CN106444765B - A kind of AGV air navigation aid of view-based access control model - Google Patents

Abstract

The invention discloses a vision-based AGV navigation method, comprising the following steps: acquiring a picture of the ground with track marks; performing grayscale processing on a reference horizontal line of the picture to obtain an array of reference horizontal lines; The array forms a track matrix; the adjacent upper and lower rows of the track matrix are XORed to obtain track edge information; the track edge information is digitally processed as the input of the AGV controller to realize the movement of the AGV control. The invention also discloses a vision-based AGV navigation system. The present invention also discloses an AGV including the above vision-based AGV navigation system. In the above-mentioned navigation method, the obtained track edge information has high precision, and the purpose of fast navigation is realized.

Description

A Vision-Based AGV Navigation Method

technical field

The invention relates to the technical field of automatic control, in particular to a vision-based AGV navigation method.

Background technique

AGV (Automated Guided Vehicle), namely "Automated Guided Vehicle", refers to the transportation equipped with electromagnetic or optical automatic guidance devices, which can travel along the specified guidance path, with safety protection and various transfer functions. car.

At present, AGV has a variety of guidance methods, including: electromagnetic guidance, laser guidance, visual guidance, etc. Because laser guidance and electromagnetic guidance are often expensive and easily affected by the complex environment of the factory, and with the development of embedded controller chips, AGVs will be more intelligent, and visual guidance will be more flexible. It is the automatic navigation of AGV. development direction.

In the existing visual navigation technology, due to the limitation of calculation speed, there is often a problem of implementation difficulty. For example, the vision-guided AGV system and method using an embedded system is costly and requires high-speed image processing capability due to the dual-camera solution; while other vision-based AGV navigation methods also use the orbit matrix method, but It is by finding all connected and coincident edge regions of the image, and the algorithm is complex.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a vision-based AGV navigation method, which can navigate quickly, has high precision, and is simple to implement.

To achieve the above object, the present invention provides a vision-based AGV navigation method, comprising the following steps:

Get a picture of the ground with track markings;

Grayscale processing is performed on the reference horizontal line of the picture to obtain a reference horizontal line array;

forming a track matrix with a plurality of the reference horizontal line arrays;

XOR processing is performed on the adjacent upper and lower rows of the track matrix to obtain track edge information;

The track edge information is digitally processed and used as the controller input of the AGV to realize the motion control of the AGV.

Compared with the above-mentioned background technology, the core of the vision-based AGV navigation method provided by the present invention is to perform grayscale processing on the reference horizontal line of the picture, thereby obtaining the reference horizontal line array; the reference horizontal line is often a horizontal straight line in the picture; this It is invented that during the walking process of the AGV, a picture will be collected at every preset time interval, and each time a picture is collected, the reference horizontal line of the picture will be gray-scaled; and a plurality of reference horizontal line arrays will be formed into a track matrix. , perform XOR processing on the track matrix to obtain the track edge information, so as to carry out digital processing and input as the controller of the AGV to realize the motion control of the AGV; it can be seen that the present invention adopts the reference horizontal line for the picture to perform grayscale The processing method is to perform grayscale processing on the reference horizontal lines of multiple pictures to obtain multiple reference horizontal line arrays, thereby forming a track matrix, which greatly simplifies the calculation process and avoids the need for continuous environmental monitoring in the prior art. Monitoring and image processing; the track edge information obtained by the above method has high precision and realizes the purpose of fast navigation.

Preferably, the step of obtaining a ground image with track marks, and obtaining the image specifically includes:

The track marks are collected at preset time intervals by the cameras installed on the AGV to obtain multiple sets of pictures.

Preferably, the range of the preset time interval is 0.3s˜0.6s.

Preferably, the step of performing grayscale processing on the reference horizontal line of the picture to obtain the reference horizontal line array specifically includes:

Every preset pixel, collect the pixel points on the reference horizontal line of the picture;

Gray value conversion is performed on each of the pixel points through a preset gray value conversion formula to obtain a reference horizontal line array.

Preferably, the step of performing gray value conversion on each of the pixel points through a preset gray value conversion formula to obtain the reference horizontal line array specifically includes:

The grayscale value h of each of the pixel points on the reference horizontal line is calculated by the formula h=(R*38+G*75+B*15)>>7; wherein, R, G and B respectively represent red , green and blue;

Compare the gray value h with the preset threshold a:

When h≤a, h takes the value 0;

When h>a, h takes the value 1;

Get an array of datum horizons.

Preferably, the step of performing digital processing on the track edge information and inputting it as the controller of the AGV to realize the motion control of the AGV specifically includes:

Obtain the track center line according to the track edge in the track edge information;

The center line of the track is input as the controller of the AGV to realize the motion control of the AGV.

Preferably, the reference horizontal line of the picture is specifically: any horizontal line in the range of 1/3 to 3/4 of the height of the picture.

Preferably, the controller of the AGV is a PID controller.

The present invention also provides a vision-based AGV navigation system, including:

A camera for taking pictures of the ground with track marks;

a grayscale processing module, configured to perform grayscale processing on the reference horizontal lines of the picture to obtain a reference horizontal line array;

a track matrix forming module, configured to form a track matrix with a plurality of the reference horizontal line arrays;

an XOR processing module for performing XOR processing on the adjacent upper and lower rows of the track matrix to obtain track edge information;

The digital processing module is used to perform digital processing on the track edge information as the input of the controller of the AGV, so as to realize the motion control of the AGV.

The present invention also provides an AGV, including the above-mentioned vision-based AGV navigation system.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative work.

1 is a schematic flowchart of a vision-based AGV navigation method provided by an embodiment of the present invention;

Fig. 2 is the mathematical transformation process that is carried out for obtaining track edge information in Fig. 1;

Fig. 3 is the PID control block diagram in Fig. 1;

4 is a front view of an AGV provided by an embodiment of the present invention;

FIG. 5 is a bottom view of FIG. 4 .

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In order to make those skilled in the art better understand the solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

Please refer to FIGS. 1 to 5. FIG. 1 is a schematic flowchart of a vision-based AGV navigation method provided by an embodiment of the present invention; FIG. 2 is a mathematical transformation process performed to obtain track edge information in FIG. 1; The PID control block diagram in 1; FIG. 4 is the front view of the AGV provided by the embodiment of the present invention; FIG. 5 is the bottom view of FIG. 4 .

The present invention provides a vision-based AGV navigation method, as shown in Figure 1 of the description, which mainly includes the following steps:

S1. Obtain a picture of the ground with track marks;

S2, performing grayscale processing on the reference horizontal lines of the picture to obtain a reference horizontal line array;

S3, forming a track matrix with a plurality of the reference horizontal line arrays;

S4, performing XOR processing on the adjacent upper and lower rows of the track matrix to obtain track edge information;

S5. Digitally process the track edge information as an input to the controller of the AGV, so as to realize the motion control of the AGV.

First of all, a track mark is set on the ground where the AGV walks, and there is a certain color contrast between the ground and the track mark. For those skilled in the art, the color contrast between the ground and the track mark should be understandable. There are no restrictions on the color of the ground and track markings, and there is no requirement for the width of the track markings.

In step S1, a picture of the ground with track marks is obtained; that is, a camera or other image acquisition device can be used to collect a picture of the ground, and the picture of the ground should have the content of the track mark.

In step S2, gray-scale processing is performed on the reference horizontal line in the picture with the content of the track marking to obtain the reference horizontal line array; because the picture is the ground situation at a certain moment, the picture can reflect the track marking at this moment relative to the ground. Location. Grayscale processing is performed on the reference horizontal line in the picture, and the reference horizontal line can be any horizontal line in the picture; that is, for a picture, only one horizontal line in the picture is processed, which can greatly simplify the processing process and improve the processing efficiency. . In the process of AGV walking, the angle of the camera or other image acquisition device should remain unchanged relative to the AGV body, and for each picture, a horizontal line with the same position should be processed to ensure that the obtained track position is accurate .

For the acquisition of the reference horizontal line array, reference may be made to the prior art, and the grayscale processing process can be calculated by using various formulas according to actual needs.

In step S3, a track matrix is formed for a plurality of the reference horizontal line arrays; obviously, during the walking process of the AGV, a picture of the ground will be collected at a preset time interval, and the track mark in each picture is relative to the ground. The position of each image will change, and for each image, one horizontal line in the image will be processed, and multiple reference horizontal line arrays corresponding to multiple images will be combined to obtain a track matrix.

In step S4, XOR processing is performed on the adjacent upper and lower rows of the track matrix to obtain track edge information; as shown in FIG. 2 of the specification. In the picture collected once, the reference horizontal line array is "00001111111110000", taking 16 pixels as an example, by collecting pictures multiple times, and performing grayscale processing on each picture, multiple reference horizontal line arrays are obtained to form a track matrix; then XOR the adjacent upper and lower rows of the track matrix; XOR the upper and lower rows of the track matrix in turn, that is, the first row is XORed with the second row, and the second row is XORed with the third row , the third row is XORed with the fourth row, and so on. As a result, the track edges will all become elements 1, and the non-edge parts will be 0, and finally the track edge information will be obtained.

In step S5, digital processing is performed on the track edge information, which is input as the controller of the AGV to realize the motion control of the AGV. That is, the input of the controller can be the subscript of the middle element in the axis array in the image; when controlling the corner of the AGV, the output is the subscript average value of the two elements 1 appearing in each row in the XOR-processed track edge information, PID closed-loop negative feedback control can be performed. As shown in the PID control block diagram in Fig. 3 of the specification. Of course, the controller of the AGV can also be other controllers. The PID controller is used in conjunction with the above method of the present invention to give full play to the characteristics of the PID controller to achieve high-precision and fast navigation.

In the above step S1, the track marks can be collected by the cameras installed on the AGV at preset time intervals to obtain multiple sets of pictures; and the range of the preset time intervals is specifically 0.3s-0.6s. As shown in FIG. 4 in the description, the camera 1 is installed at the front end of the AGV, and the distance between the camera 1 and the horizontal plane is 25° to 35°, and is set obliquely downward. The camera 1 is fixed at the centerline position of the front and top of the AGV (that is, 1/2 of the width) through the bracket to ensure the collection of track information at a certain distance in front of the AGV; the specific can be based on the body size of the AGV, the pixels of the camera 1, the control accuracy, and the AGV. The speed of the bracket is adjusted before and after the fixed position of the bracket on the center axis of the vehicle body.

In the above step S2, the steps of performing grayscale processing on the reference horizontal lines of the picture to obtain an array of reference horizontal lines specifically include:

Every preset pixel, collect the pixel points on the reference horizontal line of the picture;

Gray value conversion is performed on each of the pixel points through a preset gray value conversion formula to obtain a reference horizontal line array.

For the processing of the reference horizontal line of each picture, the basic horizontal line of the picture is binarized by the method of equal difference processing; since the reference horizontal line of the picture has different numbers of pixels according to different resolutions, in order to reduce the calculation The amount and the size of the reference horizontal line array are processed by the method of arithmetic difference. For example, using a 300,000-pixel camera with a resolution of 640*480, taking the reference horizontal line will obtain 640 pixels, and taking a point every 5 pixels by the arithmetic method will obtain 128 pixels. Pixels can be reduced or increased depending on precision and operation speed requirements.

However, for each of the pixel points, gray value conversion is performed through a preset gray value conversion formula to obtain a reference horizontal line array. In this process, the present invention preferably adopts the formula h=(R*38+G*75+B*15)>>7 to calculate the grayscale value h of each pixel point on the reference horizontal line; wherein , R, G and B represent red, green and blue respectively; then compare the gray value h with the preset threshold a: when h≤a, h takes the value 0; when h>a, The value of h is 1; thus, the array of reference horizontal lines is obtained.

Regarding the selection of the preset threshold value a, it can be tested that after the orbit is binarized, the orbit part and the ground part can be completely distinguished by 0 and 1.

h=(R*38+G*75+B*15)>>7 preferably adopted in the present invention is a 7-bit precision gray value algorithm, and this algorithm has the advantages of high precision and high speed; of course, according to actual needs , the preset gray value conversion formula can also be calculated using formulas of other precision digits, which will not be repeated in this section. In view of the above steps, grayscale processing can be performed on the collected ground pictures in the corresponding auxiliary software, that is, the RGB pictures are converted into grayscale images, and the grayscale images are converted into grayscale histograms, and the color depth is selected as 8 , and select a preset threshold value according to the distribution of the grayscale histogram, and the preset threshold value ranges from 71 to 170, preferably 120.

In step S5, the track center line can be obtained according to the track edge in the track edge information; and the track center line is input as the controller of the AGV to realize the motion control of the AGV.

In the above steps, the motion trajectory of the AGV is controlled according to the obtained track edge information, and the position deviation of the AGV is obtained by comparing the half of the subscript sum of the element where the edge is located with the half of the maximum subscript value, and the AGV angle is controlled. The selected control method is PID control. Through the method of the present invention, the horizontal offset information between the AGV and the center line of the actual track can be quickly obtained and corrected quickly. The method has high control accuracy, is simple and easy to real-time, can communicate with the automatic control system in information, and has good application prospects.

The reference horizontal line of the above-mentioned picture can be specifically: any horizontal line in the range of the picture height from 1/3 to 3/4, and the present invention is preferably a horizontal line with a height of 1/2 of the picture. Grayscale processing of the horizontal line at 1/2 the height of the picture is helpful to reduce the calculation error and improve the running accuracy.

The following describes the vision-based AGV navigation system provided by the embodiments of the present invention, and the apparatus described below can be compared with the method described above.

The present invention also provides a vision-based AGV navigation system, which mainly includes:

The camera 1 is used to obtain a picture of the ground with track marks;

a grayscale processing module, configured to perform grayscale processing on the reference horizontal lines of the picture to obtain a reference horizontal line array;

a track matrix forming module, configured to form a track matrix with a plurality of the reference horizontal line arrays;

an XOR processing module for performing XOR processing on the adjacent upper and lower rows of the track matrix to obtain track edge information;

The digital processing module is used to perform digital processing on the track edge information as the input of the controller of the AGV, so as to realize the motion control of the AGV.

An AGV provided by the present invention includes the vision-based AGV navigation system as described above, as shown in FIG. 5 of the description. Among them, the power supply 2, the motor 3, the driving wheel 4, the driver 5, the main controller 6 and the universal wheel 7 are located at the bottom of the body of the AGV, and the top of the body of the AGV is provided with a camera 1; it can be seen from the accompanying drawing 3 of the specification that the driver 5 After being controlled by the PID controller, the motor 3 is driven to rotate, thereby driving the driving wheel 4 to run. For other parts of the AGV, reference can be made to the prior art, which will not be expanded in this article.

The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the points that are different from other embodiments, and the similar parts between the various embodiments can be referred to each other. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the related part may refer to the description of the method.

Professionals may further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two, in order to clearly illustrate the possibilities of hardware and software. Interchangeability, the above description has generally described the components and steps of each example in terms of functionality. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may use different methods for implementing the described functionality for each particular application, but such implementations should not be considered beyond the scope of the present invention.

The AGV, the vision-based AGV navigation method and the system thereof provided by the present invention have been introduced in detail above. The principles and implementations of the present invention are described herein by using specific examples, and the descriptions of the above embodiments are only used to help understand the method and the core idea of the present invention. It should be pointed out that for those skilled in the art, without departing from the principle of the present invention, several improvements and modifications can also be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims (4)

1. a kind of AGV air navigation aid of view-based access control model, which comprises the steps of:

Obtain the picture with the ground of track mark；

Gray proces are carried out to the reference horizontal line of the picture, obtain reference horizontal line array；

Multiple reference horizontal line arrays are constituted into track matrix；

Exclusive or processing is carried out to adjacent two rows up and down of the track matrix, obtains rail flanges information；

Digitized processing is carried out to the rail flanges information, the controller as AGV inputs, to realize the movement of the AGV Control；

The step of reference horizontal line to the picture carries out gray proces, obtains reference horizontal line array specifically includes:

Every presetted pixel, the pixel on the reference horizontal line of the picture is acquired；

Grayvalue transition is carried out by default grayvalue transition formula each described pixel, obtains datum-plane line number Group；

It is described that grayvalue transition is carried out by default grayvalue transition formula each described pixel, obtain reference horizontal line The step of array, specifically includes:

7 each pixel on the reference horizontal line is calculated in > > by formula h=(R*38+G*75+B*15) Gray value h；Wherein, R, G and B respectively represent as red, green and blue；

The gray value h is compared with preset threshold a:

As h≤a, h value is 0；

As h > a, h value is 1；

Obtain reference horizontal line array；

Described to carry out digitized processing to the rail flanges information, the controller as AGV inputs, to realize the AGV's The step of motion control, specifically includes:

Track centre is obtained according to the rail flanges in the rail flanges information；

It is inputted the track centre as the controller of AGV, to realize the motion control of the AGV；

The reference horizontal line of the picture specifically: any one horizontal line of the picture height range in 1/3~3/4.

2. the AGV air navigation aid of view-based access control model according to claim 1, which is characterized in that described obtain has track mark The ground image shown, the step of obtaining picture, specifically include:

Camera interval prefixed time interval by being installed on the AGV is acquired track mark, obtains multiple groups Picture.

3. the AGV air navigation aid of view-based access control model according to claim 2, which is characterized in that the prefixed time interval Range is specially 0.3s~0.6s.

4. according to claim 1 to the AGV air navigation aid of view-based access control model described in 3 any one, which is characterized in that the AGV Controller be specially PID controller.