CN113156932A - Obstacle avoidance control method and system for rail flaw detection vehicle - Google Patents
Obstacle avoidance control method and system for rail flaw detection vehicle Download PDFInfo
- Publication number
- CN113156932A CN113156932A CN202011601054.5A CN202011601054A CN113156932A CN 113156932 A CN113156932 A CN 113156932A CN 202011601054 A CN202011601054 A CN 202011601054A CN 113156932 A CN113156932 A CN 113156932A
- Authority
- CN
- China
- Prior art keywords
- flaw detection
- rail
- detection vehicle
- obstacle
- rail flaw
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 85
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 36
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 22
- 239000010959 steel Substances 0.000 claims abstract description 22
- 230000004888 barrier function Effects 0.000 claims abstract description 10
- 238000012544 monitoring process Methods 0.000 claims description 11
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 239000000523 sample Substances 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 2
- 230000005540 biological transmission Effects 0.000 claims 1
- 230000008569 process Effects 0.000 abstract description 6
- 206010063385 Intellectualisation Diseases 0.000 abstract description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000000007 visual effect Effects 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, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0238—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors
- G05D1/024—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors in combination with a laser
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L23/00—Control, warning or like safety means along the route or between vehicles or trains
- B61L23/04—Control, warning or like safety means along the route or between vehicles or trains for monitoring the mechanical state of the route
- B61L23/041—Obstacle detection
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0253—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 extracting relative motion information from a plurality of images taken successively, e.g. visual odometry, optical flow
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Optics & Photonics (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a rail flaw detection vehicle obstacle avoidance control method and a rail flaw detection vehicle obstacle avoidance control system, wherein the method comprises the following steps: collecting radar data and image data of the rail flaw detection vehicle in the advancing direction, establishing an interested space area for the radar data, and removing points outside the interested space area; clustering the interested space region of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on the rail flaw detection vehicle; and processing the image data in real time through the steel rail identification model, identifying and extracting the line characteristics, and controlling the running of the rail flaw detection vehicle according to the line characteristics. The invention realizes the detection of the barrier, the identification of the line and the automatic adjustment of the running state based on the radar detection and the image identification, simplifies the manual work amount of the rail flaw detection vehicle, improves the working efficiency, and simultaneously improves the safety performance, the intellectualization and the stability of the flaw detection process of the rail flaw detection vehicle.
Description
Technical Field
The invention belongs to the technical field of rail flaw detection, and particularly relates to a rail flaw detection vehicle obstacle avoidance control method and system.
Background
At present, along with the continuous development of unmanned automobiles in the field of automobiles, people have more demands on unmanned products. The same is true in the railway industry. The technologies related to radar and cameras are widely applied to unmanned driving, and the real road condition is subjected to imaging and data processing and is combined with a control system to generate the unmanned driving function.
The double-rail ultrasonic flaw detector used in the railway flaw detection industry still mainly adopts manual driving, a driver needs to care about the front road condition all the time in a driving route, the operation time is mostly late at night, the road condition is complex, the driver needs to make different operation judgment in time according to the road condition and adjust a flaw detection system, and high-strength work also causes great pressure to corresponding workers.
Aiming at the several conditions with the most frequent occurrence in the driving route: the detection of obstacles, curves, switches, crossings and the like requires a solution to further the automation and unmanned direction of the double-rail ultrasonic flaw detector.
Disclosure of Invention
The invention provides a rail flaw detection vehicle obstacle avoidance control method and system for solving the technical problems.
In order to solve the problems, the technical scheme of the invention is as follows:
a rail flaw detection vehicle obstacle avoidance control method comprises the following steps:
collecting radar data and image data of the rail flaw detection vehicle in the advancing direction, establishing an interested space area for the radar data, and removing points outside the interested space area;
clustering the interested space region of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on the rail flaw detection vehicle;
and processing the image data in real time through the steel rail identification model, identifying and extracting the line characteristics, and controlling the running of the rail flaw detection vehicle according to the line characteristics.
In one embodiment, establishing a spatial region of interest for the radar data and culling points outside the spatial region of interest further comprises:
acquiring and analyzing radar data, establishing a grid map and performing data projection;
calculating the grid density of the grid map, reserving the dense grid according to the preset grid density, deleting the sparse grid, representing the dense grid by four points, and generating a new data set as an interested space area.
In one embodiment, clustering the spatial region of interest of the radar data by a density clustering algorithm, and identifying and extracting the contour of the obstacle further comprises:
clustering is carried out in a new data set by adopting a DBSCAN-based algorithm, and a fuzzy line segment method is applied to extract a rectangular outline of the dynamic barrier.
In one embodiment, the target tracking of the obstacle further comprises:
and carrying out target tracking on the obstacle by a multi-target hypothesis tracking method and a Kalman filter.
In one embodiment, the image data is processed in real time by a steel rail identification model, and identifying and extracting the line features further comprises:
and sequentially carrying out image denoising, edge detection and Hough transformation on the image data based on the steel rail identification model to obtain the track curvature, and identifying according to the track curvature to obtain the line characteristics.
In one embodiment, the driving control of the rail-guided vehicle according to the line characteristics further comprises:
if the line characteristics are identified and extracted to be curves or turnouts, the rail flaw detection vehicle is subjected to deceleration passing processing, and the normal speed is recovered after passing;
if the line characteristics are identified and extracted as straight lines, normal speed passing processing is carried out on the rail flaw detection vehicle;
and if the road junction characteristic is identified and extracted as the road junction, processing the image data again through the road junction identification model, and identifying and extracting the road junction characteristic, wherein if the road junction characteristic is in an open state, continuing to drive, and if the road junction characteristic is in a closed state, stopping the rail flaw detection vehicle.
In one embodiment, the method further comprises the following steps:
the video collects the pictures of the rail flaw detection vehicle in the advancing direction in real time and transmits the pictures to the monitoring intervention platform;
and receiving an intervention instruction of the monitoring intervention platform, and controlling the running of the rail flaw detection vehicle according to the intervention instruction.
An obstacle avoidance control system for a rail flaw detection vehicle, comprising: the controller is respectively connected with the radar and the camera through signals;
the controller is used for establishing an interested space area for the radar data, eliminating points outside the interested space area, clustering the interested space area of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on the rail flaw detection vehicle;
the controller is used for processing the image data in real time through the steel rail identification model, identifying and extracting line characteristics and controlling the rail flaw detection vehicle to run according to the line characteristics.
In one embodiment, the monitoring system further comprises a driving recorder, wherein the driving recorder is used for collecting the images of the advancing direction of the rail flaw detection vehicle in real time through video and transmitting the images to the monitoring intervention platform.
In one embodiment, the controller is further used for receiving an intervention command for monitoring the intervention platform and controlling the running of the rail flaw detection vehicle according to the intervention command.
Compared with the prior art, the invention has the following advantages and positive effects:
the invention can detect the barrier in the advancing direction of the rail flaw detection vehicle in real time through the radar detection based on the density clustering algorithm so as to carry out obstacle avoidance processing in advance, can accurately detect the related information of the barrier under different complex weather environments, can detect the line characteristics of the advancing direction of the rail flaw detection vehicle in real time through the camera based on the steel rail identification model so as to remind driving control, can more intuitively identify the object type and road conditions, can judge the correctness of other identification results manually, can carry out driving perception of the rail flaw detection vehicle simultaneously by matching the two, and is combined with a control system, so that the operation of different driving control according to different road sections and different road conditions under the unmanned condition can be solved, and the safety performance, the intellectualization and the stability of the flaw detection process of the rail flaw detection vehicle are greatly improved.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention.
Fig. 1 is a schematic overall flow chart of a rail flaw detection vehicle obstacle avoidance control method according to the present invention;
fig. 2 is a schematic diagram of a radar data processing flow of the rail flaw detection vehicle obstacle avoidance control method of the present invention;
FIG. 3 is a schematic diagram of a rail filter of the obstacle avoidance control method for a rail vehicle according to the present invention;
fig. 4 is a structural block diagram of an obstacle avoidance control system of a rail flaw detection vehicle according to the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description will be made with reference to the accompanying drawings. It is obvious that the drawings in the following description are only some examples of the invention, and that for a person skilled in the art, other drawings and embodiments can be derived from them without inventive effort.
For the sake of simplicity, the drawings only schematically show the parts relevant to the present invention, and they do not represent the actual structure as a product. In addition, in order to make the drawings concise and understandable, components having the same structure or function in some of the drawings are only schematically illustrated or only labeled. In this document, "one" means not only "only one" but also a case of "more than one".
The following describes in detail a method and a system for controlling obstacle avoidance of a rail vehicle according to the present invention with reference to the accompanying drawings and specific embodiments.
Referring to fig. 1, the application provides an obstacle avoidance control method for a rail flaw detection vehicle, which includes the following steps:
collecting radar data and image data of the rail flaw detection vehicle in the advancing direction, establishing an interested space area for the radar data, and removing points outside the interested space area;
clustering the interested space region of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on the rail flaw detection vehicle;
and processing the image data in real time through the steel rail identification model, identifying and extracting the line characteristics, and controlling the running of the rail flaw detection vehicle according to the line characteristics.
The present embodiment will now be described in detail, but is not limited thereto.
The embodiment is suitable for automatic operation of railway line flaw detection equipment, wherein the rail flaw detection vehicle of the embodiment is equipment for detecting rail diseases, generally runs along the existing rails, can accurately identify obstacles and line characteristics in the advancing direction under the complex working environment of a tunnel through the embodiment, and carries out real-time processing and control on the rail flaw detection vehicle according to real-time detection, so that the safety performance, the intellectualization and the stability of the flaw detection process of the rail flaw detection vehicle are greatly improved.
The embodiment specifically obtains radar data of the advancing direction of the rail flaw detection vehicle through the laser radar, and obtains image data of the advancing direction of the rail flaw detection vehicle through the camera module.
Specifically, the visual angles of the laser radar are different when the laser radar collects the obstacle points each time, the coordinate change of the collected part of the obstacle points is large, and a plurality of obstacle points are irrelevant to the tracking of the obstacle, such as a road surface, leaves, a wall body, a high-voltage power grid and the like. Too many obstacle points can affect the extraction of the peripheral rectangular outline of the obstacle, so that the original data needs to be screened, an interested space area needs to be established for radar data, and points outside the interested space area need to be removed to optimize the radar data, wherein, referring to fig. 2, the optimization process is as follows: acquiring and analyzing radar data, establishing a grid map and performing data projection; calculating the grid density of the grid map, reserving the dense grid according to the preset grid density, deleting the sparse grid, representing the dense grid by four points, generating a new data set serving as an interested space area, removing the obstacle points such as branches and the ground which are not in the interested area through the screening, only reserving the rough outline of the obstacle, and being beneficial to extracting the peripheral rectangle of the obstacle.
Referring to fig. 2, in the present embodiment, a Density Clustering algorithm is used to cluster Spatial regions of interest of radar data, identify and extract contours of obstacles, perform target tracking on the obstacles, and perform obstacle avoidance processing on a rail-probe vehicle, wherein the present embodiment clusters a new data set by using a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) Based algorithm, extracts rectangular contours of dynamic obstacles by using a fuzzy line segment method, and performs target tracking on the obstacles by using a multi-target hypothesis tracking method and a kalman filter. When the obstacle is detected, the rail flaw detection vehicle can be generally stopped and give an alarm to remind workers to carry out obstacle clearing treatment. Specifically, the DBSCAN algorithm performs clustering by using density, that is, the number of objects (points or other spatial objects) included in a certain region in a clustering space is required to be not less than a given threshold, unlike the partitioning and hierarchical clustering method, it defines clusters as a maximum set of points connected by density, can partition a region with sufficiently high density into clusters, and can find clusters of any shape in a noisy spatial database, and cluster original data by combining a variable threshold, thereby greatly improving the accuracy of clustering, simplifying the accurate extraction of a rectangular contour of a dynamic obstacle by applying a fuzzy line segment method, and finally Tracking a target by applying a multi-target Hypothesis Tracking (MHT) method and a filter.
This embodiment carries out real-time processing to image data through rail identification model, discerns and draws the line characteristic to control of traveling to the track flaw detection car according to the line characteristic, wherein, carry out image denoising, edge detection, hough transform in proper order to image data based on rail identification model, obtain the track curvature, and obtain the line characteristic according to the track curvature recognition:
first, screening is performed according to the length of the straight line, a relatively short interference line segment is removed, and then the remaining straight line is screened through an orbit filter, which is shown in fig. 3.
Assuming that the curvature range of the track is (-c, c), the curvature range is divided into n groups, i.e. the curvature range is divided into n sections on average, then, taking the left rail as an example, all the edge points G (i, j) (G (i, j) > T on the image are traversedG,TGA preset threshold value of the edge strength) to obtain the curvature G corresponding to the point1(i, j), deriving from the curvature value the corresponding packet gid as follows:
meanwhile, the weight contribution of this edge point to the curvature of the gid group can be calculated as follows:
weight=G(i,j)|sin(D1(i,j)-Θ(i,j))|t
and (4) taking values. And finally, accumulating the weights obtained by calculation and the like to the corresponding curvature groups as shown in the following formula. The weight corresponding to the right rail can be calculated similarly and added to the group corresponding to the curvature of the right rail.
w1(gid)=w1(gid)+weight
Affected by many disturbing factors in the image.
In order to make the anti-interference capability of curvature estimation stronger, the weight information in each grouping neighborhood can be combined to obtain wlrThe following formula:
from wlrThe group gid with the largest weight is selected as follows:
from the calculated curvature groups, the corresponding curvature values can be obtained as estimates for the near track curvature:
and drawing a corresponding ideal track on the image according to the curvature, wherein the ideal track is used as the line characteristic obtained by identification. Then, the embodiment can perform running control on the rail flaw detection vehicle based on the line characteristics:
if the line characteristics are identified and extracted to be curves or turnouts, the rail flaw detection vehicle is subjected to deceleration passing processing, and the normal speed is recovered after passing;
if the line characteristics are identified and extracted as straight lines, normal speed passing processing is carried out on the rail flaw detection vehicle;
and if the road junction characteristic is identified and extracted as the road junction, processing the image data again through the road junction identification model, and identifying and extracting the road junction characteristic, wherein if the road junction characteristic is in an open state, continuing to drive, and if the road junction characteristic is in a closed state, stopping the rail flaw detection vehicle.
The image processing of the embodiment utilizes a neural network algorithm to perform algorithm training through a large number of related pictures to obtain a steel rail model, so that the pictures shot by a camera are processed in real time, steel rail lines are extracted, and the identification is performed according to different characteristic points of a curve, a turnout, a crossing and other lines, so that the judgment of different conditions of the line such as a straight line, a curve, a turnout and the like is performed. And similarly, the crossing model is established to judge whether the crossing is in an open state or not.
Specifically, through machine learning in earlier stage to the crossing picture, whether be opening state according to crossing horizontal pole angle judgement, for avoiding partial horizontal pole machinery ageing, consider the crossing opening state when opening angle is greater than 80, be less than 80 for avoiding the horizontal pole transfer motion state consider closing state, the control system signal is fed back to after the image recognition, and it then stops advancing to receive the crossing closed state signal. The curve obtains the curvature radius according to the image recognition track, whether speed reduction processing is needed or not can be judged according to whether the curvature radius is smaller than 1000m, and when the control layer receives the signal recognized as the curve, the speed reduction is started to 10 km/h; and when the image identifies that the intersection point of the steel rail exists in the range of 30 meters, judging the steel rail to be a turnout, and controlling the train body to reduce the speed to 5km/h by the control layer.
Preferably, this embodiment further includes: the video collects the picture of the rail flaw detection vehicle in the advancing direction in real time, transmits the picture to the monitoring intervention platform, receives the intervention instruction of the monitoring intervention platform, and controls the running of the rail flaw detection vehicle according to the intervention instruction.
The operation of the present embodiment will now be further described with a more specific scenario.
In the in-service use, at double track formula rail flaw detector locomotive mid-mounting radar and camera to ensure not sheltering from, at the real-time collection system of radar below installation video, testing having multiple road conditions railway, set up radar scanning linear distance 30 meters, transverse distance 3 meters can include rail gauge width and the alone scope in the rail outside.
The equipment is normally driven to 15km/h, obstacles placed on the line in advance are identified, and the obstacles are set to be 0.1m2Above multiple size of multiple shape, the radar can realize multi-target synchronous identification and tracking, feeds back information to driving control operation interface when the barrier appears, shows that the red light is bright and reports to the police, and loudspeaker sound simultaneously. An operator can watch the driving environment in front of the vehicle in real time through the video acquisition system, can judge whether the vehicle is an obstacle or not, and stops the vehicle if the vehicle is not controlled by manual intervention.
The method comprises the steps that a camera shoots in real time during running of the device and processes data in a background, the background extracts the outline of the shot steel rail in a picture and draws a steel rail line, whether the outline is a curve or not is judged, and the curve is defined as the curve when the curvature radius calculated by the background in the system is smaller than 1000 meters. The shooting distance of the camera is fixed according to hardware model selection, so that the straight line distance of the shooting scene is fixed, and the curvature radius is calculated according to the curve offset distance. When the front side is identified as a curve, the deceleration control is carried out to decelerate the equipment to 10km/h for driving, after the front side is identified as a straight line, the equipment is driven for 30 m (determined by the shooting distance of a camera) to ensure that the equipment is driven out of the curve, and then the equipment is accelerated to 15km/h for driving.
The switch identification is the same as the curve identification, but the feature point reading is different. And extracting the steel rail line and calculating the curvature in curve identification, and judging the background of the image according to the characteristic that the steel rail in the turnout has a cross point after the steel rail line is extracted in turnout identification so as to judge whether the front road section is the turnout. And when the situation that the control flow is similar to the curve after the turnout is judged, the vehicle decelerates to 5km/h to drive, and accelerates after driving for 30 m after detecting the non-turnout in front.
Referring to fig. 4, the present application further provides an obstacle avoidance control system for a rail-guided vehicle based on the above embodiment, including: the controller is respectively connected with the radar and the camera through signals;
the controller is used for establishing an interested space area for the radar data, eliminating points outside the interested space area, clustering the interested space area of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on the rail flaw detection vehicle;
the controller is used for processing the image data in real time through the steel rail identification model, identifying and extracting line characteristics and controlling the rail flaw detection vehicle to run according to the line characteristics.
The controller of the embodiment adopts the programmable logic controller for carrying out data processing so as to carry out different working instructions on the equipment, and the vehicle speed in different commands is recorded in the control layer in advance, wherein the data information of the real-time encoder does not need to be collected for judgment, so that the data processing amount is saved, and the command processing speed is improved. The data processing layer that corresponds can handle radar data and camera data, classifies the different information that different equipment discerned again and gives the control layer with information transfer, improves the control layer definition, avoids taking place the information confusion, realizes clear differentiation to information such as barrier, curve, switch, crossing.
The embodiment can detect the barriers in the advancing direction of the rail flaw detection vehicle in real time through the radar detection based on the density clustering algorithm, so as to carry out obstacle avoidance processing in advance, the system can accurately detect the related information of the barriers under different complex weather environments, the line characteristics of the advancing direction of the rail flaw detection vehicle can be detected in real time through the camera based on the steel rail identification model, so as to remind to carry out driving control, the system can more visually identify the object type and the road condition, on the other hand, the correctness of other identification results can be judged manually, the system and the method are matched with each other to simultaneously carry out driving perception of the rail flaw detection vehicle, and are combined with a control system, so that the operation of different driving control according to different road sections and different road conditions under the unmanned condition can be solved, and the safety performance, the intellectualization and the stability of the flaw detection process of the rail flaw detection vehicle are greatly improved.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, it is still within the scope of the present invention if they fall within the scope of the claims of the present invention and their equivalents.
Claims (10)
1. A rail flaw detection vehicle obstacle avoidance control method is characterized by comprising the following steps:
collecting radar data and image data of a track flaw detection vehicle in the advancing direction, establishing an interested space area for the radar data, and removing points outside the interested space area;
clustering the interested space region of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on a rail flaw detection vehicle;
and processing the image data in real time through a steel rail identification model, identifying and extracting line characteristics, and controlling the running of the rail flaw detection vehicle according to the line characteristics.
2. The obstacle avoidance control method for the rail transit vehicle as claimed in claim 1, wherein the establishing a spatial region of interest for the radar data and eliminating points outside the spatial region of interest further comprises:
acquiring and analyzing the radar data, establishing a grid map and performing data projection;
and calculating the grid density of the grid map, reserving the dense grid according to the preset grid density, deleting the sparse grid, representing the dense grid by four points, and generating a new data set as the interested space area.
3. The obstacle avoidance control method for the rail transit vehicle as claimed in claim 2, wherein the clustering the spatial region of interest of the radar data by a density clustering algorithm, and the identifying and extracting the contour of the obstacle further comprises:
clustering is carried out in a new data set by adopting a DBSCAN-based algorithm, and a fuzzy line segment method is applied to extract the rectangular outline of the dynamic barrier.
4. The obstacle avoidance control method for the rail transit vehicle as claimed in claim 3, wherein the target tracking of the obstacle further comprises:
and carrying out target tracking on the obstacle by a multi-target hypothesis tracking method and a Kalman filter.
5. The obstacle avoidance control method for the rail-mounted flaw detection vehicle according to claim 1, wherein the image data is processed in real time through a steel rail identification model, and the identification and extraction of the line features further comprises:
and sequentially carrying out image denoising, edge detection and Hough transformation on the image data based on a steel rail identification model to obtain track curvature, and identifying and obtaining the line characteristics according to the track curvature.
6. The obstacle avoidance control method for the rail-probe vehicle according to claim 5, wherein the controlling of the rail-probe vehicle according to the line characteristics further comprises:
if the line characteristics are identified and extracted to be curves or turnouts, the rail flaw detection vehicle is subjected to deceleration passing processing, and the normal speed is recovered after passing;
if the line characteristics are identified and extracted to be straight lines, normal speed passing processing is carried out on the rail flaw detection vehicle;
and if the line characteristic is identified and extracted as a road junction, processing the image data again through a road junction identification model, and identifying and extracting road junction characteristics, wherein if the road junction characteristics are in an open state, the vehicle continues to run, and if the road junction characteristics are in a closed state, the vehicle is parked.
7. The obstacle avoidance control method for the rail transit vehicle as claimed in any one of claims 1 to 6, further comprising:
the video collects the pictures of the rail flaw detection vehicle in the advancing direction in real time and transmits the pictures to the monitoring intervention platform;
and receiving an intervention instruction of the monitoring intervention platform, and controlling the running of the rail flaw detection vehicle according to the intervention instruction.
8. The utility model provides a track flaw detection car keeps away barrier control system which characterized in that includes: the controller is respectively in signal connection with the radar and the camera;
the radar is used for acquiring radar data of the rail flaw detection vehicle in the advancing direction, the controller is used for establishing an interested space area for the radar data, eliminating points outside the interested space area, clustering the interested space area of the radar data through a density clustering algorithm, identifying and extracting the outline of an obstacle, carrying out target tracking on the obstacle and carrying out obstacle avoidance processing on the rail flaw detection vehicle;
the controller is used for processing the image data in real time through the steel rail identification model, identifying and extracting line characteristics, and controlling the rail flaw detection vehicle to run according to the line characteristics.
9. The obstacle avoidance control system for the rail flaw detection vehicle according to claim 8, further comprising a vehicle data recorder for video real-time acquisition of a picture of the rail flaw detection vehicle in the traveling direction and transmission to the monitoring intervention platform.
10. The obstacle avoidance control system for the rail flaw detection vehicle according to claim 8, wherein the controller is further configured to receive an intervention instruction of the monitoring intervention platform, and control the running of the rail flaw detection vehicle according to the intervention instruction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011601054.5A CN113156932A (en) | 2020-12-29 | 2020-12-29 | Obstacle avoidance control method and system for rail flaw detection vehicle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011601054.5A CN113156932A (en) | 2020-12-29 | 2020-12-29 | Obstacle avoidance control method and system for rail flaw detection vehicle |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113156932A true CN113156932A (en) | 2021-07-23 |
Family
ID=76878209
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011601054.5A Pending CN113156932A (en) | 2020-12-29 | 2020-12-29 | Obstacle avoidance control method and system for rail flaw detection vehicle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113156932A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104512434A (en) * | 2013-09-28 | 2015-04-15 | 沈阳新松机器人自动化股份有限公司 | Rail damage detection device and rail damage detection method |
CN108564525A (en) * | 2018-03-31 | 2018-09-21 | 上海大学 | A kind of 3D point cloud 2Dization data processing method based on multi-line laser radar |
CN109933075A (en) * | 2019-04-12 | 2019-06-25 | 成都宇俊盛科技有限公司 | A kind of single line laser radar survey barrier method for full-automatic unmanned rail vehicle |
CN110501719A (en) * | 2019-08-27 | 2019-11-26 | 王玉娇 | A kind of train obstacle detection method based on laser radar |
CN111985322A (en) * | 2020-07-14 | 2020-11-24 | 西安理工大学 | Road environment element sensing method based on laser radar |
CN112101128A (en) * | 2020-08-21 | 2020-12-18 | 东南大学 | Unmanned formula racing car perception planning method based on multi-sensor information fusion |
-
2020
- 2020-12-29 CN CN202011601054.5A patent/CN113156932A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104512434A (en) * | 2013-09-28 | 2015-04-15 | 沈阳新松机器人自动化股份有限公司 | Rail damage detection device and rail damage detection method |
CN108564525A (en) * | 2018-03-31 | 2018-09-21 | 上海大学 | A kind of 3D point cloud 2Dization data processing method based on multi-line laser radar |
CN109933075A (en) * | 2019-04-12 | 2019-06-25 | 成都宇俊盛科技有限公司 | A kind of single line laser radar survey barrier method for full-automatic unmanned rail vehicle |
CN110501719A (en) * | 2019-08-27 | 2019-11-26 | 王玉娇 | A kind of train obstacle detection method based on laser radar |
CN111985322A (en) * | 2020-07-14 | 2020-11-24 | 西安理工大学 | Road environment element sensing method based on laser radar |
CN112101128A (en) * | 2020-08-21 | 2020-12-18 | 东南大学 | Unmanned formula racing car perception planning method based on multi-sensor information fusion |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110296713B (en) | Roadside automatic driving vehicle positioning navigation system and single/multiple vehicle positioning navigation method | |
CN111461088B (en) | Rail transit obstacle avoidance system based on image processing and target recognition | |
CN106428558B (en) | A kind of track synthesis method for inspecting based on the dual-purpose unmanned plane of sky-rail | |
CN111198371A (en) | Forward-looking obstacle detection system | |
CN110211420B (en) | Method and system for judging whether current scene is parking lot or not and computer readable storage medium | |
CN108230254A (en) | A kind of full lane line automatic testing method of the high-speed transit of adaptive scene switching | |
CN107301776A (en) | Track road conditions processing and dissemination method based on video detection technology | |
CN106741890B (en) | A kind of high-speed railway safety detecting system based on the dual-purpose unmanned plane of empty rail | |
CN103176185A (en) | Method and system for detecting road barrier | |
CN109489584A (en) | A kind of tunnel clearance detection system and tunnel clearance recognition methods based on 3D technology | |
CN113568002A (en) | Rail transit active obstacle detection device based on laser and image data fusion | |
CN202294882U (en) | Railway locomotive shunting operation safety prevention and control system | |
CN114879160B (en) | Rail foreign matter invasion real-time monitoring method and system based on three-dimensional point cloud data | |
CN109871787A (en) | A kind of obstacle detection method and device | |
CN115113206B (en) | Pedestrian and obstacle detection method for assisting driving of underground rail car | |
CN114415171A (en) | Automobile travelable area detection method based on 4D millimeter wave radar | |
CN106494611B (en) | A kind of dual-purpose patrol unmanned machine of sky rail | |
CN113050654A (en) | Obstacle detection method, vehicle-mounted obstacle avoidance system and method for inspection robot | |
CN113799852A (en) | Intelligent active obstacle identification protection method supporting dynamic mode switching | |
CN114926422B (en) | Method and system for detecting passenger flow of getting on and off vehicles | |
Tianwen et al. | Research on obstacle detection method of urban rail transit based on multisensor technology | |
CN112508893B (en) | Method and system for detecting tiny foreign matters between double rails of railway based on machine vision | |
CN116631187B (en) | Intelligent acquisition and analysis system for case on-site investigation information | |
CN113156932A (en) | Obstacle avoidance control method and system for rail flaw detection vehicle | |
CN109116346B (en) | Radar-based human-vehicle identification access control system and method |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210723 |