communication (4G), fifth generation communication (5G), and sixth generation communication (6G) or later mobile communication systems. The network 500 is not limited to these examples. It may be a communication network or the like. Network 500 may also be a combination of these.

A user terminal 300 is a communication terminal of a user who uses the reservation service. The user terminal 300 is connected to the server 100 via the network 500 and transmits to the server 100 information regarding the use of the shared bus received from each user. The information regarding the use of the shared bus may refer to information specifying the boarding position/disembarking position required when making a reservation for the shared bus, or user information required when using the shared bus. The reservation service will be described later.

In FIG. 1, a smartphone is shown as the user terminal 300, but as the user terminal 300, any terminal can be used as long as it can realize the functions described in each embodiment described below. good. For example, the user terminal 300 can be a mobile phone (feature phone), a computer (e.g., tablet, desktop computer, laptop), a handheld computing device (as a non-limiting example, a PDA (personal digital assistant), a wearable terminal (glasses-type device). , a watch-type device, etc.) Although only two user terminals 300 are shown in FIG. , a single user may be associated with a plurality of communication terminals (for example, a smartphone, a notebook computer, etc.).

The storage device 400 stores (stores) various types of information (data) used by the operation management system 600 . Although only one storage device 400 is shown separately from the server 100 in FIG. 1, it may be integrated with the server 100. FIG. That is, the storage device 400 may be a volatile memory or non-volatile memory of the server 100 . Further, the storage device 400 may be composed of a plurality of storage devices. Storage device 400 may be connected to server 100 via a dedicated internal network different from network 500 , or may be connected to server 100 via network 500 .

Here, as an example of information stored in the storage device 400, the recommended speed used in the operation management system 600 according to one embodiment of the present invention will be described using FIGS.

Conventionally, high definition 3D map data is used to realize autonomous driving. The high-precision three-dimensional map data is data related to road information such as the number of lanes, division lines, and width, pedestrian crossings, signal stop lines, signs, building position information, and the like. Here, in order to run an autonomous vehicle, semi-static information (traffic regulations, road construction schedules, wide-area weather forecast information, etc.) and dynamic information are added to high-precision 3D map data, which is static information. A dynamic map is used that combines information that changes in real time (surrounding vehicles, pedestrians, traffic lights, etc.) and quasi-dynamic information (traffic accidents, traffic restrictions, congestion, narrow-area weather information, etc.). Autonomous vehicles refer to dynamic map data to obtain data on the road they are traveling on, and control their speed. Here, according to the operation management system 600 according to one embodiment of the present invention, the dynamic map includes a recommended speed for the traveling speed of the vehicle 200, and the vehicle 200 travels on the road at a speed according to the recommended speed. . Note that the recommended speed will be described later.

FIG. 2 is a schematic diagram of a map for explaining setting of recommended speeds according to one embodiment of the present invention. The map MAP10 may be, for example, data that divides roads according to a predetermined rule and gives identifiers (road IDentifiers) that uniquely identify roads R1-Ri. In other words, the map MAP10 may be information indicating where the sectioned road identified by the road ID is actually located. The road divisions in the map MAP10 may be the same as the road divisions included in the existing dynamic map, or may be prepared separately by the operation management side of the vehicle 200. FIG. In addition, any predetermined rule may be used to divide the roads into sections, and the roads may be divided by a predetermined length (for example, 3 m) or by a rule such as from intersection to intersection. good too. As shown in FIG. 2, in one embodiment of the present invention, a vehicle 200 (200A, 200B) travels on a sectioned road included in a map MAP10 and travels at a speed according to the recommended speed set on each sectioned road. run with

FIG. 3 is an example of the recommended speed table TB10. Note that FIG. 3 is an example, and the information stored in the recommended speed table TB10 may be more or less than this, and the road ID may be any information as long as it can identify a road segment. In the example of FIG. 3, the sectioned road R1 is identified by the road ID "m11" and the recommended speed is set to "30 km/h". Further, the section road R3 is identified by the road ID "m13", and the recommended speed is set to "35 km/h".

Here, the "recommended speed" will be explained. In order to ensure the safety of passengers and passers-by and vehicles around the vehicle, it is preferable that the autonomous vehicle comply with the legal speed limit set for each road. In particular, in the case of a shared bus, it is required to run at a speed that does not cause overturning of passengers in the vehicle. However, traveling at an unnecessarily slow speed causes traffic jams, which leads to delays in arrival times and impairs convenience for passengers. According to the operation management system according to one embodiment of the present invention, a recommended speed is recommended when traveling on each sectioned road according to the road on which the bus travels and the surrounding conditions, travel time, past operation results, etc. is set and the set recommended speed may be updated in real time. Note that, hereinafter, the recommended speed may be the upper limit speed of the divided road.

<Self-driving vehicle>
Next, with reference to FIG. 4, the hardware configuration and functional configuration of the automatic driving vehicle 200 according to one embodiment of the present invention will be described. The automated driving vehicle 200 includes an operation control section 210 and a monitoring control section 240 . These controllers may typically be realized by an electronic control unit (ECU). Here, the driving control unit 210 controls driving/running of the vehicle 200, and the monitoring control unit 240 controls processes executed in the vehicle.

The driving control unit 210 includes a driving information acquisition unit 211 , a driving force control unit 212 , a brake control unit 213 and a steering control unit 214 . The driving information acquisition unit 211 is connected to the driving control sensor group 201 and the control outside imaging unit 202, and acquires data collected by these. The driving control sensor group 201 includes sensors and the like that collect measurement data used for controlling automatic driving. Sensors, steering angle sensors, ranging sensors (LiDAR (light detection and ranging), millimeter wave radar, ultrasonic sensors, infrared sensors, etc.), IMUs (inertial measurement units), illuminance sensors, raindrop sensors, etc. . The control exterior imaging unit 207 is, for example, a camera (a monocular camera, a stereo camera, a multi-camera, etc.), and is provided with a plurality of cameras on the front, rear, left, and right of the vehicle, and captures images of the exterior of the vehicle 200 . Objects to be imaged by the control-use exterior imaging unit 207 are information necessary for automatic driving, such as lanes and obstacles. Note that the data captured by the control-use exterior imaging unit 207 may be a moving image or a still image.

Vehicle 200 further includes a driving force output device (engine/motor) 221 , brake device 222 and steering device 223 . The driving force control unit 212, the braking device 222, and the steering device 223 control the driving force output device 221, the braking device 222, and the steering device 223, respectively. The operation control unit 210 controls the vehicle 200 along a preset operation route based on the current position of the vehicle 200, the vehicle speed, the steering angle, the moving image outside the vehicle, the presence or absence of obstacles, and the like obtained from the various sensors described above. to move or stop. Operation control unit 210 generates a control signal for moving/stopping vehicle 200 on a route, and outputs the control signal to driving force control unit 212 , brake control unit 213 , and steering control unit 214 . The driving force control unit 212, the brake control unit 213, and the steering control unit 214 control the driving force output device 221, the brake device 222, and the steering device 223, respectively, based on the control signals. In addition, the existing technique may be used about control of automatic operation.

The monitor control unit 240 executes processing related to monitoring of the vehicle 200 and includes a communication control unit 241 and an in-vehicle information acquisition unit 242 . The in-vehicle information acquisition unit 242 acquires data collected by the monitoring sensor group 203 provided in the vehicle 200 . The monitoring sensor group 203 is a sensor group that collects data used for monitoring inside and outside the vehicle. Sensors for in-vehicle monitoring may be, for example, human sensors, breath sensors, temperature sensors, heat sensors, smoke sensors, and the like. Further, the sensor for monitoring outside the vehicle may be an illuminance sensor, a raindrop sensor, a ranging sensor (LiDAR, millimeter wave radar, ultrasonic sensor, infrared sensor, etc.).

In addition, the vehicle interior information acquisition unit 242 acquires data captured by the monitoring vehicle exterior image capturing unit 204 . The vehicle exterior imaging unit 204 for monitoring is, for example, a camera (a monocular camera, a stereo camera, a multi-camera, etc.), and is provided outside the vehicle 200, for example, in front, back, left, and right, and captures moving images outside the vehicle. It should be noted that the data captured by the monitoring vehicle exterior imaging unit 204 may be a moving image or a still image.

Furthermore, the in-vehicle information acquisition unit 242 acquires data captured by the in-vehicle imaging unit 205 . The in-vehicle imaging unit 205 is a camera, for example, and is provided at an entrance door in the vehicle, a place where the inside of the vehicle can be glimpsed, or the like, and images the inside of the vehicle 200 . The data captured by the in-vehicle imaging unit 205 may be a moving image or a still image. The data captured by the in-vehicle imaging unit 205 may be used to determine attributes such as seated conditions of passengers in the vehicle, number of passengers, elderly people or children. Further, for example, in the interior of the vehicle 200, in addition to the in-vehicle imaging unit 205, a speaker, a microphone, a display, and the like (not shown) may be provided as various devices.

The monitoring control unit 240 transmits the acquired various data to the server 100 . Note that, when the vehicle 200 undergoes sudden acceleration/deceleration, sharp turning (rapid steering), or the like, the monitoring control unit 240 may acquire data indicating the occurrence and transmit the data to the server 100 . The various data transmitted to the server 100 may be processed as necessary, or may be output to the display device of the operation management center without being processed. For example, the moving image data inside or outside the vehicle that is output to the display device may be used for remote monitoring by a supervisor at the operation management center. Note that these various data may be transmitted from the vehicle 200 to the server 100 every predetermined time, for example, every second.

Communication control unit 241 controls communication (data exchange) between vehicle 200 and server 100 via network 500 by communication I/F (interface) 250 . The communication control unit 241 may control communication between the vehicles 200, or control communication between the vehicle 200 and roadside equipment (infrastructure) such as roadside units and signals installed on the road. good too.

Storage unit 270 is typically realized by various recording media such as an HDD (Hard Disc Drive), an SSD (Solid State Drive), an SD card, and a flash memory, and stores various programs necessary for vehicle 200 to operate. and has a function of storing (storing) data. The storage unit 270 may also temporarily store data acquired from the various sensors described above and moving images captured by the imaging units.

In the above description, the sensor and imaging unit for collecting data used for automatic driving and the sensor and imaging unit for collecting data used for monitoring are separated, but these are the same hardware or software. may be implemented. In the above description, the operation control unit 210 and the monitoring control unit 240 are separated, but they may be realized by one control unit.

<server>
Next, the hardware configuration and functional configuration of the server 100 according to one embodiment of the present invention will be described using FIG.

(1) Server Hardware Configuration The server 100 includes a control unit 110 , a communication unit 120 , an input/output unit 130 and a storage unit 170 .

The control unit 110 is typically a processor, and includes a central processing unit (CPU), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), a microprocessor, a processor core, and a multiprocessor. (multiprocessor), ASIC (Application-Specific Integrated Circuit), FPGA (Field Programmable Gate Array), etc., logic circuits (hardware hardware) or a dedicated circuit. The server 100 preferably has a processor with high computing power for processing the large amount of data described above.

The communication unit 120 may be implemented as hardware such as a network adapter, communication software, or a combination thereof. The communication unit 120 transmits and receives various data to and from the vehicle 200 and the user terminal 300 via the network 500 .

The input/output unit 130 includes an input device (keyboard, touch panel, microphone, etc.) for inputting various operations to the server 100, and an output device (display device, speaker, etc.) for outputting processing results processed by the server 100. good.

The storage unit 170 stores various programs and data necessary for the server 100 to operate. The storage unit 170 may include, for example, a HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, and the like. Also, the storage unit 170 may include a memory that provides a work area for the control unit 110 .

(2) Server Functional Configuration The server 100 includes, as functions realized by the control unit 110, a communication control unit 111, an input/output control unit 112, an acquisition unit 113, an extraction unit 114, a generation unit 115, an update unit 116, a reservation A setting unit 140 is included. Note that each functional unit shown in FIG. 5 is not essential, and non-essential functional units may be omitted in each embodiment described below. Also, the function or processing of each functional unit may be realized by machine learning or AI within the realizable range. Of the above functional units, the reservation setting unit 140 will be described later.

Communication control unit 111 controls communication with vehicle 200 and user terminal 300 via communication unit 120 . The input/output control unit 112 controls input/output of information to/from an external device via the input/output unit 130 by wire or wirelessly. For example, the input/output control unit 112 outputs a management screen based on various data transmitted from the vehicle 200 to a display device (not shown).

The acquisition unit 113 acquires from the automatically driven vehicle 200 operation performance data collected by the automatically driven vehicle 200 in operation. Operation performance data is all information collected by various sensors and imaging units of the vehicle 200 in operation, and includes, for example, images inside and outside the vehicle, travel position, travel time, travel speed, number of passengers, type of passengers, travel It may be the weather of the time, the brightness around the road on which you are traveling, the road signs on which you are traveling, your driving history, and the like. FIG. 6 shows an example of operation performance data acquired from the vehicle 200. As shown in FIG. According to the operation management system 600 according to one embodiment of the present invention, as operation performance data, vehicle speed 21, illuminance (surrounding brightness) 22, the number of passersby, cars, and bicycles outside the vehicle 23, road signs 24 , the occurrence of sudden acceleration/deceleration/turning 25, and the type/status 26 of passengers are acquired from each vehicle 200A to 200C in motion. These data are stored in a speed database (DB) 31, an illuminance DB 32, a passerby DB 33, a sign DB 34, in association with the road ID of the sectioned road from which each vehicle 200A to 200C acquired the data, and the acquired time. It may be stored in the sudden acceleration/deceleration DB 35 and the passenger type DB 36, respectively. The acquisition unit 113 may store in the sudden acceleration/deceleration DB 35 the occurrence of sudden deceleration/acceleration when the change in the measurement value of the acceleration sensor measured by the vehicle 200 is greater than or equal to a predetermined threshold value. Further, the obtaining unit 113 may store in the sudden acceleration/deceleration etc. DB 35 the occurrence of a sharp turn when the change in the measured value of the angular velocity sensor measured by the vehicle 200 is greater than or equal to a predetermined threshold value. It should be noted that these DBs may be further associated with which vehicle the data is acquired.

The extraction unit 114 extracts incident data related to an incident that occurred during operation of the automated driving vehicle 200 and data on the automated driving vehicle 200 when the incident occurred from the data included in the operation performance data stored in the above-described DBs 31 to 36. External state data related to the external state of the vehicle are extracted and stored in the storage device 400 as learning data (accident learning data) 51 associated with each road on which an accident has occurred. An incident that occurred during operation indicates that a predetermined condition indicating that the vehicle 200 has deviated from normal operation has been satisfied, such as sudden deceleration/acceleration, sudden steering, or excessive speed. It may be a change or the like. In addition, the vehicle exterior state data regarding the vehicle exterior state is data regarding the surrounding circumstances during operation of the vehicle 200, and in addition to the data stored in the illuminance DB 32, the passerby DB 33, the sign DB 34, etc., external service may include data provided by The data provided by the external service includes, for example, traffic information such as road congestion and accidents, road data 41 such as road construction schedules, weather data 42 such as rainfall, snow cover, fine weather, and typhoons, and past data related to autonomous driving vehicles. accident (collision, contact accident, etc.), such as accident data 43 owned by an insurance company.

7A and 7B show image data outside the vehicle transmitted from the vehicle 200 as an example of data outside the vehicle. FIG. 7(a) shows an image 70a in a moving image captured in front of a vehicle 200, in which the vehicle 200 is traveling on a straight road, and pedestrians 71 to 73 riding bicycles and an oncoming vehicle 74 are present. can identify what to do. FIG. 7(b) is an image 70b similarly captured in front of the vehicle 200, in which the vehicle 200 is stopped before a T-junction 77 with a pedestrian crossing 76, and a bicycle is ahead of the vehicle 200. It can be identified that a pedestrian 75 riding a bicycle is crossing, and that a pedestrian 78 is crossing a pedestrian crossing in the left front. The acquisition unit 113 may perform image recognition on imaged data as shown in FIG. 7 and identify an object or person included in the moving image. An existing technique may be used for image recognition. An image 70b in FIG. 7B is an image when the vehicle 200 suddenly brakes. Therefore, the information included in the image 70b is stored in the learning data 51 with an incident among the learning data 50 in FIG.

FIG. 8 shows an example of learning data 51 with incident. As shown in FIG. 8, as the incident data identified by the incident data ID "ad_1", the incident "sudden braking" occurred in the time zone "9 o'clock" on the section road R1 with the road ID "m11". is stored. In addition, as data on the vehicle exterior when the accident occurred, the road shape of the divided road R1 was a "crossroad" and there was a "beware of jumping out" sign. "10 people", "5 cars", illuminance "1000 lux", and weather "light rain" are stored. The incident data identified by the incident data ID “ad_2” is the incident “sudden braking” that occurred in the time zone “16:00” on the divided road R1 with the road ID “m11” as with the incident data ID “ad_1”. , the number of passers-by in the vicinity at the time of the incident was 20, the number of cars was 3, the illuminance was 500 lux, and the weather was clear. In this way, data related to the situation around the vehicle at the time of occurrence of an accident is stored in the incident learning data 51 .

<First embodiment>
According to the first embodiment of the present invention, a prediction model for predicting the occurrence of an incident on a road in operation is generated from data on the external state of an autonomous vehicle in operation. The generating unit 115 generates a prediction model for predicting the occurrence of an incident on the road in operation from the outside state data of the automatically driven vehicle in operation using the incident learning data 51 . Specifically, the generation unit 115 uses the learning data 51 with an incident, which is data related to an incident that has occurred in the past, to set the occurrence of an incident as an data on the surrounding conditions of the vehicle) is used as an explanatory variable, and when the external condition data of a vehicle currently traveling on a sectioned road is input, it determines whether an incident will occur on the sectioned road. A prediction model for outputting data is generated (step T11).

Describe the generation of predictive models. The generation unit 115 may generate a prediction model using multiple regression analysis as machine learning. When using the accidental learning data 51 for prediction, the prediction formula can be expressed as below, for example.

The generation unit 115 sets the occurrence of an incident to "1", converts the categorical variables in the learning data 51 with an incident into numerical values by, for example, one-hot vectorization, and applies them to the above (Equation 1), Find the coefficients a, b..., j. Note that the generation unit 115 may generate a prediction model for each road segment. That is, the generation unit 115 may generate a prediction model for each road ID using data with the same road ID in the learning data 51 with incident as variables.

The update unit 116 inputs the external state data of the automatically driven vehicle in operation to the prediction model generated by the generation unit 115, and sets the road in operation in the recommended speed data according to the prediction result of the prediction model. recommended speed is updated (step T12 in FIG. 6). Each vehicle 200 continues to run referring to the updated recommended speed, and transmits each data shown in FIG. 6 to the server 100 while running.

Here, updating of the recommended speed will be described with reference to FIGS. 3 and 9. FIG. Referring to the recommended speed data (recommended speed table TB10) in FIG. 3, the recommended speed initially set for the sectioned road R1 identified by the road ID "m10" is "30 km/h". Acquisition unit 113 acquires vehicle exterior condition data around section road R1 from vehicle 200 currently traveling on section road R1. Enter state data. Then, as a result of the prediction by the prediction model, if the occurrence of an incident is predicted from the current situation around the road R1, the update unit 116 updates the recommended speed of the road R1 in the recommended speed table TB10 as shown in FIG. It may be reduced from "30km/h" to "25km/h" as follows. The time "T1" for decreasing from "30 km/h" to "25 km/h" is the time from when the vehicle 200 starts traveling on the divided road R1 to the server 100, where the server 100 sends the current vehicle exterior state data to the server 100. It may be determined according to the time required to complete a series of arithmetic processing.

As described above, according to the first embodiment of the present invention, the data (outside vehicle state data) related to the situation around the vehicle when an accident occurred in the past is learned, and the likelihood of an accident occurring in accordance with the situation around the vehicle is predicted. A predictive model is generated that Then, it becomes possible to predict in real time the occurrence of an incident on the road on which the vehicle 200 travels, and to update the recommended speed data that the vehicle 200 should refer to in real time. As a result, it is possible to realize more accurate driving assistance that ensures the safety and comfort of passengers while taking into consideration the impact of the safety of passers-by and the traveling speed of autonomous vehicles on traffic.

In the above description, the multiple regression analysis was described with the occurrence of an incident as "1", but the type of incident that occurred may be used as the objective variable. That is, a prediction model may be generated that predicts the likelihood of occurrence of each event, with sudden deceleration, sudden acceleration, and sudden turning as objective variables.

Also, in the above description, an example of using multiple regression analysis to generate a prediction model has been described, but the present invention is not limited to this. For example, Support Vector Regression, Random Forest regression, etc. may be used to generate the prediction model. Alternatively, a prediction model may be generated using an algorithm such as the k-neighborhood method, decision tree (classification tree), random forest, support vector machine, logistic regression, or the like.

Note that the learning data 50 used for generating the prediction model is preferably the latest data to some extent. Therefore, data acquired before a predetermined period from the present, such as from one month ago to the present, or from two weeks ago to the present, may be used for learning. In addition, data that does not contribute to prediction in the learning data with incident 51 may not be used in the prediction formula.

Note that in FIG. 9, at time "T2", the recommended speed for the sectioned road R1 increases from "25 km/h" to "30 km/h". In this way, if the result of the prediction by the prediction model is that no incident will occur on the road on which the vehicle is being operated, the updating unit 116 may update the speed to increase the recommended speed set for the road on which the vehicle is being operated. As a result, it is possible to prevent the vehicle 200 from running at an unnecessary low speed and causing a traffic jam.

Also, as described above, in FIG. 9, at time "T1", the recommended speed for section road R1 is reduced from "30 km/h" to "25 km/h", and at time "T3", section road R1 recommended speed has been reduced from 30km/h to 20km/h. In this way, when the prediction model predicts that an incident will occur on the road on which the vehicle is being operated, the updating unit 116 may update the recommended speed set for the road on which the vehicle is being operated to decrease. As a result, the possibility of an accident occurring can be reduced, and the vehicle 200 can be driven more safely. Determination of the recommended speed decrease and increase will be described later.

Note that the generation unit 115 may update the prediction model using data with an incident and external state data included in operation record data acquired from an automatically driven vehicle currently in operation as learning data. This will be explained using FIG. As shown in FIG. 6, vehicle speed 21, illuminance (surrounding brightness) 22, number of passers-by, cars, and bicycles outside the vehicle 23, road signs 24, occurrence of sudden acceleration/deceleration/turning 25, and types of passengers. • Operation performance data, including data relating to state 26, may be transmitted to server 100 in real time from each vehicle 200A-200C that travels. Therefore, the acquisition unit 113 of the server 100 can acquire data related to the incident that occurred in the vehicle 200 in real time, and the generation unit 115 calculates the coefficients a, b . . . It may be recalculated and updated (step T11). As a result, it is possible to generate a prediction model that is more in line with the current situation, enabling highly accurate prediction.

Note that the generation unit 115 may acquire data relating to one sectioned road from a plurality of vehicles 200A to 200C and generate a prediction model. Since the vehicle 200 is a shared bus, the route it travels is determined to some extent, and each vehicle 200 travels a certain sectioned road a plurality of times. Therefore, it is possible to acquire a large amount of actual operation data for one segmented road, and it is possible to improve the accuracy of the prediction model.

<Second embodiment>
1st Embodiment demonstrated the aspect which a prediction model is produced|generated by using data with an incident when an incident occurs as an objective variable. However, no-incident data when no incident occurs may also be used to generate the predictive model. That is, the storage device 400 further stores, as learning data, vehicle exterior state data when no incident occurs during operation of the automated driving vehicle 200 in association with each road, and the generation unit 115 stores the data such that the incident does not occur. The learned data, including cases, may be used to generate a predictive model that predicts the occurrence of incidents on roads in service.

That is, the extracting unit 114 extracts a predetermined number of vehicle exterior state data related to the exterior state of the automated driving vehicle 200 when no incident occurs from the data included in the operation performance data stored in the DBs 31 to 36, and extracts the vehicle exterior state data for each road. may be stored in the storage device 400 as learning data (no incident learning data) 52 associated with the . The generation unit 115 may assign “1” to the occurrence of an incident (with an incident) and “0” to indicate no incident, apply them to the above (Formula 1), and obtain the coefficients a, b . . . , j. For generation of the prediction model, not only (Equation 1) but also various methods including the above-described algorithm may be used.

According to the second embodiment of the present invention, it is possible to improve the prediction accuracy for the case where an incident occurs and the case where no incident occurs by including the vehicle exterior state data when no incident occurs in the learning data. .

Note that the generation unit 115 may update the prediction model by using, as learning data, vehicle exterior state data when no incident occurs, which is included in operation record data acquired from an automatically driven vehicle that is currently in operation. As described above, as shown in FIG. Operation performance data including data on passenger types/states 26 is transmitted to server 100 in real time from each vehicle 200A to 200C that is traveling. Therefore, generation unit 115 recalculates the coefficients a, b, . can be updated. As a result, it is possible to generate a prediction model that is more in line with the current situation, enabling highly accurate prediction.

<Third Embodiment>
Next, a form in which a predictive model relating to the occurrence of an incident in the vehicle is generated will be described as a third embodiment. FIG. 10 is an example of learning data used to generate a prediction model in the third embodiment. An incident in a vehicle (in-vehicle incident) refers to a state against the safety of a passenger, and may be, for example, a fall of a passenger or abrupt standing up. It should be noted that hereinafter, accidents such as sudden braking described in the first and second embodiments will be referred to as "service accidents" to distinguish them from accidents in the vehicle. The acquisition unit 113 may acquire data captured by the in-vehicle imaging unit 205 of the vehicle 200 and determine whether an incident has occurred in the vehicle. Alternatively, the occurrence of an incident in the vehicle may be determined in vehicle 200, and imaged data at that time may be transmitted from vehicle 200 to server 100 as data at the time of occurrence of the incident in the vehicle. That is, in the third embodiment, in addition to the DBs 31 to 36 shown in FIG. 6, DBs (not shown) relating to falling/standing up of passengers may be stored. The extraction unit 114 may extract data used for generating a prediction model from each DB and store it as learning data 50 .

In the example of the learning data 53 of FIG. 10, as the incident data identified by the incident data ID "ad_11", on the sectioned road R1 with the road ID "m11", during the time period "9 o'clock", the operation incident "sudden braking" occurs, and at the same time, it is memorized that an in-vehicle accident "overturn" has occurred. In addition, as data for the state outside the vehicle when an incident (incident inside the vehicle or during operation) occurs, the road shape of the divided road R1 is a "crossroad" and there is a sign "Beware of jumping out". It is remembered that the number of passers-by present was "10", the number of cars was "5", the illuminance was "1000 lux", and the weather was "light rain". Furthermore, as the vehicle interior state data, it is stored that the number of people seated in the vehicle was "3" and the number of people standing was "2". The incident data identified by the incident data ID "ad_21" is the same as the incident data ID "ad_11", on the section road R1 with the road ID "m11". Regarding "Brake", there was no incident in the car, the number of passers-by who were in the vicinity at the time of the incident was "20", the number of cars was "3", the illuminance was "500 lux", and the weather was "clear". , and that the number of people seated and the number of people standing in the car were both "0" as in-vehicle state data. Here, the "in-vehicle condition data" is data relating to the conditions inside the vehicle 200 during operation, and includes the number of passengers, the condition of the passengers (seated/standing), and the type of the passengers (elderly, child, etc.). may be data relating to In the example of the learning data 53, in this way, the learning data 53 stores in-vehicle state data regarding the state inside the vehicle together with data regarding the state outside the vehicle regarding the state around the vehicle at the time of occurrence of the accident. Incidentally, as the in-vehicle state data, the types of passengers may be further stored.

The generator 115 may use the learning data 53 to generate a prediction model according to, for example, the following prediction formula.

The generation unit 115 sets the occurrence of an in-vehicle incident and an operation incident to “1”, converts the categorical variables of the learning data 53 into numerical values by one-hot vectorization or the like, and applies them to the above (Equation 2) to obtain the coefficient a ₁ , b ₁ …, j ₁ . Note that the generation unit 115 may generate a prediction model for each road segment. That is, the generation unit 115 may generate a prediction model for each road ID using data with the same road ID in the learning data 53 as variables.

The update unit 116 inputs the external state data and the internal state data of the automatically driven vehicle in operation to the prediction model generated by the generation unit 115, and according to the prediction result of the prediction model, the recommended speed data is updated during operation. may update the recommended speeds set for the roads. That is, if the predictive model predicts that an incident will occur inside the vehicle, the recommended speed may be decreased, and if it predicts that an incident will not occur inside the vehicle, the recommended speed may be increased or maintained.

Thus, according to the third embodiment of the present invention, it is possible to predict the occurrence of an accident inside the vehicle. Therefore, it is possible to provide an operation management system that further enhances the safety of passengers. Note that the recommended speed may be updated according to the prediction result of the occurrence of an incident inside the vehicle, the prediction result of the occurrence of an incident outside the vehicle, or both.

<Fourth Embodiment>
In the first to third embodiments, a prediction model for predicting the occurrence of an incident is generated and the recommended speed is updated according to the prediction result. In contrast, in the fourth embodiment of the present invention, a prediction model for predicting the travel speed at which an incident will occur may be generated, and the recommended speed may be updated according to the predicted travel speed.

In the case of the fourth embodiment, the generator 115 may use the learning data 51 with incident to generate a prediction model according to the following prediction formula, for example.

In Equation 3, the vehicle speed is learned when an operational incident occurs. Therefore, according to the fourth embodiment, it is possible to predict the vehicle speed when it is assumed that an incident occurs with the current vehicle exterior state data. Therefore, the update unit 116 may update the recommended speed so as not to exceed the predicted running speed.

<Fifth Embodiment>
4th Embodiment demonstrated the aspect which the prediction model which predicts the driving|running|working speed at which operation|movement incidents, such as sudden braking, generate|occur|produces is produced|generated. In the fifth embodiment of the present invention, a prediction model for predicting the travel speed at which an in-vehicle incident such as a passenger falling over occurs may be generated, and the recommended speed may be updated according to the predicted travel speed.

In the case of the fifth embodiment, the generator 115 may generate a prediction model based on, for example, the following prediction formula, using only data in which an in-vehicle incident has occurred among the learning data 53 .

Equation 4 learns the vehicle speed when an in-vehicle incident occurs. Therefore, according to the fifth embodiment, the vehicle speed is calculated assuming that an in-vehicle incident occurs with the current exterior and interior state data. can be predicted. Therefore, the update unit 116 may update the recommended speed so as not to exceed the predicted running speed.

<Recommended Speed Update>
As described above, according to one embodiment of the present invention, the recommended speed is updated by increasing or decreasing it according to the prediction result. Here, the determination of the update change range will be described. Note that the update of the recommended speed described below can be applied to all of the first to fifth embodiments described above.

(1) Increase/Decrease of Predetermined Value According to an embodiment of the present invention, the update unit 116 may update the currently set recommended speed by changing it by a preset change width (predetermined value). This change width may be, for example, 5 km/h, 8 km/h, etc., but is not limited to these.

(2) Use of Coefficient According to one embodiment of the present invention, a preset coefficient may be used to update the recommended speed according to the prediction result of the prediction model. This will be described with reference to FIGS. 11 and 12. FIG. 11 and 6 are different in that coefficient optimization processing (step T13) is performed and the coefficient in step T13 is used to update the recommended speed (step T12). Hereinafter, points different from FIG. 6 will be described.

(2-1) Use of First Coefficient Updating unit 116 updates the recommended speed according to the result of prediction by the prediction model, using the first coefficient previously set according to the vehicle interior state data and the state of passengers in the vehicle. coefficients may be used.

FIG. 12 is an example of a first coefficient table TB30 relating to first coefficients set in advance and stored in the storage device 400. As shown in FIG. The first coefficient may be set according to the type of passengers in the vehicle (elderly or child), the state of the passengers (sitting or standing), and the number of passengers sitting or standing. For example, the first coefficient is set to “0.2” when “three or more elderly people are standing” and “0.15” when “three or fewer elderly people are standing.” ing. In addition, the first coefficient is set to "0.15" when "three or more children are standing". In this way, the first coefficient may be a value that indicates the likelihood (degree of risk) of occurrence of an in-vehicle accident, and the value of the first coefficient is not limited to those illustrated.

The update unit 116 updates the current in-vehicle conditions of the elderly and children included in the in-vehicle state data when it is predicted that at least one of the operational incident and the in-vehicle incident will occur as a result of the prediction of each of the prediction models described above. Depending on the state, the applicable first coefficient is selected from the first coefficient table TB30. Then, the updating unit 116 may update the recommended speed by increasing or decreasing the value obtained by multiplying the previously set recommended speed by the first coefficient as a variation width.

As described above, according to one embodiment of the present invention, the range of change in the recommended speed is determined according to the conditions inside the vehicle in addition to the results of prediction by the prediction model. Therefore, it is possible to set a recommended speed with high accuracy that is more in line with the current situation.

(2-2) Use of Second Coefficient Updating unit 116 uses the second coefficient set according to the number of occurrences of incidents over the past predetermined period to update the recommended speed according to the result of prediction by the prediction model. good.

FIG. 13 is an example of a second coefficient table TB40 relating to second coefficients set in advance and stored in the storage device 400. As shown in FIG. The second coefficient table TB40 may be prepared for each sectioned road, and may be set according to the number of operational incidents or in-vehicle incidents that occurred on the sectioned road during a past predetermined period (for example, one month or two weeks). For example, if the number of incidents is 0, the second coefficient is set to -1, and if the number of incidents is 6 to 10, the second coefficient is set to +1. . In this way, the second coefficient may be a value that corrects the recommended speed according to the likelihood of occurrence of an accident, and the value of the second coefficient is not limited to those illustrated.

The update unit 116 acquires the number of incidents that have occurred on the sectioned road during driving when it is predicted that at least one of the operation incident and the in-vehicle incident will occur as a result of the prediction of each of the prediction models described above, Select the applicable second coefficient from the two-coefficient table TB430. Then, the update unit 116 may update the recommended speed to a value obtained by adding a second coefficient to a value obtained by increasing or decreasing the previously set recommended speed by a predetermined value.

Thus, according to one embodiment of the present invention, the range of change in the recommended speed is corrected according to the likelihood of occurrence of accidents on the road in addition to the prediction result of the prediction model. Therefore, it is possible to set a recommended speed with high accuracy that is more in line with the current situation.

<Server control flow chart>
Here, a method of controlling the server 100 will be described using the flowchart of FIG. 14 . First, the acquisition unit 113 of the server 100 acquires operation record data regarding an automatically driven vehicle in operation from the automatically driven vehicle 200 (step S11). Operation performance data may be stored in each DB for each data type. The server 100 stores in the storage device 400 the incident data acquired in the past as operation performance data and the external state data regarding the external state of the autonomously driven vehicle at the time of occurrence of the incident. The acquired learning data is stored (step S12). Note that the learning data may be data within a predetermined period (for example, one month, two weeks, etc.) from the present. Next, the generation unit 115 uses the learning data to generate a prediction model for predicting the occurrence of an incident on the road in operation from the vehicle exterior state data of the automatic driving vehicle in operation (step S13). The prediction model is as described above. Then, the update unit 116 updates the recommended speed set for the road in operation in the recommended speed data according to the prediction result of the prediction model (step S14). Note that the update of the recommended speed is as described above.

Note that the generating unit 115 may update the prediction model for each road section on which the autonomous vehicle 200 travels, each time the vehicle 200 travels on the road section. That is, the vehicle 200 transmits to the server 100 the operation performance data collected by each sensor and imaging unit of the vehicle each time the road segment changes, and the generation unit 115 uses the necessary data to perform the above-described prediction. You can update the model. As a result, the prediction model can be made more in line with the current situation, and the accuracy of the prediction result can be improved.

<Sixth embodiment>
Next, a sixth embodiment of the present invention will be described. As described above, in one embodiment of the present invention, the self-driving vehicle 200 is a shared bus, and further selects a route between stops specified by a user who is a passenger from a plurality of sectioned roads. It may be an on-demand bus.

Here, the on-demand bus will be explained using FIG. FIG. 15 is an example of a dynamic map MAP20 containing data relating to stops 11-17. In the on-demand bus, the user can select from stops 11-17 the stop from which the user departs (departure point) and the stop to which the user arrives (destination). In the example of FIG. 15, it is assumed that the stop 11 is selected as the departure point and the stop 15 is selected as the destination. The server 100 may select a route connecting the bus stop 11 to the bus stop 15 from the divided roads R1 to Ri, and set the travel route. At this time, as long as the route departs from stop 11 and arrives at stop 15, any combination of road divisions may be used. Therefore, the acquisition unit 113 can acquire operation performance data for each section road included in the route each time the autonomous vehicle 200 travels. As a result, the data used to generate the prediction model can be acquired from each automated driving vehicle, and the accuracy of the prediction model can be improved.

The reservation setting unit 140 of the server 100 executes processing related to reservation of the on-demand bus, and includes a reservation reception unit 141 , a route setting unit 142 , a calculation unit 143 and a determination unit 144 .

The reservation reception unit 141 receives reservation data transmitted from the user terminal 300 . Reservation data may include data relating to designation of a departing stop and departure time as well as an arriving stop and arrival time. The route setting unit 142 sets a travel route between stops specified by the user based on the received reservation data. The calculation unit 143 calculates the travel time required to travel the travel route set based on the reservation data from the recommended speed data set for the sectioned road included in the travel route. Since the recommended speed is the upper limit speed, it is conceivable that the running speed of vehicle 200 will be slower than the recommended speed. Therefore, the calculation unit 143 may correct the value obtained by adding the recommended speed set for the divided road.

Here, the on-demand bus may be a system that allows multiple users to ride together. Therefore, depending on the reservation conditions for the first user and the second user, the first user and the second user can be efficiently transported by one vehicle 200 . Therefore, the determination unit 144 determines the first travel route and first travel time set based on the first reservation data by the first user, and the second travel route and second travel time based on the second reservation data by the second user. , it is determined whether or not the first user and the second user can board the one on-demand bus.

The above processing will be described with reference to the flowchart of FIG. 16 . The reservation accepting unit 141 acquires reservation conditions (reservation date and time, departure/destination) from the user terminal 300 (step P11). The determination unit 144 determines whether or not there is a reservation (previous contract) close to the reservation time (step P12). Note that the set reservation is stored in the storage device 400, for example, as a reservation table, and the determination unit 144 can refer to the reservation table to perform the determination in step P12. If it is determined that there is a previous contract (YES in step P12), the route setting unit 142 sets a route for going to the destination by sharing the previously contracted vehicle (step P13). The calculator 143 calculates the time required to travel the route set in step P13 using the recommended speed data (step P14).

If it is determined in step P12 that there is no prior appointment close to the reservation time (NO in step P12), the route setting unit 142 starts from the stop point of the vehicle carrying the user and goes to the destination via the departure point. A route is set (step P15). The calculator 143 calculates the time required to travel the route set in step P15 using the recommended speed data (step P16).

The determination unit 144 determines whether or not the reservation conditions are satisfied when the route set in steps P14 and P16 is traveled (step P17). That is, the determination unit 144 may determine whether or not it is possible to arrive at the destination by the arrival time specified by the user. If it is determined that the reservation conditions are satisfied (YES in step P17), the reservation setting section 140 sets the user's reservation and stores it in the reservation table (step P18). Then, the travel route, scheduled departure time, and scheduled arrival time are transmitted from the server 100 to the user terminal 300 (step P19). If it is determined in step P17 that the reservation conditions are not satisfied (NO in step P17), the reservation setting unit 140 may transmit to the user terminal 300 that the reservation is not possible (step P20).

Thus, according to the sixth embodiment of the present invention, it is determined whether or not it is possible to share a ride on the on-demand bus using the recommended speed data. As described in the first to fifth embodiments, the recommended speed is updated in real time. Therefore, according to the sixth embodiment of the present invention, it is possible to use the recommended speed data that is more accurate in line with the current situation to determine the route. It is possible to predict the time required for running

<User terminal>
Returning to FIG. 2, the hardware configuration and functional configuration of the user terminal 300 according to one embodiment of the present invention will be described.

(1) Hardware Configuration of User Terminal The user terminal 300 may include a control section 310 , a communication section 320 , a display section 330 , an input/output section 340 and a storage section 370 .

The control unit 310 is typically a processor, and includes a central processing unit (CPU), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), microprocessor, processor core, multiprocessor. (multiprocessor), ASIC (Application-Specific Integrated Circuit), FPGA (Field Programmable Gate Array), etc., logic circuits (hardware hardware) or a dedicated circuit. The control unit 310 may perform the functions and methods shown in each embodiment by reading out the program stored in the storage unit 370 and executing the code or instructions included in the read program.

The storage unit 370 stores various programs and data necessary for the user terminal 300 to operate. The storage unit 370 may include, for example, flash memory or the like. Storage unit 370 may also include a memory (RAM (Random Access Memory), ROM (Read Only Memory), etc.) that provides a work area for control unit 310 .

The communication unit 320 may be implemented as hardware such as a network adapter, communication software, or a combination thereof. The communication unit 320 may transmit and receive various data to and from the server 100 via the network 500 .

The display unit 330 is a monitor that displays data according to the display data written in the frame buffer, and may be a touch panel, a touch display, or the like, for example.

The input/output unit 340 may include an input device for inputting various operations to the user terminal 300 and an output device for outputting processing results processed by the user terminal 300 . In the input/output unit 340, the input device and the output device may be integrated, or the input device and the output device may be separated. The input device may be realized by any one of all kinds of devices capable of receiving an input operation from the user and transmitting information related to the input to the control unit 310, or a combination thereof. Input devices may include, for example, touch panels, touch displays, cameras, and microphones. The output device may output the processing result processed by the control unit 310 . The output device may include, for example, a display, touch panel, speaker, and the like.

(2) Functional Configuration of User Terminal The user terminal 300 may include a communication control section 311 , a display control section 312 , and an input/output control section 313 as functions realized by the control section 310 . Note that each functional unit shown in FIG. 2 is not essential, and non-essential functional units may be omitted in each embodiment described below. Also, the function or processing of each functional unit may be realized by machine learning or AI within the realizable range.

The communication control unit 311 may control communication with the server 100 via the network 500 by the communication unit 320 to transmit and receive various information.

The display control section 312 may control display of data on the display section 330 . For example, the display control unit 312 may cause the display unit 330 to display an on-demand bus reservation screen, which will be described later.

The input/output control unit 313 may control transmission of various information with an external device via the input/output unit 340 . For example, the input/output control unit 313 transmits various types of information to each functional unit in response to a user input operation received by an input device, and outputs various information to output devices (not shown) such as a touch panel, a monitor, and a speaker. Information from the functional unit may be transmitted. Also, the input/output control unit 313 may include a reservation information input unit 314 . The reservation information input unit 314 may receive an input of reservation for use of the on-demand bus described above from the user.

<Reservation screen on user terminal>
17(a) and (b) show an example of an on-demand bus reservation screen. FIG. 17(a) is a reservation input screen 80a, on which a date 81, a departure point 82, an arrival point 83, and the number of passengers 84 may be input. Note that these data may be input by selecting preset options. Furthermore, according to the input of the number of passengers 84, the passenger type 85 may be input to the reservation input screen 80a. When the search button 86 is selected by the user, the user terminal 300 may transmit reservation information to the server 100 .

FIG. 17(b) is a reservation detail screen 80b for a reservation set based on the reservation information, which is a display screen of the information transmitted from the server 100 in step P19 described above. As shown in the figure, when the reservation setting unit 140 determines that it is possible to share a ride with another user, information 87 indicating that there is a shared ride may be displayed. Before transitioning to the reservation details screen 80b, a list of reservations that can be set based on the reservation information may be displayed on the user terminal 300, and the user may be able to select whether or not to make a reservation.

Although the present invention has been described with reference to the drawings and examples, it should be noted that various variations and modifications will be readily apparent to those skilled in the art based on this disclosure. Therefore, it should be noted that these variations and modifications are included in the scope of the present invention. For example, the functions included in each component, each step, etc. can be rearranged so as not to be logically inconsistent, and multiple components, steps, etc. can be combined into one or divided. is. Further, the configurations shown in the above embodiments may be combined as appropriate. For example, each component described as being included in the server 100 may be distributed and implemented by a plurality of servers.

Further, the processing described above as being performed by the server 100 may be performed by the vehicle 200 and the user terminal 300 .

Furthermore, the first to sixth embodiments described above may be implemented independently, or may be implemented by combining multiple embodiments.

Further, all the data obtained from the vehicle 200 described above as explanatory variables in generating the prediction model need not necessarily be used as explanatory variables, and data that does not contribute to prediction need not be used.

Each functional unit of the server 100 and the vehicle 200 may be realized by a logic circuit (hardware) or a dedicated circuit formed in an integrated circuit (IC (Integrated Circuit) chip, LSI (Large Scale Integration)) or the like. It may be realized by software using a CPU (Central Processing Unit). Further, each functional unit may be implemented by one or more integrated circuits, and the functions of multiple functional units may be implemented by one integrated circuit. Furthermore, the server 100 described above may be realized by a plurality of server computers, and depending on the function, may be realized by calling an external platform or the like using an API (Application Programming Interface) or the like.

When each functional unit of the server 100 is implemented by software, the server 100 includes a CPU that executes instructions of a program, which is software that implements each function, and a computer (or CPU) that stores the above program and various data so that they can be read. A ROM (Read Only Memory) or a storage device (these are referred to as a "recording medium"), a RAM (Random Access Memory) for developing the above program, and the like. The object of the present invention is achieved by a computer (or CPU) reading and executing the program from the recording medium. That is, the server 100 according to the present invention functions as each component described above by the CPU executing the program loaded on the RAM. As the recording medium, a "non-transitory tangible medium" such as a semiconductor memory, a programmable logic circuit, or the like can be used. Also, the program may be supplied to the computer via any transmission medium (communication network, broadcast wave, etc.) capable of transmitting the program. The invention can also be embodied in the form of a data signal embedded in a carrier wave in which the program is embodied by electronic transmission.

Note that the above program is, for example, a script language such as ActionScript, JavaScript (registered trademark), Python, Ruby, etc., an object-oriented programming language such as C language, C ++, C #, Objective-C, Swift, Java (registered trademark), It may be implemented using a markup language such as HTML5. Furthermore, the description of "section, module, unit" in the scope of claims may be read as "means" or "circuit". For example, the communication unit can be read as communication means or a communication circuit.

Also, the program of the present disclosure may be provided to server 100 via any transmission medium (communication network, broadcast wave, etc.) capable of transmitting the program.

According to each aspect of the present disclosure described above, it is possible to operate a safe and comfortable automatic driving vehicle even when elderly people and children ride, so that the Sustainable Development Goals (SDGs) Goal 11 "Life We can contribute to the achievement of sustainable community development.

100 server (operation management device)
110 control unit 111 communication control unit 112 display processing unit 113 acquisition unit 114 extraction unit 115 generation unit 116 update unit 120 communication unit 130 input/output unit 140 reservation setting unit 141 reservation reception unit 142 route setting unit 143 calculation unit 144 determination unit 170 storage Part 200 Autonomous Driving Vehicle 201 Driving Control Sensor Group 202 Control Outside Vehicle Imager 203 Monitoring Sensor Group 204 Monitoring Outside Vehicle Imager 205 Vehicle Inside Imager 210 Operation Control Part 211 Driving Information Acquisition Part 212 Driving Force Control Part 213 Brake Control Part 214 steering control unit 221 driving force output device 222 braking device 223 steering device 240 monitoring control unit 241 communication control unit 242 in-vehicle information acquisition unit 250 communication I/F
260 display unit 270 storage unit 300 user terminal (communication terminal)
310 control unit 311 communication control unit 312 display control unit 313 input/output control unit 314 reservation information input unit 320 communication unit 330 display unit 340 input/output unit 370 storage unit 400 storage device 500 network 600 operation management system

Claims (14)

An operation management system that manages the operation of a plurality of automated driving vehicles that run by referring to recommended speed data related to recommended speeds set for each road,
An acquisition unit that acquires operation performance data collected by the automated driving vehicle in operation from the automated driving vehicle;
The incident data related to the incident that occurred during the operation of the automated driving vehicle and the vehicle exterior state data related to the exterior state of the automated driving vehicle when the incident occurred, which are included in the operation performance data. a storage unit that stores learning data associated with each road on which an abnormality has occurred;
A generation unit that uses the learning data to generate a prediction model that predicts the occurrence of the above-mentioned change on the road in operation from the external state data of the automatically-operated vehicle in operation;
an updating unit that updates the recommended speed set for the road on which the vehicle is running, in the recommended speed data, according to the result of prediction by the prediction model;
operation management system. The generation unit updates the prediction model using the abnormal data and the vehicle exterior state data included in the operation performance data acquired from the currently operating automated driving vehicle as the learning data.
The operation management system according to claim 1. The storage unit further stores, as the learning data, the data outside the vehicle when the change does not occur during operation of the autonomous vehicle in association with each road,
The generation unit generates the prediction model for predicting the occurrence of the accident on the road on which the vehicle is in operation, using the learning data including the case where the accident does not occur.
The operation management system according to claim 1 or 2. The generating unit updates the prediction model using the outside-vehicle state data when the change does not occur, which is included in the operation performance data acquired from the currently operating automated driving vehicle, as the learning data.
The operation management system according to claim 3. The update unit updates the recommended speed set for the road under operation so as to increase it when the occurrence of the change is not predicted on the road under operation as a result of the prediction by the prediction model.
The traffic control system according to any one of claims 1 to 4. The updating unit updates the recommended speed set for the road under operation to decrease when the occurrence of the change is predicted on the road under operation as a result of the prediction by the prediction model.
The traffic management system according to any one of claims 1 to 5. The acquisition unit further acquires in-vehicle state data related to the state in the vehicle of the automated driving vehicle, which is included in the operation performance data,
The update unit uses the vehicle interior state data and a first coefficient preset according to the state of passengers in the vehicle to update the recommended speed according to the prediction result of the prediction model.
The traffic management system according to any one of claims 1 to 6. The update unit uses a second coefficient set according to the number of occurrences of incidents over a past predetermined period to update the recommended speed according to the result of the prediction by the prediction model.
The traffic management system according to any one of claims 1 to 7. The recommended speed in the recommended speed data is set for each sectioned road obtained by sectioning the road according to a predetermined rule,
The generation unit updates the prediction model for each of the road divisions on which the automated driving vehicle travels, each time the road is driven.
The traffic management system according to any one of claims 1 to 8. The self-driving vehicle is an on-demand bus that runs by selecting a route between stops specified by the user from a plurality of the divided roads,
The acquisition unit acquires the operation performance data for each of the divided roads included in the route each time the automated driving vehicle travels.
The operation management system according to claim 9. a reception unit that receives from a user reservation data relating to the on-demand bus, specifying a departure stop and departure time, and an arrival stop and arrival time;
a setting unit that sets a travel route between stops specified by the user based on the reservation data;
a calculation unit that calculates the travel time required to travel the travel route set based on the reservation data from the recommended speed data set for the sectioned road included in the travel route;
The first user's and a determination unit that determines whether or not the second user can board one on-demand bus;
further comprising
The operation management system according to claim 10. An operation management device according to an operation management system that manages the operation of a plurality of automated driving vehicles that run with reference to recommended speed data related to recommended speeds set for each road,
An acquisition unit that acquires operation performance data collected by the automated driving vehicle in operation from the automated driving vehicle;
The incident data related to the incident that occurred during the operation of the automated driving vehicle and the vehicle exterior state data related to the exterior state of the automated driving vehicle when the incident occurred, which are included in the operation performance data. a storage unit that stores learning data associated with each road on which an abnormality has occurred;
A generation unit that uses the learning data to generate a prediction model that predicts the occurrence of the above-mentioned change on the road in operation from the external state data of the automatically-operated vehicle in operation;
an updating unit that updates the recommended speed set for the road on which the vehicle is running, in the recommended speed data, according to the result of prediction by the prediction model;
An operation management device with a A control method for an operation management device according to an operation management system that manages operation of a plurality of automated driving vehicles traveling with reference to recommended speed data related to recommended speeds set for each road,
The operation control device
a step of obtaining, from the automated driving vehicle, operation performance data collected by the automated driving vehicle in operation;
The incident data related to the incident that occurred during the operation of the automated driving vehicle and the vehicle exterior state data related to the exterior state of the automated driving vehicle when the incident occurred, which are included in the operation performance data. a step of storing learning data associated with each road on which an abnormality has occurred;
using the learning data to generate a prediction model that predicts the occurrence of the occurrence of the accident on the road in operation from the external state data of the autonomous vehicle in operation;
updating the recommended speed set for the road in operation in the recommended speed data according to the result of prediction by the prediction model;
A control method for an operation management device including A control program for an operation management device according to an operation management system that manages operation of a plurality of automated driving vehicles that run with reference to recommended speed data related to recommended speeds set for each road,
to the operation control device,
A function of acquiring operation performance data collected by the automated driving vehicle in operation from the automated driving vehicle;
The incident data related to the incident that occurred during the operation of the automated driving vehicle and the vehicle exterior state data related to the exterior state of the automated driving vehicle when the incident occurred, which are included in the operation performance data. A function to store learning data associated with each road on which an abnormality has occurred;
A function of generating a prediction model that predicts the occurrence of the above-mentioned change on the road in operation from the outside state data of the automatic driving vehicle in operation using the learning data;
A function of updating the recommended speed set for the road in operation in the recommended speed data according to the result of prediction by the prediction model;
A control program for an operation management device that realizes

Images