CN112211145A - Semi-automatic road sweeping method and device for road sweeper - Google Patents
Semi-automatic road sweeping method and device for road sweeper Download PDFInfo
- Publication number
- CN112211145A CN112211145A CN202010927993.2A CN202010927993A CN112211145A CN 112211145 A CN112211145 A CN 112211145A CN 202010927993 A CN202010927993 A CN 202010927993A CN 112211145 A CN112211145 A CN 112211145A
- Authority
- CN
- China
- Prior art keywords
- road
- ground
- road surface
- sweeper
- sweeping
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 241001417527 Pempheridae Species 0.000 title claims abstract description 77
- 238000010408 sweeping Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 17
- 230000003749 cleanliness Effects 0.000 claims abstract description 15
- 239000013598 vector Substances 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims description 17
- 230000011218 segmentation Effects 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 238000005265 energy consumption Methods 0.000 abstract description 3
- 230000005611 electricity Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01H—STREET CLEANING; CLEANING OF PERMANENT WAYS; CLEANING BEACHES; DISPERSING OR PREVENTING FOG IN GENERAL CLEANING STREET OR RAILWAY FURNITURE OR TUNNEL WALLS
- E01H1/00—Removing undesirable matter from roads or like surfaces, with or without moistening of the surface
- E01H1/08—Pneumatically dislodging or taking-up undesirable matter or small objects; Drying by heat only or by streams of gas; Cleaning by projecting abrasive particles
- E01H1/0827—Dislodging by suction; Mechanical dislodging-cleaning apparatus with independent or dependent exhaust, e.g. dislodging-sweeping machines with independent suction nozzles ; Mechanical loosening devices working under vacuum
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
- G01S11/12—Systems for determining distance or velocity not using reflection or reradiation using electromagnetic waves other than radio waves
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Abstract
The invention discloses a semi-automatic road sweeping method and a semi-automatic road sweeping device for a road sweeper, wherein the method comprises the following steps: collecting a color image and a depth image of a road surface in front of a vehicle through a vehicle-mounted binocular camera arranged on a road sweeper; collecting pictures under various road conditions, and making a data set of road surface cleanliness for training a convolutional neural network model; real-time road condition images acquired by the road sweeper are sent into a trained convolutional neural network for calculation, and vectors of the road surface cleanliness degree are output; detecting obstacles which obstruct the passage and appear in a passage space in the advancing direction of the road sweeper according to the depth image information; calculating the width of the road surface in front of the road sweeper according to the depth image, and guiding the road sweeper to automatically perform edge-sweeping operation; the invention greatly reduces the energy consumption of the road sweeper, improves the automation degree of the road sweeper and assists a driver to carry out road sweeping operation.
Description
Technical Field
The invention relates to the technical field of road sweepers, in particular to a semi-automatic road sweeping method and device for a road sweeper.
Background
The electric road sweeper is still in a starting stage at present, the electric road sweeper cannot be developed greatly all the time due to the limitation of battery cruising ability, the requirement of a house on the automatic operation of the road sweeper is increased day by day under the environment of the vigorous development of the automatic driving technology, and most of the existing automatic sweeping schemes need higher hardware cost, so that a method with proper cost and good algorithm stability is required to be found to improve the automation degree of the road sweeper and reduce energy consumption.
Chinese patent document No. CN106845408A discloses a street garbage recognition method in a complex environment based on DCNN, which requires first obtaining a street picture without garbage as prior knowledge, obtaining an image of garbage without background by performing comparison, and then training and classifying the garbage by using a DCNN model. The method needs prior knowledge, is suitable for being used in a fixed scene, the working scene of the road sweeper is not fixed, and the requirements on hardware cost and algorithm real-time performance are high, so that a lightweight model with good adaptability to scene changes needs to be found.
An automatic edge-patrolling method of a conventional road sweeper is disclosed, for example, in chinese patent No. CN110597250A, an automatic edge-patrolling system of a road sweeper is disclosed, a camera is used to assist an infrared sensor and an ultrasonic sensor to detect a road boundary, the method uses various sensors, hardware cost is high, and the camera is sensitive to ambient light and is difficult to adapt to outdoor operation requirements under various weather conditions.
Disclosure of Invention
The invention aims to provide a semi-automatic sweeping method and a semi-automatic sweeping device for a road sweeper, which are used for adjusting the rotating speed of a sweeping disc and the suction force of a fan according to the road surface condition, identifying obstacles in front of the road sweeper, early warning and braking, detecting the road boundary and guiding the road sweeper to automatically perform edge-sweeping operation.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
a semi-automated sweeping method for a sweeper, comprising:
acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image;
sending the obtained road surface color image to a pre-trained convolutional neural network model to obtain a current road surface cleanliness vector, and controlling the rotation speed of a sweeping disc of the road sweeper and the suction force of a fan of the road sweeper according to the cleanliness vector;
carrying out ground segmentation on the obtained pavement depth image to obtain ground points;
detecting obstacles according to the ground points, and guiding the road sweeper to perform early warning and brake according to the obstacle detection result;
and detecting the road boundary according to the ground point, and guiding the road sweeper to automatically perform edge-sweeping operation according to the road boundary detection result.
Further, the convolutional neural network model training comprises the following steps:
acquiring pictures under various road conditions, marking the acquired pictures in a classified manner, and making the classified and marked pictures into a data set; the road condition environment comprises the amount of garbage on the road surface, the type of the garbage on the road surface and the easy cleaning degree of the road surface;
and inputting the data set into a deep learning framework to build and train a convolutional neural network model.
Further, the acquiring ground points comprises the following steps:
acquiring point cloud P of camera coordinate system1;
The point cloud P1Point cloud P converted into vehicle coordinate systemc;
Setting a first height threshold and a second height threshold, points greater than the first height threshold and points less than the second height thresholdFiltering out points to obtain a ground candidate set P2;
Performing ground segmentation on the ground candidate set to obtain a ground point Pg。
Further, the obstacle detection includes the steps of:
point cloud P from the vehicle coordinate systemcAnd a ground point PgAcquiring point cloud P under vehicle coordinate system3
According to the point cloud P3Screening out obstacle candidate point PoAnd acquiring a point set of the obstacle according to the obstacle candidate points and calculating the distance coordinate of the obstacle.
Further, the road boundary detection includes:
dividing the acquired depth image into a plurality of grids according to the horizontal direction;
projecting the ground points to the grid to obtain a ground boundary point candidate set;
and clustering the candidate set of the boundary points, screening out partial outliers to obtain road boundary points, and calculating road boundary coordinates according to the boundary points.
The invention also discloses a semi-automatic road sweeping device for the road sweeper, which comprises
The binocular camera is used for acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image;
the control processing device is used for sending the obtained road surface color image to a pre-trained convolutional neural network model, obtaining a current road surface cleanliness vector, and controlling the rotating speed of a sweeping disc of the road sweeper and the suction force of a fan of the road sweeper according to the cleanliness vector; and
the system comprises a road surface depth image acquisition unit, a road surface segmentation unit and a road surface segmentation unit, wherein the road surface depth image acquisition unit is used for performing ground segmentation on the acquired road surface depth image to acquire ground points; and
the road sweeper is used for detecting obstacles according to the ground points and guiding the road sweeper to perform early warning and brake according to the obstacle detection result; and
and the road boundary detection module is used for detecting the road boundary according to the ground points and guiding the road sweeper to automatically perform edge-sweeping operation according to the road boundary detection result.
Further, the binocular camera is installed in the front of the road sweeper and is 80-100cm away from the ground.
Furthermore, a temperature control module is connected in the binocular camera and used for controlling the binocular camera to work, and the working range of the temperature control module is-30-60 degrees.
Further, the control processing device comprises an industrial personal computer and a controller, the industrial personal computer is in signal connection with the controller through a conversion module, the industrial personal computer is in signal connection with the binocular camera, and the controller is used for controlling the road sweeper to work.
Furthermore, a display is connected to the control processing device.
According to the technical scheme, the embodiment of the invention at least has the following effects:
the invention saves electricity from the speed control of the sweeping disc, identifies the quantity of the garbage to be swept by adding a visual identification technology, and further controls the rotating speed of the sweeping disc in real time, thereby achieving the function of saving electricity, detecting the front obstacle of the road sweeper and carrying out early warning or active braking; the road boundary is detected to guide the road sweeper to automatically patrol and operate, and a foundation is laid for the subsequent automatic driving of the road sweeper.
Drawings
FIG. 1 is a block flow diagram of an automatic road sweeping method in accordance with an embodiment of the present invention;
fig. 2 is a block diagram of an automatic road sweeping device according to an embodiment of the present invention.
1. The invention can automatically adjust the sweeping speed and the sucking force of the sucking disc of the road sweeper according to the cleaning degree of the road surface in the running process of the road sweeper; detecting obstacles in front of the road sweeper and carrying out early warning or active braking; detecting a road boundary to guide the road sweeper to automatically perform edge-sweeping operation; greatly reduces the energy consumption of the road sweeper, improves the automation degree of the road sweeper and assists a driver to carry out road sweeping operation.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.
In order to increase the working time of the electric road sweeper and improve the automation degree of the road sweeper, electricity is saved from the speed control of the sweeping disc, the quantity of garbage to be swept is identified by adding a visual identification technology, and the rotating speed of the sweeping disc is further controlled in real time, so that the electricity saving function is achieved, and an obstacle in front of the road sweeper is detected and early warning or active braking is carried out; the road boundary is detected to guide the road sweeper to automatically patrol and operate, and a foundation is laid for the subsequent automatic driving of the road sweeper.
The invention discloses a semi-automatic road sweeping method for a road sweeper, which comprises the following steps: step 10, acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image.
Step 20, obtaining pictures under various road conditions, marking the obtained pictures in a classified mode, and making the pictures marked in the classified mode into a data set; the road condition environment comprises the amount of garbage on the road surface, the type of the garbage on the road surface and the easy cleaning degree of the road surface.
And step 30, inputting the data set into a deep learning framework to build and train a convolutional neural network model.
And step 40, sending the obtained road surface color image to a pre-trained convolutional neural network model to obtain a current road surface cleanliness vector, and controlling the rotating speed of a sweeping disc of the road sweeper and the suction force of a fan of the road sweeper according to the cleanliness vector.
The invention discloses a semi-automatic road sweeping method for a road sweeper, which further comprises the following steps: step 10, acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image.
Step 20, acquiring point cloud P of camera coordinate system1;
Step 30, point cloud P of camera coordinate system1Point cloud P converted into vehicle coordinate systemc;
Step 40, setting a first height threshold and a second height threshold, and adding points larger than the first height threshold and points smaller than the second heightFiltering out points of a degree threshold value to obtain a ground candidate set P2;
Step 50, carrying out ground segmentation on the ground candidate set to obtain a ground point Pg。
Step 60, point cloud P according to the vehicle coordinate systemcAnd a ground point PgObtaining a point cloud P of a vehicle coordinate system3
Step 70, according to the point cloud P3Screening out obstacle candidate point PoAnd acquiring a point set of the obstacle according to the obstacle candidate points and calculating the distance coordinate of the obstacle.
And step 80, guiding the road sweeper to perform early warning and braking according to the distance coordinates of the obstacles.
The invention discloses a semi-automatic road sweeping method for a road sweeper, which further comprises the following steps: step 10, acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image.
Step 20, acquiring point cloud P of camera coordinate system1;
Step 30, point cloud P of camera coordinate system1Point cloud P converted into vehicle coordinate systemc;
Step 40, setting a first height threshold and a second height threshold, filtering out points larger than the first height threshold and points smaller than the second height threshold, and obtaining a ground candidate set P2;
Step 50, carrying out ground segmentation on the ground candidate set to obtain a ground point Pg。
Step 60, dividing the acquired depth image into a plurality of grids according to the horizontal direction;
step 70, projecting the ground points to the grid to obtain a ground boundary point candidate set;
and 80, clustering the candidate set of the boundary points, screening out partial outliers to obtain road boundary points, and calculating road boundary coordinates according to the boundary points.
And 90, guiding the road sweeper to automatically perform edge-sweeping operation according to the road boundary coordinates.
The invention also provides a semi-automatic road sweeping device based on the method, the road sweeping device comprises a binocular camera, a control processing device and a display, wherein the binocular camera is used for acquiring the current road surface image; wherein the road surface image comprises a color image and a depth image.
And the control processing device is used for sending the obtained road surface color image to a pre-trained convolutional neural network model, obtaining a current road surface cleanliness vector, and controlling the rotating speed of a sweeping disc of the road sweeper and the suction force of a fan of the road sweeper according to the cleanliness vector.
And the system is used for carrying out ground segmentation on the acquired pavement depth image to acquire ground points.
The road sweeper is used for detecting obstacles according to the ground points and guiding the road sweeper to perform early warning and brake according to the obstacle detection result;
and the road boundary detection module is used for detecting the road boundary according to the ground points and guiding the road sweeper to automatically perform edge-sweeping operation according to the road boundary detection result.
The following describes the sweeping method and apparatus in detail:
(1) binocular camera
The binocular camera is installed at the position, which is 80-100cm away from the ground, in the front of the road sweeper, and can emit infrared light to form light spots and collect color images and depth images, and the temperature control module is installed inside the camera, so that the scheme can work in an environment of-30-60 ℃ and can work under the condition of weak ambient light.
(2) Training convolutional neural network model
And acquiring pictures under various road conditions, manually classifying the shot images according to indexes of the amount of garbage on the road surface and the easy cleaning degree, and marking corresponding labels to manufacture a data set for training the convolutional neural network. The garbage types comprise common road garbage such as shredded paper scraps, various leaves, plastic bottles, broken stone blocks, plastic bags and the like, pictures are shot in different time periods of a day, and the shooting weather comprises sunny days, cloudy days, rainy days and snowy days. And dividing the created data set into a training set and a test set, wherein the training set accounts for 70% of the total data volume, and the test set accounts for 30% of the total data volume.
The convolutional neural network model used in the present invention is characterized as follows: the total number of the models is 46 layers, the models are composed of 11 inclusion modules, the models are shown in the table, the output is linear operation followed by softmax operation, the output number corresponds to road condition classification, and the models are built and trained in a deep learning framework Tensorflow.
(3) Deployed on road-sweeping machines
The trained model is packaged and deployed on a vehicle-mounted industrial personal computer, a binocular camera is automatically opened after the road sweeper is started, a front road image is collected and sent into a trained convolutional neural network for calculation, a vector of the road surface cleaning degree is output, then the road surface cleaning degree is judged according to the cleanliness vector, corresponding control signals of the rotating speed of the sweeping disc and the suction force of the fan are converted, the industrial personal computer transmits the control signals to a controller through a conversion module, and the sweeping disc and the fan are controlled to perform corresponding reactions.
(4) Ground segmentation
The binocular camera needs to be calibrated after being installed, the pose of the camera, including the installation height of the camera and the Euler angle of the pose, is acquired by using a calibration tool, converted into a rotation matrix R and a translation matrix T, stored locally, and used for converting three-dimensional point coordinates under a camera coordinate system acquired by the camera into point coordinates under a vehicle coordinate system. The scheme can realize the real-time ground segmentation function, and the method comprises the following step of acquiring point cloud P under a camera coordinate system according to a binocular camera1And converted into a point cloud P under a vehicle coordinate system through a formulac(ii) a Setting a height threshold value, filtering out points with too high or too low height, and obtaining a ground candidate point set P2(ii) a Dividing the candidate point set to obtain a ground point Pg。
Pc=RP1+T
(5) Detecting obstacles at a certain height above ground
Point cloud PcRemoving the ground point PgThen obtaining the point cloud P3:
P3=PC-Pg
Screening out a candidate point P of the obstacle according to the height coordinate, the width coordinate and the distance coordinate of the pointoCandidate points are points in a designated space in front of the vehicle, clustering is carried out according to the distances among the points, the points belonging to the same object are classified into one class, a set of corresponding object points is obtained, a noise point set and a small obstacle point set are filtered according to the number of the points, a point set of an obstacle is obtained, and the distance coordinate of the obstacle is calculated; and transmitting the detected barrier information to the controller through the conversion module, and making corresponding early warning and braking work.
(6) Guiding road sweeper to automatically patrol edge
The ground is divided to obtain ground points, the image is divided into a plurality of bar-shaped grids according to the horizontal direction, the ground points are projected into the grids, ground points close to the boundary are screened from the grids to obtain a ground boundary point candidate point set, the point sets are clustered, partial outliers are screened out to obtain road boundary points, road boundary coordinates can be obtained through calculation, whether the width of the boundary is within the range allowed by normal operation or not is continuously judged, and the road sweeper is guided to carry out edge patrol operation.
It will be appreciated by those skilled in the art that the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The embodiments disclosed above are therefore to be considered in all respects as illustrative and not restrictive. All changes which come within the scope of or equivalence to the invention are intended to be embraced therein.
Claims (10)
1. A semi-automated sweeping method for a road sweeper, comprising:
acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image;
sending the obtained road surface color image to a pre-trained convolutional neural network model to obtain a current road surface cleanliness vector, and controlling the rotation speed of a sweeping disc of the road sweeper and the suction force of a fan of the road sweeper according to the cleanliness vector;
carrying out ground segmentation on the obtained pavement depth image to obtain ground points;
detecting obstacles according to the ground points, and guiding the road sweeper to perform early warning and brake according to the obstacle detection result;
and detecting the road boundary according to the ground point, and guiding the road sweeper to automatically perform edge-sweeping operation according to the road boundary detection result.
2. A semi-automated road sweeping method according to claim 1, wherein the convolutional neural network model training comprises the steps of:
acquiring pictures under various road conditions, marking the acquired pictures in a classified manner, and making the classified and marked pictures into a data set; the road condition environment comprises the amount of garbage on the road surface, the type of the garbage on the road surface and the easy cleaning degree of the road surface;
and inputting the data set into a deep learning framework to build and train a convolutional neural network model.
3. The semi-automated sweeping method of claim 1 wherein said acquiring ground points comprises the steps of:
acquiring point cloud P of camera coordinate system1;
The point cloud P1Point cloud P converted into vehicle coordinate systemc;
Setting a first height threshold value and a second height threshold value, filtering points larger than the first height threshold value and points smaller than the second height threshold value, and acquiring a ground candidate set P2;
Performing ground segmentation on the ground candidate set to obtain a ground point Pg。
4. A semi-automated road sweeping method according to claim 3, wherein the obstacle detection comprises the steps of:
point cloud P from the vehicle coordinate systemcAnd a ground point PgAcquiring point cloud P under vehicle coordinate system3
According to the point cloud P3Screening out obstacle candidate point PoAnd acquiring a point set of the obstacle according to the obstacle candidate points and calculating the distance coordinate of the obstacle.
5. The semi-automated sweeping method of claim 1, wherein the road boundary detection comprises:
dividing the acquired depth image into a plurality of grids according to the horizontal direction;
projecting the ground points to the grid to obtain a ground boundary point candidate set;
and clustering the candidate set of the boundary points, screening out partial outliers to obtain road boundary points, and calculating road boundary coordinates according to the boundary points.
6. A be used for semi-automatic road sweeping device of road sweeper which characterized in that includes:
the binocular camera is used for acquiring a current road surface image; wherein the road surface image comprises a color image and a depth image;
the control processing device is used for sending the obtained road surface color image to a pre-trained convolutional neural network model, obtaining a current road surface cleanliness vector, and controlling the rotating speed of a sweeping disc of the road sweeper and the suction force of a fan of the road sweeper according to the cleanliness vector; and
the system comprises a road surface depth image acquisition unit, a road surface segmentation unit and a road surface segmentation unit, wherein the road surface depth image acquisition unit is used for performing ground segmentation on the acquired road surface depth image to acquire ground points; and
the road sweeper is used for detecting obstacles according to the ground points and guiding the road sweeper to perform early warning and brake according to the obstacle detection result; and
and the road boundary detection module is used for detecting the road boundary according to the ground points and guiding the road sweeper to automatically perform edge-sweeping operation according to the road boundary detection result.
7. The semi-automatic sweeping device of claim 6 wherein the binocular camera is mounted at the front of the sweeping machine, the binocular camera being 80-100cm from the ground.
8. The semi-automatic road sweeper according to claim 6, wherein a temperature control module is connected in the binocular camera and used for controlling the binocular camera to work, and the working range of the temperature control module is-30-60 degrees.
9. The semi-automatic road sweeping device of claim 6, wherein the control processing device comprises an industrial personal computer and a controller, the industrial personal computer is in signal connection with the controller through a conversion module, the industrial personal computer is in signal connection with the binocular camera, and the controller is used for controlling the road sweeping machine to work.
10. The semi-automatic road sweeping device of claim 6, wherein a display is further connected to the control processing device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010927993.2A CN112211145A (en) | 2020-09-07 | 2020-09-07 | Semi-automatic road sweeping method and device for road sweeper |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010927993.2A CN112211145A (en) | 2020-09-07 | 2020-09-07 | Semi-automatic road sweeping method and device for road sweeper |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112211145A true CN112211145A (en) | 2021-01-12 |
Family
ID=74049208
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010927993.2A Pending CN112211145A (en) | 2020-09-07 | 2020-09-07 | Semi-automatic road sweeping method and device for road sweeper |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112211145A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112965493A (en) * | 2021-02-08 | 2021-06-15 | 甘肃建投重工科技有限公司 | Control method for automatically adjusting suction and sweeping power of road cleaning sanitation vehicle |
CN114991047A (en) * | 2022-08-02 | 2022-09-02 | 山东省凯麟环保设备股份有限公司 | Cleaning method and system based on intelligent vision and bidirectional ground cleanliness judgment |
CN115057139A (en) * | 2022-07-29 | 2022-09-16 | 徐州威卡电子控制技术有限公司 | Automatic garbage can identification system and identification method used on garbage truck |
WO2023113593A1 (en) | 2021-12-15 | 2023-06-22 | Ravo B.V. | Cleaning machine with suction power control |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB808026A (en) * | 1956-03-01 | 1959-01-28 | Deighton S Patent Flue & Tube | Improvements in or relating to mechanical road sweepers |
CN105404844A (en) * | 2014-09-12 | 2016-03-16 | 广州汽车集团股份有限公司 | Road boundary detection method based on multi-line laser radar |
CN108803599A (en) * | 2018-05-31 | 2018-11-13 | 北京智行者科技有限公司 | A kind of cleaning works method based on operating mode |
CN109024417A (en) * | 2018-07-24 | 2018-12-18 | 长安大学 | A kind of Intelligent road sweeper and its roadway pollutants recognition methods and control method |
CN109288455A (en) * | 2018-09-21 | 2019-02-01 | 北京智行者科技有限公司 | Refuse sweeping method and device |
-
2020
- 2020-09-07 CN CN202010927993.2A patent/CN112211145A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB808026A (en) * | 1956-03-01 | 1959-01-28 | Deighton S Patent Flue & Tube | Improvements in or relating to mechanical road sweepers |
CN105404844A (en) * | 2014-09-12 | 2016-03-16 | 广州汽车集团股份有限公司 | Road boundary detection method based on multi-line laser radar |
CN108803599A (en) * | 2018-05-31 | 2018-11-13 | 北京智行者科技有限公司 | A kind of cleaning works method based on operating mode |
CN109024417A (en) * | 2018-07-24 | 2018-12-18 | 长安大学 | A kind of Intelligent road sweeper and its roadway pollutants recognition methods and control method |
CN109288455A (en) * | 2018-09-21 | 2019-02-01 | 北京智行者科技有限公司 | Refuse sweeping method and device |
Non-Patent Citations (3)
Title |
---|
张永博等: "激光点云在无人驾驶路径检测中的应用", 《测绘通报》 * |
王灿等: "基于三维激光雷达的道路边界提取和障碍物检测算法", 《模式识别与人工智能》 * |
马新江等: "一种基于路缘特征的点云道路边界提取方法", 《遥感信息》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112965493A (en) * | 2021-02-08 | 2021-06-15 | 甘肃建投重工科技有限公司 | Control method for automatically adjusting suction and sweeping power of road cleaning sanitation vehicle |
WO2023113593A1 (en) | 2021-12-15 | 2023-06-22 | Ravo B.V. | Cleaning machine with suction power control |
NL2030142B1 (en) * | 2021-12-15 | 2023-06-27 | Ravo B V | Cleaning machine with suction power control |
CN115057139A (en) * | 2022-07-29 | 2022-09-16 | 徐州威卡电子控制技术有限公司 | Automatic garbage can identification system and identification method used on garbage truck |
CN114991047A (en) * | 2022-08-02 | 2022-09-02 | 山东省凯麟环保设备股份有限公司 | Cleaning method and system based on intelligent vision and bidirectional ground cleanliness judgment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112211145A (en) | Semi-automatic road sweeping method and device for road sweeper | |
CN109024417B (en) | Intelligent road sweeper and road pollutant identification method and control method thereof | |
CN109460709B (en) | RTG visual barrier detection method based on RGB and D information fusion | |
CN104200657B (en) | A kind of traffic flow parameter acquisition method based on video and sensor | |
CN109797691B (en) | Unmanned sweeper and driving method thereof | |
CN103413441B (en) | Road weather condition monitoring method | |
CN103413313B (en) | The binocular vision navigation system of electrically-based robot and method | |
CN111360780A (en) | Garbage picking robot based on visual semantic SLAM | |
CN103176185B (en) | Method and system for detecting road barrier | |
CN103714363B (en) | A kind of motor vehicle exhaust smoke video identification system | |
CN108398672B (en) | Forward-tilting 2D laser radar mobile scanning-based pavement and obstacle detection method | |
CN112257623B (en) | Road surface cleanliness judgment and automatic cleaning method and automatic cleaning environmental sanitation device | |
CN109934108B (en) | Multi-target and multi-type vehicle detection and distance measurement system and implementation method | |
CN104378539B (en) | The video structural extraction of semantics video camera and its method of scene adaptive | |
CN103901890A (en) | Outdoor automatic walking device based on family courtyard and system and method for controlling outdoor automatic walking device based on family courtyard | |
EP2813973B1 (en) | Method and system for processing video image | |
CN106529404A (en) | Imaging principle-based recognition method for pilotless automobile to recognize road marker line | |
CN106592499A (en) | Snow-shoveling robot and snow-shoveling method thereof | |
CN104881661A (en) | Vehicle detection method based on structure similarity | |
CN112149595A (en) | Method for detecting lane line and vehicle violation by using unmanned aerial vehicle | |
CN112257522A (en) | Multi-sensor fusion environment sensing method based on environment characteristics | |
CN114155720B (en) | Vehicle detection and track prediction method for roadside laser radar | |
CN113538357B (en) | Shadow interference resistant road surface state online detection method | |
KR20180075985A (en) | Apparatus for autonomous steering prediction considering driving environment and method thereof | |
CN107943061A (en) | A kind of model automobile automatic Pilot experimental provision and method based on machine vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210112 |
|
RJ01 | Rejection of invention patent application after publication |