CN103886107B - Robot localization and map structuring system based on ceiling image information - Google Patents
Robot localization and map structuring system based on ceiling image information Download PDFInfo
- Publication number
- CN103886107B CN103886107B CN201410149081.1A CN201410149081A CN103886107B CN 103886107 B CN103886107 B CN 103886107B CN 201410149081 A CN201410149081 A CN 201410149081A CN 103886107 B CN103886107 B CN 103886107B
- Authority
- CN
- China
- Prior art keywords
- corner feature
- image
- module
- feature
- tracking
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to robot field, more particularly to a kind of robot localization based on ceiling image information and map structuring system.The system that the present invention is provided includes image collection module, is arranged in robot, its optical axis is perpendicular to ceiling, for being shot in robot motion to ceiling, obtains current ceiling image information；Described image processing module, for handling the ceiling image information that described image acquisition module is obtained；The characteristic extracting module, for extracting feature in the image information after the processing of described image processing module, obtains set of image characteristics；The feature tracking module, each characteristics of image that the characteristics of image for being extracted to the characteristic extracting module is concentrated carries out matched jamming, the set of image characteristics after being tracked；The memory management module, for storing the set of image characteristics that the feature tracking module is obtained, and according to described image feature set map making.The present invention facilitates map structuring.
Description
Technical field
The present invention relates to robot field, more particularly to a kind of robot localization based on ceiling image information and ground
Figure constructing system.
Background technology
With automation and intelligentized development, robot is in the industry using more and more extensively, and status is more and more important.
Robot research is to compare popular field, wherein Simultaneous Localization and Mapping now（It is synchronous
Positioning and map structuring, hereinafter referred to as SLAM）Technology is the problem of robot field is more classical, and SLAM problems can be described
For:Robot is moved in circumstances not known since a unknown position, is entered in moving process according to location estimation and map
Row self poisoning, while building increment type map on the basis of self poisoning, realizes autonomous positioning and the navigation of robot.
Because vision sensor can obtain abundant environmental information, it is used widely, that is, is regarded in SLAM in recent years
Feel SLAM.Vision SLAM can be divided into the SLAM based on monocular vision, and two kinds of binocular and multi-vision visual SLAM.But for
For mobile robot, in order to save the complexity especially data correlation problem of cost and reduction system, in many applications
The video camera of a table top forward is only equipped with, this direction is conducive to other operations of robot, such as long-range display, remote operating
And avoidance.
But the defect of this technology is：The motion of robot, which is generally approximate along camera optical axis, to be translated, therefore is obtained
Consecutive image vary less, map structuring is relatively difficult.
The content of the invention
In order to solve the above technical problems, the present invention provides a kind of robot localization and map based on ceiling image information
Constructing system, the consecutive image change of acquisition is more apparent, facilitates map structuring.
The robot localization based on ceiling image information of the present invention and map structuring system, including：Image obtains mould
Block, characteristic extracting module, feature tracking module and memory management module；
Described image acquisition module, is arranged in robot, and its optical axis is perpendicular to ceiling, in robot motion
Ceiling is shot, current ceiling image information is obtained；
Described image processing module, for handling the ceiling image information that described image acquisition module is obtained；
The characteristic extracting module, for extracting feature in the image information after the processing of described image processing module, is obtained
To set of image characteristics；
The feature tracking module, each figure that the characteristics of image for being extracted to the characteristic extracting module is concentrated
As feature carries out matched jamming, the set of image characteristics after being tracked；
The memory management module, for storing the set of image characteristics that the feature tracking module is obtained, and according to described
Set of image characteristics map making.
Wherein, described image is characterized as Corner Feature, and described image feature set is Corner Feature collection.
Wherein, the feature tracking module includes optical flow tracking module and correction module；
The optical flow tracking module, the Corner Feature collection for tracking present image by Lucas-Kanade optical flow methods
In each Corner Feature, obtain track Corner Feature collection；
The correction module, for it is described state optical flow tracking module obtain tracking Corner Feature collection be corrected.
Further, the correction module extracts Corner Feature again on current frame image, obtains new Corner Feature
Collection, concentrates each Corner Feature to be concentrated with new Corner Feature the tracking Corner Feature obtained by the optical flow tracking module
The Corner Feature that is overlapped in the range of preset window of Corner Feature as the Corner Feature collection after correction.
Or, the correction module includes the first tracking cell, the second tracking cell and judging unit：
If current frame image is image imgA, its Corner Feature integrates as A, and next two field picture is imgB；
First tracking cell, for based on Corner Feature collection A, angle point to be obtained in next two field picture imgB tracking
Feature set B；
Second tracking cell, for based on Corner Feature collection B, angle point to be obtained in current frame image imgA tracking
Feature set B2；
The judging unit, for judging whether each Corner Feature label of the Corner Feature collection A is special with the angle point
Each Corner Feature label collected in B2 is identical, if identical, takes label identical Corner Feature to add preparation Corner Feature collection
In.
Or, the correction module includes extraction unit, predicting unit, comparison unit：
The extraction unit, the Corner Feature collection B2 for extracting present image；Some Corner Feature has been in B2
The tracking Corner Feature collection B that the present image is obtained is tracked through being present in；
The predicting unit, the Corner Feature collection for predicting present image according to the Corner Feature in Corner Feature storehouse
B’；
The Corner Feature of the comparison unit, for contrasting B2, B and B ', B and B will not be appeared in B2 ' is judged as newly
Corner Feature, will be added in Corner Feature storehouse by the new Corner Feature of threshold value.
Beneficial effects of the present invention are compared with prior art：It is provided in an embodiment of the present invention to be based on ceiling image information
Robot localization and map structuring system, due to the direction of video camera to be changed to the fortune in direction, i.e. robot to ceiling
Optical axis of the direction perpendicular to video camera is moved, therefore the consecutive image change obtained is more apparent, facilitates map structuring；Further,
When tracking characteristics of image, in order to obtain more preferable tracking information, traditional tracking result is corrected, makes what tracking was obtained
Corner Feature is more stablized and accurate, and the precision of map structuring is also higher, that is, the map built is also just more accurate.
Brief description of the drawings
Fig. 1 is that the robot localization provided in an embodiment of the present invention based on ceiling image information shows with map structuring system
It is intended to；
Fig. 2 is another robot localization and map structuring based on ceiling image information provided in an embodiment of the present invention
System schematic；
Fig. 3 is the schematic diagram of correction module in Fig. 2；
Fig. 4 is another schematic diagram of correction module in Fig. 2.
Embodiment
With reference to the accompanying drawings and examples, the embodiment to the present invention is described in further detail.Implement below
Example is used to illustrate the present invention, but is not limited to the scope of the present invention.
Referring to Fig. 1, the embodiments of the invention provide a kind of robot localization based on ceiling image information and map structure
System is built, the acquisition of vision information direction of the system is towards ceiling so that the consecutive image change of acquisition greatly, is easy to map structure
Build.The system is specifically included：Image collection module 1, image processing module 2, characteristic extracting module 3 and is deposited feature tracking module 4
Store up management module 5；
Wherein, image collection module 1, is arranged in robot, and its optical axis is perpendicular to ceiling, in robot motion
When ceiling is shot, obtain current ceiling image information；
Image collection module 1 can be mounted in robot, optical axis is perpendicular to video camera of ceiling etc..
Image processing module 2, for handling the ceiling image information that image collection module 1 is obtained；
Specifically, image processing module 2 is obtained to image collection module 1 ceiling image information carry out gray proces,
Distortion correction and cutting so that characteristic extracting module 3 is convenient to extract characteristics of image.
Image in the indoor ceiling video of shooting is the coloured image based on RGB color.Such image is not
Only memory space is big, while being also not easy to be handled.Therefore need these coloured images being converted to processing more easily
Gray-scale map.
Gray-scale map is a kind of monochrome image with 256 grades of tonal gradations from black to white.Each pixel in image is with 8
Data represent, therefore the gray value of pixel is one kind in 256 kinds of gray scales between black and white.Gray-scale map only has gray scale etc.
Level, without the change of color.
In YUV color spaces, Y-component represents the brightness degree of pixel, can express figure by this brightness value
The gray value of picture.According to RGB and YUV color space conversion formula, RGB image and the corresponding relation of Y-component can be obtained, it is such as public
Shown in formula：
(1)
RGB image can be converted to by gray-scale map according to formula (1).
Further, realized because image collection module 1 obtains image by optical principle, based on existing optics skill
The limitation of art level, it is impossible to produce perfect lens, and a lens and imager keeping parallelism also are difficult to from mechanical aspects, this
A little situations can cause lens distortion.For example：Should be that the lines of straight line are bent in picture, this distortion can be to vision
SLAM systems produce influence, because vision SLAM needs the characteristics of image in image information accurate as far as possible, and the meeting pair that distorts
Extraction, tracking and map structuring on characteristics of image are impacted.For the Corner Feature of tracking, the shadow of this distortion
Sound can make the Corner Feature that traces into produce the skew relative to actual environment, although displacement is smaller, on the move can be with
Self-recision is carried out, but for the robot matched according to Corner Feature is repositioned, Corner Feature distance can be influenceed
Judge.Therefore before using video camera, image collection module can be demarcated, obtains distort matrix, the figure got
As information is corrected using distortion matrix.
Simultaneously as it is bigger apart from the more remote distortion of image center location in the image obtained, for the ease of characteristics of image
Storage and the extraction of reduction wall feature are, it is necessary to the actual picture model for feature extraction of ratio that the image range of acquisition is set
Enclose small, that is, do not carry out feature extraction and tracking processing to image peripheral part, and when characteristics of image to be preserved, Ke Yiquan
Portion is preserved.
Characteristic extracting module 3, for extracting characteristics of image in the image information after the processing of image processing module 2, is obtained
Set of image characteristics；
Characteristics of image has Corner Feature, line feature, edge feature, textural characteristics, color lump feature etc., due to Corner Feature
Extract and using fairly simple quick, also allow for storing and applied to map structuring, therefore the embodiment of the present invention is with characteristics of image
To be illustrated exemplified by Corner Feature.
Specifically, when characteristics of image is Corner Feature, characteristic extracting module 3 can be carried out using Harris angle points method
Extract.Further, can be using sub-pix angular-point detection method to logical in order to be able to obtain the Corner Feature information of higher precision
Cross Harris angle point grids to Corner Feature be modified, obtain the exact position of Corner Feature.Harris angle point grids and
Sub-pix angular-point detection method belongs to the common knowledge of those skilled in the art, therefore is no longer described in detail in the present embodiment
Journey.
Feature tracking module 4, each characteristics of image that the characteristics of image for being extracted to characteristic extracting module 3 is concentrated
Carry out matched jamming, the set of image characteristics after being tracked；
Referring to Fig. 2, the embodiment of the present invention additionally provides another robot localization based on ceiling image information and ground
Figure constructing system schematic diagram, wherein, feature tracking module 4 include optical flow tracking module 41 and correction module 42, wherein light stream with
Track module 41, for by every in Corner Feature collection cornersA in Lucas-Kanade optical flow methods tracking present image imgA
One Corner Feature information, obtains tracking Corner Feature collection cornersB.
Specifically, optical flow tracking module 41 is realized by optical flow method and tracked, and light stream is referred to：Object band optical signature position
Mobile projector express the change of image to light stream, light stream is just formed on the plane of delineation, it contains the information of moving object.
Optical flow method comprises the following steps：
1）To each Corner Feature in Corner Feature collection cornersA in present image imgA, it is obtained in next frame
Correspondence position in image imgB, calculates the luminosity gradient Ix and luminosity gradient Iy in Y-direction in X-direction.Computational methods can
To utilize difference approximation, or calculated using operator, such as sobel operators.It is public that specific calculating process belongs to those skilled in the art
General knowledge is known, not in procedure detailed.
2）Under brightness permanence condition, with the Constrained equations of pixel near default window calculation Corner Feature；
3）The above-mentioned default equation group tried to achieve is carried out being converted to matrix M；
4）It is all unusual to judge Metzler matrix obtained above, if nonsingular, is obtained by solving in image imgB
CornersA tracking Corner Feature collection cornersB.
Lucas-Kanade optical flow methods belong to the common knowledge of those skilled in the art, therefore no longer detailed in the present embodiment
Carefully repeat said process.
In actual applications, what is often occurred due to video camera and photograph is image blurring, or video camera
Cause image that in order to attractive in appearance real world images are produced with modification due to carrying video effector, these all can angle steel joint spy
The extraction and tracking levied are impacted.For example, when the serrating phenomenon of lines image generation of oblique line directions, or image are rocked
There is larger grain effect in region that should be flat in a certain two field picture suddenly, may all be extracted as Corner Feature,
Then optical flow tracking module 41 may proceed to be tracked the Corner Feature of these erroneous judgements.This erroneous judgement is most normal with the situation of oblique line
Occur, due to determining to have extracted Corner Feature on oblique line, and the difference very little of each neighborhood territory pixel on oblique line, when taking the photograph
The image information that camera is shot with video camera move when, then these Corner Features are tracked it is possible that Corner Feature
Position without changing or mobile distance is less than the situation of actual range because oblique line just Vertical Square of lines when mobile
It is little to movement, and similitude is too many on lines so that tracking mistake.That is, optical flow tracking module 41 be actually pair
The tracking of pixel in one window, it is impossible to itself judge whether pixel is characterized a little in window.
Therefore, single to have above-mentioned defect using optical flow tracking module 41, the present invention further employs straightening die
Block 42, is corrected for obtaining tracking Corner Feature collection cornersB to above-mentioned optical flow tracking module 41.
Specifically, correction module 42 extracts Corner Feature again on current frame image, obtains new Corner Feature collection, will
The angle point that the tracking Corner Feature obtained by optical flow tracking module 41 concentrates each Corner Feature to be concentrated with new Corner Feature
The Corner Feature that feature is overlapped in the range of preset window is used as the Corner Feature collection after correction.
For example, setting current frame image as imgB, obtained Corner Feature is tracked on image imgB and is integrated as cornersB, corrected
Module 42 extracts Corner Feature on image imgB again, new Corner Feature collection cornersB2 is obtained, if wicket scope is
Cornerscale, this experiment span is the 7*7 windows using Corner Feature center as window center.Corner Feature will be tracked
The Corner Feature in Corner Feature collection cornersB2 is appeared in for most in wicket near each Corner Feature in collection cornersB
The Corner Feature determined eventually, that is to say, that by each Corner Feature in tracking Corner Feature collection cornersB and new Corner Feature
After the Corner Feature that Corner Feature in collection cornersB2 is overlapped in preset window scope cornerscale is as correction
Corner Feature collection, is added in preparation Corner Feature collection cornersC.It should be noted that Corner Feature here is to change into
New Corner Feature collection cornersB2 positional information, Corner Feature label information does not change.
Or, it is the schematic diagram of correction module 42 in Fig. 2 referring to Fig. 3, wherein, it is single that correction module 42 includes the first tracking
Member 421, the second tracking cell 422 and judging unit 423：
If current frame image is image imgA, its Corner Feature integrates as A, and next two field picture is imgB.
Wherein, the first tracking cell 421, for based on Corner Feature collection A, being obtained in next two field picture imgB tracking
Corner Feature collection B.
Second tracking cell 422, for based on Corner Feature collection B, angle point to be obtained in current frame image imgA tracking
Feature set B2.
Judging unit 423, for judge Corner Feature collection A each Corner Feature label whether with Corner Feature collection B2
Each Corner Feature label is identical, if identical, takes label identical Corner Feature to add in preparation Corner Feature collection cornersC,
Prepared for next frame image trace.
Relatively stable tracking Corner Feature information can be obtained to a certain extent by above-mentioned correction module 42, because
Correction module 42 can remove partly because feature substantially or to neighbouring Corner Feature more similar Corner Feature, retain
The Corner Feature for having stronger feature participates in the tracking of next two field picture.
Or, it is another schematic diagram of correction module 42 in Fig. 2 referring to Fig. 4, correction module 42 can also include extracting
Unit 424, predicting unit 425, comparison unit 426：
Wherein, extraction unit 424, the Corner Feature collection B2 for extracting present image；Some angle point is special in B2
Levy to be already present on and track the tracking Corner Feature collection B that the present image is obtained；
Predicting unit 425, the Corner Feature collection B ' for predicting present image according to the Corner Feature in Corner Feature storehouse；
The Corner Feature of comparison unit 426, for contrasting B2, B and B ', B and B will not be appeared in B2 ' is judged as newly
Corner Feature, continues certain frame number according to new angle point rule and judges the confidence level of the Corner Feature and by by the new angle point of threshold value
Feature is added in Corner Feature storehouse.
It should be noted that Corner Feature not all in the new Corner Feature finally obtained is relative to structure
The new Corner Feature in the Corner Feature storehouse of map is built, certain position, the current institute of image collection module 1 are moved to when there is robot
When the image range of acquisition spatially produces overlapping with the image once obtained, the new angle point that above-mentioned comparison unit 426 is obtained
There may be the Corner Feature being already present in Corner Feature storehouse in feature.
Or, correction module 42 can also use pyramid LK algorithms to be tracked；Or, it is special for mobile big angle point
The angle point tracing function after parameter regulation can be used to handle two front and rear frame pictures by levying, increase image triangle
The number of plies, position is moved larger Corner Feature can also judge.After the tracking of this two field pictures has been handled, follow-up is straight
The tracing function used when line mobile image is according to linear motion is calculated.
Memory management module 5, for storing the Corner Feature information that feature tracking module 4 is obtained, and it is special according to the angle point
Reference ceases map making.
Robot localization provided in an embodiment of the present invention based on ceiling image information and map structuring system, due to inciting somebody to action
The direction of video camera is changed to optical axis of the direction of motion perpendicular to video camera of the direction to ceiling, i.e. robot, therefore obtains
Consecutive image change it is more apparent, facilitate map structuring；Further, track characteristics of image when, in order to obtain preferably with
Track information, is corrected to traditional tracking result, the Corner Feature that tracking is obtained more is stablized and accurate, map structuring
Precision is also higher, that is, the map built is also just more accurate.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, some improvement and modification can also be made, these improvement and modification
Also it should be regarded as protection scope of the present invention.
Claims (1)
1. a kind of robot localization based on ceiling image information and map structuring system, it is characterised in that the system bag
Include：Image collection module, characteristic extracting module, feature tracking module and memory management module；
Described image acquisition module, is arranged in robot, its optical axis perpendicular to ceiling, in robot motion to day
Card is shot, and obtains current ceiling image information；
Described image processing module, for handling the ceiling image information that described image acquisition module is obtained；
The characteristic extracting module, for extracting feature in the image information after the processing of described image processing module, obtains figure
As feature set；
The feature tracking module, each image that the characteristics of image for being extracted to the characteristic extracting module is concentrated is special
Levy carry out matched jamming, the set of image characteristics after being tracked；
The memory management module, for storing the set of image characteristics that the feature tracking module is obtained, and according to described image
Feature set map making；
Described image is characterized as Corner Feature, and described image feature set is Corner Feature collection；
The feature tracking module includes optical flow tracking module and correction module；
The optical flow tracking module, for passing through the current figure of Lucas-Kanade optical flow methods tracking
The Corner Feature of picture concentrates each Corner Feature, obtains tracking Corner Feature collection；
The correction module, for it is described state optical flow tracking module obtain tracking Corner Feature collection be corrected；
The correction module extracts Corner Feature again on current frame image, obtains new Corner Feature collection, will be by described
The Corner Feature that the tracking Corner Feature that optical flow tracking module is obtained concentrates each Corner Feature to be concentrated with new Corner Feature exists
The Corner Feature overlapped in the range of preset window is used as the Corner Feature collection after correction；
The correction module includes the first tracking cell, the second tracking cell and judging unit：If current frame image is image
ImgA, its Corner Feature integrates as A, and next two field picture is imgB；
First tracking cell, for based on Corner Feature collection A, Corner Feature to be obtained in next two field picture imgB tracking
Collect B；Second tracking cell, for based on Corner Feature collection B, Corner Feature to be obtained in current frame image imgA tracking
Collect B2；
The judging unit, for judge the Corner Feature collection A each Corner Feature label whether with the Corner Feature collection
Each Corner Feature label in B2 is identical, if identical, takes label identical Corner Feature to add preparation Corner Feature and concentrates；
The correction module includes extraction unit, predicting unit, comparison unit：
The extraction unit, the Corner Feature collection B2 for extracting present image；Some Corner Feature has been deposited in B2
It is to track the tracking Corner Feature collection B that the present image is obtained；
The predicting unit, the Corner Feature collection B ' for predicting present image according to the Corner Feature in Corner Feature storehouse；
The Corner Feature of the comparison unit, for contrasting B2, B and B ', B and B will not be appeared in B2 ' is judged as new angle point
Feature, will be added in Corner Feature storehouse by the new Corner Feature of threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410149081.1A CN103886107B (en) | 2014-04-14 | 2014-04-14 | Robot localization and map structuring system based on ceiling image information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410149081.1A CN103886107B (en) | 2014-04-14 | 2014-04-14 | Robot localization and map structuring system based on ceiling image information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103886107A CN103886107A (en) | 2014-06-25 |
| CN103886107B true CN103886107B (en) | 2017-10-03 |
Family
ID=50954999
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410149081.1A Expired - Fee Related CN103886107B (en) | 2014-04-14 | 2014-04-14 | Robot localization and map structuring system based on ceiling image information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103886107B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107229274B (en) * | 2016-03-24 | 2022-06-28 | 松下电器（美国）知识产权公司 | Position indication method, terminal device, self-propelled device, and program |
| CN106197427A (en) * | 2016-07-04 | 2016-12-07 | 上海思依暄机器人科技股份有限公司 | Method, device and the robot of a kind of indoor positioning navigation |
| CN106959691B (en) * | 2017-03-24 | 2020-07-24 | 联想(北京)有限公司 | Mobile electronic equipment and instant positioning and map construction method |
| WO2019057179A1 (en) * | 2017-09-22 | 2019-03-28 | 华为技术有限公司 | Visual slam method and apparatus based on point and line characteristic |
| WO2019104740A1 (en) * | 2017-12-01 | 2019-06-06 | 深圳市沃特沃德股份有限公司 | Method and system for measuring odometer compensation coefficient of vision cleaning robot |
| WO2019104741A1 (en) * | 2017-12-01 | 2019-06-06 | 深圳市沃特沃德股份有限公司 | Method and system for measuring compensating coefficient for odometer of visual robot cleaner |
| CN108181610B (en) * | 2017-12-22 | 2021-11-19 | 鲁东大学 | Indoor robot positioning method and system |
| CN108180917B (en) * | 2017-12-31 | 2021-05-14 | 芜湖哈特机器人产业技术研究院有限公司 | Top map construction method based on pose graph optimization |
| CN108665508B (en) * | 2018-04-26 | 2022-04-05 | 腾讯科技（深圳）有限公司 | Instant positioning and map construction method, device and storage medium |
| CN108888188B (en) * | 2018-06-14 | 2020-09-01 | 深圳市无限动力发展有限公司 | Sweeping robot position calibration method and system |
| CN108983769B (en) * | 2018-06-22 | 2022-06-21 | 理光软件研究所（北京）有限公司 | Instant positioning and map construction optimization method and device |
| CN110244772B (en) * | 2019-06-18 | 2021-12-03 | 中国科学院上海微系统与信息技术研究所 | Navigation following system and navigation following control method of mobile robot |
| CN112683266A (en) * | 2019-10-17 | 2021-04-20 | 科沃斯机器人股份有限公司 | Robot and navigation method thereof |
| CN111898557B (en) * | 2020-08-03 | 2024-04-09 | 追觅创新科技(苏州)有限公司 | Map creation method, device, equipment and storage medium of self-mobile equipment |
| CN115962783B (en) * | 2023-03-16 | 2023-06-02 | 太原理工大学 | Positioning method of cutting head of heading machine and heading machine |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102596517A (en) * | 2009-07-28 | 2012-07-18 | 悠进机器人股份公司 | Control method for localization and navigation of mobile robot and mobile robot using same |
| CN103680291A (en) * | 2012-09-09 | 2014-03-26 | 复旦大学 | Method for realizing simultaneous locating and mapping based on ceiling vision |
-
2014
- 2014-04-14 CN CN201410149081.1A patent/CN103886107B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102596517A (en) * | 2009-07-28 | 2012-07-18 | 悠进机器人股份公司 | Control method for localization and navigation of mobile robot and mobile robot using same |
| CN103680291A (en) * | 2012-09-09 | 2014-03-26 | 复旦大学 | Method for realizing simultaneous locating and mapping based on ceiling vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103886107A (en) | 2014-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103886107B (en) | Robot localization and map structuring system based on ceiling image information | |
| CN111462135B (en) | Semantic mapping method based on visual SLAM and two-dimensional semantic segmentation | |
| CN111983639B (en) | Multi-sensor SLAM method based on Multi-Camera/Lidar/IMU | |
| CN110070615B (en) | Multi-camera cooperation-based panoramic vision SLAM method | |
| CN107635129B (en) | Three-dimensional trinocular camera device and depth fusion method | |
| Zuo et al. | Devo: Depth-event camera visual odometry in challenging conditions | |
| CN108492316A (en) | A kind of localization method and device of terminal | |
| CN106878687A (en) | A kind of vehicle environment identifying system and omni-directional visual module based on multisensor | |
| CN113674416B (en) | Three-dimensional map construction method and device, electronic equipment and storage medium | |
| CN113888639B (en) | Visual odometer positioning method and system based on event camera and depth camera | |
| JP5833507B2 (en) | Image processing device | |
| CN206611521U (en) | A kind of vehicle environment identifying system and omni-directional visual module based on multisensor | |
| CN113223045A (en) | Vision and IMU sensor fusion positioning system based on dynamic object semantic segmentation | |
| CN112541973B (en) | Virtual-real superposition method and system | |
| CN111998862A (en) | Dense binocular SLAM method based on BNN | |
| CN117593650B (en) | Moving point filtering vision SLAM method based on 4D millimeter wave radar and SAM image segmentation | |
| CN111681275A (en) | Double-feature-fused semi-global stereo matching method | |
| CN112652020B (en) | Visual SLAM method based on AdaLAM algorithm | |
| Savinykh et al. | Darkslam: Gan-assisted visual slam for reliable operation in low-light conditions | |
| CN109883433A (en) | Vehicle positioning method in structured environment based on 360 degree of panoramic views | |
| Dai et al. | A tightly-coupled event-inertial odometry using exponential decay and linear preintegrated measurements | |
| CN114511592B (en) | Personnel track tracking method and system based on RGBD camera and BIM system | |
| Zuo et al. | Cross-modal semi-dense 6-dof tracking of an event camera in challenging conditions | |
| CN113538510A (en) | Real-time workpiece tracking and positioning device on production line | |
| CN113610001B (en) | Indoor mobile terminal positioning method based on combination of depth camera and IMU |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170828 Address after: 200082 Handan Road, Shanghai, No. 220, No. Applicant after: Zhang Wenqiang Address before: 215000, Jiangsu City, Suzhou province Longgang large village 28 west side commercial Applicant before: SUZHOU HUATIANXIONG INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20181212 Address after: 636 Zixu Road, Xukou Town, Wuzhong District, Suzhou City, Jiangsu Province Patentee after: SUZHOU RADIANT PHOTOVOLTAIC TECHNOLOGY Co.,Ltd. Address before: 200082 No. 220 Handan Road, Yangpu District, Shanghai Patentee before: Zhang Wenqiang |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171003 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |