DE102011121049A1 - Method for determining the occupancy of a parking space with at least one entrance and exit - Google Patents

Abstract

The invention relates to a method for determining the occupancy of a parking space with at least one entrance and exit (11, 12), in particular for determining the occupancy of a parking space with marked parking spaces for motor vehicles on trunk roads (9) and highways. According to the invention, image data are acquired by a vehicle driving into the parking space by a sensor unit (4), the image data representing the external appearance of the vehicle. Characteristic features of the approaching vehicle are determined by feature extraction from the image data obtained and stored in a memory at the time when the parking space is accessed. For a vehicle extending out of the parking space, image data are also acquired by a sensor unit (4) and characteristic features are extracted. By comparing the characteristic features of the outgoing vehicle with the characteristic features of all vehicles available in the memory, an identification of the outgoing vehicle takes place. In case of agreement, d. H. unambiguous identification of the vehicle, the vehicle is removed from the store. If an outgoing vehicle can not be clearly identified, it will not be considered further. Furthermore, the parking times of each vehicle in the store are continuously calculated. If the calculated parking time of a vehicle over a predetermined parking time, this vehicle is removed from the store. The occupancy of the parking space is calculated by summing the vehicles stored in the storage unit.

Description

The invention relates to a method for determining the occupancy of a parking space with at least one entrance and exit, in particular for determining the occupancy of a parking space with marked parking spaces for motor vehicles on trunk roads and highways.

A parking space is a delimited area for the temporary parking of motor vehicles with at least one entrance and exit. The area defining the parking space preferably has marked parking spaces for the motor vehicles to be parked.

The determination of the occupancy of a parking space is an inherent need for the management of parking spaces. The parking space problem on public parking spaces, especially for truck parking spaces on and on highways and service areas, is large. An impermissible parking situation, which results from overcrowding of the parking space, affects in particular safety aspects. Targeted parking management leads to better utilization of the parking space, avoids accidents and leads to cost reduction in the provision of parking space.

Public parking spaces, in particular on long-distance roads, usually permit no barrier-based access control system due to their design. This circumstance is due to both the high frequency and the free parking. On the part of the parking operator, however, there is a high level of interest in sufficiently reliable information about the current occupancy state of the parking space. Based on the knowledge of the occupancy state of parking spaces over long periods planning for additional parking space can be made more effective, more targeted and thus more cost-effective. Furthermore, there is a legitimate interest of drivers on the occupancy of lying on their route parking spaces. This applies in particular to those drivers who require guaranteed parking space to comply with legally prescribed driving and rest periods.

Privately managed parking is known, as well as in DE 60 203 689 T2 described, via an access control system, consisting of at least one barrier and often a monitoring system in the form of a camera for access control and a billing system, often a ticket system operated. This can lead to longer waiting times at the entrance or exit at high traffic volumes. A disadvantage of these systems is further that the clear knowledge of the number of vehicles within the parking space, but not on the occupied area of the individual motor vehicles is known.

From the DE 10 2005 034 372 A1 , of the DE 10 2008 029 413 A1 or Wolfgang Damm (Institute of Transport Telematics University of Dresden, "Project Stand Truck Parking Guidance Dresden", 07.12.2010, http://www.fh-erfurt.de/fhe/index.phpeID=tx_nawsecuredl&u=0&file=fileadmin/Material/Institut/Verkehr_Raum /Download/Projekte/2010/2010.12.07_Damm_Lkw-Parkleitsystem_Dresden.pdf&t=1323354531&hash=d5c7b8e56a9060eb31d812df32a7683c ) is another solution for determining the occupancy of the parking spaces within a parking space based on active and / or passive sensors for each individual or a group of parking spaces known. As active sensors laser scanners and as passive sensors different types of image sensors are used at the edge and / or over the parking spaces. As active sensors can after Martin Pozybill (Ministry of environment, nature protection and traffic Baden-Wuerttemberg, "A 8 LKW-Parkstandsanzeige" 07.12.2010, http://www.fh-erfurt.de/fhe/index.php?elD=tx_nawsecuredl&u=0&file=fileadmin /Material/Institut/Verkehr_Raum/Download/Projekte/2010/2010.12.07_Pozybill_A8_Lkw-Varkstandsanzeige.pdf & t = 1323354531 & hash = eed248f6cc9c8c0ebfc42731ab18036d Magnetic sensors below the road level and / or ultrasonic sensors, such as in the DE 10 2008 012 991 A1 described, are used. An advantage of this solution is that the occupancy of the parking space is always directly determinable. In particular, the determination of the occupancy of the parking space can take place without delay when the system is restarted. A disadvantage of these systems is the high economic investment for the sensors and high susceptibility to errors due to the large number of sensors.

Another approach is based on the "virtual barrier" at all entrances and exits of the parking area (see Martin Pozybill or DE 11 2006 003 370 T5 ). Here, the number of entering or leaving motor vehicles are determined at the entrances and exits. By counting the approaching and exiting vehicles, the current number of vehicles in the parking lot can be determined, not the parking space used. The disadvantage of this solution is that when the system is restarted, the number of motor vehicles already present in the parking space must be known and made available to the system. Furthermore, errors in the registration of the entrances or exits or in the case of traffic-impairing behavior and thus the non-registration of the motor vehicles lead to errors accumulating over time in the occupancy determination. Therefore, the occupancy error is due in regular periods to determine an inventory and to correct the system. This is associated with considerable human and logistical effort.

Finally, the reveals DE 10 2008 056 786 A1 a solution with a to be installed in each motor vehicle telematics unit. Via a communication interface of the telematics unit, the motor vehicle is partially monitored in the parking area continuously or at the entrance and exit into or out of the parking space. The disadvantage of this solution, on the one hand equipping a large number of motor vehicles with a corresponding telematics unit is necessary and on the other non-participating or not equipped with a telematics vehicles permanently distort the determination of the parking space occupancy.

With the currently known methods, the overwhelming effort results from disproportionately high investments in sensor systems, unreasonably high costs when restarting the system and / or systemically relevant errors in the determination of the occupation by unrecorded motor vehicles. At present, no comprehensively usable system is known which can be operated economically and meets the necessary requirements in terms of quality, economy and usage.

The object of the invention is to provide a method that determines the occupancy of a parking space with at least one entrance and exit with low investment costs, high reliability and sufficient accuracy.

According to the invention, this object is achieved by a method according to claim 1. For this purpose, image data are acquired by a vehicle traveling into the parking space by a sensor unit, the image data representing the external appearance of the vehicle. Characteristic features of the approaching vehicle are determined by feature extraction from the image data obtained and stored in a memory at the time when the parking space is accessed. For a vehicle leaving the parking space, image data are also acquired by a sensor unit and characteristic features are extracted. By comparing the characteristic features of the outgoing vehicle with the characteristic features of all vehicles available in the memory, an identification of the outgoing vehicle takes place. In case of agreement, d. H. unambiguous identification of the vehicle, the vehicle is removed from the store. If an outgoing vehicle can not be clearly identified, it will not be considered further. Furthermore, the parking times of each vehicle in the store are continuously calculated. If the calculated parking time of a vehicle over a predetermined parking time, this vehicle is removed from the store. The occupancy of the parking space is calculated by summing the vehicles stored in the storage unit.

It is essential for the purposes of the method according to the invention that the above-mentioned method steps in their entirety cause the method to behave independently and error-correcting without disturbance after a sufficient period of operation.

The procedure can be started without having to determine an initial occupancy of the parking space. After a sufficient time, which is longer than a predetermined parking time, the occupation of the parking space determined by the method is sufficiently accurate, whereby the accuracy increases further as the operating time progresses. Likewise, with the method, disturbances in the detection of the vehicles entering the parking space or from the parking space are corrected after a sufficient time, which is longer than the predetermined parking time. By means of the method can thus be determined with high reliability over long periods sufficiently accurate information about the occupancy of a parking space.

The method automatically corrects occupancy errors caused by unrecorded vehicles due to improper access, inadequate image data by the sensors, or inadequate identification due to a lack of characteristic features. A restart of the procedure with defined initial conditions, d. H. a predetermined initial occupancy of the parking space is generally not required.

Preferably, the sensor unit consists of at least one monocular camera, a stereo / multi-stereo camera, areal distance sensor based on modulated or structured light, light field camera and / or laser scanner. In an advantageous embodiment, the simultaneous use of complementary sensors increases the reliability of the solution.

An advantageous embodiment of the characteristic features are points, lines and / or surfaces and their location. A further advantageous embodiment consists in the extraction of the license plate number.

In addition, characteristic features such as width, length and / or height can be extracted, which allows the determination of the required footprint of a vehicle by calculation.

In a preferred embodiment of the method, in addition to the calculation of the occupancy the parking space calculated by summing the stored in the storage unit vehicles a land occupancy of the parking space from the required footprint of the stored vehicles. For this purpose, the features length and width of each vehicle are extracted from the image data and stored with the other characteristics and the time of driving the parking space. The required footprint of a vehicle is calculated from the characteristics length and width. The summation of the shelves of all stored vehicles plus a statistically determined area for spaces between the vehicles and access areas or other not suitable for parking a vehicle surfaces of a parking space results in the area occupancy of the parking space. This area occupancy is particularly advantageous when determining the occupancy of parking space without marked parking spaces.

In a further advantageous embodiment of the method, based on the captured image data, the vehicles are classified according to vehicle classes, such as trucks, semi-trailers, trucks with trailers, BUS, motorhomes, cars with trailers, cars and / or motorcycles, and stored accordingly in the memory , Due to the classification, both a vehicle class specific occupancy of the parking space can be made by summing all stored vehicles of a vehicle class as occupancy and a vehicle class specific area occupancy with the aid of the respective required footprint.

In a further advantageous embodiment of the method, the respective sensor units are provided with a communication interface and a computing and storage unit. By means of the communication unit, the data obtained can be transmitted directly to a central and / or decentralized detection point and / or users of the system and for the purpose of providing current occupancy data, the statistics collection of both general and vehicle class specific frequenting, area occupancy and parking duration and for the purpose of Occupancy forecasts are used.

• • Die Anforderungen an die Sensorik sind klein und damit preiswert. Es sind keine hohen Investitionskosten erforderlich.
• • Das System verhält sich selbstregelnd und fehlerkorrigierend.
• • Es ist grundsätzlich kein Neustart des Verfahrens mit Vorgabe definierter
• • Anfangswerte, wie einer Anfangsbelegung, nach einer Störung erforderlich.
• • Es ist keine in regelmäßigen Abständen durchzuführende Inventur des Parkraumes erforderlich.
• • Das Verfahren erlaubt die Parkraumbewirtschaftung von Parkräumen mit beliebiger Anzahl von Zu- und Ausfahrten als Stellplatz- und Flächenbelegung auch getrennt nach Fahrzeugklassen.

In summary, the method is distinguished from the methods known from the prior art by the following significant advantages:

• • The requirements for the sensors are small and therefore inexpensive. There are no high investment costs required.
• • The system behaves self-regulating and error correcting.
• • It is basically not a restart of the procedure defined with default
• • Initial values, such as an initial assignment, required after a fault.
• • No periodic inventory of parking space is required.
• • The procedure allows the parking space management of parking spaces with any number of entrances and exits as parking space and area occupancy also separated by vehicle classes.

Further details, features and advantages of the invention will become apparent from the following description of exemplary embodiments with reference to the accompanying drawings. Show it:

1 : the parking space on a trunk road with an access and an exit and marked parking spaces,

2 : the embodiment of a detection unit,

3 : the parking space on a trunk road with an access and an exit and a combined entrance and exit and marked parking spaces and

4 : the parking space with a combined entrance and exit without marked parking spaces.

The 1 shows a parking space on a trunk road 9 each with a detection unit 10 at a driveway 11 and an exit 12 with marked parking spaces for cars 13 and marked parking spaces for trucks 14 ,

The 2 shows a detection unit 10 with respect to a roadway plane 1 with a mast 3 on a foundation 2 Having a video camera 4 with a holder 5 , a laser scanner 6 with a holder 5 , a computer unit 8th with memory and a communication unit 7 ,

If a vehicle passes the driveway 11 to be a parking space by the video camera 4 and the laser scanner 6 Image data generated by the appearance of the vehicle. The image data become the computer unit 8th fed. From the image data are by means of the computer unit 8th characteristic features of the vehicle are extracted. Based on the characteristic features is a classification of the vehicle. The characteristic features, the classification and the time of capture, ie the time at which the vehicle enters the driveway 11 has happened are stored correspondingly and by means of the communication unit 7 to all computer units 8th at the entrances and exits 11 . 12 the parking space and to a central communication and computer unit, not shown. The memory contents of all computer rooms belonging to a parking space 8th as well as the central communication and computer unit are constantly synchronized. The vehicle becomes one of the marked parking spaces 13 . 14 take and initially remain on this. After a period of time t, the vehicle will leave the parking area via the exit 12 leave. Here are again image data of the external appearance of the vehicle by means of the video camera 4 and the laser scanner 6 generated and characteristic features by means of the computer unit 8th extracted. The when passing the exit 12 generated characteristic features are compared with all stored characteristic features. Can be clearly identified on the basis of the comparison of the characteristics of a vehicle, the access 11 has passed to the parking space, this vehicle is from the memory of the computer unit 8th away. The parking times of all in the memory of the computer unit 8th stored vehicles are determined continuously. When a predetermined parking time is exceeded, the vehicle is removed from the memory of the computer unit 8th away. Any removal of a vehicle from the memory of a computer unit 8th is via the communication unit 7 with all other computer units 8th and the central communication and computer unit synchronized. The occupancy of the parking space is calculated by summing the memory of the computer unit 8th stored vehicles determined.

The 3 shows a parking space on a trunk road 9 each with a detection unit 10 at a driveway 11 and an exit 12 and a combined entrance and exit 15 with marked parking spaces for cars 13 and marked parking spaces for trucks 14 , For this parking space, the procedure is equivalent to the parking space according to 1 applied. The registration unit 10 at the combined entrance and exit 15 clearly determines whether a vehicle leaves or enters the parking area.

The 4 shows a parking space without marked parking spaces with a combined entrance and exit 15 with each for the entrance and exit 11 . 12 executed registration units 10 with car vehicles 16 and the required footprint 18 as well as truck / bus vehicles 16 and required footprint 19 as well as for parking unavailable area 20 , If a vehicle passes the driveway 11 to the parking space, be through the video camera 4 and the laser scanner 6 Image data generated by the appearance of the vehicle. The image data become the computer unit 8th fed. From the image data are by means of the computer unit 8th extracted characteristic features of the vehicle, wherein the extracted characteristic features include the length and the width of the vehicle. Based on the characteristic features is a classification of the vehicle. From the characteristic features of length and width, the footprint required by the vehicle is calculated. The characteristic features, the classification, the required footprint and the time of capture, ie the time at which the vehicle enters the driveway 11 has happened are stored correspondingly and by means of the communication unit 7 to all computer units 8th at the entrances and exits 11 . 12 the parking space and to a central communication and computer unit, not shown. The memory contents of the computer units 8th and the central communication and computer unit are constantly synchronized. If a vehicle passes the exit 12 , as already happened 1 described, proceed. The occupancy of the parking space is calculated as area occupancy by a summation of the required parking spaces 18 and 19 all in the memory of the computer unit 8th stored vehicles plus the area not available for parking 20 he follows.

LIST OF REFERENCE NUMBERS

1

road plane

2

foundation

3

Mast or passage gate

4

video camera

5

bracket

6

laser scanner

7

communication unit

8th

computer unit

9

Highway

10

acquisition unit

11

driveway

12

exit

13

marked parking spaces

14

marked truck parking spaces

15

Combined access and exit

16

Vehicle BUS / truck

17

Vehicle car

18

required footprint vehicle car

19

required floor space vehicle bus / truck

20

area

t

Period

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 60203689 T2 [0005]
• DE 102005034372 A1 [0006]
• DE 102008029413 A1 [0006]
• DE 102008012991 A1 [0006]
• DE 112006003370 T5 [0007]
• DE 102008056786 A1 [0008]

Cited non-patent literature

• Wolfgang Damm (Institute of Transport Telematics University of Dresden, "Project Stand Truck Parking Guidance Dresden", 07.12.2010, http://www.fh-erfurt.de/fhe/index.phpeID=tx_nawsecuredl&u=0&file=fileadmin/Material/Institut/Verkehr_Raum /Download/Projekte/2010/2010.12.07_Damm_Lkw-Parkleitsystem_Dresden.pdf&t=1323354531&hash=d5c7b8e56a9060eb31d812df32a7683c [0006]
• Martin Pozybill (Ministry of environment, nature protection and traffic Baden-Wuerttemberg, "A 8 LKW-Parkstandsanzeige" 07.12.2010, http://www.fh-erfurt.de/fhe/index.php?elD=tx_nawsecuredl&u=0&file=fileadmin /Material/Institut/Verkehr_Raum/Download/Projekte/2010/2010.12.07_Pozybill_A8_Lab-Varkstandsanzeige.pdf & t = 1323354531 & hash = eed248f6cc9c8c0ebfc42731ab18036d [0006]

Claims (9)

Method for determining the occupancy of a parking space with at least one entrance and exit ( 11 . 12 ), wherein - image data is acquired from a vehicle traveling into the parking space and characteristic features of the external appearance of the vehicle are determined by means of feature extraction from the acquired image data, - the time of passing of the parking space by the vehicle and the determined characteristic features of the vehicle as corresponding values stored in a memory unit, the time of the exit as well as image data from a vehicle exiting the parking space detected and characteristic features of the external appearance of the vehicle by means of feature extraction from the captured image data are determined, the characteristic features of a vehicle extending from the parking space with the characteristic features of the vehicles stored in the storage unit are compared, - when identifying a vehicle stored in the storage unit as a departing vehicle, the identified vehicle from the S the parking time of each vehicle stored in the storage unit is calculated and a vehicle stored in the storage unit for more than a predetermined parking time is removed from the storage unit, and the parking space is calculated by summing the vehicles stored in the storage unit. Method according to Claim 1, characterized in that the acquisition of the image data by means of a monocular camera ( 4 ), Stereo / multi-stereo camera system, areal distance sensor based on modulated or structured light, light field camera and / or laser scanner ( 6 ) he follows. A method according to claim 1 or 2, characterized in that as characteristic features points, lines and / or surfaces and their position are extracted. A method according to claim 1 to 3, characterized in that as a characteristic feature, the vehicle license plate is extracted. A method according to claim 1 to 4, characterized in that extracted from the captured image data, the geometric dimensions width, length and / or height of the vehicle and calculates the space required for the vehicle footprint and corresponding to the characteristic features and the timing of the driving Parking space is stored by the vehicle. A method according to claim 5, characterized in that a surface occupancy of the parking space is calculated by summing the required parking areas of the parking space occupying vehicles. A method according to claim 1 to 6, characterized in that based on the detected image data of a vehicle driving into the parking space, a classification of the vehicle into a vehicle class and the vehicle class is stored corresponding to the characteristics and the time of driving of the parking space by the vehicle. A method according to claim 7, characterized in that a vehicle class specific occupancy of the parking space is calculated by respective summation of all stored vehicles of a vehicle class. A method according to claim 7 or 8, characterized in that a vehicle class specific area occupancy of the parking space is calculated by respective summation of the required parking areas of all stored vehicles of a vehicle class.

Images