1. An alignment system of an unmanned sweeper and a garbage station is characterized by comprising the unmanned sweeper (10), a buried garbage station (30), a camera and a radio frequency identification system;

the buried garbage station (30) comprises a buried garbage pit (40), a reader (50), a trunk ground (31) and a guide ground (32);

the reader (50) is positioned in front of the buried trash pit (40);

the main ground (31) and the guide ground (32) have obvious edge information, and the edge information is extracted to obtain important information in the image;

the unmanned sweeper (10) comprises a dust box (20) and a label (60);

the dust box (20) is used for filling garbage and is positioned at the rear end of the unmanned sweeper (10);

after a reader (50) in front of a buried garbage pit (40) scans a tag (60) at the front end of a dust box (20) of the unmanned sweeper, an emergency stop operation is performed;

acquiring a ground image in the buried garbage station (30) through a camera, and processing the image;

the radio frequency identification system comprises an electronic tag, a reader-writer and a computer network; the electronic tag is used for marking an object, the electronic tag exchanges data with the reader-writer through radio waves, the reader-writer can transmit a read-write command of the host to the electronic tag and then transmit data returned by the electronic tag to the host, and the data exchange and management system of the host is responsible for finishing storage, management and control of data information of the electronic tag.

2. The alignment system of the unmanned sweeper and the garbage station according to claim 1, wherein the main floor (31) is black, the guiding floor (32) is white, and when the edge information of the guiding floor (31) is processed, color features are extracted to realize detection and identification of the guiding floor.

3. The alignment system of the unmanned sweeper to the garbage station as claimed in claim 1, further comprising an upper computer system for estimating the remaining capacity of the underground garbage pit (40).

4. The method of aligning an unmanned sweeper truck with a refuse station of any one of claims 1-3, comprising the steps of:

step 1, automatically cleaning garbage in a cleaning operation area by an unmanned sweeper (10);

step 2, after the unmanned sweeper (10) detects that the dust box (20) is full of garbage, the unmanned sweeper (10) returns to the buried garbage station (30) according to the high-precision map and the navigation information;

step 3, after the unmanned sweeper (10) enters the buried garbage station (30), the unmanned sweeper enters a garbage pit alignment area in a retreating mode, a camera arranged at the rear end of the sweeper is used for acquiring a ground image in the garbage station, image processing is carried out, and after a guide ground (32) is identified, the retreating direction is continuously corrected;

step 4, after a reader (50) in front of the buried garbage pit (40) scans a label (60) on the front side of a dust box (20) of the unmanned sweeper, performing emergency stop operation, wherein the dust box (20) of the unmanned sweeper (10) is just butted with the buried garbage pit (40);

step 5, after the label information is scanned, opening the buried garbage pit (40), then opening the dust box (20) by the unmanned sweeper (10), and pouring the garbage into the buried garbage pit (40);

and 6, returning the unmanned sweeper (10) to the cleaning operation area to continue the automatic cleaning task or returning to a charging pile for charging.

5. The method for aligning the unmanned sweeper to the garbage station according to claim 4, wherein the image processing in the step 3 is to perform ground image preprocessing on the garbage station ground information collected by the camera, graying and equalizing the ground image by an image segmentation algorithm based on color region features according to the ground of the underground garbage station (30), extracting the region of interest by controlling a threshold, and segmenting the trunk region (31) and the guide region (32) to realize accurate detection of the white guide region (32).

6. The method for aligning the unmanned sweeper truck with the garbage station as claimed in claim 5, wherein the image segmentation algorithm based on the color region features firstly calls a camera of the unmanned sweeper truck to acquire video information for preparation, converts the image from RGB to HSV format, then performs color segmentation by setting a proper threshold value, extracts a white region of interest, preliminarily determines the position of the guiding ground, and then performs two-dimensional median filtering on the image to reduce the interference of speckle noise and salt and pepper noise so as to prevent the blurring of edges.

7. The alignment method of the unmanned sweeper to the garbage station according to claim 6, wherein the image segmentation further comprises performing expansion and region growing processing on the detected guiding region (32) to make the detected guiding region (32) more complete; noise interference is eliminated, the ground of the garbage station is smoothed, and isolated blocks are removed, so that interference of objects with similar colors near the guide area (32) on the detection result is prevented.

8. The method for aligning the unmanned sweeper to the garbage station according to claim 7, wherein after the guiding ground (31) is identified, the advancing direction is continuously corrected by using a Sobel operator, the boundary feature extraction is performed on the guiding area (32), and since the boundary of the guiding area (32) is a set of pixels with sharp and large-amplitude gray value of a target area of the guiding area (32), the Sobel operator is used for calculating a gray value change similarity value of the target area in an image, and then a boundary line of the target area is searched, so that the detection and the identification of the center line of the target area are realized.

9. The method for aligning an unmanned sweeper to a garbage station according to claim 8, wherein the extracted image is subjected to line detection, the edges of the guide area are subjected to line detection, non-linear edges can be eliminated, and only interesting lines are retained; the straight line detection algorithm keeps important straight line information according to parameter adjustment, can obtain the number of detected straight lines and the information of points on each straight line, can obtain the slope of the straight lines for further screening, and can calculate the center line of a guide area according to the coordinates of the road side lines after the accurate road side lines are extracted.

Background

The combination of the high-tech unmanned technology and the daily environmental sanitation sweeper is a profound innovation and practical idea without losing due to the fact that the unmanned vehicle gradually becomes the development trend of future vehicles. The application of unmanned driving in the field of environmental sanitation is a relatively closed environment, too much data and tests do not need to be accumulated, the environmental sanitation work of complex machinery can be effectively replaced, the labor intensity of environmental sanitation workers is reduced, and the cleaning efficiency is improved.

For the existing unmanned sweeper, after the dust box is filled or a circle of cleaning operation is finished, garbage can be processed only by manual dumping, and the dust box needs to be manually taken down and installed. The garbage collection mode has the advantages that the concept of unmanned and efficient cleaning of the unmanned sweeper is eliminated, and the problem of high cost caused by manual intervention cannot be solved.

Disclosure of Invention

The invention aims to provide an alignment system and method for an unmanned sweeper and a garbage station, and aims to solve the technical problem that the existing garbage collection mode cannot give up manual intervention and is high in cost.

In order to solve the technical problems, the specific technical scheme of the invention is as follows:

an alignment system and method for an unmanned sweeper and a garbage station comprises the following steps: .

An alignment system of an unmanned sweeper and a garbage station comprises the unmanned sweeper, a buried garbage station, a camera and a radio frequency identification system;

the buried garbage station comprises a buried garbage pit, a reader, a trunk ground and a guide ground;

the reader is positioned in front of the buried garbage pit;

the main ground and the guide ground have obvious edge information, and the edge information is extracted to obtain important information in the image.

The unmanned sweeper comprises a dust box and a label;

the dust box is used for filling garbage and is positioned at the rear end of the unmanned sweeper;

the reader in front of the buried garbage pit makes an emergency stop operation after scanning a tag at the front end of a dust box of the unmanned sweeper;

acquiring a ground image in the buried garbage station through a camera, and processing the image;

the radio frequency identification system comprises an electronic tag, a reader-writer and a computer network; the electronic tag is used for marking an object, the electronic tag exchanges data with the reader-writer through radio waves, the reader-writer can transmit a read-write command of the host to the electronic tag and then transmit data returned by the electronic tag to the host, and the data exchange and management system of the host is responsible for finishing storage, management and control of data information of the electronic tag.

Further, the main body ground is black, the guide ground is white, and when edge information of the guide ground is processed, color features are extracted to realize detection and identification of the guide ground.

Further, the system also comprises an upper computer system which is used for estimating the residual capacity of the underground garbage pit.

An alignment method of an unmanned sweeper and a garbage station comprises the following steps:

step 1, automatically cleaning garbage in a cleaning operation area by an unmanned sweeper;

step 2, after the unmanned sweeper detects that the dust box is full of garbage, returning the unmanned sweeper to the buried garbage station according to the high-precision map and the navigation information;

step 3, after the unmanned sweeper enters the buried garbage station, entering a garbage pit alignment area in a retreating mode, acquiring a ground image in the garbage station through a camera arranged at the rear end of the sweeper, performing image processing, and continuing to correct the retreating direction after the guiding ground is identified;

step 4, after a reader in front of the buried garbage pit scans a label on the front side of a dust box of the unmanned sweeper, performing emergency stop operation, wherein the dust box of the unmanned sweeper is just butted with the buried garbage pit;

step 5, after the label information is scanned, opening the buried garbage pit, then opening a dust box by the unmanned sweeper, and pouring the garbage into the buried garbage pit;

and 6, returning the unmanned sweeper to the cleaning operation area to continue the automatic cleaning task or returning the unmanned sweeper to the charging pile for charging.

Further, the image processing in step 3 is to perform ground image preprocessing on the ground information of the garbage station collected by the camera, graying and equalizing the ground image by an image segmentation algorithm based on color region characteristics aiming at the ground of the buried garbage station, then extracting an interested region by controlling a threshold value, and segmenting a trunk region and a guide region to realize accurate detection of a white guide region.

Further, the image segmentation algorithm based on the color region features firstly calls a camera of the unmanned vehicle to acquire video information for preparation processing, converts RGB of the image into HSV format, then performs color segmentation by setting a proper threshold value, extracts an interested white region, preliminarily determines the position of a guide ground, then performs two-dimensional median filtering on the image, reduces the interference of speckle noise and salt and pepper noise, and prevents edge blurring.

Furthermore, the detected guide area is subjected to expansion and area growth processing by image segmentation, so that the detected guide area is more complete; noise interference is also eliminated, the ground of the garbage station is smoothed, and isolated blocks are removed, so that interference of objects with similar colors near the guide area on the detection result is prevented.

Further, in the method for continuously correcting the advancing direction after the guiding ground is identified, Sobel operators are used for extracting boundary characteristics of the guiding area, and since the ground guiding area boundary refers to a set of pixels which are sharply and greatly changed in gray value of a target area of the ground guiding area, Sobel operators are used for calculating a gray value change similarity value of the target area in an image, then boundary lines of the target area are searched, and further detection and identification of the central line of the target area are achieved.

Further, the extracted image is subjected to linear detection processing, the edges of the guide area are subjected to linear detection, non-linear edges can be eliminated, and only interested straight lines are reserved; the straight line detection algorithm keeps important straight line information according to parameter adjustment, can obtain the number of detected straight lines and the information of points on each straight line, can obtain the slope of the straight lines for further screening, and can calculate the center line of a guide area according to the coordinates of the road side lines after the accurate road side lines are extracted.

The invention discloses an alignment system and method of an unmanned sweeper and a garbage station, which have the following advantages: according to the method for automatically aligning the garbage pit of the unmanned sweeper with the underground garbage station, the automatic driving and cleaning work of the unmanned sweeper can be completed, the ground image of the garbage station can be acquired through the high-definition camera at the rear of the unmanned sweeper, the trunk ground and the guide ground of the garbage station are identified according to image processing, the sweeper body is enabled to follow the guide ground to continuously correct the backward route, then the scram information is immediately sent out after the reader at the front side of the garbage pit detects the label information at the bottom of the unmanned sweeper, the alignment of the unmanned sweeper with the garbage pit is completed at this moment, the garbage dumping operation can be unmanned, and manual intervention is abandoned in the processes of garbage collection, garbage dumping and vehicle return charging.

Drawings

Fig. 1 is a schematic flow chart of an alignment method of an unmanned sweeper and a garbage station according to the present invention;

FIG. 2 is a schematic structural view of the buried garbage station of the present invention;

fig. 3 is a schematic structural view of a chassis of the unmanned sweeping vehicle of the invention;

the notation in the figure is: 10. an unmanned sweeper; 20. a dust box; 30. a buried garbage station; 31. a trunk ground; 32. guiding the ground; 40. burying a garbage pit; 50. a reader; 60. and (4) a label.

Detailed Description

In order to better understand the purpose, structure and function of the present invention, an alignment system and method for an unmanned sweeper and a garbage station according to the present invention will be described in detail with reference to the accompanying drawings.

As shown in fig. 2 and 3, the aligning system for the unmanned sweeper and the garbage station of the invention comprises an unmanned sweeper 10, a buried garbage station 30, a camera and a radio frequency identification system;

the buried garbage station 30 comprises a buried garbage pit 40, a reader 50, a trunk ground 31 and a guide ground 32; the reader 50 is positioned in front of the underground refuse pit 40; the main ground 31 and the guide ground 32 have obvious edge information, and the edge information is extracted to obtain important information in the image;

the unmanned sweeper 1 comprises a dust box 20 and a label 60; the dust box 20 is used for filling garbage and is positioned at the rear end of the unmanned sweeper 10; after a reader 50 in front of the buried garbage pit 20 scans a tag 60 at the front end of the dust box 20 of the unmanned sweeper, performing emergency stop operation;

acquiring a ground image in the buried garbage station 30 through a camera, and processing the image;

the radio frequency identification system comprises an electronic tag, a reader-writer and a computer network; the electronic tag is used for marking an object, the electronic tag exchanges data with the reader-writer through radio waves, the reader-writer can transmit a read-write command of the host to the electronic tag and then transmit data returned by the electronic tag to the host, and the data exchange and management system of the host is responsible for finishing storage, management and control of data information of the electronic tag.

As shown in fig. 1, the alignment method of the unmanned sweeper and the garbage station of the invention comprises the following steps:

step 1, the unmanned sweeper 10 automatically cleans garbage in a cleaning operation area;

step 2, after the unmanned sweeper 10 detects that the dust box 20 is full of garbage, the unmanned sweeper 10 returns to the buried garbage station 30 according to the high-precision map and the navigation information;

the ground features adopted by the buried garbage station 30 are a black main ground 31 and a white guide ground 32 as shown in fig. 2, and very obvious edge information exists between the main ground 31 and the guide ground 32, and the extraction of the edge information is helpful for acquiring important information in an image. When the side line information of the guiding ground 32 is processed, the color feature is extracted to realize the detection and identification of the guiding ground 32. The image segmentation algorithm may keep the most important straight line information according to parameter adjustment of the algorithm, and may obtain region information of the detected straight line and information of points on the region. And after the accurate ground sideline is extracted, continuously fitting and correcting the advancing route of the unmanned sweeper according to the coordinates of the road sideline.

Step 3, after the unmanned sweeper 10 enters the buried garbage station 30, acquiring a ground image in the garbage station through a camera, performing image processing, and continuously correcting the advancing direction after recognizing the guide ground 32;

the image processing means that ground images of the garbage station ground information collected by the camera are preprocessed, graying and equalization are carried out on the ground images by an image segmentation algorithm based on color region characteristics aiming at the ground of the buried garbage station 30, then an interested region is extracted by controlling a threshold value, a trunk region 31 and a guide region 32 are segmented, and accurate detection of a white guide region 32 is achieved.

The image segmentation algorithm based on the color region features firstly calls a camera of the unmanned vehicle to acquire video information for preparation processing, converts RGB (red, green and blue) of an image into HSV (hue, saturation, value) format, then performs color segmentation by setting a proper threshold value, extracts an interested white region, preliminarily determines the position of a guide ground, then performs two-dimensional median filtering on the image, reduces the interference of speckle noise and salt and pepper noise, and prevents edge blurring.

In actual use, the detected guide region 32 is further processed by expansion and region growing through image segmentation, so that the detected guide region 32 is more complete. In addition, noise interference is eliminated, the ground of the garbage station is smoothed, and isolated blocks are removed, so that interference of objects with similar colors near the guide area 32 on the detection result is prevented.

In the method for continuously correcting the advancing direction after the guiding ground 32 is identified, Sobel operators are used for extracting boundary characteristics of the guiding area, the boundary of the guiding area refers to a set of pixels which are sharply and greatly transformed in the gray value of a target area of the guiding area, the Sobel operators are used for calculating the gray value change similarity value of the target area in an image, then the boundary line of the target area is searched, and further the detection and identification of the center line of the target area are realized.

And then, performing linear detection processing on the extracted image, and performing linear detection on the edge of the guide area to remove a large number of non-linear edges and only keep interested straight lines. The straight line detection algorithm can keep the most important straight line information according to parameter adjustment, can obtain the number of detected straight lines and the information of points on each straight line, can obtain the slope of the straight lines for further screening, and can calculate the central line of a guide area according to the coordinates of the road side lines after the accurate road side lines are extracted.

Step 4, after the reader 50 in front of the buried garbage pit 40 scans the tag 60 on the front side of the dust box 20 of the unmanned sweeper, performing emergency stop operation, wherein the dust box 20 of the unmanned sweeper 10 is just butted with the buried garbage pit 40;

the reader 50 automatically recognizes the target object by radio frequency wireless signals and acquires related data, and is composed of an electronic tag, a reader/writer, and a computer network. The radio frequency identification system uses an electronic tag to identify an object, the electronic tag exchanges data with a reader-writer through radio waves, the reader-writer can transmit a read-write command of a host to the electronic tag and then transmit data returned by the electronic tag to the host, and a data exchange and management system of the host is responsible for finishing storage, management and control of data information of the electronic tag.

The automatic radio frequency identification technology is an automatic information identification technology that automatically identifies objects by radio frequency signals and captures and analyzes target data 50. Compared with the traditional identification technology, the technology has the advantages of high speed and high sensitivity, and the emergency stop operation can be immediately sent out when the reader 50 scans the information of the label 60 at the bottom of the unmanned vehicle, so that the alignment is completed.

The radio frequency automatic identification technology has the advantages of fast scanning, small volume, large capacity, strong penetrability, low cost and high reading accuracy, the automatic identification technology is more and more mature due to the advantages, the advantages of the technology are perfectly matched with the requirements of the unmanned sweeper, once the reader 50 fast scans the tag 60, a signal flow can be sent out to inform the sweeper to make an emergency stop operation, and finally high-precision alignment is completed.

As shown in fig. 2, when the reader 50 and the tag 60 are in the same vertical direction, the dust box of the unmanned sweeper fits the garbage pit most completely.

Step 5, after the label information is scanned, opening the buried garbage pit 40, then opening the dust box 20 by the unmanned sweeper 10, and pouring the garbage into the buried garbage pit 40;

and 6, returning the unmanned sweeper 10 to the cleaning operation area to continue the automatic cleaning task or returning to a charging pile for charging.

And designing an upper computer system to estimate the residual capacity of the buried garbage pit 40, and sending information to inform sanitation workers of cleaning before the buried garbage pit 40 is estimated to be full. The upper computer system mainly depends on the radio frequency automatic identification technology, and when the reader identifies the label 60 information of the unmanned sweeper once, the residual capacity is calculated according to the volume of the underground garbage pit 40, so that an early warning signal is sent out in advance, and the urban sanitation is informed to clean the garbage station.

According to the method for automatically aligning the garbage pit of the unmanned sweeper with the underground garbage station, disclosed by the invention, on the basis that the unmanned sweeper can complete automatic driving and cleaning, the ground image of the garbage station can be acquired through the high-definition camera in front of the unmanned sweeper, the trunk ground and the guide ground of the garbage station are identified according to image processing, the vehicle body is enabled to follow the guide ground to continuously correct an advancing route, then, after a reader at the front side of the garbage pit detects the label information at the bottom of the unmanned sweeper, the scram information is immediately sent out, and at the moment, the alignment of the unmanned sweeper with the garbage pit is completed.

It is to be understood that the present invention has been described with reference to certain embodiments, and that various changes in the features and embodiments, or equivalent substitutions may be made therein by those skilled in the art without departing from the spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.