METHOD FOR THE COORDINATION OF AN AUTONOMOUS VEHICLE FLEET, AND VEHICLE FLEET COORDINATION SYSTEM
Patent Information
- Application Number
- DE502020009695
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2019-07-08
- Filing Date
- 2020-07-01
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2040-07-01
AI Technical Summary
Autonomous vehicle fleets face challenges in efficient coordination due to high costs and complexity of sensors, leading to underutilization and difficulty in predicting passenger demand outside regular hours, which complicates route and charging planning.
A vehicle fleet coordination system that generates a regional environmental profile based on similar regions, using historical and real-time data to optimize route planning, battery charging, and idle time management, while adapting to spatiotemporal and event-specific factors, including machine learning for real-time adjustments.
This approach enables improved coordination of autonomous vehicles by reducing idle times, optimizing fleet deployment, and ensuring adequate service during peak events, thereby enhancing operational efficiency and passenger satisfaction.
Description
[0001] The invention relates to a method for coordinating an autonomous vehicle fleet comprising a plurality of autonomous motor vehicles in a predetermined first region by means of a vehicle fleet coordination system according to the preamble of patent claim 1. Furthermore, the invention relates to a vehicle fleet coordination system.
[0002] It is known that autonomous vehicles, especially so-called autonomous shuttles for ride-sharing or ride-hailing, are only available in small numbers at the beginning of their deployment in certain cities. This is due, among other things, to the complexity of the vehicles, but also to the high costs of the respective sensors and other technology. This means, in particular, that these vehicles should have minimal downtime and, if possible, travel with multiple passengers throughout the entire journey. However, the planning of the vehicle charging strategy and the routes must be adapted accordingly.
[0003] In this regard, it is known from the state of the art that the daily use of a large number of passengers can be predicted, including at what time and at what location demand will be particularly high. For example, it can be determined that at certain times, particularly in the afternoon, at a large factory or company, a large number of people will be likely to want to use the autonomous shuttle. However, outside of these regular times, prediction is complicated.
[0004] US 2018 / 0096606 A1 discloses a method for providing scheduling services for on-demand transportation. The method includes determining that a predictive assignment message should be transmitted to a motor vehicle, generating the predictive assignment message in response to determining that a predictive assignment should be transmitted to a motor vehicle, and transmitting the predictive assignment message to the vehicle. Generating the predictive assignment message uses one or more predictive models calculated from historical and real-time on-demand transportation service data.
[0005] US 2018 / 197 071 A1 describes embodiments for minimizing the waiting time for a driver after sending a ride request for a vehicle. An exemplary computer-implemented method includes receiving a ride request, wherein the request is for a ride from a starting location to a zone in a geographic region during a specified time window. The method further includes predicting travel demand based on a number of ride requests in the zone during the specified time window. The method further includes requesting transport of one or more vehicles into the zone in response to the predicted number of ride requests if the ride request is predicted to exceed the number of vehicles in the zone during the specified time window.
[0006] US 2019 / 197 798 A1 describes a computer system that creates a prediction of requirements for autonomous vehicles in a fleet of collectively managed autonomous vehicles based on a current state and a future event, wherein the prediction includes a predicted level of requirement and a predicted duration of the level of requirement. The system can receive status information from fleet vehicles and identify a vehicle requiring maintenance. The system can receive status information from service centers and identify a service center for servicing the vehicle.Based on the generated demand prediction, the system may determine a time at which the vehicle should be serviced at the identified service center, schedule the vehicle for service at the identified service center at the determined time, and instruct the vehicle to travel to the service center to be serviced at the determined time. In certain embodiments, the demand prediction may be generated using machine learning.
[0007] US 2018 / 209 803 A1 describes an automated ridesharing dispatch system comprising a communication interface configured to receive ride requests, each trip request including a starting point and a desired destination, and to receive a current location of the ridesharing vehicle from a mobile communication device connected to a ridesharing vehicle. The system also includes at least one processor configured to dispatch the ridesharing vehicle to pick up the multiple users and to determine a capacity status of the ridesharing vehicle based on a known passenger capacity of the ridesharing vehicle. If the capacity status is below a capacity threshold, the processor directs the ridesharing vehicle along a first route, and if the capacity threshold is reached, the processor directs the ridesharing vehicle along a second route.
[0008] The object of the present invention is to provide a method and a vehicle fleet coordination system by means of which the autonomous motor vehicles can be coordinated in an improved manner.
[0009] This object is achieved by a method and a vehicle fleet coordination system according to the independent patent claims. Advantageous embodiments are specified in the subclaims.
[0010] One aspect of the invention relates to a method for coordinating an autonomous vehicle fleet with a plurality of autonomous motor vehicles in a predetermined first region by means of a vehicle fleet coordination system, in which the vehicle fleet is coordinated as a function of a first regional environmental profile of the first region for the vehicle fleet coordination system, wherein for coordination purposes a respective position in the first region dependent on the regional environmental profile is approached autonomously by means of a respective motor vehicle of the plurality of motor vehicles.
[0011] It is provided that the first regional environment profile is generated on the basis of a second regional environment profile stored in an electronic computing device of the vehicle fleet coordination system, which second regional environment profile is generated for a second region independent of the first region.
[0012] This enables improved coordination of the autonomous vehicles in the autonomous vehicle fleet.
[0013] The reason for this is particularly that, for example, similar events take place in the second region as in the first region. A respective city can be regarded as the respective region. For example, it can be specified that the city of Berlin can be regarded as the second region and the city of Stuttgart as the first region. This means that the environmental profile for the city of Stuttgart is now generated based on the environmental profile generated for Berlin. In particular, if, for example, the autonomous vehicle fleet is to be rolled out in Stuttgart, corresponding attributes are then specified based on the environmental profile already generated for Berlin, with which the regional environmental profile can be driven. In particular, the environmental profile is characterized by corresponding events within the regions.In particular, the events have a position that is then approached by a respective autonomous vehicle.
[0014] In particular, the invention thus solves the problem that, for example, in the first region, there are a large number of events at different locations, or in other words, positions, every day, but only a portion of these events will result in a large number of driving orders for the autonomous vehicle fleet. The allocation of driving orders depends essentially on the location of the event, the time, but also primarily on the people involved. For example, after a pop concert, the probability of the autonomous vehicle fleet being used is higher than after a classical concert due to the age mix.
[0015] Furthermore, it is provided that route planning for the respective motor vehicles and / or battery charging planning for the respective motor vehicles and / or idle time planning for the respective motor vehicles is carried out as coordination using the electronic computing device. In particular, for improved coordination of the autonomous vehicle fleet, it is necessary that appropriate route planning and / or battery charging planning and / or idle time planning be carried out. This can, for example, reduce the idle times of the autonomous vehicles. By taking this into account and / or appropriate planning, improved coordination of the autonomous vehicle fleet can then be achieved.
[0016] Furthermore, requested trips from people are rejected depending on the generated initial environmental profile. In particular, it can be provided that, when planning the corresponding trips, trips can also be rejected if, for example, they would result in the required fleet size not being reached at the event location at a certain time. Alternatively or additionally, in the case of ride-sharing, larger areas could be created from which customers can be picked up and dropped off, thus reducing the number of vehicles required. This could enable improved coordination of the autonomous vehicle fleet.
[0017] According to an advantageous embodiment, the first regional environmental profile is generated based on a plurality of additional stored regional environmental profiles, which are generated for a plurality of additional regions independent of the first region. In other words, the regional environmental profile is generated depending on a plurality of additional regional environmental profiles that have already been generated. This makes it possible to access a plurality of previously generated environmental profiles, thereby generating a highly detailed regional environmental profile. This makes it possible to implement reliable coordination of the autonomous vehicles right from the start of the autonomous vehicle fleet's service provision.
[0018] In a further advantageous embodiment, the first regional environmental profile is adapted depending on regional events in the first region. In other words, it is provided that the first regional environmental profile is generated on the basis of at least the second regional environmental profile. Over time, the first regional environmental profile is then adapted accordingly depending on the regional events in the first region. In particular, it can be provided that information for these events can be retrieved, for example, from the Internet. The regional events can in particular be events that are typically collected and that are available, for example, on the Internet, for example in an event calendar, and that can then be searched via the electronic computing device.When the autonomous vehicle fleet is rolled out in the first region, the data from the second regional environmental profile can be used initially, or alternatively, manually assigned attributes can be specified to mark the events, in other words the regional occurrences. Based on these attributes, the autonomous vehicles can then be pre-planned to the vicinity of the event locations, in other words to the position. Using the actual incoming driving orders and the corresponding positions, the actual event pattern for this first region can then be learned, and thus the typical shuttle users can be assigned to the events typical for them. This allows the first environmental profile to be generated precisely, leading to improved coordination of the autonomous vehicle fleet.
[0019] Furthermore, it has proven advantageous if the first regional environmental profile is adapted temporally and / or spatially depending on the regional events. In other words, a spatiotemporal adaptation of the regional environmental profile takes place. In particular, this allows for the first region to be characterized by events. The first regional environmental profile is thus adapted temporally and / or spatially to the events taking place in the first region, which then allows the first regional environmental profile to be adapted. In particular, this enables the coordination of the autonomous vehicle fleet to be carried out in a manner adapted to the first region.
[0020] In a further advantageous embodiment, the first regional environment profile is adapted depending on the respective duration of the regional events. In other words, it can be provided that typical event lengths are "learned" using the electronic computing device. For example, an event regularly lasts two hours. The user of the autonomous vehicle therefore issues their driving request two hours after the start of the event and wants to be picked up as quickly as possible. By appropriately using the time duration, the first regional environment profile can be generated precisely. This leads to improved coordination of the autonomous vehicle fleet in the first region.
[0021] In a further advantageous embodiment, respective person profiles are generated for the respective regional events using the electronic computing device, and the first environmental profile is adapted depending on the generated person profiles. In particular, the person profile can, for example, take into account the average age of an event. Furthermore, it can be provided, for example, that the event, for example, a football match, is evaluated with the person profile to determine whether, after the football match, people regularly want to travel to the main train station of the respective first region.
[0022] This can then be taken into account in order to realize the coordination of the autonomous vehicle fleet.
[0023] It is also advantageous if the first environmental profile is adapted to the electronic computing device using machine learning, in particular using a neural network. In other words, the electronic computing device is trained, and the training process actively adapts the first environmental profile in real time. This makes it possible for the first regional environmental profile to be adapted to current events within the first region in the future, thus enabling improved coordination of the vehicle fleet.
[0024] A further aspect of the invention relates to a vehicle fleet coordination system for coordinating an autonomous vehicle fleet comprising a plurality of autonomous motor vehicles in a predetermined first region using at least one electronic computing device. The vehicle fleet coordination system is configured to implement a method according to the preceding aspect. In particular, the method is implemented using the vehicle fleet coordination system.
[0025] Advantageous embodiments of the method are considered advantageous embodiments of the vehicle fleet coordination system. The vehicle fleet coordination system has specific features that enable implementation of the method or an advantageous embodiment thereof.
[0026] Further advantages, features, and details of the invention will become apparent from the following description of a preferred embodiment and from the drawings. The features and combinations of features mentioned above in the description, as well as the features and combinations of features mentioned below in the description of the figures and / or shown alone in the single figure, can be used not only in the respective combinations specified, but also in other combinations or on their own, without departing from the scope of the invention.
[0027] The single figure shows a schematic side view of an embodiment of a vehicle fleet coordination system.
[0028] In the figure, identical or functionally identical elements are provided with the same reference numerals.
[0029] Fig. 1shows a schematic side view of an embodiment of a vehicle fleet coordination system 10. The vehicle fleet coordination system 10 is designed to coordinate an autonomous vehicle fleet 12 with a plurality of autonomous motor vehicles 14 in a predetermined first region 16. The vehicle fleet coordination system 10 has at least one electronic computing device 18. Furthermore, it can be provided, for example, that a respective autonomous motor vehicle 14 has a respective electrical energy storage device 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 electrically operated motor vehicle 14, in particular a fully electrically operated autonomous motor vehicle 14.
[0030] In the method for coordinating the autonomous vehicle fleet 12 with the plurality of autonomous motor vehicles 14, which in the present case has in particular three autonomous motor vehicles 14, the autonomous vehicle fleet 12 is coordinated in the predetermined first region 16 by means of the vehicle fleet coordination system 10 as a function of a first regional environmental profile 22 of the first region 16 for the vehicle fleet coordination system 10, wherein for the coordination purpose a respective position P in the first region 16, which position is dependent on the regional environmental profile 22, is approached autonomously by means of a respective autonomous motor vehicle 14 of the plurality of autonomous motor vehicles 14.
[0031] It is provided that the first regional environment profile 22 is generated on the basis of a second regional environment profile 24 stored in the electronic computing device 18 of the vehicle fleet coordination system 10, which second regional environment profile is generated for a second region independent of the first region 16.
[0032] The first region is, in particular, a first city, and the second region is, in particular, a second city independent of the first city. It is also self-evident that more than three autonomous vehicles 14 can be coordinated.
[0033] In particular, it can be provided that the first regional environment profile 22 is generated on the basis of a plurality of stored further regional environment profiles, which are generated for a plurality of further regions independent of the first region 16.
[0034] Furthermore, it can be provided, in particular, that the first regional environment profile 22 is adapted depending on regional events 26 in the first region 16. The regional events 26 are, in particular, events within the first region 16. These regional events 26 can be determined, for example, by evaluating an event calendar for the first region 16.
[0035] Furthermore, it can be provided that the first regional environmental profile 22 is adapted temporally and / or spatially depending on the regional events 26. In particular, a spatiotemporal adaptation of the first regional environmental profile 22 can thus be carried out.
[0036] Furthermore, it can be provided that the first regional environmental profile 22 is adapted depending on a respective time duration 28 of the regional events 26.
[0037] Furthermore, it can be provided that for the respective regional events 26 respective person profiles 30 are generated with the electronic computing device 18, and the first regional environment profile 22 is adapted depending on the generated person profiles 30.
[0038] In particular, it can be provided for this purpose that a route planning of the respective autonomous motor vehicles 14 and / or a battery charge planning of the respective autonomous motor vehicles 14, in particular a battery charge voltage of the respective electrical energy storage devices 20, and / or a downtime planning of the respective autonomous motor vehicles 14 are carried out as coordination by means of the electronic computing device 18.
[0039] Furthermore, it can be provided in particular that, depending on the generated first regional environment profile 22, requested journeys by persons are rejected.
[0040] Furthermore, it can be provided that the first regional environment profile 22 is adapted to the electronic computing device 18 by means of machine learning, in particular by means of a neural network 32.
[0041] In particular, the invention takes advantage of the fact that in a city, which in this case corresponds to the first region 16, a multitude of events 26, which can also be referred to as events, take place every day at different locations, in particular at different positions P. However, only a portion of these events 26 will result in a large number of driving orders for the autonomous vehicle fleet 12. These depend on the location of the event 26, the time, but primarily also on the people. For example, after a pop concert, due to the age mix, a higher number of driving orders for the autonomous vehicle fleet 12 is very likely to be recorded than after a classical concert.
[0042] These events can typically be collected and are available online and can be searched, for example, via the electronic computing device 18. If the autonomous vehicle fleet 12 is to be rolled out in a new city, in this case the first region 16, the data based on other cities, i.e., the second region, and the associated second regional environmental profile 24 are used. Alternatively, data can also be entered manually. The corresponding events 26 are then marked, and based on these attributes, the autonomous vehicles 14 can be pre-planned to be guided to the vicinity of the event locations. Based on the actual bookings received and the corresponding pick-up locations, the actual event pattern for this city can then be learned, and thus the typical users of the autonomous vehicle fleet 12 can be assigned to the events typical for them.In addition, typical event lengths can also be learned. For example, a particular play has a duration 28 of two hours, so the electronic computing device 18 knows that two hours after the start of the event, corresponding transport requests will arrive, which should be processed as quickly as possible. In addition, the destination can then also be used to learn where the typical customer of such an event would like to be driven. For example, it may be planned that after a football match, the respective person would often want to be driven to the main train station of the first region 16. This knowledge can then be used for the further scheduling of the autonomous vehicle fleet 12.
[0043] Based on these personal preferences of the typical autonomous vehicle users, the autonomous vehicle fleet 12 can then be positioned in such a way that the journeys are planned in advance, but also the planning regarding the battery charging of the motor vehicles 14 and thus necessary downtimes can be improved.
[0044] When planning trips, trips can also be rejected if they would result in the required fleet size not being reached at the event location, i.e., at event 26, at a specific time. Alternatively or additionally, in the case of ride-sharing, larger areas could be created from which customers are picked up and dropped off in order to reduce the number of autonomous vehicles 14 required.
[0045] Overall, the invention shows an automatic positioning of autonomous motor vehicles 14.
Claims
1. Method for coordinating an autonomous vehicle fleet (12) having a plurality of autonomous motor vehicles (14) in a predetermined first region (16) by means of a vehicle fleet coordination system (10), in which method the autonomous vehicle fleet (12) is coordinated depending on a first regional environment profile (22) of the first region (16) for the vehicle fleet coordination system (10), wherein, for coordination, a relevant position (P) in the first region (16), which position is dependent on the regional environment profile (22), is approached autonomously by means of a relevant autonomous motor vehicle (14) of the plurality of autonomous motor vehicles (14), characterized in that the first regional environment profile (22) is generated on the basis of a second regional environment profile (24), which is stored in an electronic computing device (18) of the vehicle fleet coordination system (10) and which is generated for a second region independent of the first region (16), wherein route planning of the autonomous motor vehicles (14) and / or battery charge planning of the autonomous motor vehicles (14) and / or idle time planning of the autonomous motor vehicles (14) are carried out as coordination by means of the electronic computing device (18), and wherein journeys requested by people are rejected depending on the generated first regional environment profile (22), and wherein the first region (16) is a first city and the second region is a second city that is different from the first city, the first regional environment profile (22) is adapted depending on regional events (26) in the first region (16), wherein information regarding the number or destinations of the requested journeys corresponding to the regional events is assigned to the types of regional events by the first and second environment profiles (22, 24).
2. Method according to claim 1, characterized in that the first regional environment profile (22) is generated on the basis of a plurality of stored further regional environment profiles which are generated for a plurality of further regions independent of the first region (16).
3. Method according to any of the preceding claims, characterized in that the first regional environment profile (22) is adapted temporally and / or locally depending on the regional events (26).
4. Method according to any of the preceding claims, characterized in that the first regional environment profile (22) is adapted depending on a relevant time duration (28) of the regional events (26).
5. Method according to any of the preceding claims, characterized in that person profiles (30) are generated for the relevant regional events (26) by means of the electronic computing device (18), and the first regional environment profile (22) is adapted depending on the generated person profiles (30).
6. Method according to any of the preceding claims, characterized in that the first regional environment profile (22) is adapted to the electronic computing device (18) by means of machine learning, in particular by means of a neural network (32).
7. Vehicle fleet coordination system (10) for coordinating an autonomous vehicle fleet (12) having a plurality of autonomous motor vehicles (14) in a predetermined first region (16), comprising at least one electronic computing device (18), wherein the vehicle fleet coordination system (10) is designed to carry out a method according to any of claims 1 to 6.
