Underground Parking Space Identification, Parking Guidance Method and System

Abstract

The present invention discloses a method and system for identifying parking spaces and guiding parking in an underground parking lot. By performing dot matrix processing on the underground parking lot area, the height change information of each dot at a preset sampling time #imgabs0# is collected; based on the fact that the height change information of all dots within a parking space is the same, the parking spaces in the underground parking lot area are determined, and the parking spaces in the underground parking lot can be quickly and effectively identified without being interfered by light. In addition, the present invention calculates the congestion parameter of each parking space; and based on the channel congestion parameter, the optimal parking space with the least congestion is selected, and the user is automatically guided to drive the vehicle into this parking space, which can effectively relieve the congestion in the underground parking lot.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a method and system for identifying parking spaces in underground parking lots and providing parking guidance. Background Art

[0002] In the prior art, underground parking spaces are generally identified through image acquisition. This method is highly dependent on light and is easily blocked by objects, resulting in inaccurate identification.

[0003] In addition, due to the proficiency of driving skills, drivers always fail to control parking properly. When they see a parking space, they always try to park in the current position impatiently, but ignore the difficulty of parking in the parking space itself, resulting in occupying the public passage for too long when parking, causing congestion in the underground parking lot.

[0004] Although security guards and other on-duty personnel are arranged in some complex parking spaces to use their experience to guide vehicles to park in congested parking spaces, this method is relatively effective, but it requires staff to work for a long time in the harsh environment of underground parking lots, and the labor cost is high. Relying on manual labor cannot cover all parking spaces. Therefore, how to guide drivers to quickly enter parking spaces and reduce the occupation time of public passages to alleviate the congestion of underground parking lots has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0005] The present invention provides a parking space identification and parking guidance method and system for an underground parking lot, which are used to solve the technical problems that the existing technology cannot effectively identify parking spaces and causes congestion due to parking difficulties.

[0006] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0007] A method for identifying parking spaces in an underground parking lot comprises the following steps:

[0008] The underground parking lot area is processed into a dot matrix, and the height change information of each matrix point at the preset sampling time Tj is collected;

[0009] The parking spaces in the underground parking lot area are determined based on the fact that the height change information of all the array points in the parking spaces is the same.

[0010] Preferably, the height change information of the array point position includes the change duration Ts of the last continuous height change of the array point, and determining the parking space in the underground parking area specifically includes:

[0011] Determine whether each array point is a parking point according to the change time Ts of each array point;

[0012] Among the parking points in the screening, the parking points with the same change duration Ts and adjacent positions are divided into the same point set;

[0013] For the parking points in any same point set, find the moment(s) of height change of the parking points in the same point set one or more times before. If the parking points are adjacent and the moments of height change are the same, the area formed by the adjacent parking points is a parking space.

[0014] Preferably, to determine whether each array point is a parking point according to the change duration Ts after the last height change of each array point, specifically including:

[0015] If the change duration Ts after the last height change of the array point satisfies the following formula, then it is determined that the array point is a parking point, otherwise it is a passage point:

[0016] Ts / Tj>Ty, where Ty is a preset threshold.

[0017] The acquisition of the height change information of each array point within the preset sampling time Tj is realized by a distance sensor, and the distance sensor includes a laser sensor and / or an ultrasonic sensor.

[0018] A method for guiding parking in an underground parking lot includes the following steps:

[0019] Use the above-mentioned method for identifying parking spaces in an underground parking lot to determine the positions of parking spaces in the underground parking lot;

[0020] Calculate the congestion parameters of each parking space;

[0021] And select the optimal parking space with the least congestion based on the channel congestion parameters, and guide the user to drive the vehicle into this parking space.

[0022] Preferably, the congestion parameters include the average parking time and the congestion degree; calculating the congestion parameters of each parking space includes:

[0023] For any one parking space, the following steps are performed:

[0024] Obtain the parking times Tp1, Tp2,..., Tpn of the parking space for n times of parking, and calculate its average parking time; where Tp1 represents the parking time of the first parking, Tp2 represents the parking time of the second parking, and Tpn represents the parking time of the nth parking;

[0025] Obtain the height change information of the channel points within a preset range near the parking space within a preset retrospective time before each parking, and construct an alternative channel set before each parking Among them, is the height change information of the first channel point pd before the xth parking 1 of, is the second passage point pd before the x-th parking 2 of the height change information, is the m-th passage point pd before the x-th parking m of the height change information;

[0026] Find the intersection of the alternative passage sets corresponding to all parking times of the parking space, and use the passage points corresponding to the intersection as the congested passage points of each parking space to construct the congested passage point set Dpc of the parking space;

[0027] Count the number of congested passage points in the congested passage set of each parking space respectively as the standard congestion degree of each parking space;

[0028] Real-time monitor the height changes of the congested passage points in the congested passage set of each parking space, and count the number of real-time congested passage points with non-zero height changes as the real-time congestion degree. Compare the real-time congestion degree with the standard congestion degree, and judge whether there is a vehicle near the parking space according to the comparison result;

[0029] When there is a vehicle, select the parking space with the shortest average parking time and that is idle as the optimal parking space, and guide the vehicle to park in the optimal parking space.

[0030] Preferably, obtain the parking time of the parking space for n times through the following rules:

[0031] For any parking point in the parking space where there are continuous height changes, use the time period from the start of the height change of the first parking point to the end of the height change of the last parking point in this time period as the parking time;

[0032] Preferably, the passage points within the preset range near the parking space are specifically:

[0033] During the parking process of the parking space, the dot matrix with height changes appears within the preset range of the parking space.

[0034] Preferably, compare the real-time congestion degree with the standard congestion degree, and judge whether there is a vehicle near the parking space according to the comparison result through the following formula:

[0035] When the real-time congestion degree / standard congestion degree > preset value, judge that there is a vehicle in the parking space.

[0036] A computer system includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0037] The present invention has the following beneficial effects:

[0038] 1. In the present invention, the underground parking lot area is rasterized, and the height change information of each grid point at the preset sampling time Tj is collected. Based on the fact that the height change information of all grid points within a parking space is the same, the parking spaces in the underground parking lot area are determined. Without being interfered by light, the parking spaces in the underground parking lot can be quickly and effectively identified. In addition, the present invention calculates the congestion parameter of each parking space, selects the optimal parking space with the least congestion based on the channel congestion parameter, and automatically guides the user to drive the vehicle into this parking space, which can effectively relieve the congestion in the underground parking lot.

[0039] In addition to the purposes, features and advantages described above, the present invention has other purposes, features and advantages. The following will refer to the accompanying drawings to further elaborate on the present invention in detail. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings forming a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0041] Figure 1 is the installation layout diagram of the distance sensors in the embodiment of the present invention;

[0042] Figure 2 is the schematic diagram of the parking point set in the embodiment of the present invention;

[0043] Figure 3 is the flowchart of the method for identifying parking spaces in the underground parking lot in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The following will elaborate on the embodiments of the present invention in detail with reference to the accompanying drawings. However, the present invention can be implemented in many different ways defined and covered by the claims.

[0045] Embodiment 1:

[0046] As shown in this embodiment Figure 3 a method for identifying parking spaces in an underground parking lot is disclosed, including the following steps:

[0047] S1. Rasterize the underground parking lot area, and collect the height change information of each grid point at the preset sampling time Tj;

[0048] In this embodiment, the collection of the height change information of each grid point at the preset sampling time Tj is realized by a distance sensor, and the distance sensor includes a laser sensor and / or an ultrasonic sensor. As shown in Figure 1 a distance sensor is a device that detects whether a point is occupied from top to bottom, and each device has its own device ID.

[0049] The height change information of the array point positions includes the change duration Ts of the last continuous height change of the array point:

[0050] Within a certain time Tj, data of each array point is collected. If there is no change, it is set to the zero scale. After the measured height changes, the duration Ts during which the change is maintained is recorded. If there is another change, Ts is reset to 0 and the timing restarts.

[0051] S2. Determine the parking spaces in the underground parking lot area based on the same height change information of all array points within the parking space.

[0052] S21. Determine whether each array point is a parking point according to the change duration Ts of each array point;

[0053] If the change duration Ts after the last height change of the array point satisfies the following formula, then the array point is determined to be a parking point; otherwise, it is a passage point:

[0054] Ts / Tj>Ty, where Ty is a preset threshold and is a constant.

[0055] When Ts / Tj>Ty (constant, set threshold), then the point is determined to be a parking point Pt; otherwise, it is a passage point Pd. The remaining points Pw with no change are the useless areas of the parking lot. During installation, for the detection device above the parking space, an indicator light is added. The indicator light can change color under the control of the background system. The device with the indicator light can detect height changes and can also be controlled to blink according to the ID number.

[0056] S22. Among the parking points selected, divide the parking points with the same change duration Ts and adjacent positions into the same point set;

[0057] For any parking points in the same point set, find the last or multiple height change moments of the parking points in the same point set. If the parking points are adjacent and the height change moments are the same, the area formed by the adjacent parking points is the parking space.

[0058] Such as Figure 2 , from the existing Pt points (Pt1, Pt2,..., Ptn), select Pt1, traverse the remaining Pt2 - Ptn, and select the Pt points with the same time (within a certain threshold range, which can be understood as the changes of the cars in a parking space stopping simultaneously after parking). Fit them into a point set that may be a parking area. Process the remaining Pt points in the same way to form multiple point sets, and multiple parking point set areas can be obtained.

[0059] Since there may be a situation where multiple parking spaces are parked simultaneously at the same time, each of these point sets may have multiple parking areas. Track the zeroing time for each Pt in each point set, that is, restore the measured data to its original value. When a batch of Pt points are zeroed simultaneously, it can be understood that the vehicle in that parking space has left, and then the Pt point set of this part is the real parking space. Through the special device number ID with the function of indicating lamp in the Pt point set, this point can be used for system function indication.

[0060] In addition, in this embodiment, a parking guidance method for an underground parking lot is also disclosed, including the following steps:

[0061] Use the above-mentioned parking space recognition method for the underground parking lot to determine the parking space positions in the underground parking lot;

[0062] Calculate the congestion parameters for each parking space;

[0063] The congestion parameters include the average parking time and the congestion degree; calculating the congestion parameters for each parking space includes:

[0064] For any one parking space, the following steps are all executed:

[0065] Obtain the parking times Tp1, Tp2,..., Tpn of the n parking times of the parking space, and calculate its average parking time; where, Tp1 represents the parking time of the first parking, Tp2 represents the parking time of the second parking, and Tpn represents the parking time of the nth parking;

[0066] After calculating the point set of each parking space, trigger the Ts state according to a certain point of the point set. When the parking is finally completed, that is, when all the points Ts in the point set no longer change, the single parking time Tp1 of this parking space can be calculated. Repeat this method multiple times to obtain the average parking times Tp2, Tp3... Tpn, etc. of this parking space. Obtain the average parking time Tpp of each parking space. This parking time can be dynamically updated according to the duration of each parking. For example, select the average time of the recent 10 parkings as the calculation data.

[0067] Among them, to obtain the parking times of the n parking times of the parking space, identify the Pt array of each parking space, because each parking space is composed of multiple Pt points. This Tp time is calculated from the time when a certain point in the Pt array starts to change until each point in the array stops changing (the vehicle stops stably). Specifically, it can be implemented through the following rules:

[0068] For any parking point in the parking space where there is a continuous period of height change, the duration between the start of the height change of the first parking point and the stop of the height change of the last parking point in this period is used as the parking time.

[0069] Obtain the height change information of the channel points within a preset range near the parking space during the preset retrospective time before each parking, and construct an alternative channel set before each parking. Among them, is the height change information of the first channel point pd before the x-th parking. 1 is the height change information of the second channel point pd before the x-th parking. 2 is the height change information of the m-th channel point pd before the x-th parking. m

[0070] Among them, the channel points within the preset range near the parking space are specifically: the dot matrix with height changes within the preset range of the parking space during the parking process of the parking space.

[0071] Assume that during the parking process of the parking space 1 (Pt) point set, there are occupancy changes in other channel points Pd. That is, it can be judged that during the parking of parking space 1, it has an impact on the remaining channel points. However, it is also possible that these points are triggered by other parking spaces. Therefore, we adopt the method of taking the AND of the Pd blocking arrays triggered by each parking of parking space 1, and the set of channel area points Dpc that are always blocked by this parking space can be judged.

[0072] Find the intersection of the alternative channel sets corresponding to all parkings of the parking space, and use the channel points corresponding to the intersection as the congested channel points of each parking space to construct the congested channel point set Dpc of the parking space.

[0073] Count the number of congested channel points in the congested channel set of each parking space respectively, and use it as the standard congestion degree Dpp of each parking space.

[0074] During the time periods Tp1, Tp2... Tpn of each parking of the parking space, the nearby Pd points will all be in a non-zero state (occupied by a vehicle) due to the influence of the vehicle in front of it parking. Within the Tp1 time period + backward trace TpQ = 10 seconds (constant, time adjustable), count the set of all Pd non-zero state points, and then count the point sets of the Tp2~Tpn time. Take the AND of the point sets in the recent period of time, and the Pd blocking point set Dpc that is necessarily caused by the parking of this parking space can be obtained. Sum the number of point sets Dpp. The larger the value, the greater the impact on the overall traffic. Each parking space has a blocking point set Dpc and point set Dpp during parking.

[0075] Real-time monitor the height changes of the congested channel points in the congested channel set of each parking space, and count the number of real-time congested channel points with non-zero height changes as the real-time congestion degree.

[0076] And based on the channel congestion parameters, select the optimal parking space with the least congestion, and guide the user to drive the vehicle into this parking space.

[0077] Compare the real-time congestion level with the standard congestion level, and determine whether there is a vehicle near the parking space according to the comparison result; when the real-time congestion level / standard congestion level > a preset value, it is determined that there is a vehicle in the parking space.

[0078] When there is a vehicle, select the parking space with the shortest average parking time and that is idle as the optimal parking space, and guide the vehicle to park in the optimal parking space.

[0079] In actual operation, the background system traverses all Pd points to detect their non-zero changes. When a non-zero change occurs, retrieve which point sets Dpc this Pd point belongs to. When the ratio of the non-zero positions of the Pd points in the Dpc point set to Dpp is greater than a certain threshold Dy, it can be determined that a vehicle has entered the waiting area of a certain parking space. Due to the backtracking characteristic of the TpQ value, there may be multiple parking spaces that can be parked from this area. Select the area that has not been parked and has the smallest Tpp among the above-mentioned parking spaces, and prompt the vehicle to park in by means of slow flashing of the indicator light.

[0080] In addition, a computer system is also disclosed, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0081] In summary, the present invention does not need to perform exploration and calibration on the parking lot. For example, self-modeling can complete the calculation of parking spaces and the calculation method of nearby roads, which is convenient for construction.

[0082] In addition, the present invention automatically calculates the parking spaces that can be parked around when a vehicle passes through a certain area, and guides it into the parking space that is most convenient for parking.

[0083] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An underground parking lot parking space recognition method, characterized in that, It includes the following steps: Perform dot matrix processing on the underground parking lot area, and collect the height change information of each dot at the preset sampling time Tj; Determine the parking spaces in the underground parking lot area based on the same height change information of all dots within the parking space; The height change information of the dot position includes the change duration Ts of the last continuous height change of the dot. Determining the parking spaces in the underground parking lot area specifically includes: Judge whether each dot is a parking point according to the change duration Ts of each dot; Among the parking points being screened, divide the parking points with the same change duration Ts and adjacent positions into the same point set; For the parking points in any same point set, find the previous or multiple height change moments of the parking points in the same point set. If the parking points are adjacent and the height change moments are the same, the area formed by the adjacent parking points is the parking space; Among them, the change duration Ts of the last continuous height change of the dot is obtained through the following steps: Within a certain time Tj, collect the data of each dot. If there is no change, it is at zero scale. When the measured height changes, record the duration Ts during which it maintains this change. If there is another change, Ts is reset to 0 and re-timed; Judge whether each dot is a parking point according to the change duration Ts of the last continuous height change of each dot, specifically including: If the change duration Ts of the last continuous height change of the dot satisfies the following formula, then judge that the dot is a parking point, otherwise it is a passage point: Ts / Tj>Ty, where Ty is a preset threshold.

2. The method for identifying parking spaces in an underground parking lot according to claim 1, wherein, The acquisition of the height change information of each dot at the preset sampling time Tj is realized by a distance sensor, and the distance sensor includes a laser sensor and / or an ultrasonic sensor.

3. A method for guiding parking in an underground parking lot, characterized in that, It includes the following steps: Use the underground parking lot parking space recognition method described in any one of claims 1-2 to determine the parking space position of the underground parking lot; Calculate the congestion parameter of each parking space; And select the optimal parking space with the least congestion based on the congestion parameter, and guide the user to drive the vehicle into this parking space.

4. The method for guiding parking in an underground parking lot according to claim 3, wherein The congestion parameter includes the average parking time and the congestion degree; calculating the congestion parameter of each parking space includes: For any one parking space, the following steps are all executed: Obtain the parking times Tp1, Tp2,..., Tpn of the parking space for n times of parking, and calculate its average parking time; where Tp1 represents the parking time of the first parking, Tp2 represents the parking time of the second parking, and Tpn represents the parking time of the nth parking; Obtain the height change information of the channel points within a preset range near the parking space during the preset retrospective time before each parking, and construct an alternative channel set before each parking x = 1, 2,..., n, where is the height change information of the first channel point pd before the x-th parking 1 ; is the height change information of the second channel point pd before the x-th parking 2 ; is the height change information of the m-th channel point pd before the x-th parking m ; Find the intersection of the alternative channel sets corresponding to all parkings of the parking space, and use the channel points corresponding to the intersection as the congestion channel points of each parking space to construct the congestion channel point set Dpc of the parking space; Count the number of congestion channel points in the congestion channel set of each parking space respectively as the standard congestion degree of each parking space; Real-time monitor the height change of the congestion channel points in the congestion channel set of each parking space, and count the number of real-time congestion channel points with non-zero height change as the real-time congestion degree. Compare the real-time congestion degree with the standard congestion degree, and judge whether there is a vehicle near the parking space according to the comparison result; When there is a vehicle, select the parking space with the shortest average parking time and that is idle as the optimal parking space, and guide the vehicle to park in the optimal parking space.

5. The underground parking lot parking guidance method according to claim 4, wherein, Obtain the parking time of the vehicle parking in the parking space n times, which is achieved by the following rules: For any period in which there are continuous height changes at the parking points within a parking space, the duration from the start of the height change at the first parking point to the end of the height change at the last parking point within this period is used as the parking time.

6. The underground parking lot parking guidance method according to claim 5, characterized in that, The passage points within a preset range near the parking space are specifically: During the parking process in the parking space, a dot matrix with height changes appears within the preset range of the parking space.

7. The method for guiding parking in an underground parking lot according to claim 4, characterized in that, Compare the real-time congestion degree with the standard congestion degree, and judge whether there is a vehicle near the parking space according to the comparison result, which is achieved by the following formula: When the real-time congestion degree / standard congestion degree > preset value, it is judged that there is a vehicle near the parking space.

8. A computer system, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of any one of the methods described in claims 1 to 7 above.

Images