CN109236292A - A kind of tunneling machine cutting Trajectory Planning System and method - Google Patents
A kind of tunneling machine cutting Trajectory Planning System and method Download PDFInfo
- Publication number
- CN109236292A CN109236292A CN201810738123.3A CN201810738123A CN109236292A CN 109236292 A CN109236292 A CN 109236292A CN 201810738123 A CN201810738123 A CN 201810738123A CN 109236292 A CN109236292 A CN 109236292A
- Authority
- CN
- China
- Prior art keywords
- cutting
- image
- planning
- grid
- tunneling machine
- 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.)
- Granted
Links
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21C—MINING OR QUARRYING
- E21C35/00—Details of, or accessories for, machines for slitting or completely freeing the mineral from the seam, not provided for in groups E21C25/00 - E21C33/00, E21C37/00 or E21C39/00
- E21C35/24—Remote control specially adapted for machines for slitting or completely freeing the mineral
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21C—MINING OR QUARRYING
- E21C31/00—Driving means incorporated in machines for slitting or completely freeing the mineral from the seam
- E21C31/08—Driving means incorporated in machines for slitting or completely freeing the mineral from the seam for adjusting parts of the machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Abstract
The invention discloses a kind of tunneling machine cutting Trajectory Planning System and method, system includes the image collection module being mounted in development machine main body, data processing module and control module；The method for planning track is the Grid Method environmental modeling based on regular hexagon, by cutting Region Decomposition be several sizes it is identical and with two value informations regular hexagon grid cell；Using cell decomposition using barrier as boundary, free area is decomposed into several free space units not overlapped, and the transfer from a territory element to another region is indicated by adjacent map；And logical algorithm calculates the planning of optimal path.The present invention can make cutting head of roadheader avoid dirt band automatically in the cutting course of tunnel, reduce pick abrasion, increase pick service life, accelerate cutting speed, improve cutting efficiency, to realize that the automation of development machine, efficient operation provide condition.
Description
Technical field
The present invention relates to a kind of development machines, and in particular to a kind of tunneling machine cutting Trajectory Planning System and method belong to well
Lower technical field of boring equipment.
Background technique
Coal mine roadway driving is one of important procedure in process of coal mining, and the speed of coal mine roadway driving speed will
Directly influence coal mining efficiency.In recent years, with the development of China's coal mines science and technology, though coal mine roadway driving technology and equipment
Very fast development is obtained, but since coal resources in China occurrence condition is complicated, tunnelling technology and equipment are still restriction coal mine
Highly efficient and productive key factor is exploited, the development of comprehensive pick lags far behind fully mechanized mining, and efficient mechanicalization driving is to guarantee that mine realizes height
Produce efficient necessary condition, and the developing direction of tunnelling technology.
Unmanned driving refers to underground coal mine digging device when not needing artificial direct intervention, by driving ring
The Intellisense in border, according to the self-service completion excavation operation of the program of regulation.In order to realize unmanned digging device operation track and
The accurate observing and controlling of posture needs to realize accurate pose measurement and automatic deviation correction in the presence of a harsh environment, nobody adopts in the prior art
Pick equipment track contexture by self is to plan the cutting track of digging device according to the performance indicator of setting, goes out its cutting
Regular end face.And the method for this path planning is roughly divided into two kinds, one is the global roads based on environment priori Complete Information
Diameter planning, also known as static or segregation reasons；Another kind is the local paths planning based on sensor information, also known as dynamic
Perhaps the shortcomings that online path planning is static or segregation reasons be can not Real-time Feedback subsurface environment, adaptability is poor, right
In the bad coal working face of geological conditions, this method receives serious limitation, and its accuracy is dependent on operator's
Working experience needs operator rule of thumb to judge coal-rock interface and spoil and adjusts cutting path, and there are hysteresis quality, drivings
Process is easy to appear deviation；And the sector planning path based on multi-sensor information, it needs to design multisensor parameter and comes to disconnected
Face formation hardness situation of change is described, but obtained description information is not comprehensive, and path planning is easy to cause mistake occur,
And the parameter for integrating multiple sensors is needed to be analyzed in information integration process, when some sensor degradation will result in instead
The information of feedback is not complete, so that analysis result be caused mistake occur.
Summary of the invention
In order to overcome various deficiencies of the existing technology, the present invention provide a kind of tunneling machine cutting Trajectory Planning System with
Method, can Real-time Feedback underground coal petrography situation, and description information is comprehensive, and path planning is based on multi-grid Map building, accurately
Du Genggao can effectively avoid tunneling machine cutting spoil, reduce the abrasion and destruction of pick, improve cutting speed.
To solve the above-mentioned problems, a kind of tunneling machine cutting Trajectory Planning System of the present invention, which is characterized in that including installation
Image collection module, data processing module and control module in development machine main body；Described image processing module includes anti-
Blasting shell body is regarded with the intracorporal CCD camera of explosion-proof shell, the explosion-proof casing of the video camera front end of CCD is mounted on equipped with high light transmission
Window is equipped with flame-proof type searchlight above CCD camera；The data processing module include embedded microprocessor system with
PLC control system, the image information of shooting is delivered to microprocessor system by CCD camera, after microprocessor system will be handled
Information be delivered to PLC control system as input variable signal；The control module includes being connected with cutting arm of tunneling machine
Electrohydraulic proportional control valve, the PLC control system output signal to the movement of electrohydraulic proportional control valve control cutting arm.
In order to reduce the influence that the vibration in development machine work process generates system, described image processing module, data
Processing module, control module pass through mounting bracket and vibration isolator is fixed in development machine main body.
A kind of tunneling machine cutting method for planning track, which comprises the following steps:
The first step, the image for treating cutting tunnel cutting section in advance carry out feature extraction, describe coal by BoW model
With dirt band image, coal and dirt band are distinguished using support vector machines training, the coal petrography texture information that will acquire tunnel cutting section is made
For learning sample；
Second step, CCD camera shooting image carry out Digital Image Processing, will be in its textural characteristics and learning sample
Texture model library compares, and then obtains section coal petrography distribution situation, and on this basis, using Grid Method to tunnel section
Carry out Map building；
Third step is horizontally divided into several territory elements to grating map using cell decomposition；Upper and lower two grid point value is not
With quantity Z > 3 when, difference faceted boundary line is split；
4th step, using the grid cell of zone boundary as the beginning and end of path planning, make the walking of exercise machine
Path covers all grids in each territory element, and the path planning route in each region is respectively created, and selects to repeat to cover
The least one group of route of rate is the optimum path planning route for being used as the territory element；
5th step, the optimum programming route that all areas are successively obtained according to four-step method.
Subregion is carried out to entire grid map, the shape of barrier in region can be made more regular, while in each area
Piecemeal carries out path planning in domain, and for the planning of compared to whole grid map directapath, operand is substantially reduced, and improves and is
It unites the response time, therefore cutting speed can effectively improve.
Further, grid cell used by the grating map of Map building is carried out to tunnel section in second step to be positive
Hexagon.Compared with square grid, regular hexagon grid map has bigger selectivity, square grid in the movement direction
Figure can only carry out the movement of four direction when moving between unit, and moving direction is available there are six regular hexagon tools, this
So that path planning has more flexibility and variability；And it is more accurate specific on the shape description to barrier.
Further, coal and dirt band image are described by BoW model, specific assorting process is as follows in the first step:
1) feature vector in training sample image block is extracted；
2) visual dictionary is constructed using K-Means algorithm；
3) training SVM classifier.
Further, for image captured by CCD in second step, image category is judged by following three step:
1) SIFT feature of image is first extracted；
2) image is expressed as numerical value histogram vector with the word in vocabulary；
3) classified by SVM classifier to spoil and coal body.
Compared with prior art, cutting paths planning method of the present invention, the cutting based on institute's captured in real-time are disconnected
The image in face carries out Grid Method modeling, while carrying out region division to grating map according to certain constraint condition, in each area
Optimum path planning is carried out in domain respectively, compared to existing paths planning method, it is disconnected that cutting can not only be reacted in real time, comprehensively
The case where face, additionally it is possible to optimal cutting path quickly be cooked up according to section situation, the time of information processing is shortened, improve
Cutting speed effectively prevents abrasion and destruction of the cutting spoil to pick during tunneling machine cutting, is also avoided that cutting cash
Dangerous temperature actuation gas caused by stone.
Detailed description of the invention
Fig. 1 is schematic structural view of the invention；
Fig. 2 is image collection module structural schematic diagram of the present invention；
Fig. 3 is present system schematic diagram；
Fig. 4 is that territory element boundary selects schematic diagram in the specific embodiment of the invention；
Fig. 5 is the path planning schematic diagram in the specific embodiment of the invention in territory element.
In figure: 1, development machine main body, 2, image collection module, 2-1, explosion-resistant enclosure, 2-2, high light transmission form, 2-3, CCD
Video camera, 2-4, flame-proof type searchlight, 3, data processing module, 4, vibration isolator, 5, mounting rack, 6, control module；
001, barrier one；002, barrier two；003, barrier three；004, barrier four；
01, region A；02, region B.
Specific embodiment
The present invention is described in detail with reference to the accompanying drawing.
As shown in Figure 1 to Figure 3, a kind of tunneling machine cutting Trajectory Planning System, which is characterized in that including being mounted on development machine
Image collection module 2, data processing module 3 and control module 6 in main body 1；Described image processing module 2 includes explosion-proof shell
Body 2-1 is equipped with high saturating with the intracorporal CCD camera 2-3 of explosion-proof shell, the explosion-proof casing of the front end video camera 2-3 of CCD is mounted on
Flame-proof type searchlight 2-4 is installed above light window 2-2, CCD camera 2-3；The data processing module 3 includes embedded
The image information of shooting is delivered to microprocessor system, micro- place by microprocessor system and PLC control system, CCD camera 2-3
Treated information as input variable signal is delivered to PLC control system by reason device system；The control module 6 includes and pick
The electrohydraulic proportional control valve being connected into machine cutting arm, the PLC control system output signal to electrohydraulic proportional control valve control and cut
Cut the movement of arm.
In order to reduce the influence that the vibration in development machine work process generates system, described image processing module 2, data
Processing module 3, control module 6 are fixed in development machine main body 1 by mounting bracket 5 and vibration isolator 4.
A kind of tunneling machine cutting method for planning track, which comprises the following steps:
The first step, the image for treating cutting tunnel cutting section in advance carry out feature extraction, describe coal by BoW model
With dirt band image, coal and dirt band are distinguished using support vector machines training, the coal petrography texture information that will acquire tunnel cutting section is made
For learning sample；
The BOW model of image indicates the histogram that the feature vector of all image blocks in i.e. image obtains.It is specific sorted
Journey is as follows:
1) feature vector in training sample image block is extracted
Image and gaussian kernel function are carried out convolution first by SIFT algorithm, are obtained Gaussian difference scale space, are passed through extreme value
To primarily determine position and the scale where key point, this step is cursorily to detect the position of point of interest, therefore obtain for point detection
To point of interest in contain a large amount of garbage, then these key points will accurately be positioned, to obtain its ruler
The information such as degree, direction, specific steps include point and the removal skirt response that precise interpolation positioned, filtered out low contrast, thus
The precise position information of required point of interest is obtained.In next step to each point of interest distribution direction and scale, each is special
Sign just has four parameters, the horizontal coordinate of central point, vertical coordinate, scale and the direction of central point.Final step is just
It is that feature is described, when feature being described using SIFT description, each feature will be expressed as the spy of 128 dimensions
Vector is levied, the neighborhood window of 16x 16 is taken using key point as the center of circle first, the window is then divided into 4x4 sub-regions, every
The gradient accumulated value of 8 directions (0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 °) is calculated in sub-regions, it is every in this way
A feature can be indicated with the vector that 4x4x8=128 is tieed up.Feature is described according to the method can be to avoid scale
Transformation, rotationally-varying influence.
2) visual dictionary is constructed using K-Means algorithm
K-Means algorithm is a kind of indirect clustering method based on similarity measurement between sample, by n in vector space
Characteristic point is divided into specified k class according to variance within clusters and the smallest principle, as shown in the formula of lower section.
Wherein SI(i=1,2 ... k) indicate that central point is μiIth cluster classification, xjExpression belongs to classification siData
Point, the specific steps of k-means clustering method include: that (1) randomly selects k initial center；(2) each data and cluster are calculated
N data are assigned to according to minimum distance principle using k initial center as in the cluster classification of representative by the distance at center；(3)
Center calculation is carried out to the k classification newly formed according to the result of previous step, obtains new cluster centre；(4) step (2) are repeated
(3), until result restrains.Then visual dictionary is constructed as " word " using k cluster centre.In this way when treating
When image of classifying carries out classification processing, feature extraction and description equally are carried out to it, it then will be in these features and visual dictionary
Word matched, most like word is found by certain method of discrimination, picture each in this way may be expressed as
The set of vision word probability of occurrence can be correspondingly indicated in a manner of statistic histogram by approximation, therefore, BoW model
Can be counted as the histogram based on independent characteristic indicates.
3) training SVM classifier,
After application k-means algorithm generates vision word, each image will be indicated by a word bag, herein
On the basis of classifier is trained, achieve the purpose that distinguish image category.
One classifier of good performance will not only meet the correctness of classification, also make the distance between the two sides of hyperplane
It is sufficiently large, with { xi,yi, i=1,2 ... n indicate the sample set of linear separability, the label of classification is indicated with y ∈ { 1, -1 },
If the dimension of the linear space is d dimension, then, linear discriminant function can indicate are as follows:
G (x)=ω x+b
Classifying face, that is, hyperplane equation can be expressed as:
ω x+b=0
Decision function can be with is defined as:
F (x)=sgn (ω x+b)
Enable the sample nearest from classifying face | g (x) |=1, then can then calculate class interval 2 | | ω | |, therefore,
If guaranteeing that the interval of two classes is maximum, then need to make | | ω | | or | | ω | |2It is minimum.According to analysis above, it is contemplated that after
Continuous convenience of calculation, linear separability support vector machines can be attributed to following optimization problem:
Need exist for the condition met are as follows:
yi(w·xi+ b) >=1, i=1,2 ... n
However many problems are all linearly inseparables in reality, need to increase loose item ζi>=0, it needs to meet:
yi(w·xi+b)-1+ζi≥0
It is calculated to simplify, can convert the Solve problems of above-mentioned optimal classification surface under conditions of upper establishment, ask
Following formula minimum:
Wherein, C is constant, and value is bigger, indicates bigger to the punishment of mistake classification.Here Lagrange multiplier is introduced to come
It is solved, the form of available following linear separability antithesis:
Constraint condition are as follows:
Wherein, αiIt may be there are three types of value:
(1)αi=0；(2) 0 < αi< C (；3)αi=C
The vector x for meeting (2)iReferred to as standard supporting vector, the vector for meeting condition (3) are known as boundary supporting vector.
And to the only supporting vector for determining that optimal hyperlane and decision function work.
Since only and in training sample supporting vector is related with the inner product operation of sample to be sorted for the discriminant classification function, because
This it is only necessary to know that the space inner product operation, the optimum linearity classification problem in this feature space can be solved.It needs exist for
It is applied to Mercer condition: enabling K (x, x) to indicate arbitrary symmetric function, it is abundant necessity of the inner product of some feature space
Condition is, for arbitraryAndHave
According to the Hilbert principle in Statistical Learning Theory, as long as there is a kind of operation to meet Mercer condition, it can
To replace inner product here to use.The dot product in above-mentioned optimal classification surface, obtained majorized function are replaced with inner product K (x, x) are as follows:
By above analysis discussion, the basic thought of SVM can be summarized are as follows: by define interior Product function appropriate into
The input space is transformed to a higher dimensional space, seeks optimal classification surface in this new higher dimensional space by row nonlinear transformation.
Second step, CCD camera shooting image carry out Digital Image Processing, will be in its textural characteristics and learning sample
Texture model library compares, and then obtains section coal petrography distribution situation, and on this basis, using Grid Method to tunnel section
Carry out Map building；
It is specific as follows, for image captured by CCD, image category is judged by following three step:
1) SIFT feature of image is first extracted；
2) image is expressed as numerical value histogram vector with the word in vocabulary；
3) classified by SVM classifier to spoil and coal body.
Carrying out grid cell used by the grating map of Map building to tunnel section is regular hexagon.With square grid
Lattice are compared, and regular hexagon grid map has bigger selectivity in the movement direction, when square grid figure moves between unit
The movement of four direction can only be carried out, and regular hexagon tool is available there are six moving direction, this has more path planning
There are flexibility and variability；And it is more accurate specific on the shape description to barrier.
Assuming that drift section maximum length is L, maximum width W, each hexagonal grid side length is a, then grid number is(size of a is judged that a selection is smaller to make cutting more accurate as the case may be, but cutting is imitated
Rate reduces, and selection is larger can to accelerate cutting speed, but can make accuracy decline).Then in each grid carry out coal petrography judgement,
If coal body area is greater than spoil area, the grid is determined for coal, demarcates grid point value i=1；It is on the contrary then be determined as dirt band, it marks
It is set to 0.And such as figure coordinate system is established, any region of whole cross section can be indicated with two bit arrays (x, y).The barrier
The black grid in object i.e. figure is hindered to represent spoil, grid point value i=0；White grid represents coal body, grid point value i=1 in figure.Lower section
Three barriers, wherein barrier one be regular shape, barrier two be regular shape, barrier three be irregular shape,
Third step is horizontally divided into several territory elements to grating map using cell decomposition；Upper and lower two grid point value is not
With quantity Z > 3 when, difference faceted boundary line is split；
By taking some cutting section modeling grating map in Fig. 4 as an example, in Fig. 4 in X=29 and two row grid of X=30, i
(29,1)=0, i (30,1)=1, upper and lower two grid point value is different, and such case is present in y=1, and 2 ... 8,14,15,16,
When 17, so the quantity Z=12 > 3 different from two grid point values above and below two row of X=30 for X=29, then basis has differences value
Faceted boundary, carry out horizontal partition between the boundary of (29,8) and (30,8) and the boundary of (29,14) and (30,14),
Horizontal partition is carried out between the boundary and map boundary line of (30,17) and (29,17), entire map is divided into several with relatively regular
The region of figure.And regional value k is assigned to each grid.
4th step, using the grid cell of zone boundary as the beginning and end of path planning, make the walking of exercise machine
Path covers all grids in each territory element, and the path planning route in each region is respectively created, and selects to repeat to cover
The least one group of route of rate is the optimum path planning route for being used as the territory element；
Specific steps are as follows: from the lower left corner (1,1) setting in motion, be laterally moved, and by the cutting shape of cutting grid
State value is changed to cutting h=1 by the h=0 to cutting, when encountering the i.e. i=0 of dirt band, is moved upwards, until in the region
The h of grid becomes 1, subsequently into next region.
By taking Fig. 4 and Fig. 5 as an example, A01 and the boundary region B02 intersection in region, the grid cell on the region boundary A01 is by a left side
It is A01~A05 to the right side, the grid cell on the region boundary B02 is B01-B06 from left to right；Entering area from region B02 unit
When the A01 unit of domain, can by A01-A05 any one as 01 unit cutting starting point, cellular zone is reached with the shortest distance to elder generation
Lower left corner grid in domain, is laterally moved, and the cutting state value of cutting grid is changed to cut by the h=0 to cutting
H=1 is cut, when encountering the i.e. i=0 of dirt band, is moved upwards, until the h of grid becomes 1 in the region, subsequently under
One region is compared using A01-A05 as the cutting repetitive rate in five kinds of paths of starting point, i.e., empty cutting grid number selects it
In a kind of least path planning scheme as the region,
5th step, the optimum programming route that all areas are successively obtained according to four-step method.
Subregion is carried out to entire grid map, the shape of barrier in region can be made more regular, while in each area
Piecemeal carries out path planning in domain, and for the planning of compared to whole grid map directapath, operand is substantially reduced, and improves and is
It unites the response time, therefore cutting speed can effectively improve.
Claims (6)
1. a kind of tunneling machine cutting Trajectory Planning System, which is characterized in that the image including being mounted on development machine main body (1) obtains
Modulus block (2), data processing module (3) and control module (6)；Described image processing module (2) includes explosion-proof casing (2-1)
Be mounted on the intracorporal CCD camera of explosion-proof shell (2-3), the explosion-proof casing of front end video camera (2-3) of CCD is equipped with high light transmission
Form (2-2) is equipped with flame-proof type searchlight (2-4) above CCD camera (2-3)；The data processing module (3) includes
The image information of shooting is delivered to microprocessor by embedded microprocessor system and PLC control system, CCD camera (2-3)
Treated information as input variable signal is delivered to PLC control system by system, microprocessor system；The control module
It (6) include the electrohydraulic proportional control valve being connected with cutting arm of tunneling machine, the PLC control system outputs signal to electric-hydraulic proportion control
The movement of valve control cutting arm processed.
2. tunneling machine cutting Trajectory Planning System according to claim 1, which is characterized in that described image processing module
(2), data processing module (3), control module (6) are fixed on development machine main body by mounting bracket (5) and vibration isolator (4)
(1) on.
3. a kind of tunneling machine cutting method for planning track, which comprises the following steps:
The first step, the image for treating cutting tunnel cutting section in advance carry out feature extraction, and coal and folder are described by BoW model
Cash image distinguishes coal and dirt band using support vector machines training, will acquire the coal petrography texture information of tunnel cutting section as
Practise sample；
The image that second step, CCD camera are shot carries out Digital Image Processing, by the texture in its textural characteristics and learning sample
Model library compares, and then obtains section coal petrography distribution situation, and on this basis, is carried out using Grid Method to tunnel section
Map building；
Third step is horizontally divided into several territory elements to grating map using cell decomposition；Upper and lower two grid point value is different
When quantity Z > 3, difference faceted boundary line is split；
4th step, using the grid cell of zone boundary as the beginning and end of path planning, make the walking path of exercise machine
All grids in each territory element are covered, the path planning route in each region is respectively created, and select to repeat coverage rate most
One group of few route is the optimum path planning route for being used as the territory element；
5th step, the optimum programming route that all areas are successively obtained according to four-step method.
4. tunneling machine cutting method for planning track according to claim 3, which is characterized in that tunnel section in second step
Carrying out grid cell used by the grating map of Map building is regular hexagon.
5. tunneling machine cutting method for planning track according to claim 4, which is characterized in that pass through BoW mould in the first step
Type describes coal and dirt band image, and specific assorting process is as follows:
1) feature vector in training sample image block is extracted；
2) visual dictionary is constructed using K-Means algorithm；
3) training SVM classifier.
6. tunneling machine cutting method for planning track according to claim 5, which is characterized in that for CCD institute in second step
The image of shooting judges image category by following three step:
1) SIFT feature of image is first extracted；
2) image is expressed as numerical value histogram vector with the word in vocabulary；
3) classified by SVM classifier to spoil and coal body.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810738123.3A CN109236292B (en) | 2018-07-06 | 2018-07-06 | Cutting track planning method for heading machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810738123.3A CN109236292B (en) | 2018-07-06 | 2018-07-06 | Cutting track planning method for heading machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109236292A true CN109236292A (en) | 2019-01-18 |
| CN109236292B CN109236292B (en) | 2021-01-05 |
Family
ID=65071921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810738123.3A Active CN109236292B (en) | 2018-07-06 | 2018-07-06 | Cutting track planning method for heading machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109236292B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112288029A (en) * | 2020-11-06 | 2021-01-29 | 电子科技大学 | Method for classifying vehicle tracks in urban road network |
| CN112901195A (en) * | 2021-01-21 | 2021-06-04 | 中国矿业大学 | Automatic section forming control system and method for cantilever longitudinal shaft type heading machine |
| CN114352301A (en) * | 2021-12-09 | 2022-04-15 | 中国煤炭科工集团太原研究院有限公司 | Method and device for determining cutting track of boom-type roadheader and electronic equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496004A (en) * | 2011-11-24 | 2012-06-13 | 中国矿业大学(北京) | Coal-rock interface identifying method and system based on image |
| CN104318254A (en) * | 2014-11-13 | 2015-01-28 | 中国矿业大学(北京) | Quick coal and rock recognition method based on DCT low-frequency component characteristics |
| CN205189863U (en) * | 2015-11-30 | 2016-04-27 | 郑东风 | Supplementary rectifying system in entry driving machine position and arbitrary section location visual system of cut |
| CN205330670U (en) * | 2015-12-01 | 2016-06-22 | 中国矿业大学(北京) | Coal petrography recognition device based on study of dependence tolerance |
| CN106761738A (en) * | 2016-12-15 | 2017-05-31 | 中国矿业大学 | The boom-type roadheader and method in cut path can be automatically planned based on machine vision |
| CN107014379A (en) * | 2017-05-25 | 2017-08-04 | 中国矿业大学 | A kind of development machine absolute space apparatus for detecting position and posture and method |
| CN107120116A (en) * | 2017-07-03 | 2017-09-01 | 中国矿业大学 | A kind of coal mining machine roller automatic height regulating device and method based on image recognition |
| US20180073358A1 (en) * | 2016-09-09 | 2018-03-15 | The Mosaic Company | Rotary boring mining machine inertial steering system |
-
2018
- 2018-07-06 CN CN201810738123.3A patent/CN109236292B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496004A (en) * | 2011-11-24 | 2012-06-13 | 中国矿业大学(北京) | Coal-rock interface identifying method and system based on image |
| CN104318254A (en) * | 2014-11-13 | 2015-01-28 | 中国矿业大学(北京) | Quick coal and rock recognition method based on DCT low-frequency component characteristics |
| CN205189863U (en) * | 2015-11-30 | 2016-04-27 | 郑东风 | Supplementary rectifying system in entry driving machine position and arbitrary section location visual system of cut |
| CN205330670U (en) * | 2015-12-01 | 2016-06-22 | 中国矿业大学(北京) | Coal petrography recognition device based on study of dependence tolerance |
| US20180073358A1 (en) * | 2016-09-09 | 2018-03-15 | The Mosaic Company | Rotary boring mining machine inertial steering system |
| CN106761738A (en) * | 2016-12-15 | 2017-05-31 | 中国矿业大学 | The boom-type roadheader and method in cut path can be automatically planned based on machine vision |
| CN107014379A (en) * | 2017-05-25 | 2017-08-04 | 中国矿业大学 | A kind of development machine absolute space apparatus for detecting position and posture and method |
| CN107120116A (en) * | 2017-07-03 | 2017-09-01 | 中国矿业大学 | A kind of coal mining machine roller automatic height regulating device and method based on image recognition |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112288029A (en) * | 2020-11-06 | 2021-01-29 | 电子科技大学 | Method for classifying vehicle tracks in urban road network |
| CN112901195A (en) * | 2021-01-21 | 2021-06-04 | 中国矿业大学 | Automatic section forming control system and method for cantilever longitudinal shaft type heading machine |
| CN114352301A (en) * | 2021-12-09 | 2022-04-15 | 中国煤炭科工集团太原研究院有限公司 | Method and device for determining cutting track of boom-type roadheader and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109236292B (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2021254666B2 (en) | Method and system for online monitoring and optimization of mining and mineral processing operations | |
| AU2011221435B2 (en) | System and method for terrain analysis | |
| Chen et al. | Machine learning-based classification of rock discontinuity trace: SMOTE oversampling integrated with GBT ensemble learning | |
| CN109386291B (en) | Method and device for planning cutting path of heading machine and cutting control system of heading machine | |
| CN109236292A (en) | A kind of tunneling machine cutting Trajectory Planning System and method | |
| CA2873816C (en) | Systems and methods for processing geophysical data | |
| CN106194184A (en) | The miner improved and method | |
| McKinnon et al. | Automatic identification of large fragments in a pile of broken rock using a time-of-flight camera | |
| US20200356929A1 (en) | Mining System | |
| CN112668109B (en) | Method for establishing cutting route model of fully-mechanized coal mining face | |
| US20240060419A1 (en) | Method and system for automated rock recognition | |
| CN115200540A (en) | Mine roadway deformation monitoring and early warning method and system | |
| CN110717618A (en) | Submarine hydrothermal sulfide resource evaluation and prediction method based on multi-index comprehensive elements | |
| Manzoor et al. | Rock fragmentation variations with increasing extraction ratio in sublevel caving: a case study | |
| Silversides et al. | Detection of geological structure using gamma logs for autonomous mining | |
| Ghosh | Applications, Promises and Challenges of Artificial Intelligence in Mining Industry: A Review | |
| CN112347841B (en) | Anti-collision application method of gangue filling tamping mechanism based on deep learning | |
| Valencia et al. | Blasthole Location Detection Using Support Vector Machine and Convolutional Neural Networks on UAV Images and Photogrammetry Models | |
| Mousavi et al. | A comparative study of kriging and simulation-based methods in classifying ore and waste blocks | |
| CN117408123B (en) | Space-time propulsion health state evaluation system and method for mining and transportation equipment | |
| Varian | Development of a wet muck database and drawpoint spill hazard susceptibility tool for an operating cave mine | |
| Afradi et al. | PREDICTION OF TBM PENETRATION RATE OF WATER CONVEYANCE TUNNELS IN IRAN USING MODERN METHODS | |
| Letshwiti | Improving Highwall Monitoring Through Fracture Identification in Open-Pit Mines Using Image Segmentation | |
| Khakulov et al. | Improving the Prospective Design of Mining Based on the Promotion Modern Software and Hardware and Technical Means of Monitoring and Control | |
| CN116821770A (en) | Intelligent prediction and classification method for deformation of surrounding rock of roadway in fault area |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |