DE102020002997B4 - Method for predicting the probability of at least one free parking space for motor vehicles in a rest area, as well as a parking space prediction system - Google Patents

Abstract

Method for predicting a probability (12) of at least one free parking space (14) for motor vehicles (16, 18) in a rest area (20) using a parking space prediction system (10), in which a parameter characterizing a motor vehicle (16) is recorded and transmitted to an electronic computing device (22) external to the motor vehicle of the parking space prediction system (10), wherein the probability (12) of at least one free parking space (14) for motor vehicles (16, 18) is predicted by means of the electronic computing device (22) external to the motor vehicle as a function of the transmitted characterizing parameter, wherein a vehicle dynamics parameter (v, P) of the motor vehicle (16) is recorded as a parameter characterizing the motor vehicle (16) and the probability (12) of at least one free parking space (14) for motor vehicles (16, 18) is predicted as a function of the recorded vehicle dynamics parameter (v, P), characterized in that a low probability (12) of a free parking space (14) is predicted if the motor vehicle (16) has not stopped at the rest area (20) and has left the rest area (20) after a specified time period (t) and a high probability (12) of a free parking space (14) is predicted if the motor vehicle (16) has stopped at the rest area (20) and has not left the rest area (20) after a specified time period (t).

Description

The invention relates to a method for predicting the probability of at least one free parking space for motor vehicles in a rest area using a parking space prediction system according to the preamble of claim 1. The invention further relates to a parking space prediction system.

It is known that, for example, in the European Union, truck drivers must take breaks at precisely defined times. For instance, these truck drivers must take a break after four hours of driving. Therefore, information about rest areas available to truck drivers is important during the planning phase of a given journey.

The EP 2 953 111 A1 This concerns a method and a device for determining available parking spaces in truck parking areas and notifying truck drivers. At least one central server and a multitude of trucks participating in the method are connected via a wireless, bidirectional communication device using a server module and a truck module carried in each truck. The central server stores and/or generates position data and information on the capacity of truck parking areas. Position data of the parking positions is recorded in a parked truck and transmitted to the server. From this position data of multiple trucks, the server determines the number of parked trucks relevant to the method and, accordingly, the number of occupied parking spaces in a specific truck parking area. By comparing this with the server's capacity data, the server calculates the probability of available parking spaces and communicates this information to drivers of trucks relevant to the method.

The US 2014/0 266 800 A1 This describes an architecture that uses crowd-sourced parking information to calculate the probability of finding parking spaces at specific street segments, parking lots, and/or within larger geographic areas. The crowd-sourced parking information can be derived from geolocation traces. This approach employs a location trace mining method to calculate the probability of finding parking spaces at specific street segments, parking lots, and/or within larger geographic areas. The location traces can be analyzed to classify parking spaces as public, private, and semi-private. Furthermore, the location traces can be analyzed to derive the times and dates on which a vehicle is permitted to park at a particular location.

The DE 10 2016 003 262 A1 This relates to a method for determining available parking spaces for a vehicle, in particular a truck, wherein information determined about available parking spaces in a specified parking area is displayed to a vehicle occupant. Data is collected on the specified parking area, the positions of vehicles driving in the vicinity of the parking area, the positions of vehicles parked in the specified parking area, and the expected parking duration of the parked vehicles. Based on the collected data, currently available or potentially available parking spaces in the specified parking area are determined, as are possible upcoming consecutive parking maneuvers of vehicles driving in the vicinity of the parking area and vehicles parked in the parking area. The determined information is then displayed in the vehicle.

The object of the present invention is to provide a method and a parking space prediction system by means of which an improved prediction for at least one free parking space within a rest area for vehicles can be realized.

This problem is solved by a method and a parking space prediction system according to the independent claims. Advantageous embodiments are specified in the dependent claims.

One aspect of the invention relates to a method for predicting the probability of at least one free parking space for a motor vehicle in a rest area by means of a parking space prediction system, in which a parameter characterizing a motor vehicle is recorded and transmitted to a motor vehicle-external electronic computing device of the parking space prediction system, wherein, by means of the motor vehicle-external electronic computing device, the probability of at least one free parking space for motor vehicles is predicted as a function of the transmitted characterizing parameter.

It is planned that a vehicle dynamics parameter will be recorded as a parameter characterizing the vehicle, and that the probability of at least one free parking space for motor vehicles will be predicted depending on the recorded vehicle dynamics parameter.

This makes it possible to reliably predict the probability of at least one parking space for other vehicles based on the vehicle dynamics parameter.

Furthermore, a low probability of finding a free parking space is predicted if the vehicle did not stop at the rest area and left after a specified period of time. In particular, the driver will only visit the rest area if they intend to take a break. If the vehicle drives through the rest area, it can be assumed that no free parking space was available at that time. Therefore, the probability of finding a free parking space can be considered low.

Furthermore, a high probability of finding a free parking space is predicted if the vehicle has stopped at the rest area and left after a further predetermined period. This further predetermined period could, for example, be five minutes. If the vehicle remains within the rest area for longer than five minutes, it can be assumed that a break has been taken. This is because a free parking space was available for the vehicle. This information can then be transmitted to the vehicle's external electronic computer, enabling a corresponding prediction of available parking spaces at the rest area.

In other words, the system is designed to capture not only positional data but also the vehicle's dynamic parameters. For example, the vehicle's speed, its current position (especially while moving), and so-called "driver card" information can be used. This allows, for instance, the detection of when a vehicle has entered, or is about to enter, an area with potential parking spaces. Three possibilities can then be determined: If the vehicle does not stop within the rest area and leaves again within a short time, it can be concluded that no parking spaces are available. If the vehicle stops and its break begins, the start of the break can be determined. If the vehicle has stopped and the end of the break is imminent, the end of the break can be anticipated.

In particular, the probability of finding at least one free parking space can be determined by analyzing the vehicle's speed over time, especially across a large number of other vehicles, and by using GPS position data. Furthermore, additional "Driver Card" information can be utilized, collected over time from a large number of vehicles. This data is then used to learn, based on fleet data, which parking spaces are occupied at which times. This allows for predictions of whether and when a parking space will be available at a specific rest area or service station. This data is then forwarded to individual vehicles in the fleet to optimize route planning and is continuously updated using additional fleet data. The Driver Card information includes details about the vehicle's journey and the driver's activity. Specifically, the start of the journey and any breaks can be stored on the Driver Card.

Preferably, the vehicle is designed as a truck. In particular, if a legally mandated rest period is required for the truck driver, the parking space prediction system will allow the driver to reliably determine whether a parking space is available for a corresponding break at the rest area to be accessed in the future.

In this context, a rest area is understood to mean, in particular, a parking area for trucks. For example, such a rest area can be a service station on a motorway or a truck stop on a rural road. A rest area is defined, in particular, by the possibility of a longer stay, for example, 30 minutes.

According to an advantageous embodiment, the vehicle's speed is recorded as a driving dynamics parameter during its passage through the rest area. If, for example, the speed does not change noticeably during passage, and especially if it remains high, it can be concluded that no parking space is available for the vehicle user. The user thus "passes through" the rest area and must, for example, seek out the next rest area. This can be used to record that the rest area was full at a certain time and No free parking space was available for additional vehicles. This information can then be evaluated and transmitted to the parking space prediction system. This allows for a reliable prediction of the probability of finding at least one free parking space.

It is also advantageous to record the vehicle's current position as a driving dynamics parameter while it is passing through the rest area. In particular, the position can also be recorded before entering the rest area. This allows conclusions to be drawn about whether a vehicle will soon take a break. This, in turn, can be used to reliably predict the probability of finding an available parking space within the rest area.

It is also advantageous to record a user's driving time as a vehicle dynamics parameter. Particularly with trucks, the corresponding driving times of the vehicle driver are recorded. This can be done, for example, using a "driver card." This data can then be used to predict where each driver will take a break. This information can then be incorporated into the relevant algorithm, enabling the future determination of available parking spaces.

Furthermore, it has proven advantageous to transfer the specific probability to a large number of other motor vehicles. In particular, this large number of motor vehicles can also be referred to as a fleet. The corresponding probability of finding an available parking space can then be transferred to the fleet, enabling it to reliably plan its route and its rest periods along that route.

In a further advantageous implementation, the probability of finding a parking space is determined based on a time of day, day of the week, and/or location within the rest area. Furthermore, factors such as holiday periods, construction work, or waiting times at border crossings can also be taken into account. Additionally, driving bans can be considered. This allows for a reliable prediction of available parking spaces at a given rest area at different times of day and days of the week.

Another aspect of the invention relates to a parking space prediction system for predicting the probability of at least one free parking space for motor vehicles in a rest area, comprising at least one electronic computing device external to the motor vehicle, wherein the parking space prediction system is configured to carry out a method according to the preceding aspect. In particular, the method is carried out using the parking space prediction system.

Advantageous implementations of the procedure are to be considered advantageous implementations of the parking space prediction system. The parking space prediction system possesses tangible features that enable the implementation of the procedure and an advantageous implementation thereof.

Further advantages, features, and details of the invention will become apparent from the following description of preferred embodiments and from the drawings. The features and combinations of features mentioned above in the description, as well as those mentioned below in the figure description and/or shown in the figures alone, can be used not only in the combinations specified, but also in other combinations or individually, without departing from the scope of the invention.

• 1 eine schematische Draufsicht auf eine Ausführungsform eines Parkplatzvorhersagesystems; und
• 2 ein schematisches Zeit-Geschwindigkeits-Diagramm eines Kraftfahrzeugs.

This shows:

• 1 a schematic top view of an embodiment of a parking space prediction system; and
• 2 A schematic time-speed diagram of a motor vehicle.

In the figures, identical or functionally equivalent elements are provided with the same reference symbols.

1 Figure 1 shows a schematic top view of an embodiment of a parking space prediction system 10. The parking space prediction system 10 is designed to predict a probability 12 for at least one free parking space 14 for motor vehicles 16, 18 in a rest area 20 and has at least one vehicle-external electronic computing device 22.

In the procedure for predicting the probability 12 for at least one parking space 14 for motor vehicles 16, 18 of the parking space prediction system 10, a parameter characterizing a motor vehicle 16 is recorded and transmitted to a motor vehicle-external electronic computing device 22 of the parking space prediction system 10, wherein the probability 12 for at least one free parking space 14 for motor vehicles 16, 18 is predicted by means of the motor vehicle-external electronic computing device 22.

It is intended that a vehicle dynamics parameter v, P of the vehicle 16 is recorded as a parameter characterizing the vehicle 16 and that, depending on the recorded vehicle dynamics parameter v, P, the probability 12 for at least one free parking space 14 for vehicles 16, 18 is predicted.

In particular, it is shown here that a first motor vehicle 16 has already entered the rest area 20. Another motor vehicle 18 is still on a road, for example, a motorway. Based on the vehicle dynamics parameter v, P of motor vehicle 16, the probability 12 of whether a free parking space 14 can be found within the rest area 20 can now be determined. This can then, for example, be transmitted to the other motor vehicle 18, so that it knows whether a free parking space 14 is available within the rest area 20 when the other motor vehicle 18 enters the rest area 20.

Motor vehicles 16 and 18 are specifically designed as trucks.

In particular, it may be provided that a speed v is recorded as a vehicle dynamics parameter v, P during a passage of the motor vehicle 16 within the rest area 20. Furthermore, a current position P of the motor vehicle 16 during a passage of the motor vehicle 16 within the rest area 20 may be recorded as a vehicle dynamics parameter v, P. It may also be provided that the travel times of a user of the motor vehicle 16 are recorded as vehicle dynamics parameters v, P.

The specific probability 12 can in particular be transferred to a large number of other motor vehicles 18, which can in particular be referred to as a fleet, wherein, for example, the specific probability 12 is further determined depending on a time of day and/or a day of the week and/or a location of the rest area 20.

It may also be provided that a low probability is predicted if the motor vehicle 16 has not stopped in rest area 20 and has left rest area 20 after a predetermined time interval t. This is particularly relevant in the 2 depicted.

It has also proven advantageous to predict a high probability 12 of a free parking space 14 if the vehicle 16 has stopped in the rest area 20 and has not left the rest area 20 after a further predetermined time interval t. It can also be provided that the dwell time of the vehicle 16 within the rest area 20 is taken into account when predicting the probability 12.

In particular, a fleet management system can thus be used to generate an estimation algorithm that estimates the probability 12 that a free parking space 14 is available at rest area 20 or whether rest area 20 is full. The algorithm is supplied with relevant data from vehicles 16 and 18 in real time, so that the probability 12 can also be determined in real time.

For this purpose, the speed v, especially over a period of time, is considered, as well as the position P, which can be a global navigation satellite system position. Furthermore, the position P can be unpersonalized. So-called "driver card" information can also be taken into account to determine, for example, that vehicle 16 has entered rest area 20. Additionally, this "driver card" information can be used to estimate when vehicle 16 will leave rest area 20. The probability prediction can be updated in real time and further refined, for example, based on feedback from the respective users of vehicles 16 and 18. This algorithm is implemented on a backend, which is the vehicle-external electronic computing device 22. Vehicles 16 and 18 then communicate with this backend.

2 shows a schematic time-velocity diagram. 2 This illustrates, in particular, the case where motor vehicle 16 enters rest area 20 at time t1 with an initial speed v1. Motor vehicle 16 decelerates to speed v2 within rest area 20. At time t2, which is particularly close to time t1, for example, only five minutes later, motor vehicle 16 accelerates back to speed v1. This can be interpreted, in particular, as indicating that no free parking space 14 was available within rest area 20. Motor vehicle 16 was only in rest area 20 for a short time and therefore found a correspondingly free parking space. Parking space 14 was sought but not found, which is why at time t2 it drove back onto, for example, the motorway and accelerated again.

In particular, it can thus be shown that the motor vehicle 16 did indeed enter rest area 20, but did not stop and left rest area 20 again in a short time, which suggests that there was no free parking space.

In another, not in 2 In the case described, for example, vehicle 16 may have entered rest area 20, which can be determined in particular based on its position P, stopped, which in turn can be determined based on its speed v, and the "Driver Card" information can be used to predict that a break will soon be taken. In particular, the start of the break can be determined.

In another case, it can be determined that, for example, vehicle 16 is on break and the legally required break time is about to expire. When vehicle 16 "starts moving," which can be determined in particular based on the speed v of vehicle 16, the end of the break can be determined.

In another scenario, it may be stipulated that the vehicle 16 is below a predetermined threshold for breaks, for example, that a break must be taken within the next fifteen minutes. If the vehicle 16 then fails to enter a corresponding rest area 20, which can be determined, for example, based on the vehicle 16's position P and speed v, a full rest area 20 can be identified.

In particular, if different data from different motor vehicles 16, 18 regarding available parking spaces 14 have been collected over a specified period and transmitted to the vehicle-external electronic computing device 22, this device can be trained and thus made available as fleet data for a fleet of motor vehicles 16, 18. This allows for an estimation of the probability 12 of when, at what time, corresponding rest areas 20 will be overcrowded or have available parking spaces 14. This information can then be incorporated into the route planning for the motor vehicles 16, 18.

The electronic computing device 22 can also be supplied with relevant information over time, so that the estimate of the probability 12 for a free parking space 14 can be updated and transmitted to the motor vehicles 16, 18.

Overall, the invention demonstrates a real-time prediction of available parking spaces 14 through the use of “on-board” data and the use of an “off-board” learning algorithm within the electronic computing device 22.

Claims (8)

Method for predicting a probability (12) of at least one free parking space (14) for motor vehicles (16, 18) in a rest area (20) using a parking space prediction system (10), in which a parameter characterizing a motor vehicle (16) is recorded and transmitted to an electronic computing device (22) external to the motor vehicle of the parking space prediction system (10), wherein the probability (12) of at least one free parking space (14) for motor vehicles (16, 18) is predicted by means of the electronic computing device (22) external to the motor vehicle as a function of the transmitted characterizing parameter, wherein a vehicle dynamics parameter (v, P) of the motor vehicle (16) is recorded as a vehicle dynamics parameter (v, P) of the motor vehicle (16) and the probability (12) of at least one free parking space (14) for motor vehicles (16, 18) is predicted as a function of the recorded vehicle dynamics parameter (v, P), characterized in that a low probability (12) of a free parking space (14) is predicted if the motor vehicle (16) has not stopped at the rest area (20) and has left the rest area (20) after a specified time period (t) and a high probability (12) of a free parking space (14) is predicted if the motor vehicle (16) has stopped at the rest area (20) and has not left the rest area (20) after a specified time period (t). Procedure according to Claim 1 , characterized in that a speed (v) is recorded as a vehicle dynamic parameter (v, P) during a passage of the motor vehicle (16) within the rest area (20). Procedure according to Claim 1 or 2 , characterized in that a current position (P) of the motor vehicle (16) is recorded as a vehicle dynamic parameter (v, P) during a passage of the motor vehicle (16) within the rest area (20). A method according to one of the preceding claims, characterized in that as Driving dynamics parameters (v, P) and driving times of a user of the motor vehicle (16) are recorded. Method according to one of the preceding claims, characterized in that the determined probability (12) is transferred to a plurality of further motor vehicles (16, 18). Method according to one of the preceding claims, characterized in that the specific probability (12) is determined depending on a time of day and/or a day of the week and/or a location of the rest area (20). Method according to one of the preceding claims, characterized in that a holding time of the motor vehicle (16) within the rest area (20) is taken into account when predicting the probability (12). Parking space prediction system (10) for predicting a probability (12) of at least one free parking space (14) for motor vehicles (16, 18) in a rest area (20), with at least one motor vehicle-external electronic computing device (22), wherein the parking space prediction system (10) is used to carry out a procedure according to one of the Claims 1 until 7 is trained.