Method for dispatching autonomous fleet and fleet dispatching system
By using the first regional environmental archives and other regional environmental archives in the fleet scheduling system in combination with regional events and personnel archives for scheduling optimization, the problem of low scheduling efficiency of independent fleets is solved, and more efficient vehicle arrival and user satisfaction is achieved.
Patent Information
- Application Number
- CN202080048512.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-07-08
- Filing Date
- 2020-07-01
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2040-07-01
AI Technical Summary
The prior art is difficult to effectively dispatch autonomous motor vehicles, especially when predicting passenger demands in a complex manner, resulting in low efficiency and high cost in vehicle parking time and driving route planning.
By using the first regional environmental archives in the fleet scheduling system, generating environmental archives in other regions, combining regional events and personnel archives for scheduling optimization, and adjusting scheduling strategies using electronic computing devices and machine learning to ensure that the autonomous fleet can accurately reach the activity location. And improve user satisfaction.
More effective scheduling of the independent fleet is achieved, reducing vehicle parking time and battery charging time, and improving the overall efficiency and user satisfaction of the fleet.
Smart Images

Figure CN114096805B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for dispatching an autonomous vehicle fleet having a plurality of autonomous motor vehicles in a first predetermined area by means of a vehicle fleet dispatching system according to the preamble of claim 1. The invention also relates to a vehicle fleet dispatching system. Background Art
[0002] It is known that autonomous vehicles, in particular so-called autonomous shuttles for “carpooling” or “private rides”, are initially available in only a small number in some cities. This is due in particular to the complexity of the vehicles, but also to the high costs of the corresponding sensor devices and other technical equipment. This means in particular that the vehicles should only have short parking times and should generally have as many passengers as possible on the way. However, the planning of the charging strategy for the vehicles and the driving routes must also be adjusted accordingly.
[0003] For this purpose, it is known from the prior art to be able to predict the daily use of many passengers, when and where there will be a high demand. For example, it can be determined that at certain times, especially in the afternoon, for example in large factories or large enterprises, many people who may want to use autonomous shuttles can be found. However, beyond the specified period, the prediction becomes complicated.
[0004] US 2018 / 0096606 A1 discloses a method for providing a deployment service for on-demand transportation. The method comprises: determining that a predictive dispatch message should be transmitted to a motor vehicle, generating a predictive dispatch message in response to “determining that a predictive dispatch message should be transmitted to a motor vehicle”, and transmitting the predictive dispatch message to the vehicle. For the generation of the predictive dispatch message, one or more prediction models calculated from historical and real-time on-demand transportation service data are used. Summary of the invention
[0005] The object of the present invention is to provide a method and a fleet dispatching system, by means of which autonomous motor vehicles can be dispatched better.
[0006] This object is achieved by a method and a fleet dispatching system according to the independent claims. Advantageous embodiments are described in the dependent claims.
[0007] One aspect of the present invention relates to a method for dispatching an autonomous vehicle fleet having a plurality of autonomous motor vehicles in a predetermined first area with the aid of a vehicle fleet dispatching system, wherein the autonomous vehicle fleet is dispatched based on a first area environment profile of the first area for the vehicle fleet dispatching system, and wherein, for the dispatching, corresponding autonomous motor vehicles among the autonomous motor vehicles are autonomously driven to corresponding locations in the first area associated with the area environment profile.
[0008] It is provided that the first regional environment file is generated based on the second regional environment file stored in the electronic computing device of the fleet dispatching system, and the second regional environment file is generated for a second region independent of the first region.
[0009] This allows for improved maneuvering of autonomous vehicles that can execute autonomous fleets.
[0010] This is particularly in view of the following background, that is, for example, similar activities take place in the second region as in the first region. The corresponding city can be regarded as the corresponding region. For example, it can be provided that the city of Berlin can be regarded as the second region and the city of Stuttgart can be regarded as the first region. It is therefore provided that, based on the environmental profile generated for Berlin, an environmental profile for the city of Stuttgart is now generated. In particular, when an autonomous vehicle fleet is to be deployed in Stuttgart, for example, corresponding characteristics are set based on the environmental profile that has been generated for Berlin, whereby driving according to the regional environmental profile is possible. In particular, the environmental profile is characterized by the corresponding activities in the region. In particular, the locations of the activities are then reached by the corresponding autonomous motor vehicles accordingly.
[0011] Thus, the invention solves the problem that there are many events in a first region, for example, every day at different locations, in other words, locations, but only some of these events result in a large number of driving requests for the autonomous fleet. The issuance of driving requests is essentially related to the location and time of the event, but also mainly to the corresponding persons. For example, after a pop concert, the probability of using the autonomous fleet is higher than that of a classical concert because the audience is of mixed ages.
[0012] According to an advantageous embodiment, the first regional environment profile is generated based on a plurality of stored other regional environment profiles, which are generated for a plurality of other regions that are independent of the first region. In other words, the regional environment profile is generated based on a plurality of other regional environment profiles that have already been generated. This allows the use of a plurality of already generated environment profiles, thereby already generating very detailed and specific regional environment profiles. This allows the reliable dispatching of autonomous vehicles already at the start of the autonomous fleet service undertaking.
[0013] In another advantageous embodiment, the first regional environment file is adjusted according to regional events in the first region. In other words, it is provided that the first regional environment file is generated based on at least the second regional environment file. Over time, the first regional environment file is then adjusted accordingly according to regional events in the first region. In particular, it can be provided that event-related information can be obtained, for example, from the Internet. Regional events can be activities in particular, which are generally integrated and provided, for example, on the Internet, such as in an event calendar, and can then be searched by an electronic computing device. When the autonomous vehicle fleet is then to be deployed in the first region, the data of the second regional environment file can first be used, or alternatively, the corresponding attributes can also be set manually, and the activities, in other words, regional events are marked with these attributes, and then the autonomous motor vehicle can be pre-planned to the vicinity of the event location, in other words, the location, based on these attributes. Through the actually arriving driving commissions and the locations adapted thereto, the actual characteristics of the activities can then be learned for the first region. Typical users of the shuttle are therefore also assigned to activities that are typical for them. This allows a first environment profile to be created precisely, which leads to improved autonomous fleet scheduling.
[0014] It has also proven to be advantageous if the first regional environment profile is adapted in time and / or location to regional events. In other words, in particular a spatiotemporal adaptation of the regional environment profile occurs. In particular, this allows an activity-related characterization to be performed for the first region. Thus, the first regional environment profile is adapted in time and / or location to events occurring in the first region, whereby the first regional environment profile can then be adapted. In particular, this allows the dispatching of the autonomous vehicle fleet to be performed in an adaptive manner to the first region.
[0015] In another advantageous embodiment, the first regional environment profile is adjusted according to the corresponding duration of the regional event. In other words, it can be provided that the typical activity duration of the event is "learned" by means of an electronic computing device. For example, the event generally lasts two hours. The user of the autonomous motor vehicle therefore issues his driving request two hours after the start of the event and wants to be picked up as quickly as possible. Therefore, by using the duration accordingly, the first regional environment profile can be accurately generated. This leads to improved autonomous vehicle fleet scheduling within the first region.
[0016] In another advantageous embodiment, a corresponding personnel profile is generated for the corresponding regional event by means of an electronic computing device, and the first environment profile is adjusted according to the generated personnel profile. In particular, the personnel profile can take into account, for example, the average age of the personnel at the event. It can also be provided, for example, that an event, such as a football match, is evaluated using the personnel profile as follows, that is, after the football match, people generally want to take a bus to the main railway station in the corresponding first region. This can then be taken into account in order to enable the dispatching of the autonomous vehicle fleet.
[0017] It has also proven to be advantageous if the route planning of the respective motor vehicles and / or the battery charging planning of the respective motor vehicles and / or the parking time planning of the respective motor vehicles are carried out as a schedule by means of an electronic computing device. In particular, better autonomous fleet scheduling requires the execution of corresponding route planning and / or battery charging planning and / or parking time planning. Thus, for example, the parking time of the autonomous motor vehicles can be shortened. Thus, better autonomous fleet scheduling can be achieved by this consideration or corresponding planning.
[0018] According to another advantageous embodiment, a requested trip of a person is rejected based on the first environmental profile generated. In particular, it can be provided that when planning the corresponding trip, trips can also be rejected if, for example, the required fleet size cannot be achieved at the event location at the specified time. As an alternative or in addition, a wider area can be created for picking up and dropping off customers in the case of "carpooling" so that only a small number of motor vehicles are required. This makes it possible to achieve improved autonomous fleet scheduling.
[0019] It is also advantageous to adjust the first environment profile by means of machine learning of the electronic computing device, in particular by means of a neural network. In other words, it is provided that the electronic computing device learns and actively adjusts the first environment profile in real time through the learning process. This allows the first regional environment profile to be adapted to current events in the first region in the future as well, thus allowing better fleet scheduling.
[0020] Another aspect of the present invention relates to a fleet dispatching system for dispatching an autonomous fleet of vehicles having a plurality of autonomous motor vehicles within a predetermined first region, the fleet dispatching system having at least one electronic computing device, wherein the fleet dispatching system is designed to execute the method according to the above-mentioned aspect. In particular, the method is executed by means of the fleet dispatching system.
[0021] An advantageous embodiment of the method is regarded as an advantageous embodiment of a fleet dispatching system. The fleet dispatching system has subject features which allow the method or an advantageous embodiment thereof to be carried out.
[0022] Further advantages, features and details of the invention are apparent from the following description of preferred embodiments and in conjunction with the figures. The features and feature combinations mentioned above in the description and the features and feature combinations mentioned below in the description of the figures and / or shown individually in the single figure may be used not only in the respectively indicated combination, but also in other combinations or individually without exceeding the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Here, the single figure shows a schematic side view of an embodiment of a fleet dispatching system.
[0024] In the figures, identical or functionally identical components are provided with the same reference symbols. DETAILED DESCRIPTION
[0025] FIG. 1 shows a schematic side view of an embodiment of a vehicle fleet dispatching system 10. The vehicle fleet dispatching system 10 is designed to dispatch an autonomous vehicle fleet 12 having a plurality of autonomous motor vehicles 14 in a predetermined first region 16. The vehicle fleet dispatching system 10 has at least one electronic computing device 18. It can also be provided, for example, that the respective autonomous motor vehicle 14 has a respective storage battery 20, which can also be referred to as a battery. In other words, it can be provided that the autonomous motor vehicle 14 is an at least partially electric motor vehicle 14, in particular a fully electric autonomous motor vehicle 14.
[0026] In a method for dispatching an autonomous vehicle fleet 12 having a plurality of autonomous motor vehicles 14, in particular three autonomous motor vehicles 14, the autonomous vehicle fleet 12 is dispatched within a predetermined first area 16 by means of a vehicle fleet dispatching system 10 based on a first area environment profile 22 for the first area 16 of the vehicle fleet dispatching system 10, wherein for the dispatching, corresponding autonomous motor vehicles 14 of the autonomous motor vehicles 14 autonomously drive to a corresponding position P within the first area 16 associated with the area environment profile 22.
[0027] It is provided that the first regional environment file 22 is generated based on the second regional environment file 24 stored in the electronic computing device 18 of the fleet dispatching system 10 , and the second regional environment file is generated for a second region that is independent of the first region 16 .
[0028] The first region is in particular a first city, and the second region is in particular a second city that is independent of the first city. In addition, it is obvious that more than three autonomous motor vehicles 14 can also be dispatched.
[0029] In particular, it may be provided that the first regional environment profile 22 is generated on the basis of a plurality of stored further regional environment profiles, which were generated for a plurality of further regions that are independent of the first region 16 .
[0030] In particular, it can also be provided that the first regional environment profile 22 is adjusted as a function of regional events 26 in the first region 16. Regional events 26 are in particular activities in the first region 16. Regional events 26 can be determined, for example, by evaluating an event calendar for the first region 16.
[0031] It can also be provided that the first regional environment profile 22 is adjusted in time and / or location as a function of the regional event 26. In particular, a spatiotemporal adjustment of the first regional environment profile 22 can thus be carried out.
[0032] Provision can also be made for the first regional environment profile 22 to be adjusted as a function of the respective duration 28 of the regional event 26 .
[0033] It can also be provided that, for the corresponding regional event 26 , a corresponding personnel file 30 is generated by the electronic computing device 18 , and the first regional environment file 22 is adjusted according to the generated personnel file 30 .
[0034] In particular, it may be provided that route planning of the respective autonomous motor vehicle 14 and / or battery charging planning of the respective autonomous motor vehicle 14 , in particular battery charging planning of the respective battery 20 and / or parking time planning of the respective autonomous motor vehicle 14 are carried out as scheduling items by means of the electronic computing device 18 .
[0035] In particular, it can also be provided that a requested trip of a person is rejected based on the generated first regional environment profile 22 .
[0036] Furthermore, it may be provided that the first regional environment profile 22 is adapted by means of machine learning, in particular a neural network 32 , of the electronic computing device 18 .
[0037] The present invention makes use of the fact that in a city, which corresponds here to the first region 16, many events 26, which can also be called events, take place every day at different locations, especially at different positions P, but only some of these events 26 result in a large number of driving requests for the autonomous vehicle fleet 12. They are related to the location and time of the event 26, but mainly also to the people. For example, after a pop concert, more driving requests for the autonomous vehicle fleet 12 will probably occur compared to a classical concert, because the audience is of a more diverse age group.
[0038] The events can generally be summarized and provided on the Internet and can be searched by, for example, an electronic computing device 18. When an autonomous fleet 12 is to be deployed in a new city, here the first region 16, data based on another city, namely the second region and the second region environment file 24 to which it belongs is used. Alternatively, manual data entry can also be performed. The corresponding events 26 are then marked, and based on these attributes, the autonomous motor vehicles 14 can then be guided to the vicinity of the event location in a pre-planned manner. Through the orders that actually arrive and the pick-up locations adapted thereto, the actual characteristics related to the event can then be learned for the city and the typical users of the autonomous fleet 12 can therefore also be assigned to the typical events for them. In addition, the typical event duration can also be known / learned. For example, a certain play has a duration of two hours 28, so the electronic computing device 18 knows that the corresponding driving commission will arrive two hours after the start of the event, and they should be processed as quickly as possible. In addition, it can then be known through the driving destination where the typical customers of such an event want to take the bus. For example, it can be stipulated that after a football match, the corresponding people usually want to take the bus to the main railway station of the first region 16. This understanding can in turn be used for further scheduling of the autonomous vehicle fleet 12 .
[0039] Autonomous vehicle fleet 12 may then be positioned based on the personal preferences of typical autonomous vehicle users so that trips are pre-planned and planning regarding battery charging and required parking times for vehicles 14 is improved.
[0040] In this case, it is also possible to reject trips during trip planning that would result in the required fleet size not being available at the event location, ie event 26, at the specified time. Alternatively or additionally, a wider area for picking up and dropping off customers can be created in the case of "carpooling", so that fewer autonomous vehicles 14 need to be involved.
[0041] In summary, the present invention demonstrates the automatic positioning of an autonomous vehicle 14 .
Claims
1. A method for dispatching an autonomous vehicle fleet (12) having a plurality of autonomous motor vehicles (14) in a predetermined first area (16) by means of a vehicle fleet dispatching system (10), wherein: The autonomous vehicle fleet (12) is dispatched according to a first regional environment file (22) of a first region (16) for the vehicle fleet dispatching system (10), wherein for the dispatching, a corresponding autonomous motor vehicle (14) among the plurality of autonomous motor vehicles (14) autonomously drives to a corresponding position (P) in the first region (16) associated with the regional environment file (22), The invention is characterized in that the first regional environment file (22) is generated based on a second regional environment file (24) stored in an electronic computing device (18) of the fleet dispatching system (10), and the second regional environment file is generated for a second region that is independent of the first region (16). wherein a route planning of the corresponding autonomous motor vehicle (14) and / or a battery charging planning of the corresponding autonomous motor vehicle (14) and / or a parking time planning of the corresponding autonomous motor vehicle (14) is performed as a schedule with the aid of an electronic computing device (18), and a requested trip of a person is rejected based on a generated first regional environment profile (22), The first region (16) is a first city, the second region is a second city unrelated to the first city, the first region environment file (22) is adjusted according to a regional event (26) in the first region (16), wherein actual characteristics related to the regional event (26) are learned for the first region through actually arriving driving requests and positions adapted thereto, and thus typical users of the autonomous vehicle fleet (12) are assigned to corresponding regional events.
2. The method according to claim 1, characterized in that: The first region environment file (22) is generated based on a plurality of stored other region environment files, wherein the other region environment files are generated for a plurality of other regions that are independent of the first region (16).
3. The method according to claim 1 or 2, characterized in that: The first regional environment profile (22) is adjusted in terms of time and / or location according to the regional event (26).
4. The method according to claim 1 or 2, characterized in that: The first regional environment profile (22) is adjusted according to the corresponding duration (28) of the regional event (26).
5. The method according to claim 1 or 2, characterized in that: A corresponding personnel file (30) is generated for a corresponding regional event (26) by means of an electronic computing device (18), and the first regional environment file (22) is adjusted according to the generated personnel file (30).
6. The method according to claim 1 or 2, characterized in that: The first regional environment profile (22) is adjusted by means of machine learning of an electronic computing device (18).
7. The method according to claim 6, characterized in that: The first regional environment profile (22) is adjusted with the aid of a neural network (32).
8. A fleet dispatching system (10) for dispatching an autonomous fleet (12) having a plurality of autonomous motor vehicles (14) within a predetermined first area (16), comprising at least one electronic computing device (18), wherein: The fleet dispatching system (10) is designed to execute the method according to any one of claims 1 to 7.
