CN107992036B

CN107992036B - Method and device for planning vehicle access path in intelligent parking garage and storage medium

Info

Publication number: CN107992036B
Application number: CN201711155275.2A
Authority: CN
Inventors: 范衠; 姚利; 谢红辉; 朱贵杰; 李冲; 王宇鹏
Original assignee: Shantou University
Current assignee: Shantou University
Priority date: 2017-11-20
Filing date: 2017-11-20
Publication date: 2021-03-16
Anticipated expiration: 2037-11-20
Also published as: CN107992036A

Abstract

The embodiment of the invention discloses a method for planning a route of a vehicle stored and taken in an intelligent parking garage, which provides a modeling mode for the environment of the intelligent parking garage, realizes the route planning problem of an AGV by combining morphological expansion operation of image processing on the basis of the modeling, and solves the problems of deadlock, collision conflict and the like in the route planning process based on a dynamic time window method. The problem of environment modeling of the intelligent parking garage is solved, a map is not required to be constructed by using a sensor, and the map similar to the environment of the parking garage can be manually generated and used for AGV work. The embodiment of the invention also provides a device and a storage medium for planning the access vehicle path in the intelligent parking garage. The system has the advantages of high vehicle access automation degree, high cost performance, high safety and reliability and the like.

Description

Method and device for planning vehicle access path in intelligent parking garage and storage medium

Technical Field

The invention relates to the technical field of parking and taking vehicles in an intelligent parking lot, in particular to a path planning method, a device and a storage medium for AGV vehicle access in an intelligent parking garage.

Background

Along with rapid development of science and technology and economy, continuous expansion of urban scale, rapid increase of urban population density, rapid increase of automobile holding quantity, insufficient supply of urban parking spaces and the like, the problems of urban traffic jam and difficult parking spaces are increasingly serious. Under this background, how to use limited land resources to the maximum extent to provide more convenient services for city development, and provide necessary parking spaces for increasing automobiles is a problem to be solved by many current large-city planners, and a mobile intelligent parking garage based on agvs (automated Guided vehicle) is also in force, and can well solve the problem of the parking space, save large land resources, improve parking efficiency, reduce parking accidents, and the like, and many companies are also currently engaged in research and development of related technologies.

At present, the main technology is a guidance technology, for example, a magnetic guidance method is adopted for AGV navigation and path planning, and a certain track and a magnetic stripe of a route need to be set in a parking lot. Or the laser radar is adopted to construct the indoor map to provide the map for the AGV, and the two methods need great investment and consumption.

Disclosure of Invention

The embodiment of the invention aims to solve the technical problem of providing a method, a device and a storage medium for planning the route of a vehicle to be stored and taken in an intelligent parking garage, which can effectively avoid competition and conflict of route resources and smoothly finish the tasks of storing, taking and parking the vehicle in a short time.

In order to solve the technical problem, the embodiment of the invention provides a method for planning a route of a car stored and taken in an intelligent parking garage, which comprises the following steps:

s1: modeling each element in the parking garage by a gray value image method to obtain a binary grid map;

s2: performing morphological expansion on the grid map to generate a parking potential energy map for path planning;

s3: and solving resource competition of multiple parking paths by adopting a dynamic time window method, and selecting an optimal path from the parking potential energy map.

Further, the step S1 specifically includes the following steps:

marking each element in the parking garage into the gray level image with the corresponding number of gray level values, and preprocessing the gray level image by adopting a threshold value segmentation method to obtain the binary grid map.

Further, the step S2 specifically includes the following steps:

and detecting the grid map and judging whether the grid map is correct or not.

Further, the step S2 specifically includes the following steps:

judging whether the inlet and the outlet of the grid map are both one and only one, and whether the coordinates of the inlet and the outlet are both on the boundary and are not coincident;

if yes, judging whether the inlet, the outlet and the parking spaces are communicated with each other, and communicating each parking space with the aisle.

Furthermore, the step of judging whether the inlet, the outlet and the parking space are communicated with each other or not further comprises the following steps:

defining a 3 x 3 structural element;

and performing expansion processing on the inlet, the outlet and the parking space of the binary grid map by using the structural body elements, wherein if the expanded position elements are 1, the expanded position elements are in four-way communication, and otherwise, the binary grid map is wrong.

Further, the step S2 further includes:

starting morphological expansion at the entrance of the parking garage, and continuously increasing the potential energy value of the parking garage at the position expanded each time;

removing the pixel points of the last morphological expansion from each expansion result, detecting whether an obstacle exists at the expansion position, and enabling the obstacle and the last expansion pixel not to participate in the next morphological expansion;

and continuously performing morphological expansion and increasing the potential energy value of each pixel position to generate the parking potential energy map.

Further, the optimal path is selected from the parking potential energy map by the following method:

the optimal path is from the parking space potential energy position to the exit along the path of potential energy decline.

Correspondingly, the embodiment of the present invention further provides an apparatus for accessing vehicle path planning in an intelligent parking garage, including a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

Accordingly, the embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program, and the computer program realizes the steps of the method when being executed by a processor.

The embodiment of the invention has the following beneficial effects: the optimal path planning can be carried out on the AGV according to the path searching strategy of the potential energy reduction by obtaining the parking potential energy diagram and the vehicle taking potential energy diagram of the intelligent garage.

1. According to the invention, the modeling of an intelligent garage is explained, and various elements in the parking lot space are marked when the grid map is used for modeling, so that a gray level graph is generated and is used for generating a potential energy graph through later morphological expansion processing.

2. In the invention, from the perspective of image processing, how to generate the potential energy diagram is explained, and how to plan the optimal path for the AGV by using the mode of potential energy reduction in the potential energy diagram is explained.

3. The invention provides a method for distributing priority to each AGV when a plurality of AGVs work, and introducing a dynamic time window method to avoid the problem of competition and conflict of path resources of the plurality of AGVs after an optimal path is planned.

Drawings

FIG. 1 is a schematic diagram of elements in an intelligent parking garage according to the present invention;

FIG. 2 is a diagram of a process for generating a potential energy map in conjunction with morphological processing in accordance with the present invention;

FIG. 3 is a potential energy diagram generated after the map shown in FIG. 2 is morphologically processed;

FIG. 4 is a schematic diagram of various element visualizations of an embodiment of the invention;

FIG. 5 is a potential energy diagram of the embodiment of the invention of FIG. 4;

figure 6 is a flow chart of generating a potential energy map.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings.

Aiming at the problem of planning the path of an AGV (automatic guided vehicle) in an intelligent parking garage at present, the invention generates different gray level map environments according to different elements from the angle of modeling the intelligent parking garage to the inlet, the outlet, the barrier, the passageway and the parking space of the parking garage from the angle of image processing, selects a proper structural body from a morphological expansion operation mode, and generates a corresponding parking potential energy map and a corresponding vehicle-taking potential energy map from an original gray level map of the intelligent parking garage to complete the path planning of parking and vehicle-taking, wherein the path is a path along which the potential energy is reduced. And the problem of path resource competition and conflict when a plurality of AGVs work simultaneously is considered, and the resource conflict is effectively avoided by adopting a dynamic time window method.

The embodiment of the invention provides a method for planning an AGV transportation vehicle path in an intelligent parking garage, wherein a potential energy diagram is obtained by using an image processing method, a specific implementation process of morphological processing and potential energy diagram generation is shown in FIG. 4, and the specific embodiment is explained by combining with FIG. 5 as follows:

(1) the map is regarded as an image I with 5 grey levels_src: the gray value of the entrance I is 0, the gray value of the exit E is 1, the gray value of the parking space P is 2, the gray value of the passageway X is 3, the gray value of the obstacle B is 4, each part is divided by threshold processing, and the gray values of the 5 levels are divided into binary images according to the following threshold dividing method:

an inlet:

and (4) outlet:

parking spaces:

passageway:

obstacle:

(2) judging whether the map is correct:

for the generated binary map, it is necessary to determine whether an element of the map has an error, which is a preprocessing operation for performing a later morphological operation, and determine whether there is only one entry and one exit of the map, where the entry and the exit are both on a boundary and do not overlap. Meanwhile, whether the parking spaces of the generated map are available or not needs to be judged, and the judgment method is to use one structural element to detect whether each parking space is communicated with the aisle or not. Define a 3 x 3 structure se:

0	1	0
			1	1	1
0	1	0

use of I_{x_dilate}＝imdilate(I_xSe) expansion of the feasible corridor area with the structure, wherein I_{x_dilate}Shows the result of swelling, I_xIs a two-value diagram of the passageway, and se is a cross-shaped knot shown in the upper tableAnd (5) constructing elements. The rationality judgment of the map is the preprocessing of generating a corresponding potential energy map on the map.

(3) A step of obtaining a potential energy map using morphological operations:

potential value generation analysis: according to the AGV movement rules, and taking fig. 1 as an example, with the horizontal direction as X-axis and the vertical direction as Y-axis, then from the entrance, the robot moves the first step, reaching the farthest points (0,3), (1,4), (0,5), which are 4 neighborhoods of the entrance position, and the corresponding positions are labeled 1 in the figure; the farthest points that can be reached by the robot moving two steps are (0,2), (1,3) (2,4), which are 4 neighbors of the farthest point reached by the first step, and the corresponding position is labeled 2 in the above figure; and so on, the farthest point reached in the nth step is 4 neighborhoods of the farthest points reached in the nth-1 st step, and the corresponding position is marked as n in the figure. The results obtained are shown in FIG. 3.

3.1, firstly obtaining a feasible passageway of the AGV in the intelligent parking area, I_{feasible_region}Is a feasible region binary image with the gray value less than 4:

I_{feasible_region}＝I＜4

3.2 the furthest position the robot reaches from the entry position one step is 4 connections (0,3), (1,4), (0,5) of the entry position (0,4), so still using a cross-shaped structure se, using a pair of structures I_IIs subjected to primary expansion, I_{curr_step}I.e. the current original image I_IResults of primary expansion:

I_{curr_step}＝imdilate(I_I,se)

to ensure that the AGV remains within the feasible region after inflation, a calculation is therefore required:

I_{curr_step}＝I_{curr_step}∩I_{feasibale_region}

wherein I_{curr_step}Is an iterative process of the currently feasible region after dilation, and I_{feasible_region}The AND operation is used to remove infeasible regions that do not participate in the next expansion. I.e. I_{curr_point}Point with middle pixel 1:

I_{[1]curr_point}＝I_{curr_step}-I_I

3.3, the farthest position reached by the AGV in two steps is equivalent to the position obtained by subtracting the first expansion from the two successive expansions, and the feasible region is detected after each expansion:

I_{curr_step}＝imdilate(I_I,se)∩I_{feasible_region}

I_{curr_step}＝imdilate(I_{last_step},se)∩I_{feasible_region}

I_{[2]curr_po int}＝I_{curr_step}-I_{last_step}

(4) by analogy, the step that the AGV walks n steps to the farthest position is equivalent to the step pair I_ICarry out the nth dilation minus the pair I_IPosition after n-1 th expansion (after each expansion, the result graph is compared with I_{feasible_region}Performing an and operation to ensure that the robot is in a feasible area), I_{[n]curr_point}To indicate the feasible region that the AGV could reach by n steps:

I_{[n]curr_point}＝I_{curr_step}-I_{last_step}

(5) the I obtained in each step_{[i]curr_point}Representing the feasible region reached by the ith step of AGV walking, multiplying the feasible region by the corresponding step number I, and superposing the feasible region and the corresponding step number I to obtain a potential energy diagram I_{energy_map}：

The specific implementation result is shown in fig. 5, and a potential energy map with zero potential energy at the exit can be obtained by using the same method and used for searching a path from the parking space to the exit.

(6) The sequence of potential energy descending along the coordinate position according to the path planning strategy of the parking potential energy diagram in the intelligent parking garage is the optimal path, so that the optimal path can be planned for the AGV for taking the vehicle by the same method for planning the path for taking the vehicle along the sequence of potential energy descending.

Path search strategy: according to the illustration in fig. 3, the farthest position that the nth step can reach is marked, the minimum number of steps needed from each point to the entrance position is marked on the figure, for example, the minimum number of steps from the parking space number (5,1) at the upper right corner to the entrance position is 10, if the step number marked on each position is regarded as a potential energy, the figure 3 is called a potential energy diagram, and then the optimal path from the parking space to the entrance position is the potential energy of all positions on the path and the minimum path, for example: the potential energy sequence of the parking space (5,1) is 10 → (5,0), the potential energy is 9 → (4,0), the potential energy is 8 → (3,0), the potential energy is 7 → (2,0), the potential energy is 6 → (1,0), the potential energy is 5 → (0,0), the potential energy is 4 → (0,1), the potential energy is 3 → (0,2), the potential energy is 2 → (0,3), the potential energy is 1 → the entrance position (0,4) and the potential energy is 0, and the path from the parking space to the entrance is searched by the method that the parking robot walks in a direction smaller than the potential energy of the current position every step and finally reaches the entrance position with the potential energy of 0.

If a plurality of AGVs operate simultaneously in the process, the priority is allocated to each AGV, and the problems of competition and conflict of path resources are solved by combining a dynamic time window method, so that the vehicles can be accessed quickly and efficiently.

The embodiment of the invention also provides a device for planning the access vehicle path in the intelligent parking garage, which can be computing equipment such as a desktop computer, a notebook computer, a palm computer, a cloud server and the like. The device for image similarity comparison may include, but is not limited to, a processor, and a memory. It will be understood by those skilled in the art that the schematic illustration is merely a means for accessing vehicle path plans in an intelligent parking garage and does not constitute a limitation on accessing vehicle path plans in an intelligent parking garage, and may include more or fewer components than those shown, or some components in combination, or different components, for example, the means for accessing vehicle path plans in an intelligent parking garage may also include input and output devices, network access devices, buses, etc.

The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like, said processor being the control center for the access vehicle path planning in said one intelligent parking garage, and various interfaces and lines are used to connect the various parts of the access vehicle path planning in the whole one intelligent parking garage.

The memory can be used for storing the computer program and/or the module, and the processor can realize various functions of accessing the vehicle path planning in the intelligent parking garage by running or executing the computer program and/or the module stored in the memory and calling the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.

The module/unit for accessing the vehicle path planning in the intelligent parking garage can be stored in a computer readable storage medium if the module/unit is realized in the form of a software functional unit and is sold or used as an independent product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A method for planning vehicle access paths in an intelligent parking garage is characterized by comprising the following steps:

s2: starting morphological expansion at the entrance of the parking garage, and continuously increasing the potential energy value of the parking garage at the position expanded each time;

carrying out morphological expansion continuously and increasing the potential energy value of each pixel position to generate a parking potential energy map;

2. The method for planning vehicle parking and taking paths in an intelligent parking garage according to claim 1, wherein said step S1 comprises the following steps:

marking each element in the parking garage into the gray value image with the corresponding number of gray values, and preprocessing the gray value image by adopting a threshold segmentation method to obtain the binary grid map.

3. The method for planning vehicle parking and taking paths in an intelligent parking garage according to claim 2, wherein said step S2 comprises the following steps:

and detecting the grid map and judging whether the grid map is correct or not.

4. The method according to claim 3, wherein said step S2 further comprises the following steps:

5. The method according to claim 4, wherein said determining whether the entrance, exit and parking spaces are four-way connected further comprises the steps of:

defining a 3 x 3 structural element;

6. The method for planning vehicle parking and taking paths in an intelligent parking garage according to claim 5, wherein the optimal path is selected from the parking potential energy map by the following steps:

the optimal path is obtained from the potential energy position of the parking space to the exit along the path of potential energy reduction.

7. An apparatus for accessing vehicle path planning in an intelligent parking garage, comprising a memory, a processor and a computer program stored in said memory and executable on said processor, characterized in that said processor, when executing said computer program, implements the steps of the method according to any of claims 1 to 6.

8. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.