CN112000094A - Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method - Google Patents
Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method Download PDFInfo
- Publication number
- CN112000094A CN112000094A CN202010696179.4A CN202010696179A CN112000094A CN 112000094 A CN112000094 A CN 112000094A CN 202010696179 A CN202010696179 A CN 202010696179A CN 112000094 A CN112000094 A CN 112000094A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- binocular
- transmission line
- line
- identification
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000005540 biological transmission Effects 0.000 title claims abstract description 37
- 238000007689 inspection Methods 0.000 claims abstract description 52
- 238000005259 measurement Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 13
- 230000000007 visual effect Effects 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000005516 engineering process Methods 0.000 claims abstract description 6
- 239000000725 suspension Substances 0.000 claims description 16
- 230000004888 barrier function Effects 0.000 claims description 11
- XYDVHKCVOMGRSY-UHFFFAOYSA-N 4-(4-benzylphenyl)-1,3-thiazol-2-amine Chemical compound S1C(N)=NC(C=2C=CC(CC=3C=CC=CC=3)=CC=2)=C1 XYDVHKCVOMGRSY-UHFFFAOYSA-N 0.000 claims description 9
- 101000928034 Homo sapiens Proteasomal ubiquitin receptor ADRM1 Proteins 0.000 claims description 9
- 102100036915 Proteasomal ubiquitin receptor ADRM1 Human genes 0.000 claims description 9
- 101001050286 Homo sapiens Jupiter microtubule associated homolog 1 Proteins 0.000 claims description 8
- 102100023133 Jupiter microtubule associated homolog 1 Human genes 0.000 claims description 8
- 230000009471 action Effects 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 238000002474 experimental method Methods 0.000 claims description 7
- 125000006850 spacer group Chemical group 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 4
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 3
- 239000010931 gold Substances 0.000 claims description 3
- 229910052737 gold Inorganic materials 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 abstract description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000004392 development of vision Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02G—INSTALLATION OF ELECTRIC CABLES OR LINES, OR OF COMBINED OPTICAL AND ELECTRIC CABLES OR LINES
- H02G1/00—Methods or apparatus specially adapted for installing, maintaining, repairing or dismantling electric cables or lines
- H02G1/02—Methods or apparatus specially adapted for installing, maintaining, repairing or dismantling electric cables or lines for overhead lines or cables
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Electromagnetism (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method, and belongs to the technical field of electric power operation robots. In the prior art, the research on hardware identification of the inspection robot is concentrated on a monocular identification and ranging system, and the method has the disadvantages of excessively complicated algorithm, poor real-time performance and low positioning precision. The system implementation of the invention comprises the following steps: acquiring images along the high-voltage transmission line; carrying out SVM model training by using the acquired images; establishing a preset measuring point database of the obstacle hardware by using an obstacle real-time distance measuring technology based on target characteristics; a binocular camera is adopted to build a visual processing platform, and image acquisition is carried out along a line; recognizing the obstacles by using a binocular camera-based SVM multi-cycle multi-classification obstacle recognition method; calling corresponding measuring points from a preset measuring point library according to the types of the obstacles to carry out binocular vision distance measurement; and the movement of the inspection robot is controlled by combining the type of the obstacle and the ranging result, so that the obstacle is accurately crossed.
Description
Technical Field
The invention relates to a single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method, and belongs to the technical field of electric power operation robots.
Background
The high-voltage overhead transmission line is exposed outside for a long time and is easily influenced by natural disasters, bird damage, continuous tension and the like, so that the high-voltage overhead transmission line is damaged by abrasion, corrosion, strand breakage and the like. Once damage occurs, if the damage is not discovered and eliminated in time, large-area power failure and huge economic loss can be caused. The inspection robot is used for replacing manpower to inspect the overhead high-voltage transmission line, so that the influence of the manual inspection on inspection precision can be avoided, inspection blind spots are eliminated, the inspection efficiency and the inspection precision are improved, the line maintenance quality is ensured, the inspection intensity and risk are greatly reduced, the operation maintenance cost can be effectively reduced, the labor intensity of workers is reduced, the economic and social benefits are taken into consideration, and the inspection robot has important significance in the aspects of adapting to market demands and the like.
The inspection robot inevitably meets the blocking of hardware obstacles such as various suspension clamps, spacers, vibration dampers and the like when walking on the line, the premise of smoothly crossing the hardware obstacles is to accurately identify the front obstacle object and carry out corresponding obstacle crossing action, and the accurate identification of the line hardware obstacles is one of the key technologies of the inspection robot research.
At present, conventional sensors such as infrared sensors and ultrasonic sensors are mostly adopted to detect line obstacles in the inspection process of the inspection robot, but the conventional sensors are influenced by the environment, the climate and the like, the obstacle identification precision of the conventional sensors is not high, and thus the inspection robot is easy to make misjudgment so as to cause accidents.
With the continuous development of vision technology, the obstacle detection and identification method based on machine vision is gradually used in the inspection of inspection robots, and the current common obstacle detection and identification method is basically a monocular identification distance measurement system. For example: an algorithm for extracting depth by means of a single camera is provided in a document 'monocular distance measurement algorithm in visual navigation of a high-voltage line patrol robot' published by Chengli and Wukungping, the characteristic that the line patrol robot images far and near along a lead is utilized, a small hole imaging model is combined, the distance of an obstacle is obtained, the measurement precision meets the requirement, but the algorithm is too complicated, and the instantaneity is poor.
The patent provides a monocular and binocular combined high-voltage transmission line hardware fitting online identification and positioning system and method aiming at the problems that accuracy is low in sensor identification and real-time performance is poor due to the fact that an algorithm in a monocular identification distance measurement system is complex.
Disclosure of Invention
The invention aims to provide a single-and-double-purpose combined high-voltage transmission line hardware fitting online identification and positioning system and method, which are used for solving the problems of hardware fitting identification and distance measurement in the operation process of an inspection robot of an air high-voltage transmission line.
One of the invention contents of the invention is to design a single-double-purpose combined high-voltage transmission line hardware fitting online identification and positioning system, which comprises: the system comprises a ground base station, a cloud server, communication equipment, a control center, inspection equipment, a control panel, an environment sensing module, a sensor system, an execution motor and a driving module;
the ground base station is connected with the line patrol robot through a network bridge and communicates with the line patrol robot; a binocular camera of the control center captures images; the vision processing platform processes the image and transmits the image to an ARM1 in the control panel; the ARM1 receives information of the sensor system and the control center and transmits the information to the ARM 2; the ARM2 receives information of the environment sensing module and the ARM 1; ARM2 sends out obstacle crossing instruction to actuating motor and drive module, and inspection robot accomplishes obstacle crossing action, continues to patrol the line.
Preferably, the visual processing platform specifically refers to an embedded single-module supercomputer Jetson TX 2; the ARM1 collects signal information around the inspection robot in the control panel; ARM2 sends the action command in the control panel, and the control patrols line robot and realizes crossing the barrier.
The invention also discloses a monocular and binocular combined overhead high-voltage transmission line identification and positioning method, which is an online identification and positioning system of hardware fittings of a monocular and binocular combined high-voltage transmission line and comprises the following implementation steps:
s1, acquiring an image along a high-voltage power transmission line;
s2, carrying out SVM model training by using the image acquired in the step S1;
s3, establishing a preset measurement point database of the barrier hardware fitting by using a barrier real-time distance measurement technology based on target characteristics;
s4, a binocular camera is adopted to build a visual processing platform, and image acquisition is carried out along a line;
s5, identifying the type of the obstacle in the image in the step S4 through an SVM multi-cycle multi-classification obstacle identification method based on a binocular camera;
s6, calling corresponding measuring points from a preset measuring point database of the obstacle hardware fitting according to the types of the obstacles to carry out binocular vision distance measurement;
and S7, controlling the motion of the line inspection robot by combining the type of the obstacle and the distance measurement result, and realizing accurate obstacle crossing of the line inspection robot.
In step S2, preferably, the training of the SVM model involves four recognition models, which are respectively suitable for four hardware fittings, i-type suspension clamp, ii-type suspension clamp, spacer and damper, on the line, and four SVM classifiers are designed according to the four trained recognition models.
In step S3, preferably, the real-time distance measurement uses five distance measurement points, and obtains the preset point-taking rules of various obstacles with the upper left corner of the image recognition rectangular frame as the origin.
In step S3, preferably, the appearance and image characteristics of four hardware fittings on the high-voltage transmission line are studied, a binocular ranging experiment is performed on each hardware fitting, and a ranging sensitive point is studied, so that a measurement point database is preset according to the obstacle hardware fitting for establishing the inspection robot.
The binocular camera in the step S4 is a left camera and a right camera, specifically, a ZED binocular 3D stereo camera, which is located on the inspection robot.
The SVM multi-cycle multi-classification obstacle identification method based on the binocular camera in the step S5 comprises the following steps:
s5.1, carrying out image preprocessing on the image collected in the step S4;
s5.2, extracting the characteristics of the image preprocessed in the step S5.1;
s5.3, entering the extracted features into a series SVM multi-classifier, and preliminarily identifying whether obstacles exist or not;
and S5.4, importing the output results of the series SVM multi-classifiers into a data fusion module for fusion to obtain a final recognition result.
In step S5.3, preferably, if both the two classifiers have no obstacle, the next picture is identified, and if any one of the two classifiers shows an obstacle, the speed is reduced.
In step S5.4, the final recognition result is the type of the front obstacle.
The system and the method for identifying and positioning the hardware fittings of the high-voltage transmission line in a single-eye and double-eye combined manner have the advantages that the line inspection robot can accurately identify and accurately position various hardware fittings on the overhead transmission line in a classified manner, and the automatic and accurate obstacle crossing of the line inspection robot is realized by combining the types of obstacles and distance measurement results.
Drawings
Fig. 1 is a schematic diagram of a principle of a monocular and binocular combined overhead high-voltage transmission line identification and positioning method;
FIG. 2 is a binocular camera-based SVM multi-cycle multi-classification obstacle recognition method;
fig. 3 is a hardware composition schematic diagram of a single-double-eye combined high-voltage transmission line hardware online identification and positioning system.
The reference numerals include: 100-ground base station, 101-cloud server, 102-communication equipment, 103-control center, 104-control board, 105-sensor system, 106-execution motor and drive, 107-environment perception, 108-inspection equipment.
Detailed Description
The invention is described in further detail below with reference to the following figures and detailed description:
a monocular and binocular combined overhead high voltage transmission line identification and positioning method is shown in a combined figure 1, a combined figure 2 and a combined figure 3, and the implementation process comprises the following steps:
s1, acquiring an image along a high-voltage power transmission line;
s2, carrying out SVM model training by using the image acquired in the step S1;
s3, establishing a preset measurement point database of the barrier hardware fitting by using a barrier real-time distance measurement technology based on target characteristics;
s4, a binocular camera is adopted to build a visual processing platform, and image acquisition is carried out along a line;
s5, identifying the type of the obstacle in the image in the step S4 through an SVM multi-cycle multi-classification obstacle identification method based on a binocular camera;
s6, calling corresponding measuring points from a preset measuring point database of the obstacle hardware fitting according to the types of the obstacles to carry out binocular vision distance measurement;
and S7, controlling the motion of the line inspection robot by combining the type of the obstacle and the distance measurement result, and realizing accurate obstacle crossing of the line inspection robot.
The training of the SVM model in the step S2 relates to four recognition models in total, and the training is respectively suitable for four hardware fittings, namely a suspension clamp I type, a suspension clamp II type, a spacer and a damper, on a line, and four SVM classifiers are designed according to the four trained recognition models.
In step S3, the real-time obstacle ranging technique based on the target features is implemented as follows:
in order to prevent the selected distance measuring points from deviating from the obstacle, the proper distance measuring points need to be selected at the positions where the obstacle area of the line hardware is large and the parallax map generation effect is optimal for distance measurement.
The distance precision is high when the number of the taken distance measurement points is large, but the response of real-time output distance information is slow; the distance precision will be reduced when the number of the distance measuring points is small. Through multiple point taking experimental tests, it is finally determined that the most suitable distance measuring points are less than or equal to 5, the distance measuring precision requirement is met, and the distance information can be output in real time. The method is characterized in that two common suspension clamps of a suspension clamp I type and a suspension clamp II type are arranged according to different aluminum wire sections and different voltages, external characteristic analysis and distance measurement experiments are carried out on four obstacles of the suspension clamp I type, the suspension clamp II type, a spacer and a damper, it is determined that the distance measurement points of the suspension clamp and the damper are 5, and the distance measurement points of the novel suspension clamp and the spacer are 4.
And after the preset rule of the distance measuring point of each obstacle is determined, establishing a preset measuring point database of the obstacle hardware fitting. During ranging, when the system identifies the obstacle species, the corresponding preset point coordinates are selected for ranging. The upper left corner of the image recognition rectangular frame is used as an origin (x, y), the width is w, and the height is h, so that the preset point-taking rule of various obstacles is as follows:
i type of suspension clamp gets a coordinate: p1(x + w/3, y +37h/60), P2(x + w/2, y +37h/60), P3(x +2w/3, y +37h/60), P4(x +5w/12, y +4h/5), P5(x +7w/12, y +4 h/5);
suspension clamp II type point coordinates: p1(x +7w/12, y +11h/30), P2(x +7w/12, y +17h/30), P3(x +11w/12, y +17h/30), P4(x + w/3, y +19 h/30);
and (3) taking point coordinates of the spacers: p1(x +7w/30, y +2h/5), P2(x +11w/15, y +2h/5), P3(x +7w/30, y +2h/3), P4(x +11w/15, y +2 h/3);
and (3) coordinate of a vibration damper taking point: p1(x +4w/11, y +13h/23), P2(x +8w/11, y +13h/23), P3(x +6w/11, y +15h/23), P4(x +4w/11, y +18h/23), P5(x +8w/11, y +18 h/23).
The distance measuring points are based on the identification frame, when the line obstacle is identified from far to near, the identification frame is changed from small to large, and the corresponding width w and height h are changed along with the change, so that the positions of the acquired five distance measuring points are also changed correspondingly, and the distance measuring points can adapt to the changed identification frame.
Since the inspection robot inevitably shakes while walking on the line, the taken point deviates from the preset fixed position and an error data value is generated. In the process of manual point-taking distance measurement experiment, the same data value 16000 is generated as long as the taken distance measurement point is not in the profile of the disparity map, so after the point-taking distance measurement is well performed, if the error value occurs, all distance measurement values are added and subtracted by 16000n (n is the number of times of occurrence of 16000), finally, the sum of the accurate distance measurement values is divided by the corresponding number of the point-taking points, the obtained average value is the distance of the obstacle, and the algorithm is high in distance measurement speed and strong in real-time performance.
In step S3, preferably, the appearance and image characteristics of four hardware fittings on the high-voltage transmission line are studied, a binocular ranging experiment is performed on each hardware fitting, and a ranging sensitive point is studied, so that a measurement point database is preset according to the obstacle hardware fitting for establishing the inspection robot.
The binocular camera in the step S4 is a left camera and a right camera, specifically, a ZED binocular 3D stereo camera, which is located on the inspection robot.
As shown in fig. 2, the method for identifying an obstacle based on the binocular camera SVM multi-cycle multi-classification in step S5 mainly includes the following steps:
s5.1, carrying out image preprocessing on the image collected in the step S4;
s5.2, extracting the characteristics of the image preprocessed in the step S5.1;
s5.3, entering the extracted features into a series SVM multi-classifier, and preliminarily identifying whether obstacles exist or not;
and S5.4, importing the output results of the series SVM multi-classifiers into a data fusion module for fusion to obtain a final recognition result.
In step S5.3, preferably, if both the two classifiers have no obstacle, the next picture is identified, and if any one of the two classifiers shows an obstacle, the speed is reduced.
In step S5.4, the identification is the type of the preceding obstacle as a result.
A high tension transmission line gold utensil on-line identification positioning system that single mesh combines, its constitutional structure includes: ground basic station, cloud ware, communications facilities, control center, equipment of patrolling and examining, control panel, environmental perception module, sensor system, actuating motor and drive module combine fig. 1, fig. 2, fig. 3 to show, and the concrete implementation process includes:
s8, the ground base station 100 is connected with the line patrol robot through a network bridge and communicates with the line patrol robot;
s9, capturing images by a binocular camera of a vision processing platform in the control center 103;
s10, processing the image by the visual processing platform and transmitting the image to an ARM1 of the control panel 104;
s11, transmitting information of the ARM1 comprehensive sensor system 105 and the control center 103 to ARM 2;
s12, sending an obstacle crossing instruction to the execution motor and drive module 106 by the information of the ARM2 comprehensive environment sensing module 107 and the ARM 1;
and S13, the line inspection robot finishes obstacle crossing action and continues to inspect the line.
The visual processing platform in the step S10 is specifically an embedded single-module supercomputer Jetson TX 2; the ARM1 collects signal information around the inspection robot in the control panel 104.
And the ARM2 in the step S11 sends an action instruction in the control panel to control the inspection robot to realize obstacle crossing.
The inventor has set up emulation transmission line environment in the open air, carries out the overall arrangement according to actual line environment structure, and the main gold utensil obstacle on this circuit has: the inspection robot can walk, cross obstacles, climb and the like on a simulation line, and meets the environment of experimental requirements.
And a binocular camera is arranged outside the arm in the walking direction of the inspection robot and connected with a vision processing platform for experiment. During the experiment, the inspection robot detects the obstacle at a position about 1.5 meters away from the obstacle, the deceleration in the step S5.3 is executed, the real-time distance from the obstacle to the inspection robot is continuously measured by slowly advancing, and the relative error of the distance measurement is about 20 percent. Along with the line patrol robot is closer to the barrier, the error of real-time distance measurement is smaller and smaller, the error reaches the minimum value about 0.5m, and the relative error is 4% at the moment. When the distance is 0.5m, the inspection robot starts to perform obstacle crossing behavior, and the obstacle crossing process is as follows:
when the inspection robot detects that the distance between the inspection robot and the obstacle is 0.5m, the state of the middle arm is changed from the state of not contacting the electric wire to the state of clamping the electric wire;
after the middle arm is clamped, the front arm is changed from a clamping state to a non-contact wire state;
after the front arm passes through the barrier, the front arm is changed from the non-contact wire state to the wire clamping state;
after the clamping of the front arm is finished, the middle arm is changed from a clamping state to a non-contact wire state;
after the middle arm passes through the barrier, the state of the middle arm is changed from the state of not contacting the electric wire to the state of clamping the electric wire;
after the middle arm is clamped, the rear arm is changed from a clamping state to a non-contact wire state;
after the rear arm passes through the barrier, the rear arm is changed from the state of not contacting the electric wire to the state of clamping the electric wire;
after the clamping of the rear arm is finished, the middle arm is changed from a clamping state to a non-contact wire state;
and the inspection robot finishes the obstacle crossing action and continues to perform inspection work along the power transmission line.
It will be apparent to those skilled in the art that modifications and equivalents may be made in the embodiments and/or portions thereof without departing from the spirit and scope of the present invention.
Claims (10)
1. The utility model provides a high tension transmission line gold utensil on-line identification positioning system that monocular and binocular combine which characterized in that includes: the system comprises a ground base station, a cloud server, communication equipment, a control center, inspection equipment, a control panel, an environment sensing module, a sensor system, an execution motor and a driving module;
the ground base station is connected with the line patrol robot through a network bridge and communicates with the line patrol robot; a binocular camera of the control center captures images; the vision processing platform processes the image and transmits the image to an ARM1 in the control panel; the ARM1 receives information of the sensor system and the control center and transmits the information to the ARM 2; the ARM2 receives information of the environment sensing module and the ARM 1; ARM2 sends out obstacle crossing instruction to actuating motor and drive module, and inspection robot accomplishes obstacle crossing action, continues to patrol the line.
2. The system for on-line identification and positioning of the high-voltage transmission line hardware fittings combined in single and double eyes according to claim 1, wherein the visual processing platform is specifically an embedded single-module supercomputer Jetson TX 2; the ARM1 collects signal information around the inspection robot in the control panel.
3. The system for identifying and positioning the hardware of the high-voltage transmission line in a monocular-binocular mode according to claim 1, wherein ARM2 sends an action instruction in a control board to control a line inspection robot to achieve obstacle crossing.
4. A single-double-eye combined overhead high-voltage transmission line hardware online identification and positioning method is characterized by comprising the following implementation steps of:
s1, acquiring an image along a high-voltage power transmission line;
s2, carrying out SVM model training by using the image acquired in the step S1;
s3, establishing a preset measurement point database of the barrier hardware fitting by using a barrier real-time distance measurement technology based on target characteristics;
s4, a binocular camera is adopted to build a visual processing platform, and image acquisition is carried out along a line;
s5, identifying the type of the obstacle in the image in the step S4 through an SVM multi-cycle multi-classification obstacle identification method based on a binocular camera;
s6, calling corresponding measuring points from a preset measuring point database of the obstacle hardware fitting according to the types of the obstacles to carry out binocular vision distance measurement;
and S7, controlling the motion of the line inspection robot by combining the type of the obstacle and the distance measurement result, and realizing accurate obstacle crossing of the line inspection robot.
5. The monocular and binocular combined overhead high voltage transmission line recognition and positioning method according to claim 4, wherein the training of the SVM model in the step S2 involves four recognition models which are respectively applicable to four hardware fittings of a suspension clamp I type, a suspension clamp II type, a spacer and a damper on the line, and four SVM classifiers are designed according to the trained four recognition models.
6. The monocular and binocular combined overhead high voltage transmission line identification and positioning method as claimed in claim 4, wherein the real-time ranging in the step S3 uses five ranging points, and the preset point-taking rules of various obstacles are obtained by taking the upper left corner of the image identification rectangular frame as an origin.
7. The method according to claim 4, wherein in step S3, four hardware shapes and image characteristics on the high voltage transmission line are studied, and a binocular ranging experiment is performed on each hardware to study a ranging sensitive point, so that a measurement point database is preset according to the establishment of the obstacle hardware of the inspection robot.
8. The monocular and binocular combined overhead high voltage transmission line identification and positioning method as claimed in claim 4, wherein in step S4, the binocular cameras are left and right cameras, specifically, a ZED binocular 3D stereo camera, which is located on the line patrol robot.
9. The monocular and binocular combined overhead high voltage transmission line recognition and positioning method as claimed in claim 4, wherein the binocular camera based SVM multicycle multiclass obstacle recognition method in step S5 comprises the steps of:
s5.1, carrying out image preprocessing on the image collected in the step S4;
s5.2, extracting the characteristics of the image preprocessed in the step S5.1;
s5.3, entering the extracted features into a series SVM multi-classifier, and preliminarily identifying whether obstacles exist or not;
and S5.4, importing the output results of the series SVM multi-classifiers into a data fusion module for fusion to obtain a final recognition result.
10. The monocular and binocular combined overhead high voltage transmission line identification and positioning method as recited in claim 9, wherein in step S5.3, if both the two classifiers have no obstacle, the identification of the next picture is continued, and if any one of the classifiers shows an obstacle, the speed is reduced; in step S5.4, the final recognition result is the type of the front obstacle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010696179.4A CN112000094A (en) | 2020-07-20 | 2020-07-20 | Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010696179.4A CN112000094A (en) | 2020-07-20 | 2020-07-20 | Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112000094A true CN112000094A (en) | 2020-11-27 |
Family
ID=73466628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010696179.4A Pending CN112000094A (en) | 2020-07-20 | 2020-07-20 | Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112000094A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283496A (en) * | 2021-05-21 | 2021-08-20 | 国网宁夏电力有限公司中卫供电公司 | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation |
| CN113946154A (en) * | 2021-12-20 | 2022-01-18 | 广东科凯达智能机器人有限公司 | Visual identification method and system for inspection robot |
| CN114743010A (en) * | 2022-06-13 | 2022-07-12 | 山东科技大学 | Ultrahigh voltage power transmission line point cloud data semantic segmentation method based on deep learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4905151A (en) * | 1988-03-07 | 1990-02-27 | Transitions Research Corporation | One dimensional image visual system for a moving vehicle |
| CN109190712A (en) * | 2018-09-21 | 2019-01-11 | 福州大学 | A kind of line walking image automatic classification system of taking photo by plane based on deep learning |
| CN110687904A (en) * | 2019-12-09 | 2020-01-14 | 广东科凯达智能机器人有限公司 | Visual navigation routing inspection and obstacle avoidance method for inspection robot |
| CN110977973A (en) * | 2019-12-11 | 2020-04-10 | 国电南瑞科技股份有限公司 | Automatic obstacle crossing device of overhead transmission line inspection robot |
-
2020
- 2020-07-20 CN CN202010696179.4A patent/CN112000094A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4905151A (en) * | 1988-03-07 | 1990-02-27 | Transitions Research Corporation | One dimensional image visual system for a moving vehicle |
| CN109190712A (en) * | 2018-09-21 | 2019-01-11 | 福州大学 | A kind of line walking image automatic classification system of taking photo by plane based on deep learning |
| CN110687904A (en) * | 2019-12-09 | 2020-01-14 | 广东科凯达智能机器人有限公司 | Visual navigation routing inspection and obstacle avoidance method for inspection robot |
| CN110977973A (en) * | 2019-12-11 | 2020-04-10 | 国电南瑞科技股份有限公司 | Automatic obstacle crossing device of overhead transmission line inspection robot |
Non-Patent Citations (1)
| Title |
|---|
| 王吉岱: "基于双目视觉技术的高压输电线路巡检机器人在线测距", 《科学技术与工程》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283496A (en) * | 2021-05-21 | 2021-08-20 | 国网宁夏电力有限公司中卫供电公司 | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation |
| CN113283496B (en) * | 2021-05-21 | 2022-07-29 | 国网宁夏电力有限公司中卫供电公司 | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation |
| CN113946154A (en) * | 2021-12-20 | 2022-01-18 | 广东科凯达智能机器人有限公司 | Visual identification method and system for inspection robot |
| CN114743010A (en) * | 2022-06-13 | 2022-07-12 | 山东科技大学 | Ultrahigh voltage power transmission line point cloud data semantic segmentation method based on deep learning |
| CN114743010B (en) * | 2022-06-13 | 2022-08-26 | 山东科技大学 | Ultrahigh voltage power transmission line point cloud data semantic segmentation method based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112000094A (en) | Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method | |
| CN112990310B (en) | Artificial intelligence system and method for serving electric robot | |
| CN111958592B (en) | Image semantic analysis system and method for transformer substation inspection robot | |
| CA2950791C (en) | Binocular visual navigation system and method based on power robot | |
| Zhu et al. | Online camera-lidar calibration with sensor semantic information | |
| CN108890652B (en) | Transformer substation inspection robot and transformer substation equipment inspection method | |
| WO2018108098A1 (en) | Autonomous operation method for live working robot based on multi-sensor information fusion | |
| CN107146247A (en) | Automobile assistant driving system and method based on binocular camera | |
| CN110298330B (en) | Monocular detection and positioning method for power transmission line inspection robot | |
| CN109885091B (en) | Unmanned aerial vehicle autonomous flight control method and system | |
| CN112528979A (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN113900436B (en) | Inspection control method, inspection control device, inspection control equipment and storage medium | |
| CN112379605B (en) | Bridge crane semi-physical simulation control experiment system and method based on visual servo | |
| CN114359865A (en) | Obstacle detection method and related device | |
| EP3961583A3 (en) | Method for detecting obstacle, electronic device, roadside device and cloud control platform | |
| CN116866520B (en) | AI-based monorail crane safe operation real-time monitoring management system | |
| CN114529811A (en) | Rapid and automatic identification and positioning method for foreign matters in subway tunnel | |
| CN117115249A (en) | Container lock hole automatic identification and positioning system and method | |
| CN109977884B (en) | Target following method and device | |
| CN206416179U (en) | A kind of motion target tracking positioning and grasping system based on binocular vision | |
| CN113298044B (en) | Obstacle detection method, system, device and storage medium based on positioning compensation | |
| CN114905512A (en) | Panoramic tracking and obstacle avoidance method and system for intelligent inspection robot | |
| CN110516551B (en) | Vision-based line patrol position deviation identification system and method and unmanned aerial vehicle | |
| Fu et al. | Vision based navigation for power transmission line inspection robot | |
| Wang et al. | Hardware Identification & Distance Estimation for an High-Voltage Transmission Line Inspection Robot |
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 |