CN108007472A - Measure the method and system of vision sweeping robot odometer penalty coefficient - Google Patents
Measure the method and system of vision sweeping robot odometer penalty coefficient Download PDFInfo
- Publication number
- CN108007472A CN108007472A CN201711252871.2A CN201711252871A CN108007472A CN 108007472 A CN108007472 A CN 108007472A CN 201711252871 A CN201711252871 A CN 201711252871A CN 108007472 A CN108007472 A CN 108007472A
- Authority
- CN
- China
- Prior art keywords
- sweeping robot
- vision
- camera
- odometer
- ceiling
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C22/00—Measuring distance traversed on the ground by vehicles, persons, animals or other moving solid bodies, e.g. using odometers, using pedometers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C25/00—Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Manufacturing & Machinery (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention relates to vision sweeping robot, specially a kind of method and system for measuring vision sweeping robot odometer penalty coefficient, wherein method are to calculate penalty coefficient according to formula according to after the displacement that the photo shot when moving calculates mobile actual displacement and odometer calculates in specified environment according to vision sweeping robot.Using vision sweeping robot, the sensor of itself calculates penalty coefficient to the present invention, improves calculating speed and precision.
Description
Technical field
The present invention relates to vision sweeping robot factory testing field, especially relates to a kind of measurement vision sweeper
The method and system of device people's odometer penalty coefficient.
Background technology
Odometer is a kind of critical component on vision sweeping robot, for measuring the displacement of vision sweeping robot and shifting
Dynamic speed.Odometer is typically provided with encoder, according to encoder rotate angle and wheel circumference calculating move away from
From.Because of factors such as environmental factor or structures, displacement distance and actual range that odometer calculates have a certain distance, generally exist
Odometer will be demarcated before dispatching from the factory, calculate penalty coefficient and the distance that odometer calculates is compensated, be allowed to examine
The distance of survey is accurate.
Present vision sweeping robot is all that manually calibrated and calculated goes out penalty coefficient before dispatching from the factory, and is so measured
Speed is slow, and precision is nor very high.
The content of the invention
The main object of the present invention can fast automatic calibration vision sweeping robot odometer method to provide one kind.
The present invention propose it is a kind of measure vision sweeping robot odometer penalty coefficient method, the ground of measuring environment with
Ceiling be it is horizontally disposed, the described method includes：
Vision sweeping robot mobile a distance in the test environment, while camera collection ceiling image；
According to the image of collection, the actual displacement L2 moved of computation vision sweeping robot；
Obtain the displacement L1 of the sweeping robot of odometer record；
The displacement L1 of the sweeping robot recorded with the actual displacement L2 moved of sweeping robot divided by odometer, calculates
Go out penalty coefficient.
Further, the step of image according to collection, the displacement L2 that computation vision sweeping robot is actual to be moved
Including：
The picture gathered according to vision sweeping robot, carries out space three-dimensional reconstruct by the characteristic point in picture, obtains institute
The three dimensional space coordinate of characteristic point is stated, and then obtains the three-dimensional information of the image；
According to the three-dimensional information of described image, the actual displacement L2 moved of the vision sweeping robot is calculated.
Further, the three-dimensional information according to described image, it is actual mobile to calculate the vision sweeping robot
Displacement L2 the step of include：
Obtain scale number T1 of the displacement distance in three dimensions of vision sweeping robot；
The corresponding dimensional information s of each scale is calculated according to default rule；
Dimensional information s is multiplied by the scale number T1 of three dimensions with the displacement distance of vision sweeping robot, is calculated
The actual displacement L2 moved of vision sweeping robot.
Further, described the step of calculating each scale corresponding dimensional information s according to default rule, includes：
Camera is obtained to the scale number T2 of ceiling plane；
According to known environment information and the vision sweeping robot parameter of itself, camera is calculated apart from ceiling
Actual height H；
With camera apart from ceiling actual height H divided by camera to the scale number T2 of ceiling plane, calculate
Go out dimensional information s.
Further, the scale number T2 for obtaining camera to ceiling plane includes step：
Estimate ceiling plane using RANSAC plane fittings, obtain normal n；
According to formula T2=nT* camera is calculated to scale the number T2, wherein n of ceiling plane in xTRepresent that normal n exists
The vector in the reconstruction space, x represent coordinate of the camera in the three dimensions.
Further, it is described penalty coefficient is calculated after include step：
Multiple penalty coefficients are repeatedly calculated, calculate the average value of multiple penalty coefficients.
Further, described the step of repeatedly calculating multiple penalty coefficients, includes：
Vision sweeping robot calculates multiple penalty coefficients on the ground of unlike material after movement respectively.
The present invention also proposes a kind of system for measuring vision sweeping robot odometer penalty coefficient, the ground of measuring environment
Be with ceiling it is horizontally disposed, the system comprises：
Mobile device, a distance is moved for vision sweeping robot in the test environment, while camera is adopted
Collect ceiling picture；
Actual displacement device, for the picture according to collection, calculates the actual displacement L2 moved of vision sweeping robot；
Odometer gearshift, for obtaining the displacement L1 of odometer record；
Penalty coefficient device, for the machine of sweeping the floor recorded with the actual displacement L2 moved of sweeping robot divided by odometer
The displacement L1 of people, is calculated penalty coefficient.
Further, the actual displacement device includes：
Module is built, for the picture gathered according to vision sweeping robot, the characteristic point in picture is subjected to space three
Dimension reconstruct, obtains the three dimensional space coordinate of the characteristic point, and then obtain the three-dimensional information of the image；
Computing module, for the three-dimensional information according to described image, it is actual mobile to calculate the vision sweeping robot
Displacement L2.
Further, the computing module includes：
Mobile scale submodule, for obtaining scale number T1 of the displacement distance in three dimensions of vision sweeping robot；
Length information submodule, for calculating the corresponding dimensional information s of each scale according to default rule；
Calculation formula submodule, for being multiplied by with the displacement distance of vision sweeping robot in the scale number T1 of three dimensions
Dimensional information s, is calculated the actual displacement L2 moved of vision sweeping robot.
Further, the length information submodule includes：
Scaling units are obtained, for obtaining camera to the scale number T2 of ceiling plane；
Computed altitude unit, for according to known environment information and the vision sweeping robot parameter of itself, calculating
Actual height H of the camera apart from ceiling；
Calculation formula unit, for according to camera apart from ceiling actual height H divided by camera to ceiling
The scale number T2 of plane, is calculated dimensional information s.
Further, the acquisition scaling units include：
Normal subelement, for using RANSAC plane fittings estimation ceiling plane, obtaining normal n；
Calculate apart from subelement, for according to formula T2=nT* x, is calculated camera to the scale of ceiling plane
Number T2, wherein nTRepresent vectors of the normal n in the reconstruction space, x represents coordinate of the camera in the three dimensions.
Further, the system of the measurement vision sweeping robot odometer penalty coefficient further includes：
Multiple computing device, for calculating the distance of normal n and vision robot's camera of sweeping the floor, obtains T2.
Further, the repeatedly computing device includes：
Earth module, multiple compensation are calculated for vision sweeping robot on the ground of unlike material after movement respectively
Coefficient.
Compared with prior art, the beneficial effects of the invention are as follows：Using vision sweeping robot, the sensor of itself can
To be automatically performed the calculating of penalty coefficient, the calibration efficiency of vision sweeping robot is improved, reduces the time manually demarcated, is improved
Stated accuracy.
Brief description of the drawings
The step of Fig. 1 is the method for the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is illustrated
Figure；
The step of Fig. 2 is the method for the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is illustrated
Figure；
The step of Fig. 3 is the method for the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is illustrated
Figure；
The step of Fig. 4 is the method for the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is illustrated
Figure；
The step of Fig. 5 is the method for the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is illustrated
Figure；
Fig. 6 is the structural representation of the system of the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention
Figure；
Fig. 7 is the actual displacement of the system of the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention
The structure diagram of device；
Fig. 8 is the computing module of the system of the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention
Structure diagram；
Fig. 9 is the length information of the system of the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention
The structure diagram of submodule；
Figure 10 is that the calculating of the system of the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is high
Spend the structure diagram of unit；
Figure 11 is that the structure of the system of the measurement vision sweeping robot odometer penalty coefficient of one embodiment of the invention is shown
It is intended to.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
With reference to Fig. 1, the method that the present invention one measures vision sweeping robot odometer penalty coefficient is proposed, measuring environment
Ground and ceiling are horizontally disposed, and the distance of the two planes is known, and measuring method includes step：
S1, vision sweeping robot mobile a distance in the test environment, while camera collection ceiling figure
Piece；
S2, the picture according to collection, the actual displacement L2 moved of computation vision sweeping robot；
S3, the displacement L1 for obtaining odometer record；
The displacement L1 of S4, the sweeping robot recorded with the actual displacement L2 moved of sweeping robot divided by odometer, meter
Calculation draws penalty coefficient.
In the present embodiment, vision sweeping robot moves in the environment, floor and ceiling be it is horizontally disposed, i.e.,
Plate and ceiling are smooth planes, and without the gradient, radian, inclination, vision sweeping robot can be swept the floor with normal movement, vision
Machine carries vision system, and vision system is exactly to replace human eye with machine to measure and judge.Vision system refers to pass through machine
Device vision products (i.e. image-pickup device, is divided to two kinds of CMOS and CCD) will be ingested target and be converted into picture signal, send to specially
Image processing system, according to the information such as pixel distribution and brightness, color, is transformed into digitized signal；Vision system is to this
A little signals carry out various computings to extract clarification of objective, and then control the device action at scene according to the result of differentiation.Can
For measurement distance, vision sweeping robot is provided with vision system and can measure the distance of front obstacle, obtain cleaning
Cartographic information of environment etc..In this embodiment, picture is gathered according to measuring environment and vision system, measures vision and sweep the floor
The actual moving displacement L2 of a distance of robot movement, then reads the odometer in this segment distance vision sweeping robot
The displacement L1 calculated, with L2 divided by L1, it is possible to calculate penalty coefficient.
With reference to Fig. 2, further, the step S2 includes：
S21, the picture gathered according to vision sweeping robot, carry out space three-dimensional reconstruct by the characteristic point in picture, obtain
To the three dimensional space coordinate of the characteristic point, and then obtain the three-dimensional information of the image；
S22, according to the three-dimensional information, calculate the actual displacement L2 moved of the vision sweeping robot.
In the present embodiment, vision sweeping robot during cleaning, take pictures by vision system, and extracts spy according to the photo of shooting
Point is levied, carries out space three-dimensional reconstruct according to characteristic point, the two-dimension picture that will be shot is converted into the picture of three-dimensional, passes through video camera
Demarcate to establish effective imaging model, solve the inside and outside parameter of video camera, can thus combine the matching result of image
The three-dimensional point coordinate in space is obtained, so as to achieve the purpose that to carry out three-dimensional reconstruction, three dimensions is built up, then vision is swept the floor
The position after position and movement before robot movement is labeled in the three dimensions, and calculating vision in the coordinate-system sweeps
The actual displacement distance L2 of floor-washing robot.Three dimensions is established, more accurately obtains the position of vision sweeping robot, is calculated
Actual displacement distance L2 out is more accurate.
With reference to Fig. 3, further, step S22 includes：
S221, scale number T1 of the displacement distance in three dimensions for obtaining vision sweeping robot；
S222, calculate the corresponding dimensional information s of each scale according to default rule；
S223, with the displacement distance of vision sweeping robot be multiplied by dimensional information s in the scale number T1 of three dimensions, calculates
Draw the actual displacement L2 moved of vision sweeping robot.
In the present embodiment, in three dimensions, each dimension has a scale every a distance, the scale it is corresponding away from
From just referring to dimensional information s.In the three-dimensional coordinate system, scale the number T1, Ran Houzai of the movement of vision sweeping robot are obtained
According to the length information s of each scale number, L2 can be calculated according to the two amounts.Specifically, L2 is equal to T1 and s and multiplies
Product.Wherein the size of s can be calculated according to some of default rule and user preset parameters.
It is further, described that the corresponding length information s S222 of each scale are calculated according to default rule with reference to Fig. 4
The step of include：
S2221, obtain camera to the scale number T2 of ceiling plane；
S2222, according to known environment information and the vision sweeping robot parameter of itself, calculate camera apart from day
The actual height H of card；
S2223, with camera apart from ceiling actual height H divided by camera to ceiling plane scale number T2,
Dimensional information s is calculated.
In the present embodiment, in three dimensions, camera is a point, and residing for ceiling is a plane, is calculated
The scale number of camera and ceiling plane.In addition, according to known environment, can calculate between camera and ceiling
Actual distance value.I.e. the corresponding actual distance value of scale number can calculate, then, the corresponding physical length of each scale
Value can be obtained significantly.With the scale quantity of actual height H of the camera away from ceiling divided by camera to ceiling plane
T2, you can to calculate s.In known environment, height H1 of the ceiling apart from ground is known, vision sweeping robot
It is parameter-embedded in, camera height H2 residing on vision sweeping robot is known, then, vision sweeping robot
When placing on the ground, the actual distance value of camera and ceiling is H=H1-H2.Then H divided by T2 is used, ruler is calculated
Spend information s.
With reference to Fig. 5, further, step S2221 includes：
S22211, using RANSAC plane fittings estimate ceiling plane, obtain normal n；
S22212, according to formula T2=nT* x obtains T2, obtains camera to scale the number T2, wherein n of ceiling planeT
Represent vectors of the normal n in the reconstruction space, x represents coordinate of the camera in the three dimensions.
In the present embodiment, RANSAC is the abbreviation of Random Sample Consensus, it is comprising abnormal according to one group
The sample data set of data, calculates the mathematical model parameter of data, obtains the algorithm of effective sample data.RANSAC algorithms pass through
It is usually used in computer vision, the match point of a pair of of camera and the meter of fundamental matrix is solved the problems, such as at the same time in stereoscopic vision field
Calculate.Camera has certain angle, containing the point in many ceiling planes in the three dimensions of foundation, using RANSAC into
Row fitting, it is possible to calculate plane of the ceiling plane in straight three dimensions, the plane in three dimensions put down by ceiling
The normal in face is indicated with n.Then ceiling is calculated to vision sweeping robot according to normal in the three dimensions
The distance of camera, for the distance, that is, camera to ceiling in the distance T2 of three dimensions, which is in three-dimensional coordinate system
Scale number, rather than actual distance value.Specifically, normal n and camera to ceiling between the distance T2 of three dimensions
Correspondence be：T2=nT* x, in the formula, x is coordinate of the camera in the three dimensions, nTRepresent normal this three
Vector in dimension space.
Further, it is described penalty coefficient is calculated after include step：
S5, repeatedly calculate multiple penalty coefficients, calculates the average value of multiple penalty coefficients.
In the present embodiment, after penalty coefficient is calculated, inevitably there are error for the penalty coefficient.And the error
It is random error, further to reduce error, the best way is exactly that many experiments calculate, and the step of above-mentioned S1-S4 is repeated several times
Suddenly multiple penalty coefficients are drawn, calculate the average value of multiple penalty coefficients, using the average value as final penalty coefficient, are improved
The precision of the detecting distance of vision sweeping robot.
Further, the step of repeatedly calculating multiple penalty coefficients includes：
Vision sweeping robot calculates multiple penalty coefficients on the ground of unlike material after movement respectively.
In the present embodiment, because vision sweeping robot will face different families, therefore when measuring penalty coefficient,
Consider various factors, it is therefore desirable to consider different friction coefficient.General home-use floor include concrete floor, ceramic tile,
The different floor of the various materials such as timber floor, carpet.Measurement draws compensation system on the floor of unlike material different coefficients of friction
After number, calculating is averaged, finally draws the penalty coefficient of a standard.
With reference to Fig. 6, the present invention also proposes a kind of system for measuring vision sweeping robot odometer penalty coefficient, wherein,
The ground of measuring environment is horizontally disposed with ceiling, which includes：
Mobile device 1, a distance is moved for vision sweeping robot in the test environment, while camera is adopted
Collect ceiling picture；
Actual displacement device 2, for the picture according to collection, calculates the actual displacement moved of vision sweeping robot
L2；
Odometer gearshift 3, the displacement L1 calculated for obtaining odometer；
Penalty coefficient device 4, for the sweeper recorded with the actual displacement L2 moved of sweeping robot divided by odometer
The displacement L1 of device people, is calculated penalty coefficient.
In the present embodiment, vision sweeping robot moves in the environment, floor and ceiling be it is horizontally disposed, i.e.,
Plate and ceiling are smooth planes, can be swept the floor without the gradient, radian, inclination vision sweeping robot with normal movement, vision
Machine carries vision system, and vision system is exactly to replace human eye with machine to measure and judge.Vision system refers to pass through machine
Device vision products (i.e. image-pickup device, is divided to two kinds of CMOS and CCD) will be ingested target and be converted into picture signal, send to specially
Image processing system, according to the information such as pixel distribution and brightness, color, is transformed into digitized signal；Vision system is to this
A little signals carry out various computings to extract clarification of objective, and then control the device action at scene according to the result of differentiation.Can
For measurement distance, the mobile device 1 of vision sweeping robot is when controlling the movement of vision sweeping robot, while control regards
Feel is taken pictures, and gathers the picture of ceiling.In this embodiment, actual displacement device 2 is according to measuring environment and vision system
System collection picture, measures the actual moving displacement L2 of a distance of vision sweeping robot movement, then odometer displacement
Device 3 reads the displacement L1 that the odometer in this segment distance vision sweeping robot calculates, penalty coefficient device 4 with L2 divided by
L1, it is possible to calculate penalty coefficient.
With reference to Fig. 7, further, the actual displacement device 2 includes：
Module 21 is built, for the picture gathered according to vision sweeping robot, the characteristic point in picture carries out space three
Dimension reconstruct, obtains the three dimensional space coordinate of the characteristic point, and then obtain the three-dimensional information of the image；
Computing module 22, it is actual mobile for calculating the vision sweeping robot according to the three-dimensional information of described image
Displacement L2.
In the present embodiment, vision sweeping robot during cleaning, take pictures by vision system, builds photograph of the module 21 according to shooting
Piece extracts characteristic point, carries out space three-dimensional reconstruct according to characteristic point, the two-dimension picture that will be shot is converted into the picture of three-dimensional, leads to
Camera calibration is crossed to establish effective imaging model, solves the inside and outside parameter of video camera, can thus combine image
Matching result obtains the three-dimensional point coordinate in space, so as to achieve the purpose that to carry out three-dimensional reconstruction, builds up three dimensions, then will
The position after position and movement before the movement of vision sweeping robot is labeled in the three dimensions, and computing module 22 is in the coordinate
The actual displacement distance L2 of vision sweeping robot is calculated in system.Three dimensions is established, vision is more accurately obtained and sweeps
The position of floor-washing robot, the actual displacement distance L2 calculated are more accurate.
With reference to Fig. 8, further, the computing module 22 includes：
Mobile scale submodule 221, for obtaining scale number T1 of the movement in three dimensions of vision sweeping robot；
Length information submodule 222, for calculating the corresponding dimensional information s of each scale according to default rule；
Calculation formula submodule 223, for the displacement distance of vision sweeping robot three dimensions scale number T1
Dimensional information s is multiplied by, dimensional information L2 is calculated.
In the present embodiment, in three dimensions, each dimension has a scale every a distance, the scale it is corresponding away from
From just referring to dimensional information s.In the three dimensions, mobile scale submodule 221 obtains the mark of vision sweeping robot movement
Number of degrees T1, then length information submodule 222 calculate the corresponding length s of each scale, calculation formula submodule 223 further according to
The length information s of each scale number, L2 can be calculated according to the two amounts.Specifically, L2 is equal to T1 and s and product.Its
The size of middle s can be calculated according to some of default rule and user preset parameters.
With reference to Fig. 9, further, the length information submodule 222 includes：
Scaling units 2221 are obtained, for obtaining camera to the scale number T2 of ceiling plane；
Computed altitude unit 2222, for according to known environment information and the vision sweeping robot parameter of itself, meter
Calculate actual height H of the camera apart from ceiling；
Calculation formula unit 2223, for camera apart from ceiling actual height H divided by camera to ceiling
The scale number T2 of plane, is calculated dimensional information s.
In the present embodiment, in three dimensional coordinate space, camera is a point, and residing for ceiling is a plane, meter
Calculate scaling units 2221 and calculate camera and the scale number of ceiling plane.In addition, according to known environment, computed altitude list
Member 2222 can calculate the actual distance value between camera and ceiling.That is the corresponding actual distance value of scale number
To calculate, then, the corresponding actual (tube) length angle value of each scale can be obtained significantly.Calculation formula unit 2223 uses camera
The scale quantity T2 of actual height H divided by camera away from ceiling to ceiling plane, you can to calculate s.Known
In environment, height H1 of the ceiling apart from ground be it is known, vision sweeping robot it is parameter-embedded in, camera is in vision
Residing height H2 is known on sweeping robot, then, when vision sweeping robot is placed on the ground, camera and day
The actual distance value of card is H=H1-H2.Then H divided by T2 is used, dimensional information s is calculated.
With reference to Figure 10, further, the calculating scaling units 2222 include：
Normal subelement 22221, for using RANSAC plane fittings estimation ceiling plane, obtaining normal n；
Calculate apart from subelement 22222, for according to formula T2=nT* x obtains T2, obtains camera to ceiling plane
Scale number T2, wherein nTRepresent vectors of the normal n in the reconstruction space, x represents seat of the camera in the three dimensions
Mark.
In the present embodiment, RANSAC is the abbreviation of Random Sample Consensus, it is comprising abnormal according to one group
The sample data set of data, calculates the mathematical model parameter of data, obtains the algorithm of effective sample data.RANSAC algorithms pass through
It is usually used in computer vision, the match point of a pair of of camera and the meter of fundamental matrix is solved the problems, such as at the same time in stereoscopic vision field
Calculate.Camera has certain angle, contains the point in many ceiling planes, normal subelement in the three dimensions of foundation
22221 are fitted using RANSAC, it is possible to calculate plane of the ceiling plane in straight three dimensions, the plane is three
N is indicated in dimension space.Then calculate and ceiling is calculated according to normal in the three dimensions apart from subelement 22222
To the distance of vision sweeping robot camera, in the distance T2 of three dimensions, which is the distance, that is, camera to ceiling
Scale number in three-dimensional coordinate system, rather than actual distance value.Specifically, normal n and camera to ceiling in three-dimensional
Correspondence between the distance T2 in space is：T2=nT* x, in the formula, x is coordinate of the camera in the three dimensions,
nTRepresent vector of the normal in the three dimensions.
With reference to Figure 11, further, the system of the measurement vision sweeping robot odometer penalty coefficient further includes：
Multiple computing device 5, for calculating the distance of normal n and vision robot's camera of sweeping the floor, obtains T2.
In the present embodiment, after penalty coefficient is calculated, inevitably there are error for the penalty coefficient.And the error
It is random error, further to reduce error, the best way is exactly that many experiments calculate, in multiple computing device multiplicating
Multiple penalty coefficients are calculated in the device operation stated, and calculate the average value of multiple penalty coefficients, using the average value as finally
Penalty coefficient, improve vision sweeping robot detecting distance precision.
Further, multiple computing device 5 includes：
Earth module 51, multiple benefits are calculated for vision sweeping robot on the ground of unlike material after movement respectively
Repay coefficient.
In the present embodiment, because vision sweeping robot will face different families, therefore when measuring penalty coefficient,
Consider various factors, it is therefore desirable to consider different friction coefficient.General home-use floor include concrete floor, ceramic tile,
The different floor of the various materials such as timber floor, carpet.Earth module 51 measures on the floor of unlike material different coefficients of friction
After drawing penalty coefficient, calculating is averaged, finally draws the penalty coefficient of a standard.
The foregoing is merely the preferred embodiment of the present invention, is not intended to limit the scope of the invention, every utilization
The equivalent structure or equivalent flow shift that description of the invention and accompanying drawing content are made, it is related to be directly or indirectly used in other
Technical field, be included within the scope of the present invention.
Claims (10)
1. it is a kind of measure vision sweeping robot odometer penalty coefficient method, it is characterised in that the ground of measuring environment with
Ceiling be it is horizontally disposed, the described method includes：
Vision sweeping robot mobile a distance in the test environment, while camera collection ceiling image；
According to the image of collection, the actual displacement L2 moved of computation vision sweeping robot；
Obtain the displacement L1 of the sweeping robot of odometer record；
The displacement L1 of the sweeping robot recorded with the actual displacement L2 moved of sweeping robot divided by odometer, is calculated benefit
Repay coefficient.
2. the method for vision sweeping robot odometer penalty coefficient as claimed in claim 1, it is characterised in that the basis
The step of image of collection, the displacement L2 that computation vision sweeping robot is actual to be moved, includes：
The picture gathered according to vision sweeping robot, carries out space three-dimensional reconstruct by the characteristic point in picture, obtains the spy
The three dimensional space coordinate of point is levied, and then obtains the three-dimensional information of the image；
According to the three-dimensional information of described image, the actual displacement L2 moved of the vision sweeping robot is calculated.
3. the method for measurement vision sweeping robot odometer penalty coefficient as claimed in claim 2, it is characterised in that described
According to the three-dimensional information of described image, calculate the vision sweeping robot it is actual move displacement L2 the step of include：
Obtain scale number T1 of the displacement distance in three dimensions of vision sweeping robot；
The corresponding dimensional information s of each scale is calculated according to default rule；
Dimensional information s is multiplied by the scale number T1 of three dimensions with the displacement distance of vision sweeping robot, vision is calculated
The actual displacement L2 moved of sweeping robot.
4. the method for measurement vision sweeping robot odometer penalty coefficient as claimed in claim 3, it is characterised in that described
The step of calculating each scale corresponding dimensional information s according to default rule includes：
Camera is obtained to the scale number T2 of ceiling plane；
According to known environment information and the vision sweeping robot parameter of itself, reality of the camera apart from ceiling is calculated
Height H；
With camera apart from ceiling actual height H divided by camera to the scale number T2 of ceiling plane, ruler is calculated
Spend information s.
5. the method for measurement vision sweeping robot odometer penalty coefficient as claimed in claim 4, it is characterised in that described
Obtaining the scale number T2 of camera to ceiling plane includes step：
Estimate ceiling plane using RANSAC plane fittings, obtain normal n；
According to formula T2=nT* camera is calculated to scale the number T2, wherein n of ceiling plane in xTRepresent that normal n is heavy at this
The vector in space is built, x represents coordinate of the camera in the three dimensions.
6. it is a kind of measure vision sweeping robot odometer penalty coefficient system, it is characterised in that the ground of measuring environment with
Ceiling be it is horizontally disposed, the system comprises：
Mobile device, a distance, while camera collection day are moved for vision sweeping robot in the test environment
Card picture；
Actual displacement device, for the picture according to collection, calculates the actual displacement L2 moved of vision sweeping robot；
Odometer gearshift, for obtaining the displacement L1 of odometer record；
Penalty coefficient device, for the sweeping robot that is recorded with the actual displacement L2 moved of sweeping robot divided by odometer
Displacement L1, is calculated penalty coefficient.
7. measurement as claimed in claim 6 is regarding the system of sweeping robot odometer penalty coefficient, it is characterised in that the reality
Border gearshift includes：
Module is built, for the picture gathered according to vision sweeping robot, the characteristic point in picture is subjected to space three-dimensional weight
Structure, obtains the three dimensional space coordinate of the characteristic point, and then obtains the three-dimensional information of the image；
Computing module, for the three-dimensional information according to described image, calculates the actual position moved of the vision sweeping robot
Move L2.
8. measurement as claimed in claim 7 is regarding the system of sweeping robot odometer penalty coefficient, it is characterised in that the meter
Calculating module includes：
Mobile scale submodule, for obtaining scale number T1 of the displacement distance in three dimensions of vision sweeping robot；
Length information submodule, for calculating the corresponding dimensional information s of each scale according to default rule；
Calculation formula submodule, for being multiplied by scale in the scale number T1 of three dimensions with the displacement distance of vision sweeping robot
Information s, is calculated the actual displacement L2 moved of vision sweeping robot.
9. measurement as claimed in claim 8 is regarding the system of sweeping robot odometer penalty coefficient, it is characterised in that the length
Degree message sub-module includes：
Scaling units are obtained, for obtaining camera to the scale number T2 of ceiling plane；
Computed altitude unit, for according to known environment information and the vision sweeping robot parameter of itself, calculating shooting
Actual height H of the head apart from ceiling；
Calculation formula unit, for according to camera apart from ceiling actual height H divided by camera to ceiling plane
Scale number T2, dimensional information s is calculated.
10. measurement as claimed in claim 9 is regarding the system of sweeping robot odometer penalty coefficient, it is characterised in that described
Obtaining scaling units includes：
Normal subelement, for using RANSAC plane fittings estimation ceiling plane, obtaining normal n；
Calculate apart from subelement, for according to formula T2=nT* x, is calculated camera to the scale number T2 of ceiling plane,
Wherein nTRepresent vectors of the normal n in the reconstruction space, x represents coordinate of the camera in the three dimensions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711252871.2A CN108007472A (en) | 2017-12-01 | 2017-12-01 | Measure the method and system of vision sweeping robot odometer penalty coefficient |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711252871.2A CN108007472A (en) | 2017-12-01 | 2017-12-01 | Measure the method and system of vision sweeping robot odometer penalty coefficient |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108007472A true CN108007472A (en) | 2018-05-08 |
Family
ID=62056164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711252871.2A Pending CN108007472A (en) | 2017-12-01 | 2017-12-01 | Measure the method and system of vision sweeping robot odometer penalty coefficient |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108007472A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110262283A (en) * | 2019-06-11 | 2019-09-20 | 远形时空科技（北京）有限公司 | A kind of the vision robot's emulation platform and method of more scenes |
| CN110595480A (en) * | 2019-10-08 | 2019-12-20 | 瓴道(上海)机器人科技有限公司 | Navigation method, device, equipment and storage medium |
| CN110888421A (en) * | 2018-08-21 | 2020-03-17 | 广州弘度信息科技有限公司 | Robot, and motion self-adaption method and system of robot |
| CN111053498A (en) * | 2018-10-17 | 2020-04-24 | 郑州雷动智能技术有限公司 | Displacement compensation method of intelligent robot and application thereof |
| CN111671360A (en) * | 2020-05-26 | 2020-09-18 | 深圳拓邦股份有限公司 | Sweeping robot position calculating method and device and sweeping robot |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007156576A (en) * | 2005-11-30 | 2007-06-21 | Mitsubishi Heavy Ind Ltd | Method and device for adjusting odometry(wheel range finder) parameter for traveling carrier |
| US20090024251A1 (en) * | 2007-07-18 | 2009-01-22 | Samsung Electronics Co., Ltd. | Method and apparatus for estimating pose of mobile robot using particle filter |
| CN102376089A (en) * | 2010-12-09 | 2012-03-14 | 深圳大学 | Target correction method and system |
| CN105631844A (en) * | 2014-10-28 | 2016-06-01 | 上海大学 | Image camera calibration method |
| CN106289327A (en) * | 2016-11-08 | 2017-01-04 | 成都通甲优博科技有限责任公司 | A kind of difference moves robot speedometer scaling method |
-
2017
- 2017-12-01 CN CN201711252871.2A patent/CN108007472A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007156576A (en) * | 2005-11-30 | 2007-06-21 | Mitsubishi Heavy Ind Ltd | Method and device for adjusting odometry(wheel range finder) parameter for traveling carrier |
| US20090024251A1 (en) * | 2007-07-18 | 2009-01-22 | Samsung Electronics Co., Ltd. | Method and apparatus for estimating pose of mobile robot using particle filter |
| CN102376089A (en) * | 2010-12-09 | 2012-03-14 | 深圳大学 | Target correction method and system |
| CN105631844A (en) * | 2014-10-28 | 2016-06-01 | 上海大学 | Image camera calibration method |
| CN106289327A (en) * | 2016-11-08 | 2017-01-04 | 成都通甲优博科技有限责任公司 | A kind of difference moves robot speedometer scaling method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110888421A (en) * | 2018-08-21 | 2020-03-17 | 广州弘度信息科技有限公司 | Robot, and motion self-adaption method and system of robot |
| CN110888421B (en) * | 2018-08-21 | 2022-07-12 | 广州哨马智能装备科技有限公司 | Robot, and motion self-adaption method and system of robot |
| CN111053498A (en) * | 2018-10-17 | 2020-04-24 | 郑州雷动智能技术有限公司 | Displacement compensation method of intelligent robot and application thereof |
| CN110262283A (en) * | 2019-06-11 | 2019-09-20 | 远形时空科技（北京）有限公司 | A kind of the vision robot's emulation platform and method of more scenes |
| CN110595480A (en) * | 2019-10-08 | 2019-12-20 | 瓴道(上海)机器人科技有限公司 | Navigation method, device, equipment and storage medium |
| CN111671360A (en) * | 2020-05-26 | 2020-09-18 | 深圳拓邦股份有限公司 | Sweeping robot position calculating method and device and sweeping robot |
| CN111671360B (en) * | 2020-05-26 | 2021-11-16 | 深圳拓邦股份有限公司 | Sweeping robot position calculating method and device and sweeping robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108007472A (en) | Measure the method and system of vision sweeping robot odometer penalty coefficient | |
| US9965870B2 (en) | Camera calibration method using a calibration target | |
| CN106651794B (en) | A kind of projection speckle bearing calibration based on virtual camera | |
| CN106871787B (en) | Large space line scanning imagery method for three-dimensional measurement | |
| CA2731680C (en) | System for adaptive three-dimensional scanning of surface characteristics | |
| EP2829842A1 (en) | Method and system for volume determination using a structure from motion algorithm | |
| CN101681519B (en) | Measurement device and measurement method | |
| CN106595528A (en) | Digital speckle-based telecentric microscopic binocular stereoscopic vision measurement method | |
| CN102387374A (en) | Device and method for acquiring high-precision depth map | |
| Hui et al. | A novel line scan camera calibration technique with an auxiliary frame camera | |
| CN107167093A (en) | The combined type measuring system and measuring method of a kind of laser line scanning and shadow Moire | |
| CN107144241A (en) | A kind of binocular vision high-precision measuring method compensated based on the depth of field | |
| CN108362205A (en) | Space ranging method based on fringe projection | |
| CN102012213B (en) | New method for measuring foreground height through single image | |
| Wan et al. | A study in 3d-reconstruction using kinect sensor | |
| Deng et al. | Registration of multiple rgbd cameras via local rigid transformations | |
| Wang et al. | Measurement and analysis of depth resolution using active stereo cameras | |
| Furukawa et al. | Interactive shape acquisition using marker attached laser projector | |
| Pankaj et al. | 3-D imaging techniques and review of products | |
| CN108680182A (en) | Measure the method and system of vision sweeping robot odometer penalty coefficient | |
| WO2019104741A1 (en) | Method and system for measuring compensating coefficient for odometer of visual robot cleaner | |
| CN109597005A (en) | The measuring system for space magnetic field of view-based access control model positioning | |
| CN107392955A (en) | A kind of depth of field estimation device and method based on brightness | |
| CN114663486A (en) | Building height measurement method and system based on binocular vision | |
| Dias et al. | Combining intensity and range images for 3D modelling |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190904 Address after: Room 402, 4th floor, Kanghe Sheng Building, New Energy Innovation Industrial Park, No. 1 Chuangsheng Road, Nanshan District, Shenzhen City, Guangdong Province, 518000 Applicant after: Shenzhen Infinite Power Development Co., Ltd. Address before: 518000 Nanshan District, Shenzhen, Guangdong Province, 1079 Nanhai Avenue, Shekou, Garden City Digital Building B Block 503, 602 Applicant before: SHENZHEN WOTE WODE CO., LTD. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180508 |