KR102473140B1 - Parking control system using parking robot based on artificial intelligence - Google Patents

Abstract

The present invention relates to a parking control system using a parking robot based on artificial intelligence. According to the present invention, in the parking control system using a parking robot based on artificial intelligence, the parking spot and expected exit time of a vehicle parked in a parking lot are set as input coefficients, and the parking spot of a vehicle entering the parking lot is an output coefficient A learning unit that learns the learning model set to, a database unit that stores the license plate, average parking time, entry time, and expected exit time of the currently parked vehicle for each vehicle, and when a vehicle enters the parking lot, the parking lot entrance A vehicle recognition unit that recognizes the vehicle number and predicts the exit time using a camera installed in the vehicle, determines the parking spot of the currently entered vehicle by applying the number of currently parked vehicles expected to exit at the same time to the learning model It includes a determination unit and a control unit that controls to move the entering vehicle to the determined parking spot using the parking robot.
As described above, according to the present invention, congestion between vehicles entering or leaving the same time zone is eliminated by determining a parking spot in consideration of vehicle information that has entered the parking lot and vehicle information parked in the parking lot through artificial intelligence. In addition, the inconvenience of double parking and the burden of vehicle damage can be reduced by using a parking robot.

Description

Parking control system using parking robot based on artificial intelligence {PARKING CONTROL SYSTEM USING PARKING ROBOT BASED ON ARTIFICIAL INTELLIGENCE}

The present invention relates to a parking control system using a parking robot based on artificial intelligence. It relates to a parking control system using a parking robot based on artificial intelligence to park a corresponding vehicle in a determined parking spot using one or more parking robots.

With the development of the industry and the increase in the number of vehicles, the parking problem of vehicles has also become a serious situation. In particular, due to the ever-increasing number of vehicles, the shortage of parking spaces is more serious not only in downtown areas but also in apartments, general residential areas, and shopping malls without parking facilities such as underground parking lots and parking towers.

Due to the lack of such parking space, so-called double parking, in which parking is overlapped with a vehicle already parked in a parking area, is routinely performed. However, if the driver who parked first needs to get out, the driver of the double-parked vehicle has to come out and move the vehicle, or the driver who parked first pushes and moves the double-parked vehicle, resulting in vehicle damage or personal injury. Various problems such as can occur, which is a cause of disputes between borrowers.

More specifically, if a driver who parked first pushes a double-parked vehicle to move it, there is no external means to properly stop the vehicle after the double-parked vehicle has been pushed and movement has begun, so the force is incorrectly applied and the vehicle collides with another vehicle. A collision or, in severe cases, a human accident may occur.

And, when vehicles use a public parking lot, it is difficult for the borrowers to grasp an empty parking space in real time, and accordingly, there is a problem in that the congestion level in the parking lot increases and the time required to park is increased.

In addition, due to the increasing number of vehicles, the area of a parking spot in a parking lot is very narrow. Therefore, there is a problem that damage to the vehicle such as a moon cock may occur when the owner parks the vehicle and gets in or out of the vehicle.

The background technology of the present invention is disclosed in Korean Patent Publication No. 10-1943628 (published on January 30, 2019).

The technical problem to be achieved by the present invention is to determine a parking spot in consideration of information about a vehicle entering a parking lot and information about a vehicle parked in the parking lot through artificial intelligence, and to park the vehicle in the determined parking spot using one or more parking robots. It is about a parking control system using a parking robot based on artificial intelligence that allows

According to an embodiment of the present invention for achieving this technical problem, in a parking control system using an artificial intelligence-based parking robot, a parking spot and expected exit time of a vehicle parked in a parking lot are set as input coefficients, and A learning unit that learns a learning model that sets the parking spot of an entering vehicle as an output coefficient, a database unit that stores the license plate, average parking time, entry time, and expected exit time of the currently parked vehicle for each vehicle, When a vehicle enters the parking lot, a vehicle recognition unit that recognizes the vehicle number and predicts the exit time using a camera installed at the entrance of the parking lot, and applies the number of currently parked vehicles that are expected to leave at the same time to the learning model A decision unit for determining a parking spot for an entered vehicle, and a control unit for controlling the moving of the entering vehicle to the determined parking spot using a parking robot.

The learning unit may be trained to determine a parking spot of a currently entered vehicle such that a distance from a currently parked vehicle having the same expected exit time zone as an expected exit time of a vehicle currently entering the parking lot is increased.

The learner determines that, when there are one or more previously parked vehicles having the same expected exit time zone as the expected exit time of the entered vehicle, the sum of distance differences between the current positions of the one or more pre-parked vehicles among vacant parking spaces is You can decide on the largest parking spot.

The learning unit determines a parking spot farther from the exit as the average parking time of the entered vehicle is longer when there is no previously parked vehicle having the same expected exit time zone as the expected exit time of the entered vehicle , the shorter the average parking time, the closer the exit to the parking spot.

When there are a plurality of parking robots and each parking robot leaves a vehicle at the same time, a movement line for leaving the parking robot may be learned so that a collision does not occur between the parking robots.

The parking robot is implemented with a smaller height and narrower width than the vehicle pallet for mounting the vehicle, and when there is a vehicle entering and exiting, it can move under the vehicle pallet and move the vehicle pallet on which the vehicle is mounted in orthogonal operation.

As described above, according to the present invention, by determining a parking spot in consideration of vehicle information that has entered the parking lot and vehicle information parked in the parking lot through artificial intelligence, congestion between vehicles entering or leaving the same time zone is eliminated.

In addition, by using a parking robot to park a vehicle in or out of a parking lot, it is possible to reduce the hassle of double parking and the burden of vehicle damage, and to prevent vehicle damage such as a door knock due to a narrow parking space in the parking lot. can

1 is a diagram for explaining a parking control system using a parking robot based on artificial intelligence according to an embodiment of the present invention.
2A is a diagram for explaining a parking robot according to an embodiment of the present invention.
2B is a diagram for explaining a state in which a vehicle is mounted on a parking robot according to an embodiment of the present invention.
3 is a configuration diagram showing the configuration of a parking control system using a parking robot based on artificial intelligence according to an embodiment of the present invention.
4 is a flowchart illustrating a parking control method using an artificial intelligence-based parking control system according to an embodiment of the present invention.
FIG. 5 is a diagram for explaining step S420 of FIG. 4 .
FIG. 6A is a diagram for explaining step S440 of FIG. 4 .
FIG. 6B is a diagram for explaining parking spot determination in step S440 of FIG. 4 .

Hereinafter, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice with reference to the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. In addition, in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a certain component is said to "include", it means that it may further include other components without excluding other components unless otherwise stated.

Then, with reference to the accompanying drawings, an embodiment of the present invention will be described in detail so that those skilled in the art can easily practice it.

Hereinafter, a parking control system 100 using a parking robot 300 based on artificial intelligence according to an embodiment of the present invention will be described using FIGS. 1 to 2B.

1 is a diagram for explaining a parking control system using a parking robot based on artificial intelligence according to an embodiment of the present invention.

As shown in FIG. 1, the parking control system 100 using the parking robot 300 according to an embodiment of the present invention is connected to the camera 200 and the parking robot 300 through a network.

First, the parking control system 100 receives a vehicle number through the camera 200 installed at the entrance of the parking lot, and the camera 200 connects to the network by wire or wirelessly and transmits an image of a vehicle entering the parking lot. can be implemented as

And, the parking robot 300 is implemented as a device for moving a vehicle entering the parking lot to the parking spot determined by the parking control system 100 .

Hereinafter, the parking robot 300 according to an embodiment of the present invention will be described using FIGS. 2A and 2B.

2A is a diagram for explaining a parking robot according to an embodiment of the present invention, and FIG. 2B is a diagram for explaining a state in which a vehicle is mounted on the parking robot according to an embodiment of the present invention.

As shown in FIGS. 2A and 2B , the parking robot 300 is implemented with a lower height and narrower width than the vehicle pallet 310 for mounting the vehicle 400 thereon.

If there is a vehicle 400 entering or exiting the parking lot, the parking robot 300 may move under the vehicle pallet 310, lift the vehicle pallet 310 on which the vehicle 400 is mounted, and park the vehicle. Then, the parking robot 300 lowers its height and comes out from under the vehicle pallet 310. At this time, the parking robot 300 may move by orthogonally driving.

That is, the parking control system 100 receives the number of the vehicle 400 entering the parking lot through the camera 200 installed at the entrance of the parking lot, determines the parking spot by using the information of the vehicle 400, and parks The robot 300 controls the vehicle 400 to be parked in the determined parking spot.

Hereinafter, the configuration of the parking control system 100 using the parking robot 300 according to an embodiment of the present invention will be described with reference to FIG. 3 .

3 is a configuration diagram showing the configuration of a parking control system using a parking robot based on artificial intelligence according to an embodiment of the present invention.

As shown in FIG. 3, the parking control system 100 using a parking robot according to an embodiment of the present invention includes a learning unit 110, a database unit 120, a vehicle recognition unit 130, a determination unit 140, and A control unit 150 is included.

First, the learning unit 110 sets a parking spot of a vehicle parked in a parking lot and an expected exit time as input coefficients and trains a learning model that sets a parking spot of a vehicle entering the parking lot as an output coefficient.

At this time, the learning unit 110 trains the learning model to determine the parking spot of the vehicle entering the parking lot so that the distance from the parked vehicle having the same expected exit time zone as the expected exit time of the vehicle entered the parking lot is increased.

Next, the database unit 120 stores license plates of vehicles parked in the parking lot, average parking time, entry time, and expected exit time for each vehicle.

At this time, the database unit 120 stores the average parking time history for vehicles that have previously used the corresponding parking lot, and uses the vehicle number parked in the corresponding parking lot, the entry time and the average parking time previously used to determine the corresponding parking time. Predict and store the expected departure time of the vehicle.

Next, when the vehicle enters the parking lot, the vehicle recognition unit 130 recognizes the vehicle number using a camera installed at the entrance of the parking lot and predicts the departure time of the vehicle.

Here, the vehicle recognition unit 130 recognizes the license plate of the vehicle through the image taken by the camera, records the parking lot entry time of the vehicle, and uses the average parking time of the corresponding vehicle number stored in the database unit 120 to Estimate the estimated time of departure of the vehicle.

Next, the determination unit 140 determines the parking spot of the entered vehicle by applying the number of vehicles expected to exit in the same time period to the learning model.

At this time, the decision unit 140 applies the learning model learned using artificial intelligence to learn to determine the parking spot of a vehicle that has entered the parking lot so as to move away from a vehicle parked in the same parking lot with the currently entered vehicle and the expected departure time. do.

Next, the control unit 150 uses the parking robot to control the entry vehicle to move to the determined parking spot.

At this time, when there are a plurality of parking robots 300 entering and exiting the parking lot, the controller 150 may control the moving lines of the plurality of parking robots 300 not to overlap each other.

Hereinafter, a parking control method using a parking control system will be described using FIGS. 4 to 6B.

4 is a flowchart illustrating a parking control method using a parking control system according to an embodiment of the present invention.

First, the learning unit 110 learns a learning model that outputs a parking spot of a vehicle entering the parking lot by inputting a parking spot of a vehicle already parked in the parking lot and an expected exit time (S410).

At this time, if there is a parked vehicle having the same expected exit time zone as the expected exit time of the vehicle entering the parking lot, the learning unit 110 determines that the parking spot of the vehicle currently entering the parking lot has the same expected exit time zone. Train the learning model to determine a point far from

In more detail, if there is one or more previously parked vehicles having the same expected exit time zone as the expected exit time of the vehicle currently entering the parking lot, the learning unit 110 determines whether the parking spot of the vehicle currently entering the parking lot is empty. The learning model is trained so that the sum of the difference between the current position of one or more pre-parked vehicles and the distance among seats is the largest.

In addition, if there is no previously parked vehicle having the same expected exit time zone as the expected exit time of the vehicle entering the parking lot, the learning unit 110 determines the location farther from the exit as the average parking time of the vehicle entering the parking lot is longer. A parking spot is determined, and the learning model is trained to determine a parking spot closer to the exit as the average parking time is shorter.

Next, the database unit 120 stores license plates, average parking time, entry time, and expected exit time for each vehicle parked in the corresponding parking lot (S420).

At this time, the database unit 120 stores the vehicle number and average parking time of the vehicle with respect to a vehicle having a history of using the corresponding parking lot in the past.

In addition, the database unit 120 predicts and stores the expected exit time from the entry time and pre-stored average parking time according to the vehicle number currently using the corresponding parking lot.

Hereinafter, vehicles parked in the corresponding parking lot stored in the database unit 120 will be described using FIG. 5 .

FIG. 5 is a diagram for explaining step S420 of FIG. 4 .

As shown in FIG. 5, assuming that the average parking time of the vehicle number "23 or 4355" that has previously used the parking lot is 9 hours, the "23 or 4355" vehicle enters at 9:00 am and arrives at 6:00 pm , and vehicle “23 or 4355” is currently parked at location A-7.

In addition, the average parking time of vehicles with vehicle number "11th 9874" who have used the parking lot before is 1 hour, and the entry time is 12:00 PM, so the expected departure time is 1:00 PM, and the current "11th 9874" vehicle is parked at C-1, and the average parking time of the vehicle number "93 1065" is 5 hours, and the entry time is 1:00 PM, so the expected departure time is 6:00 PM, and the current vehicle "93 1065" is parked at B-1, and the average parking time of the vehicle with license plate number "83 bar 5127" is 3 hours, and the entry time is 2:00 pm, so the expected exit time is 5:00 pm, and the current vehicle "83 bar 5127" is parked at D-4.

That is, the database unit 120 stores the average parking time, entry time, expected exit time, and vehicle location of each vehicle number for each vehicle parked in the corresponding parking lot.

Next, when a vehicle enters the parking lot, the vehicle recognition unit 130 recognizes the license plate number of the currently entered vehicle through a camera installed at the entrance and predicts an expected exit time (S430).

At this time, the vehicle recognition unit 130 may recognize the number of the vehicle through the video and image of the entering and exiting vehicle captured by the camera, and simultaneously record the time when the recognized vehicle entered the parking lot.

In addition, the vehicle recognition unit 130 predicts the expected exit time by using the average parking time of the corresponding vehicle number in the parking lot pre-stored in the database unit 120.

Next, the determination unit 140 determines the parking spot of the currently entered vehicle by applying the number of vehicles expected to exit in the same time period to the learning model (S440).

At this time, if the occupancy rate of vehicles parked in the parking lot is less than 50% of the total available parking spaces, the determining unit 140 may randomly assign seats without using a learning model based on artificial intelligence. .

In addition, the determination unit 140 may predict the number of vehicles to be taken out for each time period through the pre-stored expected exit time of the vehicles parked in the corresponding parking lot from the database unit 120 .

For example, assuming that the expected departure time of a currently entering vehicle is 6:00 and there is a parked vehicle leaving within 6:05, the determining unit 140 determines the previously parked vehicle for the currently entering vehicle's parking spot. The parking spot is determined through a learning model to park far away from the vehicle.

At this time, the range of the expected departure time zone may be set to 5 minutes, and the set minute interval may be changed.

For example, assuming that the expected departure time of one vehicle is 1:00 PM, the range of the same expected departure time zone may include other vehicles leaving the vehicle between 1:00 PM and 1:05 PM.

In addition, the determination unit 140 determines a parking spot through a learning model so that a vehicle currently entering the parking lot is parked at a point where the sum of distance differences from vehicles having the same expected exit time is the largest.

Hereinafter, parking spot determination will be described using FIGS. 6A and 6B.

FIG. 6A is a diagram for explaining step S440 of FIG. 4 , and FIG. 6B is a diagram for explaining parking spot determination in step S440 of FIG. 4 .

As shown in FIG. 6A, a vehicle 500a currently entering a parking lot is placed on a vehicle pallet 310, and a vehicle 500b having the same expected exit time zone as the previously parked vehicle 500a, 500c, 500d) Assuming that there are three units, the determining unit 140 determines the parking spot of the vehicle 500a entered from among ⓐ to ⓙ that are vacant in the corresponding parking lot.

Here, in the parking lot of FIG. 6A, since the occupancy rate of parked vehicles is 50% or more of the total available parking spots, the determining unit 140 may determine the parking spot using an artificial intelligence-based learning model.

In addition, the determination unit 140 learns to park the vehicle 500a at the parking spot where the sum of the difference between the location and the distance of the vehicles 500b, 500c, and 500d having the same expected departure time slot is the largest with respect to the parking spot of the vehicle 500a entering the parking lot. It is determined through the model, and the entered vehicle 500a may be parked in one of the empty seats ⓐ to ⓙ.

As shown in FIG. 6B , the determination unit 140 determines the ⓐ spot where the sum of the difference between the positions and distances of the vehicles 500b, 500c, and 500d having the same expected departure time zone with respect to the parked spot of the entered vehicle 500a is the largest. The parking spot is determined through the learning model so that it is parked in the parking lot.

And, if there is no previously parked vehicle having the same expected exit time zone as the expected exit time of the vehicle currently entering the parking lot, the determination unit 140 determines the parking spot according to the average parking time.

For example, assuming that the time the vehicle entered the parking lot is 1 o'clock, the expected exit time zone is 3 o'clock, and there is no parked vehicle having the same expected exit time zone as the vehicle entering the parking lot, the determination unit 140 Among vehicles parked in a parking lot, it learns to determine a parking spot farther from the exit compared to vehicles with an average parking time of 1 hour.

model can be trained. In addition, the determination unit 140 determines a parking spot through a learning model so as to determine a parking spot at a location close to the exit compared to a vehicle having an average parking time of 5 hours among vehicles parked in the parking lot.

Next, the control unit 150 controls the parking robot to move the vehicle entering the parking lot to the parking spot determined through the learning model from the determination unit 140 (S450).

That is, when there are a plurality of parking robots 300 and each parking robot 300 unloads one or more previously parked vehicles having the same expected exit time zone, the control unit 150 determines that the movement lines between the parking robots 300 overlap. Control not to collide.

As described above, according to an embodiment of the present invention, congestion between vehicles entering or exiting the same time zone is eliminated by determining a parking spot in consideration of vehicle information that has entered the parking lot and vehicle information parked in the parking lot through artificial intelligence. In addition, by using a parking robot to park a vehicle in or out of a parking lot, it is possible to reduce the hassle of double parking and the burden of vehicle damage, and to prevent vehicle damage such as a door knock due to a narrow parking space in the parking lot. can

Although the present invention has been described with reference to the embodiments shown in the drawings, this is only exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical scope of protection of the present invention should be determined by the technical spirit of the appended claims.

100: parking control system, 110: database unit,
120: vehicle recognition unit, 130: learning unit,
140: control unit, 200: camera,
300: parking robot, 310: vehicle pallet,
400: vehicle, 500a: parking lot entry vehicle
500b, 500c, 500d: parked vehicles with the same expected exit time

Claims (6)

In the parking control system using a parking robot based on artificial intelligence,
A learning unit for learning a learning model by setting the parking spot of a vehicle parked in the parking lot and the expected time to exit the parking lot as input coefficients and setting the parking spot of a vehicle entering the parking lot as an output coefficient;
A database unit for storing vehicle license plates, average parking time, entry time, and expected exit time for each vehicle of the currently parked vehicle for the corresponding parking lot;
When a vehicle enters the parking lot, a vehicle recognition unit that recognizes the license plate number of the entered vehicle using a camera installed at the entrance of the parking lot and predicts the exit time;
A decision unit determining the parking spot of the currently entered vehicle by applying the number of currently parked vehicles expected to leave at the same time to the learning model; and
And a control unit for controlling to move the entry vehicle to the determined parking spot using a parking robot,
The learning unit,
A parking control system that is learned to determine the parking spot of a vehicle entering the parking lot so that the distance from the parked vehicle having the same expected exit time zone as the expected exit time of the vehicle entered the parking lot is farther away. delete According to claim 1,
The learning unit,
If there are one or more previously parked vehicles having the same expected exit time zone as the expected exit time of the entered vehicle,
A parking control system for determining a parking spot having the largest sum of distance differences from the positions of one or more previously parked vehicles among vacant parking spots. According to claim 1,
The learning unit,
If there is no previously parked vehicle having the same expected exit time zone as the expected exit time of the entered vehicle,
The parking control system for determining a parking spot at a location farther from the exit as the average parking time of the entered vehicle is longer, and determining a parking spot at a location closer to the exit as the average parking time is shorter. According to claim 1,
The learning unit,
When there are a plurality of parking robots and each parking robot leaves the vehicle at the same time,
A parking control system that learns a movement line for exiting the parking robot so that a collision does not occur between each parking robot. According to claim 1,
The parking robot,
It is implemented with a smaller height and narrower width than a vehicle pallet for mounting a vehicle,
Parking control system that moves under the vehicle pallet when there is a vehicle entering and exiting and moves the vehicle pallet with the vehicle in orthogonal operation.

Images