Vehicle dispatching system, vehicle dispatching management method

Abstract

The present disclosure relates to a vehicle dispatch system, a vehicle dispatch management method. The vehicle dispatch system of the present disclosure includes: a plurality of travel service vehicles including one or more autonomous vehicles that autonomously travel according to a vehicle dispatch plan; and at least one processor that executes a vehicle dispatch plan generation process. The vehicle dispatch plan generation process includes the following processes: acquiring a running design domain of the one or more autonomous vehicles; acquiring, for a running partition of the one or more autonomous vehicles, a prediction of running environment information for a period up to a future prescribed period; determining, based on the running design domain and the prediction of the running environment information, an inoperable vehicle among the one or more autonomous vehicles that is predicted to not satisfy the running design domain in the current vehicle dispatch plan; and updating the vehicle dispatch plan according to vehicle information or running information of the inoperable vehicle.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a technology of generating a vehicle dispatch plan of a vehicle. In particular, it relates to a technology of coping with a case where an autonomous vehicle to be dispatched will not satisfy a design domain of operation. BACKGROUND

[0002] Patent Literature 1 discloses an autonomous vehicle that accommodates at least two different travel modes, which include a first travel mode configured for autonomous travel and a second travel mode configured for guidance by a pilot vehicle. Further, it is disclosed that the second travel mode can be associated with a second geographical area defined as not guaranteeing autonomy, and that the autonomous vehicle can travel in the second geographical area by the second travel mode.

[0003] In addition to this, as documents indicating the state of the art in this technical field, there are Patent Literatures 2 to 6 below.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT LITERATURE

[0006] Patent Literature 1: Japanese Patent Application Publication No. 2019-510670

[0007] Patent Literature 2: Japanese Patent Application Publication No. 2021-163369

[0008] Patent Literature 3: Japanese Patent Application Publication No. 2021-043896

[0009] Patent Literature 4: Japanese Patent Application Publication No. 2020-106525

[0010] Patent Literature 5: International Publication No. WO 2020 / 196084

[0011] Patent Literature 6: International Publication No. WO 2020 / 196086

[0012] In recent years, various technologies and businesses related to the operation of MaaS (Mobility as a Service) vehicles for the provision of mobility services have attracted attention. In the operation of MaaS vehicles, a vehicle dispatch plan is generally given in accordance with the content of the mobility service to be provided, the request of a user, and the like, and the operation of MaaS vehicles based on the vehicle dispatch plan is performed. Here, an autonomous vehicle that travels autonomously in accordance with the vehicle dispatch plan can be included in the MaaS vehicles.

[0013] Further, in an autonomous vehicle, an operational design domain (ODD) that is a condition of an environment in which autonomous travel can be continued is determined in accordance with the vehicle or the autonomous driving system applied. When the ODD becomes not satisfied, the autonomous vehicle typically acts in a manner of seeking a driving operation from a driver in the vehicle or safely stopping the vehicle in a case where there is no driver in the vehicle. In this case, in a case where the autonomous vehicle without a driver in the vehicle has stopped, and in a case where a main reason that the ODD will not be satisfied belongs to a relatively long period, a person needs to be dispatched to recover the autonomous vehicle that has stopped.

[0014] An autonomous vehicle without a driver in the vehicle is also assumed as an autonomous vehicle that can be included in a MaaS vehicle. In the use of the MaaS vehicle, a response such as dispatching a person to recover the autonomous vehicle that has stopped can cause an influence on passengers due to a stop of operation, an increase in labor cost due to the dispatched person, and the like. On the other hand, an autonomous vehicle with a wide ODD generally has a high price of sensors and systems mounted thereon, and it is difficult to be actively adopted. Therefore, a technique for reducing the frequency of recovery and the labor of recovery is sought.

[0015] According to the technology disclosed in Patent Literature 1, an autonomous vehicle can be guided from a region in which the ODD is not satisfied (second geographical region) to a region in which the ODD is satisfied by a lead vehicle. That is, as long as the lead vehicle is prepared in advance, the labor of recovery can be reduced, and a person does not need to be dispatched. However, in the technology disclosed in Patent Literature 1, the region in which the ODD is not satisfied needs to be determined in advance, and a case in which the ODD is not satisfied due to a dynamic factor cannot be dealt with. In addition, from a cost aspect, it is also undesirable to have the lead vehicle reside in the region in which the ODD is not satisfied. SUMMARY

[0016] In view of the above-described problems, one object of the present disclosure is to provide a technique capable of reducing the frequency of recovery and the labor of recovery of an autonomous vehicle.

[0017] A first disclosure relates to a vehicle dispatch system.

[0018] A first disclosed vehicle dispatch system includes: a plurality of MaaS vehicles; at least one processor; and a program memory coupled with the at least one processor, storing a plurality of executable instructions. The plurality of executable instructions are configured to cause the at least one processor to perform a vehicle dispatch plan generation process that generates a vehicle dispatch plan for the plurality of MaaS vehicles. Further, the plurality of MaaS vehicles includes one or more autonomous vehicles that autonomously travel in accordance with the vehicle dispatch plan. Also, the vehicle dispatch plan generation process includes: obtaining an operational design domain for the one or more autonomous vehicles; obtaining, for an operational partition of the one or more autonomous vehicles, a prediction of an operational environment information for a period up to a prescribed future period; determining, based on the operational design domain and the prediction of the operational environment information, an inoperable vehicle among the one or more autonomous vehicles that is predicted to not satisfy the operational design domain in the current vehicle dispatch plan; and a plan update process that updates the vehicle dispatch plan in accordance with vehicle information or operational information of the inoperable vehicle.

[0019] A second disclosure relates to a vehicle dispatch system having the following features in addition to the vehicle dispatch system of the first disclosure.

[0020] The one or more autonomous vehicles are configured to implement the following steps: stopping when becoming not to satisfy the operational design domain; after stopping, following, in response to any one of the plurality of MaaS vehicles passing by as a preceding vehicle, the preceding vehicle; and resuming the autonomous travel when the operational design domain is satisfied by the following.

[0021] A third disclosure relates to a vehicle dispatch system having the following features in addition to the vehicle dispatch system of the second disclosure.

[0022] The plan update process includes: selecting, in accordance with the vehicle information or operational information, one or more recovery vehicles from the plurality of MaaS vehicles to update the vehicle dispatch plan in such a way that the one or more recovery vehicles pass through a location where the inoperable vehicle is predicted to not satisfy the operational design domain as the preceding vehicle.

[0023] A fourth disclosure relates to a vehicle dispatch system having the following features in addition to the vehicle dispatch system of the third disclosure.

[0024] The one or more recovery vehicles are manually driven vehicles or autonomous vehicles that satisfy the operational design domain in the updated vehicle dispatch plan.

[0025] A fifth disclosure relates to a vehicle dispatch system having the following features in addition to the vehicle dispatch system of the first to fourth disclosures.

[0026] The plan update processing includes updating the vehicle dispatch plan in a manner that, according to the vehicle information or operation information, the non-operable vehicles before the vehicle dispatch are replaced with manually driven vehicles or autonomous driving vehicles that satisfy the operation design domain.

[0027] The sixth disclosure relates to a vehicle dispatch system with respect to the second to fourth disclosures, and further has the following features.

[0028] The plan update processing includes updating the vehicle dispatch plan in a manner that, according to the vehicle information or operation information, each of a set of the non-operable vehicles before the vehicle dispatch that are predicted to not satisfy the operation design domain at the same location in the current vehicle dispatch plan is replaced with at least one of a manually driven vehicle or an autonomous driving vehicle that satisfies the operation design domain.

[0029] The seventh disclosure relates to a vehicle dispatch management method for a plurality of MaaS vehicles that operate based on a vehicle dispatch plan. Here, the plurality of MaaS vehicles includes one or more autonomous driving vehicles that autonomously travel according to the vehicle dispatch plan.

[0030] The vehicle dispatch management method of the seventh disclosure includes acquiring an operation design domain for the one or more autonomous driving vehicles, acquiring a prediction of an operation environment information for a period up to a future period for an operation zone of the one or more autonomous driving vehicles, determining, based on the operation design domain and the prediction of the operation environment information, a non-operable vehicle of the one or more autonomous driving vehicles that is predicted to not satisfy the operation design domain in the current vehicle dispatch plan, and updating the vehicle dispatch plan according to vehicle information or operation information of the non-operable vehicle.

[0031] The eighth disclosure relates to a vehicle dispatch management method with respect to the vehicle dispatch management method of the seventh disclosure, and further has the following features.

[0032] The one or more autonomous driving vehicles are configured to implement the following steps: stopping when the operation design domain is not satisfied, in response to any one of the plurality of MaaS vehicles passing as a preceding vehicle after stopping, the one or more autonomous driving vehicles performing a following travel of the preceding vehicle, and resuming the autonomous travel when the operation design domain is satisfied by the following travel. Furthermore, the updating of the vehicle dispatch plan includes updating the vehicle dispatch plan in a manner that, according to the vehicle information or operation information, one or more recovery vehicles are selected from the plurality of MaaS vehicles, the one or more recovery vehicles pass as the preceding vehicle from a location where the non-operable vehicle is predicted to not satisfy the operation design domain.

[0033] The ninth disclosure relates to the vehicle dispatch management method according to the seventh disclosure or the eighth disclosure, and further has the following feature.

[0034] The updating of the vehicle dispatch plan includes updating the vehicle dispatch plan in a manner that replaces the non-operable vehicles before the vehicle dispatch with manual driving vehicles or autonomous driving vehicles that satisfy the operational design domain, based on the vehicle information or the operational information.

[0035] The tenth disclosure relates to the vehicle dispatch management method according to the seventh disclosure, and further has the following feature.

[0036] The one or more autonomous driving vehicles are configured to implement the steps of stopping when becoming not to satisfy the operational design domain, performing follow-up travel with respect to any one of the plurality of MaaS vehicles as a preceding vehicle in response to the preceding vehicle passing after the stopping, and resuming the autonomous travel when the operational design domain is satisfied by the follow-up travel. Further, the updating of the vehicle dispatch plan includes updating the vehicle dispatch plan in a manner that replaces at least one of the non-operable vehicles before the vehicle dispatch with a manual driving vehicle or an autonomous driving vehicle that satisfies the operational design domain, with respect to each of a set of the non-operable vehicles before the vehicle dispatch that is predicted to not satisfy the operational design domain at the same location as the current vehicle dispatch plan.

[0037] Effects of Invention

[0038] According to the present disclosure, non-operable vehicles in a current vehicle dispatch plan are determined based on a prediction of an operational design domain (ODD: Operational Design Domain) and operational environment information for autonomous driving vehicles included in a plurality of MaaS vehicles. Then, the vehicle dispatch plan is updated based on vehicle information or operational information of the non-operable vehicles. Thereby, it is possible to cope with non-operable vehicles predicted during a period up to a prescribed future period in the current vehicle dispatch plan, and thus it is possible to reduce the frequency of recovery of autonomous driving vehicles and the amount of work for recovery. Furthermore, it is possible to seek a reduction in the cost of operation. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 is a conceptual diagram for explaining an outline of a vehicle dispatch system of the present embodiment.

[0040] Figure 2 is a conceptual diagram for explaining non-operable vehicles determined in a vehicle dispatch plan generation process.

[0041] Figure 3is a conceptual diagram for describing an action when the autonomous vehicle of the present embodiment becomes not to satisfy the ODD.

[0042] Figure 4 is a conceptual diagram showing a first example of the vehicle dispatch plan updated by the update of the vehicle dispatch plan. Figure 2 is a conceptual diagram showing an example of the vehicle dispatch plan updated by the update of the vehicle dispatch plan shown in

[0043] Figure 5 is a conceptual diagram showing an example of the vehicle dispatch plan.

[0044] Figure 6 is a conceptual diagram showing a running status achieved when the vehicle dispatch plan is updated by the third example of the update of the vehicle dispatch plan.

[0045] Figure 7 is a block diagram showing a schematic configuration of a server.

[0046] Figure 8 is a block diagram showing a schematic configuration of a vehicle dispatch plan generation process.

[0047] Figure 9 is a block diagram showing a schematic configuration of an autonomous vehicle.

[0048] Figure 10 is a block diagram showing a schematic configuration of a process executed by an autonomous travel control device.

[0049] Figure 11 is a flowchart showing a vehicle dispatch plan generation process.

[0050] Figure 12 is a flowchart showing a process of generating a vehicle dispatch plan that becomes an initial value.

[0051] Figure 13 is a flowchart showing a process executed by an autonomous travel control device.

[0052] Explanation of Reference Signs:

[0053] 1: MaaS vehicle

[0054] 2: Server

[0055] 3: Vehicle dispatch station

[0056] 6: Running zone

[0057] 7: Vehicle dispatch route

[0058] 10: Vehicle dispatch system

[0059] 100: Autonomous travel control device

[0060] 110: Sensor

[0061] 120: travel control device

[0062] 150: communication device

[0063] 200: information processing section

[0064] 210: memory

[0065] 211: program

[0066] 212: instruction

[0067] 213: data

[0068] 220: processor

[0069] 250: communication device. DETAILED DESCRIPTION

[0070] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In the case where a numerical value of the number, the quantity, the amount, the range, and the like of each element is mentioned in the embodiment shown below, unless it is specifically mentioned or it is obviously determined in principle that the numerical value is mentioned, the idea of the present disclosure is not limited to the mentioned numerical value. In addition, as for the constitution and the like explained in the embodiment shown below, unless it is specifically mentioned or it is obviously determined in principle that the constitution is mentioned, it is not necessarily essential in the idea of the present disclosure. Note that the same reference numerals are attached to the same or equivalent portions in each drawing, and the repeated explanation is appropriately simplified or omitted.

[0071] 1. Vehicle dispatch system

[0072] Referring to Figure 1 A summary of the vehicle dispatch system 10 of the present embodiment will be described. In the vehicle dispatch system 10 of the present embodiment, a vehicle dispatch plan of a plurality of MaaS vehicles 1 is generated. By the vehicle dispatch plan, a destination, a transfer station, a vehicle dispatch route, a vehicle dispatch time, a passing or arrival time at each place, and the like are given for each of the plurality of MaaS vehicles 1. Then, each of the plurality of MaaS vehicles 1 operates based on the generated vehicle dispatch plan.

[0073] The plurality of MaaS vehicles 1 include one or more autonomous vehicles that autonomously travel in accordance with a vehicle dispatch plan. Here, the plurality of MaaS vehicles 1 can include various autonomous vehicles that differ in form and specifications. For example, autonomous vehicles without a driver, autonomous vehicles that can be switched to manual driving, autonomous vehicles equipped with many sensors and having high detection capabilities for the surrounding environment, and autonomous vehicles equipped with only a minimum of sensors and having low detection capabilities for the surrounding environment, and the like can be included. Note that manual driving vehicles can be included in the plurality of MaaS vehicles 1. The operation of the manual driving vehicles is performed by a driver driving the vehicle in accordance with the generated vehicle dispatch plan.

[0074] Each of the plurality of MaaS vehicles 1 is dispatched, for example, from any of a plurality of vehicle dispatch stations located at specific locations. Furthermore, each of the plurality of MaaS vehicles 1 returns to any of the plurality of vehicle dispatch stations after the operation ends. However, the plurality of MaaS vehicles 1 can also be dispatched from a prescribed parking place such as a parking lot.

[0075] In Figure 1 In the vehicle dispatch system 10 illustrated in the figure, a process of generating a vehicle dispatch plan (hereinafter also referred to as "vehicle dispatch plan generation process") is executed in the server 2. The server 2 is configured to be able to mutually transmit and receive information with the plurality of MaaS vehicles 1 via a communication network 4. For example, the communication network 4 is constituted by a base station that can wirelessly communicate with the plurality of MaaS vehicles 1 and the Internet. In this case, the server 2 and the base station are connected to the Internet.

[0076] The server 2 acquires, from the plurality of MaaS vehicles 1 via the communication network 4, information (hereinafter also referred to as "vehicle information") regarding each of the plurality of MaaS vehicles 1. Here, as the vehicle information, the form, the specifications, the travel location, the state of the vehicle body (wetness, freezing, and the like), the travel state (vehicle speed, acceleration, and the like), the surrounding environment (preceding vehicle, road surface, weather, air temperature, and the like), the state of the mounted sensors (presence or absence of obstruction of the camera image, presence or absence of optical axis shift, and the like), the mounting position, the autonomous travel path, and the like of the vehicle are exemplified. In particular, the vehicle information includes information on an operational design domain (ODD) of each of one or more autonomous vehicles (hereinafter also simply referred to as "autonomous vehicle") included in the plurality of MaaS vehicles 1. As the information on the ODD, the brightness at which autonomous travel can be continued, the state of the road surface at which autonomous travel can be continued, the weather state at which autonomous travel can be continued, the detection accuracy of the sensor at which autonomous travel can be continued, the road shape at which autonomous travel can be continued, the category of the surrounding obstacle at which autonomous travel can be continued, and the like are exemplified.

[0077] Then, in the server 2, vehicle dispatch plan generation processing is executed based on the acquired information, and the server 2 transmits a vehicle dispatch plan to the plurality of MaaS vehicles 1 via the communication network 4. In this sense, the server 2 can also be referred to as a "vehicle dispatch plan device".

[0078] Further, an automatic driving vehicle without a driver generally acts in a manner to safely stop when it will not satisfy the ODD. At this time, in a case where the main reason that the ODD will not be satisfied belongs to a relatively long period, the automatic driving vehicle needs to be recovered. For example, in a case of an automatic driving vehicle that will not satisfy the ODD in a heavy rain condition, a case where the weather changes to be heavy rain during operation; a case of an automatic driving vehicle that will not satisfy the ODD at night without coping with dark places, a case where the sun sets during operation to be night; a case where the ODD will not be satisfied due to narrowing of a road width due to construction, appearance of a fallen object; and a case of an automatic driving vehicle premised on a car-only road where the ODD will not be satisfied due to detection of a pedestrian or a bicycle on a road during travel, and the like.

[0079] When the automatic driving vehicle is recovered, dispatch of a person as a driver can cause an influence on passengers due to stop of operation, increase of labor cost due to dispatch of a person, and the like. Therefore, it is desirable to reduce the frequency of such recovery and the labor for recovery.

[0080] Therefore, in the vehicle dispatch system 10 of the present embodiment, the server 2 acquires a prediction of operation environment information for a period until a predetermined period in the future for each operation zone of the automatic driving vehicle. Among the acquired prediction of operation environment information, information of the current operation environment can be included. Further, in the vehicle dispatch plan generation processing, based on the ODD and the prediction of operation environment information for each of the automatic driving vehicles, an automatic driving vehicle predicted to not satisfy the ODD in the current vehicle dispatch plan (hereinafter also referred to as "non-operable vehicle") is determined from among the automatic driving vehicles.

[0081] Here, the operation zone of the automatic driving vehicle refers to, for example, a region in which operation of the automatic driving vehicle is scheduled. The operation zone can be a specific region given in advance, or a region given according to an operation route given by an operation plan (for example, a region including the entire operation route).

[0082] Furthermore, the operating environment information refers to environmental information related to the operation of multiple MaaS vehicles 1. Specifically, the operating environment information includes information related to the ODD (Operating Device Designation) of the autonomous vehicles. Examples of operating environment information include road traffic information (road shape, road surface condition, information on vehicles involved in parking accidents, etc.), sunlight information, weather information, and information on the occurrence of special events (falling objects, appearance in dark places, etc.). Server 2 can be configured to acquire operating environment information at various times up to a specified period. It should be noted that server 2 can be configured to receive this information via communication network 4 from... Figure 1 Other devices not shown in the diagram acquire operating environment information. For example, they can acquire operating environment information by communicating with servers that send road traffic information or servers that send weather information.

[0083] Furthermore, operational information refers to information related to operation based on the vehicle scheduling plan. For example, operational information includes vehicle scheduling routes given by the vehicle scheduling plan, and the scheduled arrival times for each location along those routes. Server 2 can obtain operational information either from the autonomous vehicle via communication network 4 or by referring to the current vehicle scheduling plan.

[0084] Reference Figure 2 The unoperable vehicles identified during the vehicle scheduling plan generation process are explained. Figure 2 This is a conceptual example illustrating a vehicle scheduling plan at a given moment for a running partition 6 that includes multiple paths 5. Figure 2 In the vehicle scheduling plan, the vehicle scheduling routes 7a, 7b, 7c and 7d for each of the four MaaS vehicles 1 (1a, 1b, 1c and 1d) are shown.

[0085] Now, assume that each of the four MaaS vehicles 1 is an autonomous vehicle. That is, each of the four MaaS vehicles 1 autonomously drives along its own vehicle scheduling route. Furthermore, assume that based on predictions of driving environment information, the prediction is... Figure 2 If a situation occurs at location 8 that will not meet the ODD (Operational Data Definition), then autonomous vehicles 1a and 1b, whose vehicle scheduling routes include location 8, will be identified as inoperable vehicles. However, if a situation occurs at location 8 that will not meet the ODD after a predetermined time when autonomous vehicle 1a or 1b is predicted to pass through location 8 in the vehicle scheduling plan, then autonomous vehicle 1a or 1b may not be identified as inoperable vehicles.

[0086] As an example of the prediction of the occurrence of a situation in which the automated vehicle will not satisfy the ODD, the following situation can be cited. The place where it will rain heavily is predicted based on the weather forecast at each time as the prediction of the running environment information. The timing at which it will become nighttime during the running of the automated vehicle is predicted based on the scheduled time of sunset as the prediction of the driving environment information. The place where a shadow is cast is predicted based on the azimuth of the sun at each time as the prediction of the driving environment information and the information of the buildings in the running zone 6. The place where fog is generated, the place where hot air is generated due to exhaust gas, gas generated from a manhole, is predicted based on the humidity and the air temperature at each time as the prediction of the running environment information. The appearance of a falling object, a pedestrian, a bicycle is predicted by acquiring detection information of a sensor equipped in a vehicle that is traveling.

[0087] Note that it can be determined for each of the automated vehicles whether it will not satisfy the ODD. For example, in Figure 2 , it is assumed that it is predicted based on the prediction of the driving environment information that it will rain heavily at the place 8. Further, it is assumed that the specifications of the automated vehicle la are high, the ODD will be satisfied even if it rains heavily, and the specifications of the automated vehicle lb are low, the ODD will not be satisfied when it rains heavily. At this time, the automated vehicle lb is determined to be a non-running vehicle, and on the other hand, the automated vehicle la is not determined to be a non-running vehicle.

[0088] Thus, in the vehicle dispatch system 10 of the present embodiment, based on the prediction of the ODD and the running environment information for each of the automated vehicles, a non-running vehicle is determined in the current vehicle dispatch plan.

[0089] Then, in the vehicle dispatch system 10 of the present embodiment, in the running plan generation process, a process of updating the vehicle dispatch plan based on the vehicle information or the running information of the non-running vehicle is executed (hereinafter also referred to as "plan update process"). Thereby, it is possible to cope with the non-running vehicle predicted in the period until the future prescribed period in the current vehicle dispatch plan, and thus it is possible to reduce the frequency of the collection of the automated vehicles and the effort of the collection. Hereinafter, several examples of the update of the vehicle dispatch plan achieved by the plan update process are shown.

[0090] 1-1. First Example

[0091] First, in the first example, the automated vehicle is configured to act as follows when it becomes not to satisfy the ODD. With reference to Figure 3 , the action of the automated vehicle of the first example when it becomes not to satisfy the ODD is described. Figure 3The diagram illustrates the situation when autonomous vehicle 1b arrives at location 8 where the ODD (Operational Development Domain) will not be met. At this point, autonomous vehicle 1b first acts by safely parking. For example, autonomous vehicle 1b stops on the shoulder while simultaneously assessing the traffic conditions of surrounding vehicles. Next, in response to any of the multiple MaaS vehicles 1 passing as a lead vehicle, autonomous vehicle 1b follows the lead vehicle. Figure 3 The diagram illustrates a scenario where MaaS vehicle 1e passes by as a lead vehicle, and autonomous vehicle 1b follows behind. Then, once autonomous vehicle 1b has satisfied the ODD (Optical Design Domain) through following the lead vehicle, it resumes autonomous driving. Figure 3 The diagram shows the autonomous vehicle 1b resuming autonomous driving after leaving location 8, which would no longer meet the ODD, by following another vehicle.

[0092] With this configuration of autonomous vehicles, even if they stop without meeting the ODD (Open Design Principle), the autonomous vehicle can still be led by any one of the multiple MaaS (Multi-Action Service) vehicles 1, following it until the ODD is met, thus detaching from the stopped state. In this sense, it can be considered as an autonomous vehicle that has already stopped (…). Figure 3 The MaaS vehicle 1 (shown as an autonomous vehicle 1b) passes ahead of the MaaS vehicle 1 ( Figure 3 The MaaS vehicle 1e shown is a "recycled vehicle" of a parked autonomous vehicle.

[0093] In the first example, based on the vehicle information or operation information of the inoperable vehicles, one or more reclaimed vehicles are selected from multiple MaaS vehicles 1, and the vehicle scheduling plan is updated in such a way that one or more reclaimed vehicles act as lead vehicles to pass through the location where the predicted inoperable vehicles will not meet the ODD.

[0094] Figure 4 This means that through the first example... Figure 2 The diagram shown is a conceptual illustration of an example of an updated vehicle scheduling plan. Now, assume that autonomous vehicles 1a and 1b are identified as inoperable vehicles predicted to not meet the ODD at location 8. Figure 4 In the updated vehicle dispatch plan shown, MaaS vehicle 1e, which was assigned to vehicle dispatch route 7e from location 8, is re-dispatch as a recovery vehicle.

[0095] Here, the MaaS vehicle le is selected in such a way that the ODD is satisfied in the vehicle dispatch route 7e passing through the location 8. For example, in a case where the autonomous vehicles la and lb are determined as non-operable vehicles due to a prediction that heavy rain will occur at the location 8, autonomous vehicles in the plurality of MaaS vehicles 1 that satisfy the ODD even in the condition of heavy rain are selected. Or a manual driving vehicle is selected. As the autonomous vehicle that satisfies the ODD even in the condition of heavy rain, an autonomous vehicle to which a camera is added, an autonomous vehicle on which a plurality of LiDARs (capable of detecting the influence due to rain according to reflection characteristics) that irradiate laser beams of mutually different wavelengths are mounted are exemplified.

[0096] As the autonomous vehicle that can be selected for other conditions and that satisfies the ODD, the following examples can be given. However, a manual driving vehicle can also be selected for other conditions. As the autonomous vehicle that satisfies the ODD even in a condition where a dark place occurs, an autonomous vehicle to which a LiDAR is added, an autonomous vehicle on which a night vision camera is mounted are exemplified. As the autonomous vehicle that satisfies the ODD even in a condition where a fallen object occurs, an autonomous vehicle on which a LiDAR that irradiates a laser beam obliquely downward, a LiDAR that is disposed at a low position of the vehicle, or a high-density LiDAR is mounted is exemplified. As the autonomous vehicle that satisfies the ODD even in a condition where a pedestrian or a bicycle is detected, an autonomous vehicle on which a high-density LiDAR is mounted, an autonomous vehicle on which an infrared thermograph is mounted is exemplified.

[0097] In a case where the vehicle dispatch system 10 of the present embodiment is applied, these autonomous vehicles can be prepared as the plurality of MaaS vehicles 1 in accordance with conditions that can be assumed in the operation zone 6. In particular, the preparation can be made as necessary and sufficient in consideration of a balance with costs.

[0098] Note that the prediction of the operation environment information is also made at a location other than the location 8 of the vehicle dispatch route 7e, and the MaaS vehicle 1 is selected in such a way that the ODD is satisfied in the vehicle dispatch route 7e. However, this can also be made in such a way that the MaaS vehicle 1 that satisfies the ODD at least at the location 8 is selected as a recovery vehicle, and a vehicle dispatch route that satisfies the ODD is given to the selected recovery vehicle, and at least the vehicle dispatch route passing through the location 8.

[0099] Further, the vehicle dispatch route 7e of the MaaS vehicle le is given in such a manner that the MaaS vehicle le passes through the location 8 as a preceding vehicle of the autonomous vehicles la and lb. That is, at least the vehicle dispatch route 7e of the MaaS vehicle le is given in such a manner that the MaaS vehicle le passes through the location 8 later than the predetermined time at which the autonomous vehicles la and lb pass through the location 8. This can be performed by acquiring the predetermined time at which the autonomous vehicles la and lb pass through the location 8 as the vehicle information or the operation information of the autonomous vehicles la and lb. Also, in order to shorten the stop time of the autonomous vehicles la and lb, the vehicle dispatch route 7e can also be given in such a manner that the MaaS vehicle le passes through the location 8 immediately after the predetermined time at which the autonomous vehicles la and lb pass through the location 8.

[0100] Thus, by updating the vehicle dispatch plan to dispatch the MaaS vehicle le, the autonomous vehicles la and lb can perform follow-up travel of the MaaS vehicle le at the location 8. Then, the autonomous vehicles la and lb can resume autonomous travel after passing through the location 8 by follow-up travel. Note that cases where a plurality of non-operable vehicles are determined, and cases where locations predicted to not satisfy the ODD are not on the same path 5, and cases where times predicted to not satisfy the ODD are separated from each other are also considered. In such cases, the vehicle dispatch plan can be updated to dispatch a plurality of recovery vehicles in accordance with the vehicle information or the operation information of the plurality of non-operable vehicles. In particular, the vehicle dispatch plan can be given in such a manner that one recovery vehicle recovers a plurality of non-operable vehicles, and an efficient vehicle dispatch plan can be given.

[0101] As explained above, according to the first example, a recovery vehicle can be caused to pass through a location predicted to not satisfy the ODD of a non-operable vehicle as a preceding vehicle. Thereby, even if the non-operable vehicle becomes to not satisfy the ODD, the non-operable vehicle can pass through the location predicted to not satisfy the ODD by follow-up travel of the recovery vehicle, and can resume autonomous travel. Further, the frequency and the effort of recovery of the autonomous vehicle can be reduced. In particular, the dispatch of the recovery vehicle can be performed as necessary and sufficiently with respect to the predicted non-operable vehicle, and thus the dispatch of the recovery vehicle can be implemented at low cost.

[0102] 1-2. Second Example

[0103] In the second example, the vehicle dispatch plan is updated in such a manner that the non-operable vehicle before vehicle dispatch is replaced with a manual driving vehicle or an autonomous vehicle capable of satisfying the ODD in accordance with the vehicle information or the operation information of the non-operable vehicle.

[0104] For example, assume that the current vehicle dispatch plan is Figure 2The vehicle scheduling plan shown depicts autonomous vehicles 1a, 1b, 1c, and 1d before scheduling. Furthermore, it is assumed that autonomous vehicles 1a and 1b are predicted to be inoperable vehicles that will not meet the ODD at location 8. In this case, in the second example, the vehicle scheduling plan is updated by replacing autonomous vehicles 1a and 1b with manually driven vehicles or autonomous vehicles that can meet the ODD at location 8.

[0105] For example, if autonomous vehicles 1a and 1b are determined to be inoperable due to a predicted dark area at location 8, they can be replaced with manually driven vehicles or autonomous vehicles that meet the ODD even in the dark. Alternatively, if autonomous vehicles 1a and 1b are determined to be inoperable due to a predicted heavy rain at location 8, they can be replaced with manually driven vehicles or autonomous vehicles that meet the ODD even in the heavy rain.

[0106] By updating the vehicle scheduling plan in this way, a vehicle scheduling plan can be implemented for vehicles that are predicted to be inoperable and will not meet the ODD before the vehicle scheduling is completed, either manually driven or MaaS vehicles that can meet the ODD. This reduces the frequency and time required for autonomous vehicles to be retrieved.

[0107] It should be noted that the first and second examples can also be combined. For example, consider using the vehicle scheduling plan change implemented by the first example to deal with vehicles that are not operational after scheduling, and using the vehicle scheduling plan change implemented by the second example to deal with vehicles that were not operational before scheduling.

[0108] 1-3. The third case

[0109] First, in the third example, the autonomous vehicle is configured to perform the actions described in the first example when it becomes unsatisfactory in terms of the ODD (see [reference]). Figure 3 Perform the same action.

[0110] In the third example, based on the vehicle information or operation information of the inoperable vehicles, the vehicle scheduling plan is updated as follows: for each of the sets of inoperable vehicles that are predicted to fail to meet the ODD in the current vehicle scheduling plan, at least one inoperable vehicle is replaced with a manually driven vehicle or an autonomous vehicle that can meet the ODD.

[0111] The following is for reference Figure 5 Let's explain the third example. Figure 5 and Figure 2 This is also a conceptual diagram representing an example of a vehicle scheduling plan. Now, suppose the current vehicle scheduling plan is... Figure 5The vehicle dispatch plan shown, the autonomous vehicles 1a, 1b, 1c, 1d, 1e, and 1f are before vehicle dispatch. Further, it is assumed that, according to the prediction of the running environment information and the ODD of each of the autonomous vehicles, it is predicted that a situation will occur in which the autonomous vehicles 1a and 1b will not satisfy the ODD at the location 8a, and a situation will occur in which the autonomous vehicles 1c and 1e will not satisfy the ODD at the location 8b. That is, in Figure 5 In the example shown, the autonomous vehicles 1a, 1b, 1c, and 1e are determined to be non-operable vehicles. In particular, in Figure 5 In the example shown, the sets of non-operable vehicles in which the locations at which it is predicted that the ODD will not be satisfied are the same are the autonomous vehicles 1a and 1b (the same at the location 8a) and the autonomous vehicles 1c and 1e (the same at the location 8b).

[0112] At this time, in the third example, the vehicle dispatch plan is updated in such a way that at least one of the autonomous vehicles 1a or 1b is replaced with a manual driving vehicle or an autonomous vehicle that can satisfy the ODD at the location 8a, and at least one of the autonomous vehicles 1c or 1e is replaced with a manual driving vehicle or an autonomous vehicle that can satisfy the ODD at the location 8b.

[0113] By updating the vehicle dispatch plan in this way, the vehicle dispatch plan for the non-operable vehicles that have been replaced can be achieved by a manual driving vehicle or a MaaS vehicle 1 that can satisfy the ODD. On the other hand, the non-operable vehicles that have not been replaced can escape from the locations at which the ODD will not be satisfied by following the replaced manual driving vehicles or autonomous vehicles that can satisfy the ODD at the locations at which the ODD will not be satisfied, and can resume autonomous driving. Furthermore, the frequency and the amount of time of recovery of the autonomous vehicles can be reduced.

[0114] In Figure 6 the running situations achieved when the vehicle dispatch plan shown is updated in such a way that the autonomous vehicle 1a and the autonomous vehicle 1e are replaced by the third example are shown in Figure 5 the running situations achieved when the vehicle dispatch plan shown is updated in such a way that the autonomous vehicle 1a and the autonomous vehicle 1e are replaced by the third example are shown in Figure 6 As shown, it can be seen that, by updating the vehicle dispatch plan in this way, the autonomous vehicle 1b can escape from the location 8a by following the autonomous vehicle 1a, and the autonomous vehicle 1c can escape from the location 8b by following the autonomous vehicle 1e. Further, the autonomous vehicles 1a, 1d, 1e, and 1f can satisfy the ODD in the updated vehicle dispatch plan.

[0115] Note that it can also be assumed that different situations are predicted to occur at each of the locations at which it is predicted that the ODD will not be satisfied. For example, in Figure 5The first example is a case where it is predicted that heavy rain will occur at the location 8a, it is predicted that darkness will occur at the location 8b, or a situation where the autonomous vehicle cannot satisfy the ODD occurs at the location 8a, a situation where the autonomous vehicle can satisfy the ODD if it is a high-specification autonomous vehicle occurs at the location 8b, and the like. In this case, the MaaS vehicle 1 that replaces the set of non-operable vehicles can be selected in accordance with the respective situations of the locations that are predicted to not satisfy the ODD. In particular, the MaaS vehicle 1 that is necessary and sufficient can be selected in consideration of the cost.

[0116] Note that the first example and the third example can be combined. For example, the non-operable vehicle after the vehicle dispatch is addressed by the change in the vehicle dispatch plan realized by the first example, and the non-operable vehicle before the vehicle dispatch is addressed by the change in the vehicle dispatch plan realized by the third example.

[0117] 2. Configuration

[0118] Hereinafter, the outline configuration of the server (vehicle dispatch plan device) 2 and the autonomous vehicle in the vehicle dispatch system 10 of the present embodiment will be described.

[0119] 2-1. Configuration of the server (vehicle dispatch plan device)

[0120] Figure 7 is a block diagram that shows the outline configuration of the server (vehicle dispatch plan device) 2. The server 2 is provided with an information processing section 200 and a communication device 250. The information processing section 200 is configured to be able to mutually transmit and receive information with the communication device 250. For example, it is configured by electrical connection through a cable, connection through an optical line, or the like.

[0121] The information processing section 200 is a computer that includes a memory 210 and a processor 220. The memory 210 is coupled to the processor 220 and stores a plurality of executable instructions 212 and various data 213 required for execution of processing. The instructions 212 are given by a program 211. In this sense, the memory 210 can also be referred to as a "program memory".

[0122] Execution of various processing based on the data 213 is realized by the processor 220 acting in accordance with the instructions 212. In particular, execution of the vehicle dispatch plan generation processing is realized. Then, the information processing section 200 transmits the vehicle dispatch plan generated by execution of the vehicle dispatch plan generation processing to the communication device 250.

[0123] The communication device 250 performs transmission and reception of various information by communicating with devices outside the server 2. That is, the communication device 250 is a device for performing communication via the communication network 4. The vehicle dispatch plan acquired from the information processing section 200 is transmitted to the plurality of MaaS vehicles 1 by the communication device 250. Further, information received by the communication device 250 is delivered to the information processing section 200. As the information received by the communication device 250, exemplified are vehicle information of the plurality of MaaS vehicles 1 (including information of ODDs), and prediction of the operation environment information of the operation zone of the plurality of MaaS vehicles 1. The information acquired by the information processing section 200 from the communication device 250 is stored as data 213 in the storage 210.

[0124] Next, the outline of the vehicle dispatch plan generation processing will be described with reference to Figure 8 The vehicle dispatch plan generation processing is constituted by the non-operable vehicle determination processing P210 and the plan update processing P220.

[0125] In the non-operable vehicle determination processing P210, processing of determining non-operable vehicles in the current vehicle dispatch plan based on the information of ODDs and the prediction of the operation environment information for each of the plurality of MaaS vehicles 1 is executed. That is, non-operable vehicles in the current vehicle dispatch plan are given as the processing result of the non-operable vehicle determination processing P210. This can be given, for example, by ID information given individually for each of the plurality of MaaS vehicles 1. Also, information of the place where the ODD will not be satisfied and the time when the ODD will not be satisfied for the determined non-operable vehicles can be included in the processing result of the non-operable vehicle determination processing P210.

[0126] In the plan update processing P220, processing of updating the vehicle dispatch plan according to the vehicle information or the operation information of the non-operable vehicles determined by the non-operable vehicle determination processing P210 is executed. By the plan update processing P220, for example, the update of the vehicle dispatch plan of the above-described first example, second example, or third example, or a combination thereof is performed. Then, the updated vehicle dispatch plan is given as the processing result of the vehicle dispatch plan generation processing.

[0127] Note that, when the vehicle dispatch plan generation processing is initially executed for the plurality of MaaS vehicles 1, in the plan update processing P220, a vehicle dispatch plan that becomes an initial value is first generated. The processing of the generation of the vehicle dispatch plan that becomes an initial value will be described later.

[0128] 2-2. Configuration of an autonomous vehicle

[0129] Figure 9is a block diagram showing a schematic configuration of an autonomous vehicle included in a plurality of MaaS vehicles 1. The autonomous vehicle is provided with an autonomous travel control device 100, a sensor 110, a travel control device 120, and a communication device 150. The autonomous travel control device 100 is configured to be able to mutually transmit and receive information with the sensor 110, the travel control device 120, and the communication device 150. Typically, electrical connection is made by a wire harness. As other configurations, connection by wireless communication, connection by an optical communication line, and the like are exemplified. Further, the communication device 150 is configured to be able to mutually transmit and receive information with the sensor 110.

[0130] The sensor 110 detects information on a driving environment of the autonomous vehicle, and outputs the detected information. The detected information output from the sensor 110 is transmitted to the autonomous travel control device 100 and the communication device 150. Typically, the sensor 110 includes a surrounding environment detection sensor that detects information on a surrounding environment (preceding vehicle, white line, obstacle, and the like) of the autonomous vehicle, and a travel state detection sensor that detects information on a travel state (vehicle speed, acceleration, yaw rate, and the like) of the autonomous vehicle. As the surrounding environment detection sensor, a camera, a millimeter wave radar, a LiDAR, and the like are exemplified. As the travel state detection sensor, a wheel speed sensor, a G sensor, a gyro sensor, and the like are exemplified. In addition to these, as the sensor 110, a GPS receiver that acquires a position by GPS is exemplified.

[0131] Note that each of the autonomous vehicles can be provided with different sensors 110 according to the specifications. For example, each of the autonomous vehicles can be appropriately provided with a night vision camera, an infrared thermal imager, a high-density LiDAR, and the like according to the specifications.

[0132] The autonomous travel control device 100 executes a process of autonomous travel based on acquired information, and outputs a control signal. The autonomous travel control device 100 acquires at least a vehicle dispatch plan generated in the server 2 from the communication device 150, and executes a process of autonomous travel according to the vehicle dispatch plan. The control signal output from the autonomous travel control device 100 is transmitted to the travel control device 120. Further, the autonomous travel control device 100 can also output information on a control state, operation information to the communication device 150 in conjunction with execution of the process of autonomous travel. Typically, the autonomous travel control device 100 is realized by an ECU (Electronic Control Unit).

[0133] The communication device 150 performs transmission and reception of various information by communicating with devices outside the autonomous vehicle. The communication device 150 communicates with the server 2 at least via the communication network 4, and receives a vehicle dispatch plan generated in the server 2. Also, the communication device 150 can also be configured to communicate with the server 2 or other MaaS vehicles 1 in the surroundings, and receive operation information of the other MaaS vehicles 1 so as to be able to detect approach and passing by as a preceding vehicle of the other MaaS vehicles 1 in the autonomous travel control device 100. Furthermore, the communication device 150 at least transmits information of the ODD of the autonomous vehicle to the server 2. In addition to this, as information received by the communication device 150, map information, road traffic information, and the like are exemplified. As information transmitted by the communication device 150, vehicle information, operation information, and the like are exemplified.

[0134] The travel control device 120 performs processing of travel control. The travel control device 120 performs travel control in accordance with a control signal acquired from the autonomous travel control device 100, thereby realizing autonomous travel of the autonomous vehicle. The travel control device 120 is realized by, for example, an ECU that controls the operation of a set of actuators equipped in the autonomous vehicle. As the set of actuators equipped in the autonomous vehicle, an actuator that drives a power device (an internal combustion engine, an electric motor, or the like), an actuator that drives a brake mechanism, an actuator that drives a steering mechanism, and the like are exemplified.

[0135] Next, the outline configuration of the processing performed by the autonomous travel control device 100 will be described with reference to Figure 10 The outline configuration of the processing performed by the autonomous travel control device 100 will be described with reference to

[0136] In the outside environment recognition processing P110, recognition of the outside environment of the surroundings of the autonomous vehicle is performed from the detection information of the sensor 110 and the map information. As the processing performed in the outside environment recognition processing P110, self-position estimation, integration of the detection information and the self-position estimation result with the map information, and the like are exemplified. The outside environment recognition processing P110 can employ a preferable known technique.

[0137] In the ODD departure sensing processing P120, it is determined whether or not the ODD is satisfied from the detection information of the sensor 110 and the map information. Also, the recognition result obtained by the outside environment recognition processing P110 can be used in the determination of whether or not the ODD is satisfied.

[0138] In the travel plan processing P130, travel plan such as judgment of travel action, generation of autonomous travel path is performed based on the recognition result of the outside environment recognition processing P110. Here, in the travel plan processing P130, either of normal class processing P131 or abnormal class processing P132 is executed according to the sensing result of the ODD departure sensing processing P120. The normal class processing P131 is processing executed when it is assumed that the sensing result of the ODD departure sensing processing P120 satisfies the ODD, and in the normal class processing P131, processing for continuing autonomous travel in accordance with the vehicle dispatch plan is executed. The abnormal class processing P132 is processing executed when it is assumed that the sensing result of the ODD departure sensing processing P120 does not satisfy the ODD, and in the abnormal class processing P132, processing for performing travel in accordance with the travel plan in the case where the ODD is not satisfied is executed. In the abnormal class processing P132, processing of the action described in the above is performed. In the abnormal class processing P132, approach of another MaaS vehicle 1 and detection as a preceding vehicle are detected based on the operation information of the other MaaS vehicle 1 acquired as communication information. Figure 3

[0139] In the control amount calculation processing P140, a control signal giving a control amount (for example, a control amount of acceleration, braking, and steering) of travel control is generated based on the processing result of the travel plan processing P130. Typically, an autonomous travel path is given as the processing result of the travel plan processing P130, and a control signal giving a control amount in a manner to travel along the autonomous travel path is generated. The generated control signal is output from the autonomous travel control device 100.

[0140] 3. Processing

[0141] Hereinafter, the processing executed in the server (vehicle dispatch plan device) 2 and the autonomous vehicle in the vehicle dispatch system 10 of the present embodiment will be described.

[0142] 3-1. Vehicle dispatch plan generation processing

[0143] Figure 11 is a flowchart showing the vehicle dispatch plan generation processing executed in the server (vehicle dispatch plan device) 2 by the processor 220. Figure 11 The processing of the flowchart shown in FIG. 10 is repeatedly executed at every prescribed processing cycle.

[0144] In step S100, the processor 220 acquires a prediction of the operation environment information at each time in the period until the future prescribed period.

[0145] After step S100, step S110 is entered. Here, the processing of step S110 to step S140 is executed for each of the autonomous vehicles included in the plurality of MaaS vehicles 1 (i = 1, 2,..., N, N is the total number of autonomous vehicles).​

[0146] In step S110, the processor 220 acquires information of the autonomous driving vehicle as the object (vehicle information, operation information). Here, information of the ODD of the autonomous driving vehicle as the object is acquired.

[0147] After step S110, the process proceeds to step S120.

[0148] In step S120, the processor 220 determines whether the autonomous driving vehicle as the object will not satisfy the ODD at each time in the period until a prescribed period in the future based on the prediction of the operation environment information and the ODD of the autonomous driving vehicle as the object.

[0149] In the case where the ODD will not be satisfied (step S130: Yes), the autonomous driving vehicle as the object is set as the non-operable vehicle (step S140).

[0150] After the execution of the process of steps S110 to S140 is completed for each of the autonomous driving vehicles (i = 1, 2,..., N), the process proceeds to step S150.

[0151] In step S150, the processor 220 determines whether there is an autonomous driving vehicle set as the non-operable vehicle through the execution of the process of steps S110 to S140.

[0152] In the case where there is an autonomous driving vehicle set as the non-operable vehicle (step S150: Yes), the process proceeds to step S160. In the case where there is no autonomous driving vehicle set as the non-operable vehicle (step S150: No), the vehicle dispatch plan is not updated and the present process ends.

[0153] In step S160 (plan update process), the processor 220 updates the vehicle dispatch plan according to the vehicle information or the operation information of the non-operable vehicle. The update of the vehicle dispatch plan is performed, for example, in the update of the vehicle dispatch plan of the first example, the second example, or the third example described above or a combination thereof.

[0154] After step S160, the present process ends.

[0155] Thus, the vehicle dispatch plan generation process is executed by the processor 220 in the server (vehicle dispatch plan device) 2. Further, thus, the vehicle dispatch management method of the plurality of MaaS vehicles operated based on the vehicle dispatch plan is realized by the vehicle dispatch system 10 of the present embodiment.

[0156] 3-2. Generation of vehicle dispatch plan to be initial value

[0157] The server (vehicle dispatch plan device) 2, when initially performing the vehicle dispatch plan generation processing on the plurality of MaaS vehicles 1, first, generates a vehicle dispatch plan that becomes an initial value and outputs. Here, the vehicle dispatch plan that becomes the initial value can be generated based on a prediction of the operation environment information, the vehicle information, or the operation information of the plurality of MaaS vehicles 1. Hereinafter, the processing of generating the vehicle dispatch plan that becomes the initial value based on the prediction of the operation environment information, the vehicle information, or the operation information of the plurality of MaaS vehicles 1 will be described.

[0158] Now, it is assumed that the plurality of MaaS vehicles 1 include: upper-level vehicles that are manual driving vehicles or automatic driving vehicles that are high in specifications and can satisfy the ODD in the operation zone; and lower-level vehicles that are automatic driving vehicles that will sometimes not satisfy the ODD depending on the situation of the operation zone. Here, generally, the upper-level vehicles are high in cost, and the lower-level vehicles are low in cost. Therefore, it is desirable to generate a vehicle dispatch plan that utilizes the upper-level vehicles to the minimum necessary. Therefore, at the time of generation of the vehicle dispatch plan that becomes the initial value, first, the upper-level vehicles are assigned operations that are high in importance. For example, the upper-level vehicles are assigned operations of emergency transport and passenger transport. Then, the lower-level vehicles are assigned other operations.

[0159] The above is assumed as a premise, and the processing of generating the vehicle dispatch plan that becomes the initial value will be described with reference to Figure 12 The processing of generating the vehicle dispatch plan that becomes the initial value will be described. Here, it is assumed that the automatic driving vehicles included in the plurality of MaaS vehicles 1 are configured to implement the action described in Figure 3 in the server (vehicle dispatch plan device) 2 by the processor 220. The processing of the flowchart shown in Figure 12 is executed.

[0160] In step S200, the processor 220 generates a vehicle dispatch plan for the upper-level vehicles.

[0161] After step S200, the processing proceeds to step S210.

[0162] In step S210, the processor 220 lists path candidates for the lower-level vehicles from the departure place, the destination, and the map information of the lower-level vehicles.

[0163] After step S210, the processing proceeds to step S220.

[0164] In step S220, the processor 220 extracts a place where the ODD will not be satisfied for each of the path candidates listed in step S210. Here, the extracted place includes not only a place where the ODD will not be satisfied due to a static factor (for example, information given by map information) but also a place predicted to be a place where the ODD will not be satisfied based on a prediction of the running environment information.

[0165] After step S220, the process proceeds to step S230.

[0166] In step S230, the processor 220 determines whether there is a path among the path candidates listed in step S210 that does not have a place where the ODD will not be satisfied extracted in step S220.

[0167] In a case where there is a path that does not have a place where the ODD will not be satisfied (step S230: Yes), the processor 220 generates a vehicle dispatch plan in such a way that the path is taken as the vehicle dispatch route of the lower-level vehicle (step S270), and the process ends. Note that, in a case where there are a plurality of paths that do not have a place where the ODD will not be satisfied, the best path in terms of the shortest travel distance, travel time, or the like can be taken as the vehicle dispatch route.

[0168] In a case where there is no path that does not have a place where the ODD will not be satisfied (step S230: No), the process proceeds to step S240.

[0169] In step S240, the processor 220 removes a place where the ODD will not be satisfied extracted in step S220 that is present in the vehicle dispatch route of the upper-level vehicle. This is because, if a place where the ODD will not be satisfied is present in the vehicle dispatch route of the upper-level vehicle, it is expected that the lower-level vehicle will be able to escape from the place where the ODD will not be satisfied by following the upper-level vehicle at the place.

[0170] After step S240, the process proceeds to step S250.

[0171] In step S250, the processor 220 again determines whether there is a path among the path candidates listed in step S210 that does not have a place where the ODD will not be satisfied extracted.

[0172] In a case where there is a path that does not have a place where the ODD will not be satisfied (step S250: Yes), the processor 220 generates a vehicle dispatch plan in such a way that the path is taken as the vehicle dispatch route of the lower-level vehicle (step S270), and the process ends.

[0173] In the case where there is no path to a place where the ODD will not be satisfied (step S250: No), the processor 220 generates a vehicle dispatch plan by replacing the lower-level vehicle with a vehicle that can travel on any path (step S260). For example, the lower-level vehicle is replaced with an upper-level vehicle. Alternatively, the lower-level vehicle is replaced with an autonomous vehicle that can satisfy the ODD according to the situation that occurs. After step S260, the process ends.

[0174] By generating the vehicle dispatch plan that becomes the initial value as explained above, a vehicle dispatch plan can be given in which the prediction of the operation environment information, the vehicle information, or the operation information of the plurality of MaaS vehicles 1 is considered in the vehicle dispatch plan that becomes the initial value. Furthermore, a vehicle dispatch plan can be given in which the upper-level vehicle is utilized to the necessary minimum limit, and thus a low-cost vehicle dispatch plan can be generated. Furthermore, optimization of the cost can be sought in the vehicle dispatch plan that is updated later.

[0175] Note that, in addition to the generation of the vehicle dispatch plan that becomes the initial value explained in Figure 12 above, a scheme in which the following modifications are made can also be adopted.

[0176] One is a case where the lower-level vehicle includes a plurality of autonomous vehicles of different specifications. In this case, in step S220, the extraction of the place where the ODD will not be satisfied can also be performed for each specification. Furthermore, in step S230 or step S250, the determination of whether there is a path to a place where the ODD will not be satisfied can also be performed for each specification. Then, in step S230 or step S250, in the case where it is determined that there is a path to a place where the ODD will not be satisfied (step S230: Yes or step S250: Yes), the vehicle dispatch plan can be generated in such a manner that an autonomous vehicle of the specification with the lowest allocation cost is assigned. A lower-cost vehicle dispatch plan can be generated by adopting a scheme in which such modifications are made.

[0177] Another is for a case where there is no path that does not satisfy the ODD in step S250 (step S250: No). In this case, the following processing can be executed instead of or in addition to the processing of step S260. First, a location that needs to be removed from the locations that will not satisfy the ODD is extracted. The location that needs to be removed is, for example, a location where there is a path that does not satisfy the ODD if the location is removed. Next, a vehicle in the upper-level vehicle that can pass through the location that needs to be removed is selected, and a vehicle whose required cost (required time, presence or absence of a passenger, and the like) is the smallest. Here, as to whether the location that needs to be removed can be passed through, for example, for a vehicle that performs emergency transport, whether the target time can be caught up with when passing through becomes an index, and for a passenger-carrying vehicle, whether the location can be passed through in an interval where there is no passenger becomes an index. Then, the selected upper-level vehicle is corrected to a vehicle dispatch route that passes through the location that needs to be removed to generate a vehicle dispatch plan. By adopting such a modified scheme, the frequency of replacing a lower-level vehicle with a more costly vehicle can be reduced. Furthermore, reduction of cost can be sought.

[0178] 3-3. Processing performed by the autonomous travel control device

[0179] Figure 13 is a flowchart showing processing performed by the autonomous travel control device 100 in the automated vehicle. Figure 13 The processing of the flowchart shown is repeatedly performed at each prescribed processing cycle.

[0180] In step S300, the autonomous travel control device 100 performs the outside environment recognition processing P110.

[0181] After step S300, the processing proceeds to step S310.

[0182] In step S310, the autonomous travel control device 100 performs the ODD departure sensing processing P120.

[0183] When it is assumed that the sensing result of the ODD departure sensing processing P120 satisfies the ODD (step S320: Yes), the processing proceeds to step S330 (normal class processing P131). When it is assumed that the sensing result of the ODD departure sensing processing P120 does not satisfy the ODD (step S320: No), the processing proceeds to step S341 and step S342 (abnormal class processing P132).

[0184] In step S330, the autonomous travel control device 100 generates a travel plan in a manner that the vehicle continues autonomous travel in accordance with the vehicle dispatch plan based on the recognition result of the outside environment recognition processing P110.

[0185] In step S341, the autonomous travel control device 100 executes processing of causing the automated driving vehicle to safely stop. After the automated driving vehicle stops, waiting is performed until the recovery vehicle passes.

[0186] Then, processing of causing the automated driving vehicle to follow the recovery vehicle in response to the recovery vehicle passing is executed (step S342). Note that when the ODD is satisfied by the follow-up travel (step S320: YES), the processing of step S330 is executed, and autonomous travel is restarted again.

[0187] 4. Effects

[0188] As explained above, according to the present embodiment, the non-operable vehicle in the current vehicle dispatch plan is determined based on the prediction of the ODD and the operation environment information for the automated driving vehicles included in the plurality of MaaS vehicles 1. Then, processing of updating the vehicle dispatch plan according to the vehicle information or the operation information of the non-operable vehicle is executed. Thereby, it is possible to cope with the non-operable vehicle predicted in the current vehicle dispatch plan until the future of the prescribed period, and it is possible to reduce the frequency of recovery of the automated driving vehicle, the effort of recovery. Further, it is possible to seek reduction of the cost of operation.

[0189] Note that the vehicle dispatch system 10 of the present embodiment can also be configured to limit the object of performing the generation and the update of the vehicle dispatch plan to the recovery vehicle that performs the recovery of the automated driving vehicle, with reference to the first example of the update of the vehicle dispatch plan. That is, in this case, the vehicle dispatched by the vehicle dispatch system 10 is the recovery vehicle, and in addition, the automated driving vehicle that is the object of recovery can not be the vehicle dispatched by the vehicle dispatch system 10. In particular, the automated driving vehicle that is the object of recovery is not limited to the vehicle that performs the prescribed operation such as a bus, a taxi, and the like, and can be a vehicle that goes to an arbitrary set destination. However, the server 2 needs to be configured to be able to acquire the vehicle information (information including the ODD), the operation information from the automated driving vehicle as explained above. When so configured, the server 2 can also be referred to as a "recovery plan device".

Claims

1. A vehicle dispatch system, characterized by, Comprising: a plurality of ride service vehicles; at least one processor; a program memory coupled to the at least one processor, storing a plurality of executable instructions, the plurality of executable instructions configured to cause the at least one processor to execute a vehicle dispatch plan generation process that generates a vehicle dispatch plan for the plurality of ride service vehicles, the plurality of ride service vehicles including one or more autonomous vehicles that autonomously travel according to the vehicle dispatch plan, the vehicle dispatch plan generation process including the following processes: acquiring a running design domain for the one or more autonomous vehicles, the running design domain including a road shape in which autonomous travel can continue according to the vehicle dispatch plan; acquiring, for a running zone of the one or more autonomous vehicles, a prediction of a running environment information for a period up to a prescribed period in the future; determining, based on the running design domain and the prediction of the running environment information, an inoperable vehicle among the one or more autonomous vehicles that is predicted to not satisfy the running design domain in the current vehicle dispatch plan; and a plan update process that updates the vehicle dispatch plan according to vehicle information or running information of the inoperable vehicle, the one or more autonomous vehicles configured to implement the following steps: stopping when the running design domain is not satisfied; after stopping, following a preceding vehicle in the plurality of ride service vehicles in response to the preceding vehicle passing by; and resuming the autonomous travel when the running design domain is satisfied by the following.

2. The vehicle dispatch system according to claim 1, wherein the plan update process includes updating the vehicle dispatch plan by selecting one or more recovery vehicles from the plurality of ride service vehicles according to the vehicle information or running information, and causing the one or more recovery vehicles to pass through a location where the inoperable vehicle is predicted to not satisfy the running design domain as the preceding vehicle.

3. The vehicle dispatch system according to claim 2, wherein the one or more recovery vehicles are manually driven vehicles or autonomous vehicles that satisfy the running design domain in the updated vehicle dispatch plan.

4. The vehicle dispatch system according to any one of claims 1 to 3, wherein the plan update process includes updating the vehicle dispatch plan by replacing the inoperable vehicle before vehicle dispatch with a manually driven vehicle or an autonomous vehicle that satisfies the running design domain according to the vehicle information or running information.

5. The vehicle dispatch system according to any one of claims 1 to 3, wherein the plan update process includes updating the vehicle dispatch plan by replacing the inoperable vehicle before vehicle dispatch with a manually driven vehicle or an autonomous vehicle that satisfies the running design domain according to the vehicle information or running information. ​ ​ ​ updating the vehicle dispatch plan in such a manner that at least one of the non-operable vehicles is replaced with a manually driven vehicle or an autonomous vehicle that satisfies the operational design domain.

6. A vehicle dispatch management method for a plurality of ride-hailing vehicles that are operated based on a vehicle dispatch plan, the vehicle dispatch management method comprising: the plurality of ride-hailing vehicles including one or more autonomous vehicles that autonomously travel according to the vehicle dispatch plan; obtaining an operational design domain for the one or more autonomous vehicles, the operational design domain including a road shape along which the one or more autonomous vehicles can continue to autonomously travel according to the vehicle dispatch plan; obtaining, for an operational zone of the one or more autonomous vehicles, a prediction of an operational environment information for a period until a future prescribed period; determining, based on the operational design domain and the prediction of the operational environment information, a non-operable vehicle of the one or more autonomous vehicles that is predicted to not satisfy the operational design domain in the current vehicle dispatch plan; and updating the vehicle dispatch plan based on vehicle information or operational information of the non-operable vehicle.

7. The vehicle dispatch management method according to claim 6, wherein the updating of the vehicle dispatch plan includes, based on the vehicle information or operational information, updating the vehicle dispatch plan in such a manner that one or more recovery vehicles are selected from the plurality of ride-hailing vehicles, and the one or more recovery vehicles pass through a location where the non-operable vehicle is predicted to not satisfy the operational design domain as a preceding vehicle.

8. The vehicle dispatch management method according to claim 6 or 7, wherein the updating of the vehicle dispatch plan includes, based on the vehicle information or operational information, updating the vehicle dispatch plan in such a manner that the non-operable vehicle before vehicle dispatch is replaced with a manually driven vehicle or an autonomous vehicle that satisfies the operational design domain.

9. The vehicle dispatch management method according to claim 6, wherein the one or more autonomous vehicles are configured to: stop when becoming not to satisfy the operational design domain; perform follow-up travel with respect to a preceding vehicle of the plurality of ride-hailing vehicles after stopping; and resume the autonomous travel when the follow-up travel satisfies the operational design domain, and the updating of the vehicle dispatch plan includes, based on the vehicle information or operational information, updating the vehicle dispatch plan in such a manner that one or more recovery vehicles are selected from the plurality of ride-hailing vehicles, and the one or more recovery vehicles pass through a location where the non-operable vehicle is predicted to not satisfy the operational design domain as the preceding vehicle. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ updating the vehicle dispatch plan in such a way that, for each of the set of non-operable vehicles predicted to be at a location that will not satisfy the operational design domain in the current vehicle dispatch plan, at least one of the non-operable vehicles is replaced with a manually driven vehicle or an autonomous vehicle that satisfies the operational design domain.

Images