CN110444102A - Map constructing method, device and unmanned equipment - Google Patents
Map constructing method, device and unmanned equipment Download PDFInfo
- Publication number
- CN110444102A CN110444102A CN201810408970.3A CN201810408970A CN110444102A CN 110444102 A CN110444102 A CN 110444102A CN 201810408970 A CN201810408970 A CN 201810408970A CN 110444102 A CN110444102 A CN 110444102A
- Authority
- CN
- China
- Prior art keywords
- map
- coordinate system
- unmanned equipment
- under
- switching
- 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
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
- G09B29/003—Maps
- G09B29/005—Map projections or methods associated specifically therewith
Abstract
The disclosure proposes a kind of map constructing method, device and unmanned equipment, is related to unmanned technical field.A kind of map constructing method of the disclosure comprises determining that the environmental form of unmanned equipment local environment, environmental form include indoor bright light environments, outdoor bright light environments or dark surrounds；Map is constructed using associated map structuring mode according to environmental form；When environmental form changes, switches the map structuring mode used and calibrate the map of switching front and back.The method can be switched in time associated map structuring mode when environment changes to construct map, and fully take into account detector scale, position it is inconsistent caused by map structuring fracture the problem of, executing Map calibration makes the map linking of switching front and back, improves map structuring to the adaptability of environmental change.
Description
Technical field
This disclosure relates to unmanned technical field, especially a kind of map constructing method, device and unmanned equipment.
Background technique
SLAM (Simultaneous Localization and Mapping, instant positioning and map structuring) refers to and takes
When carrying the robots of certain sensors and being moved in foreign environment, in the premise for not being similar to this prior information of environmental map
Under, only by the observation of sensors towards ambient to current environment the dynamic construction cartographic model in the way of a kind of increment, while
It is determined during map structuring from the position in map.Therefore, it is based on SLAM technology, even if unmanned equipment is complete
It can still not complete to build figure and positioning, and realize independent navigation and path planning in the foreign environment of prior information entirely.
The environment of unmanned device side pair is diversification and complication, in this environment robot is had
Standby capacity of will, reliable and with robustness SLAM system is indispensable one of key technology.
Summary of the invention
Inventors have found that relevant SLAM system relies on single sensing data, GMapping, base such as based on laser
In the ORB (Oriented FAST and Rotated BRIEF, rapid characteristic points extract and description algorithm)-of monocular vision
SLAM, the RGBD-SLAM for being based on RGBD (Red Green Blue and Depth, RGB and depth) camera.When operation ring
When border meets some requirements, system can preferably complete map structuring and location tasks.Such as some relative quiescent
Indoor scene under carry out two-dimensional map building task, if robot be equipped with the sufficiently high laser radar of precision, by
Some SLAM systems, such as GMapping, this task can largely be completed by robot.But work as SLAM
When system needs long-play, when especially facing scene changes in the process of running, such as light conditions, the open degree of environment
When changing, the variation issue for handling this scene will be unable to.And in actual application, the variation of scene is very general
Time.Therefore, the map structuring technology under more scenes is an importance of the SLAM from theory to practical application.
The purpose of some embodiments of the present disclosure is to improve map structuring to the adaptability of environment.
According to some aspects of the disclosure, a kind of map constructing method is proposed, comprising: determine ring locating for unmanned equipment
The environmental form in border, environmental form include indoor bright light environments, outdoor bright light environments or dark surrounds；It is used according to environmental form
Associated map structuring mode constructs map；When environmental form changes, switch the map structuring mode used and school
The map of quasi- switching front and back.
Optionally, map structuring mode includes following at least two: in the case where environmental form is indoor bright light environments,
Map is constructed according to the depth data of depth camera detection；In the case where environmental form is outdoor bright light environments, according to vision
The image data of detector detection constructs map；In the case where environmental form is dark surrounds, detected using laser detector
Range data construct map.
Optionally, the map of calibration switching front and back includes: and is incited somebody to action according to the continuity of switching moment unmanned equipment pose
Under the map unification to the same coordinate system that switching front and back constructs.
Optionally, calibration switching front and back map include: obtain in the map structuring mode that switching moment stops nobody
Final pose of the steer under global coordinate system, coordinate original is in position when global coordinate system is started using unmanned equipment
Point；Obtain initial pose of the unmanned equipment under local coordinate system in the map structuring mode that switching moment opens；Root
According to final pose and initial pose, the coordinate system transformation matrix that initial pose is transformed to final pose is determined；According to coordinate system
Transformation matrix will be under the map that constructed after switching unification to global coordinate system.
Optionally, according to final pose and initial pose, determine that the coordinate system that initial pose is transformed to final pose becomes
Changing matrix includes:
According to formula
Tglobal=Tconversion·Tlocal
Determine transformation matrix of coordinates Tconversion, wherein TglobalFor position of the unmanned equipment under global coordinate system
Appearance, TlocalFor pose of the unmanned equipment under local coordinate system；Transformation matrix of coordinates TconversionIncluding spin matrix
RconversionWith translation vector tconversion, according to formula
cglobal=Rconversion·clocal+tconversion
Determine spin matrix RconversionWith translation vector tconversion, cglobalIt is unmanned equipment in global coordinate system
Under coordinate, clocalFor coordinate of the unmanned equipment under local coordinate system.
Optionally, the map of calibration switching front and back further include: when unmanned equipment enters outdoor bright light environments: according to
The coordinate of unmanned equipment determines the image data of visual detector detection in the map structuring mode that switching moment stops
Scale factor；According to the map under scale factor and image data building local coordinate system.
Optionally, map constructing method further include: if environmental form when unmanned equipment starts is outdoor bright ring
Border then determines the scale factor of image data according to the detection data of range sensor；According to scale factor and image data structure
Build the map under local coordinate system.
Optionally, map constructing method further include: if environmental form when unmanned equipment starts is outdoor bright ring
Border then constructs the map under local coordinate system according to image data；When environment changing occurs, according to what is opened in switching moment
The coordinate of unmanned equipment determines the scale factor of image data in map structuring mode；It is corrected according to scale factor according to figure
As the scale for the map that data construct, the map under global coordinate system is generated, and the map constructed before and after switching is unified to complete
Under office's coordinate system.
The method can be switched in time associated map structuring mode when environment changes to construct map, and
Fully take into account detector scale, position it is inconsistent caused by map structuring fracture the problem of, execute before Map calibration makes switching
Map linking afterwards, improves map structuring to the adaptability of environmental change.
According to other aspects of the disclosure, a kind of map structuring device is proposed, comprising: environment determination module is configured
For the environmental form of the unmanned equipment local environment of determination, environmental form include indoor bright light environments, outdoor bright light environments or
Dark surrounds；Map structuring module is configured as constructing map using associated map structuring mode according to environmental form；It cuts
Block is changed the mold, is configured as switching the map structuring mode used when environmental form changes；Calibration module is configured as
The map of calibration switching front and back.
Optionally, map structuring mode includes following at least two: in the case where environmental form is indoor bright light environments,
Map is constructed according to the depth data of depth camera detection；In the case where environmental form is outdoor bright light environments, according to vision
The image data of detector detection constructs map；In the case where environmental form is dark surrounds, detected using laser detector
Range data construct map.
Optionally, calibration module is configured as the continuity according to switching moment unmanned equipment pose, before switching
Under the map unification to the same coordinate system constructed afterwards.
Optionally, calibration module is configured as: obtaining unmanned in the map structuring mode that switching moment stops set
The standby final pose under global coordinate system, coordinate origin is in position when global coordinate system is started using unmanned equipment；It obtains
Take initial pose of the unmanned equipment under local coordinate system in the map structuring mode that switching moment opens；According to final
Pose and initial pose determine the coordinate system transformation matrix that initial pose is transformed to final pose；According to coordinate system transformation square
Battle array will be under the map that constructed after switching unification to global coordinate system.
Optionally, according to final pose and initial pose, determine that the coordinate system that initial pose is transformed to final pose becomes
Changing matrix includes: according to formula
Tglobal=Tconversion·Tlocal
Determine transformation matrix of coordinates Tconversion, wherein TglobalFor position of the unmanned equipment under global coordinate system
Appearance, TlocalFor pose of the unmanned equipment under local coordinate system；Transformation matrix of coordinates TconversionIncluding spin matrix
RconversionWith translation vector tconversion, according to formula
cglobal=Rconversion·clocal+tconversion
Determine spin matrix RconversionWith translation vector tconversion, cglobalIt is unmanned equipment in global coordinate system
Under coordinate, clocalFor coordinate of the unmanned equipment under local coordinate system.
Optionally, calibration module is also configured to when unmanned equipment enters outdoor bright light environments: according to switching
The coordinate of unmanned equipment determines the scale of the image data of visual detector detection in the map structuring mode that moment stops
The factor；According to the map under scale factor and image data building local coordinate system.
Optionally, if environmental form when map structuring module is additionally configured to the starting of unmanned equipment is outdoor bright
Environment then determines the scale factor of image data, according to scale factor and image data according to the detection data of range sensor
Construct the map under local coordinate system.
Optionally, if environmental form when calibration module is also configured to the starting of unmanned equipment is outdoor bright ring
Border then constructs the map under local coordinate system according to image data；When environment changing occurs, according to what is opened in switching moment
The coordinate of unmanned equipment determines the scale factor of image data in map structuring mode；It is corrected according to scale factor according to figure
As the scale for the map that data construct, the map under global coordinate system is generated, and the map constructed before and after switching is unified to complete
Under office's coordinate system.
According to the other aspect of the disclosure, a kind of map structuring device is proposed, comprising: memory；And it is coupled to and deposits
The processor of reservoir, processor are configured as based on being stored in the instruction execution of memory above any one map structuring side
Method.
This device can be switched in time associated map structuring mode when environment changes to construct map, and
Fully take into account detector scale, position it is inconsistent caused by map structuring fracture the problem of, execute before Map calibration makes switching
Map linking afterwards, improves map structuring to the adaptability of environmental change.
According to the further aspects of the disclosure, a kind of computer readable storage medium is proposed, be stored thereon with computer journey
Sequence instruction, realizes the step of executing above any one map constructing method when which is executed by processor.
By executing the instruction on such computer readable storage medium, can switch in time when environment changes
Map is constructed to associated map structuring mode, and fully takes into account detector scale, the inconsistent caused map in position
The influence of fracture is constructed, executing Map calibration makes the map linking of switching front and back, improves adaptation of the map structuring to environmental change
Ability.
In addition, proposing a kind of unmanned equipment, comprising: environmental sensor is configured according to some aspects of the disclosure
To detect environment locating for unmanned equipment；At least two in laser detector, depth camera or visual detector, matched
It is set to and provides detection data for building map；With above any one map structuring device.
The unmanned equipment can detect environment locating for unmanned equipment, switch in time when environment changes
Map is constructed to associated map structuring mode, and fully takes into account detector scale, the inconsistent caused map in position
The problem of building fracture, executing Map calibration makes the map linking of switching front and back, improves adaptation of the map structuring to environmental change
Ability.
Detailed description of the invention
Attached drawing described herein is used to provide further understanding of the disclosure, constitutes a part of this disclosure, this public affairs
The illustrative embodiments and their description opened do not constitute the improper restriction to the disclosure for explaining the disclosure.In the accompanying drawings:
Fig. 1 is the flow chart of some embodiments of the map constructing method of the disclosure.
Fig. 2 is the flow chart of some embodiments of Map calibration in the map constructing method of the disclosure.
Fig. 3 is the flow chart of other embodiments of Map calibration in the map constructing method of the disclosure.
Fig. 4 is the schematic diagram of some embodiments of the map structuring device of the disclosure.
Fig. 5 is the schematic diagram of other embodiments of the map structuring device of the disclosure.
Fig. 6 is the schematic diagram of the other embodiment of the map structuring device of the disclosure.
Fig. 7 is the schematic diagram of some embodiments of the unmanned equipment of the disclosure.
Specific embodiment
Below by drawings and examples, the technical solution of the disclosure is described in further detail.
The flow chart of some embodiments of the map constructing method of the disclosure is as shown in Figure 1.
In a step 101, the environmental form of unmanned equipment local environment is determined.In some embodiments, Ke Yigen
Environmental form division is carried out according to the wide degree of environment, light levels, such as environmental form is divided into indoor bright light environments, outdoor
Bright light environments and dark surrounds determine that unmanned equipment is in any environment by sensor.In some embodiments,
Object detecting apparatus can be cooperated to determine environmental form by light sensor equipment.
In a step 102, map is constructed using associated map structuring mode according to environmental form.
Since laser detector will receive interference in the case where illumination is strong, it is more suitable for dark surrounds, therefore in environment
In the case that type is dark surrounds, map is constructed using the range data of laser detector detection, it is such as real using GMapping
Existing map structuring.
Detection process can preferably be executed using camera in the stronger situation of light.Currently used camera includes clapping
The camera of two dimensional image is taken the photograph, and camera (such as binocular camera, 3D camera, RGBD camera, kinect phase with depth finding function
Machine etc.).
It the characteristics of depth information can be directly acquired based on depth camera, indoors can be more using depth camera under environment
Add direct and accurate acquisition range data, therefore in the case where environmental form is indoor bright light environments, according to depth camera
The depth data of detection constructs map, such as realizes map structuring using RGBD-SLAM.
In an outdoor environment, since environment is excessively open, the depth finding ability of depth camera is limited, therefore is more applicable in
In the camera using shooting two dimensional image, map is constructed by image procossing, map structuring is such as realized using ORB-SLAM.
In step 103, when environmental form changes, switch the map structuring mode used and calibrate switching front and back
Map.In some embodiments, different since the position of different its detector of map structuring mode is not fully identical
Map structuring mode possesses the coordinate system of itself, therefore is the seamless interfacing of map after realization switching, needs to calibrate switching front and back
Map.It in some embodiments, can be according to the continuity of switching moment unmanned equipment pose, by the building of switching front and back
Map it is unified under the same coordinate system.
The method can be switched in time associated map structuring mode when environment changes to construct map, and
The influence for fully taking into account detector scale, the inconsistent caused map structuring fracture in position, before execution Map calibration makes switching
Map linking afterwards, improves map structuring to the adaptability of environmental change.
In some embodiments, scene classification algorithm and object manipulator scene based on convolutional neural networks can be used
The Bayesian filter of detection optimizes, and realization efficiently and accurately detects indoor and outdoor, dark three kinds of environment.Environment measuring
Model obtains the image of current environment, and the image of current environment is then input to the environment classifier based on convolutional neural networks
In, environment classification is obtained as a result, being that environment classification result adds time and sky between consecutive image finally by Bayesian filter
Between correlation information, promote the stability and accuracy of classification results.
By such method, it can be improved environmental form and determine accuracy, reducing error influences, and avoids map structuring side
The switching repeatedly of formula, reduces operand, also improves the accuracy of map structuring.
The flow chart of some embodiments of Map calibration is as shown in Figure 2 in the map constructing method of the disclosure.
In step 201, obtain in the map structuring mode that switching moment stops unmanned equipment in world coordinates
Final pose under system, coordinate origin is in position when global coordinate system is started using unmanned equipment.
In some embodiments, used local coordinate system is as global coordinate system when can be to start；At another
In embodiment, the local coordinate used when starting first can be subjected to scale calibration, guarantee local coordinate system and full-size(d) phase
Global coordinate system is used as after symbol.
In step 202, obtain in the map structuring mode that switching moment opens unmanned equipment in local coordinate
Initial pose under system.
Final pose due to the continuity of unmanned equipment state, in the map structuring mode that switching moment closes
Should be identical or very much like with the initial pose in the map structuring mode of unlatching, therefore on the basis of the pose of switching moment
It can obtain the transformation relation of switching front and rear coordinate system.
In step 203, according to final pose and initial pose, the coordinate that initial pose is transformed to final pose is determined
It is transformation matrix.
In some embodiments, final pose is indicated with world coordinates, initial pose with the Local Coordinate Representations after switching,
According to formula
Tglobal=Tconversion·Tlocal
It can determine transformation matrix of coordinates Tconversion, wherein TglobalIt is unmanned equipment under global coordinate system
Pose, TlocalFor pose of the unmanned equipment under local coordinate system；Transformation matrix of coordinates TconversionIncluding spin matrix
RconversionWith translation vector tconversion, according to formula
cglobal=Rconversion·clocal+tconversion
It can determine spin matrix RconversionWith translation vector tconversion, cglobalIt is sat for unmanned equipment in the overall situation
Coordinate under mark system, clocalFor coordinate of the unmanned equipment under local coordinate system.
It in step 204, will be under the map that constructed after switching unification to global coordinate system according to coordinate system transformation matrix.In
It, can gradually will be with the unmanned equipment of the coordinate record under local coordinate system with the building of map in some embodiments
Pose is converted to global coordinate system, can also with above obtained in coordinate system transformation matrix correction switching after local coordinate
System relies on world coordinates to carry out map structuring.
By such method, the continuity that switching moment unmanned equipment pose can be utilized, by switching front and back structure
The map built is unified under global coordinate system, to construct continuous map under global coordinate system, can also make nobody drive
The positioning for sailing equipment is more accurate.
When in some embodiments, due to using visual pattern building map, the scale of visual pattern is indefinite, therefore meeting
Influence the ratio of map structuring.The flow chart of other embodiments of Map calibration such as Fig. 3 in the map constructing method of the disclosure
It is shown.
In step 301, judge whether to switch to outdoor bright light environments by dark surrounds or indoor bright light environments.If switching
Target be outdoor bright light environments, it is determined that map will be constructed using the image data of visual detector detection, and execute step
302。
In step 302, it is determined according to the coordinate of the unmanned equipment in the map structuring mode that switching moment stops
The scale factor of the image data of visual detector detection.In some embodiments, before the handover if using laser detector
The range data of detection constructs map, then determines detection range according to the detection result of laser detector；Before the handover if adopted
With the mode of the depth data building map detected according to depth camera, is then determined and visited according to the distance measurement result of depth camera
Ranging from.Detection range can be the height, the distance apart from barrier or z-axis coordinate of unmanned equipment.It is visited in vision
Survey in the image data of device detection, using the detection range demarcate unmanned equipment height or apart from identical barrier away from
From so as to obtain the scale factor of image data.
In step 303, according to the map under scale factor and image data building local coordinate system, so that this is locally sat
Mark system has the absolute measure for meeting truth.
It in step 304, will be under the map that constructed after switching unification to global coordinate system.
By such method, can make up for it visual pattern cannot directly obtain the deficiency of range information, before realizing switching
Scale afterwards is unified, improves the accuracy and continuity of map structuring.
It in some embodiments, can not be with vision figure if the startup environment of unmanned equipment is outdoor bright light environments
As directly acquiring scale factor.In some embodiments, range sensor detection range data be can use, and then obtain scale
The factor, thus make building local coordinate system under map size meet truth as far as possible, even if unmanned equipment not
In the case where carrying out environment changing, the accurate map of scale bar can be also constructed.
In some embodiments, in the case where the startup environment of unmanned equipment is outdoor bright light environments, when nobody
When steer switches to dark surrounds or indoor bright light environments, the continuity of switching moment unmanned equipment pose can use
Property, the scale of the global coordinate system before the range information correction switching obtained according to the map structuring mode switched to, to make
The scale of global coordinate system is more accurate, and then the map structuring mode switched to is unified to building ground under global coordinate system
Figure.In some embodiments, in the bright ring in outdoor before can also being corrected according to the scale factor of the global coordinate system after correction
The map constructed under border, to improve the accuracy of each stage map.
The schematic diagram of some embodiments of the map structuring device of the disclosure is as shown in Figure 4.
Environment determination module 401 can determine the environmental form of unmanned equipment local environment.In some embodiments,
Environmental form division can be carried out according to wide degree, the light levels of environment, such as environmental form is divided into indoor bright ring
Border, outdoor bright light environments and dark surrounds, determine that unmanned equipment is in any environment by sensor.In some realities
It applies in example, object detecting apparatus can be cooperated to determine environmental form by light sensor equipment.
Map structuring module 402 can construct map using associated map structuring mode according to environmental form.Due to
Laser detector will receive interference in the case where illumination is strong, be more suitable for dark surrounds, therefore be dark ring in environmental form
In the case where border, map is constructed using the range data of laser detector detection.Depth can be directly acquired based on depth camera
The characteristics of information, more can directly and accurately obtain range data using depth camera under environment indoors, therefore in ring
In the case that border type is indoor bright light environments, map is constructed according to the depth data of depth camera detection.In an outdoor environment,
Since environment is excessively open, the detectivity of depth camera is limited, therefore is more applicable for the camera using shooting two dimensional image,
Map is constructed by image procossing.
Switching module 403 can switch the map structuring mode used, comprising: black when environmental form changes
Map is constructed using the range data of laser detector detection in dark situation；It is visited in outdoor bright light environments according to visual detector
The image data of survey constructs map；Map is constructed according to the depth data of depth camera detection in bright light environments indoors.Work as nothing
When environmental form locating for people's steer changes, corresponding map structuring mode is switched to.
Calibration module 404 can calibrate the map that map structuring mode switches front and back when environmental form changes.By
Not fully identical in the position of different its detector of map structuring mode, different map structuring modes possesses the seat of itself
Mark system, therefore be the seamless interfacing of map after realization switching, need to calibrate the map of switching front and back.In some embodiments, may be used
With according to the continuity of switching moment unmanned equipment pose, the map constructed before and after switching is unified to the same coordinate system
Under.
This device can be switched in time associated map structuring mode when environment changes to construct map, and
Fully take into account detector scale, position it is inconsistent caused by map structuring fracture the problem of, execute before Map calibration makes switching
Map linking afterwards, improves map structuring to the adaptability of environmental change.
In some embodiments, environment determination module 401 include perception data acquiring unit, convolutional neural networks unit and
Bayesian filter optimizes unit, can use environment classification algorithm and object manipulator environment measuring based on convolutional neural networks
Bayesian filter optimization realize indoor and outdoor, dark three kinds of environmental forms are efficiently and accurately detected.Environment determines mould
Block 401 obtains the image in current environment by perception data acquiring unit, and the image of current environment is then input to convolution
In the environment classifier of neural network unit, environment classification is obtained as a result, being environment classification finally by Bayesian filter unit
As a result time and space correlation information between consecutive image is added, the stability and accuracy of classification results are promoted.
Such device can be improved environmental form and determine accuracy, and reducing error influences, and avoid map structuring mode
Switch repeatedly, reduce operand, also improves the accuracy of map structuring.
In some embodiments, calibration module 404 can obtain switching moment stop map structuring mode in nobody
Final pose of the steer under global coordinate system, and obtain in the map structuring mode that switching moment opens nobody drive
Sail initial pose of the equipment under local coordinate system.
Final pose due to the continuity of unmanned equipment state, in the map structuring mode that switching moment closes
Should be identical or very much like with the initial pose in the map structuring mode of unlatching, therefore on the basis of the pose of switching moment
It can obtain the transformation relation of switching front and rear coordinate system.
In some embodiments, it can be determined according to above-mentioned final pose and initial pose and be transformed to initial pose finally
The coordinate system transformation matrix of pose.Such as, final pose is indicated with world coordinates, and initial pose is with the local coordinate table after switching
Show, according to formula
Tglobal=Tconversion·Tlocal
Determine transformation matrix of coordinates Tconversion, wherein TglobalFor position of the unmanned equipment under global coordinate system
Appearance, TlocalFor pose of the unmanned equipment under local coordinate system；Transformation matrix of coordinates TconversionIncluding spin matrix
RconversionWith translation vector tconversion, according to formula
cglobal=Rconversion·clocal+tconversion
Determine spin matrix RconversionWith translation vector tconversion, cglobalIt is unmanned equipment in global coordinate system
Under coordinate, clocalFor coordinate of the unmanned equipment under local coordinate system.
It will be under the map that constructed after switching unification to global coordinate system according to coordinate system transformation matrix.In some embodiments
In, can with map building gradually by with the pose of the unmanned equipment of the coordinate record under local coordinate system convert to
It, can also be with the local coordinate system after coordinate system transformation matrix correction switching obtained in above, based on complete under global coordinate system
Office's coordinate carries out map structuring.
Such device can utilize the continuity of switching moment unmanned equipment pose, the ground that will be constructed before and after switching
Under figure unification to global coordinate system, so that continuous map is constructed under global coordinate system, the unmanned equipment that can also make
Positioning it is more accurate.
In some embodiments, when the targeted manner of switching is that the image data detected using visual detector constructs map
When, since the scale of visual pattern is indefinite, calibration module 404 is first according to the nothing in the map structuring mode that switching moment stops
The coordinate of people's steer determines the scale factor of the image data of visual detector detection, then according to scale factor and image
Data construct the map under local coordinate system, so that the local coordinate system has the absolute measure for meeting truth, and will cut
Under the map unification to global coordinate system constructed after changing.Such device, which can make up for it visual pattern, cannot directly obtain distance letter
The deficiency of breath realizes that the scale of switching front and back is unified, improves the accuracy and continuity of map structuring.
In some embodiments, when the startup environment of unmanned equipment is outdoor bright light environments, map structuring module
It can use range sensor detection range data, and then obtain scale factor, make the map scale under the local coordinate system of building
It is very little to meet truth as far as possible, even if can also be constructed compared with subject in the case where unmanned equipment is without environment changing
True map.
In some embodiments, in the case where the startup environment of unmanned equipment is outdoor bright light environments, work as environment
When determining module determines that unmanned equipment switches to dark surrounds or indoor bright light environments by outdoor bright light environments, calibration module
The continuity that can use switching moment unmanned equipment pose is believed according to the distance that the map structuring mode switched to obtains
The scale of global coordinate system before breath correction switching to keep the scale of global coordinate system more accurate, and then will switch to
Map is constructed under the unification to global coordinate system of map structuring mode.In some embodiments, calibration module can also be according to correction
The map constructed under outdoor bright light environments before the scale factor amendment of global coordinate system afterwards, thus with improving each stage
The accuracy of figure.
The structural schematic diagram of one embodiment of disclosure map structuring device is as shown in Figure 5.Map structuring device includes
Memory 501 and processor 502.Wherein: memory 501 can be disk, flash memory or other any non-volatile memory mediums.
Memory is used to store the instruction in the above corresponding embodiment of map constructing method.Processor 502 is coupled to memory
501, it can be used as one or more integrated circuits to implement, such as microprocessor or microcontroller.The processor 502 is for holding
The instruction stored in line storage can be improved map structuring to the adaptability of environmental change.
It in some embodiments, can be as shown in fig. 6, map structuring device 600 includes memory 601 and processor
602.Processor 602 is coupled to memory 601 by BUS bus 603.The map structuring device 600 can also be connect by storage
Mouthfuls 604 are connected to external memory 605 to call external data, can also be connected to by network interface 606 network or
The other computer system (not shown) of person.It no longer describes in detail herein.
In this embodiment, it is instructed by memory stores data, then above-metioned instruction is handled by processor, can be improved
Adaptability of the map structuring to environmental change.
In another embodiment, a kind of computer readable storage medium, is stored thereon with computer program instructions, this refers to
The step of enabling the method realized in map constructing method corresponding embodiment when being executed by processor.Those skilled in the art answer
Understand, embodiment of the disclosure can provide as method, apparatus or computer program product.Therefore, the disclosure can be used completely hard
The form of part embodiment, complete software embodiment or embodiment combining software and hardware aspects.Moreover, the disclosure can be used
The computer that one or more wherein includes computer usable program code can be with non-transient storage medium (including but not
Be limited to magnetic disk storage, CD-ROM, optical memory etc.) on the form of computer program product implemented.
The schematic diagram of some embodiments of the unmanned equipment of the disclosure is as shown in Figure 7.Map structuring device 70 can be with
For above any one map structuring device.Laser detector 72 can provide detection number under dark surrounds for building map
Bright light environments for building map detection data can be provided indoors according to, depth camera 73, visual detector 74 can be in outdoor
Bright light environments provide detection data for building map.
Environmental sensor 71 can determine environmental classes locating for unmanned equipment according to wide degree, the light levels of environment
Type.In some embodiments, environmental form may include indoor bright light environments, outdoor bright light environments and dark surrounds.Some
In embodiment, environmental sensor 71 may include light sensor equipment and object detecting apparatus, and in some embodiments, distance is visited
Measurement equipment can be laser detector 72 or depth camera 73, and light sensor equipment can be visual detector 74.At another
In embodiment, environmental sensor 71 can also be image acquisition equipment, can acquire unmanned equipment local environment, use
The Bayesian filter of scene classification algorithm and object manipulator scene detection based on convolutional neural networks optimizes, and realizes to room
Interior, outdoor, dark three kinds of environment are efficiently and accurately detected.In some embodiments, visual detector 74 can be arranged
To continue working state, detection data is provided for building map in outdoor bright light environments, and provide number for environmental form judgement
According to basis, data basis is provided for environmental form judgement in other environmental forms, is carried to reduce unmanned equipment
Equipment.
The unmanned equipment can detect environment locating for unmanned equipment, switch in time when environment changes
Map is constructed to associated map structuring mode, and fully takes into account detector scale, the inconsistent caused map in position
The problem of building fracture, executing Map calibration makes the map linking of switching front and back, improves adaptation of the map structuring to environmental change
Ability.
The disclosure is reference according to the method for the embodiment of the present disclosure, the flow chart of equipment (system) and computer program product
And/or block diagram describes.It should be understood that each process in flowchart and/or the block diagram can be realized by computer program instructions
And/or the combination of the process and/or box in box and flowchart and/or the block diagram.It can provide these computer programs to refer to
Enable the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to generate
One machine so that by the instruction that the processor of computer or other programmable data processing devices executes generate for realizing
The device for the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
So far, the disclosure is described in detail.In order to avoid covering the design of the disclosure, it is public that this field institute is not described
The some details known.Those skilled in the art as described above, completely it can be appreciated how implementing technology disclosed herein
Scheme.
Disclosed method and device may be achieved in many ways.For example, can by software, hardware, firmware or
Person's software, hardware, firmware any combination realize disclosed method and device.The step of for the method it is above-mentioned
Sequence is merely to be illustrated, and the step of disclosed method is not limited to sequence described in detail above, unless with other sides
Formula illustrates.In addition, in some embodiments, the disclosure can be also embodied as recording program in the recording medium, these
Program includes for realizing according to the machine readable instructions of disclosed method.Thus, the disclosure also covers storage for executing
According to the recording medium of the program of disclosed method.
Finally it should be noted that: above embodiments are only to illustrate the technical solution of the disclosure rather than its limitations；To the greatest extent
Pipe is described in detail the disclosure referring to preferred embodiment, it should be understood by those ordinary skilled in the art that: still
It can modify to the specific embodiment of the disclosure or some technical features can be equivalently replaced；Without departing from this public affairs
The spirit of technical solution is opened, should all be covered in the claimed technical proposal scope of the disclosure.
Claims (19)
1. a kind of map constructing method, comprising:
Determine that the environmental form of unmanned equipment local environment, the environmental form are bright including indoor bright light environments, outdoor
Environment or dark surrounds；
Map is constructed using associated map structuring mode according to the environmental form；
When the environmental form changes, switches the map structuring mode used and calibrate the map of switching front and back.
2. according to the method described in claim 1, wherein, the map structuring mode includes following at least two:
In the case where environmental form is indoor bright light environments, map is constructed according to the depth data of depth camera detection；
In the case where environmental form is outdoor bright light environments, map is constructed according to the image data of visual detector detection；
In the case where environmental form is dark surrounds, map is constructed using the range data of laser detector detection.
3. according to the method described in claim 2, wherein, the map of calibration switching front and back includes:
The continuity of unmanned equipment pose according to switching moment, the map constructed before and after switching is unified to same seat
Under mark system.
4. according to the method described in claim 3, wherein, the map of calibration switching front and back includes:
Obtain most final position of the unmanned equipment described in the map structuring mode that switching moment stops under global coordinate system
Appearance, coordinate origin is in the position when global coordinate system is started using the unmanned equipment；
Obtain initial bit of the unmanned equipment under local coordinate system described in the map structuring mode that switching moment opens
Appearance；
According to the final pose and the initial pose, the coordinate that the initial pose is transformed to the final pose is determined
It is transformation matrix；
It will be under the map that constructed after switching unification to the global coordinate system according to the coordinate system transformation matrix.
5. according to the method described in claim 4, wherein, described according to the final pose and the initial pose, determining will
The coordinate system transformation matrix that the initial pose is transformed to the final pose includes:
According to formula
Tglobal=Tconversion·Tlocal
Determine transformation matrix of coordinates Tconversion, wherein TglobalFor pose of the unmanned equipment under global coordinate system,
TlocalFor pose of the unmanned equipment under local coordinate system；Transformation matrix of coordinates TconversionIncluding spin matrix
RconversionWith translation vector tconversion, according to formula
cglobal=Rconversion·clocal+tconversion
Determine spin matrix RconversionWith translation vector tconversion, cglobalIt is the unmanned equipment in global coordinate system
Under coordinate, clocalFor coordinate of the unmanned equipment under local coordinate system.
6. according to method described in claim 3,4 or 5, wherein the map of calibration switching front and back further include: when the nothing
When people's steer enters outdoor bright light environments:
The coordinate of unmanned equipment according to the map structuring mode stopped in switching moment determines the visual detection
The scale factor of the image data of device detection；
According to the map under the scale factor and described image data building local coordinate system.
7. according to the method described in claim 2, further include:
If environmental form when the unmanned equipment starts is outdoor bright light environments,
The scale factor of described image data is determined according to the detection data of range sensor；
According to the map under the scale factor and described image data building local coordinate system.
8. the method according to claim 2 or 7, further includes:
If environmental form when the unmanned equipment starts is outdoor bright light environments, part is constructed according to image data and is sat
Map under mark system；
When environment changing occurs, the coordinate of unmanned equipment according to the map structuring mode opened in switching moment
Determine the scale factor of described image data；
According to the scale for the map that scale factor amendment is constructed according to image data, the map under global coordinate system is generated,
And it will be under the map unification to the global coordinate system that constructed before and after switching.
9. a kind of map structuring device, comprising:
Environment determination module is configured to determine that the environmental form of unmanned equipment local environment, the environmental form include
Indoor bright light environments, outdoor bright light environments or dark surrounds；
Map structuring module is configured as constructing map using associated map structuring mode according to the environmental form；
Switching module is configured as switching the map structuring mode used when the environmental form changes；
Calibration module is configured as the map of calibration switching front and back.
10. device according to claim 9, wherein the map structuring mode includes following at least two:
In the case where environmental form is indoor bright light environments, map is constructed according to the depth data of depth camera detection；
In the case where environmental form is outdoor bright light environments, map is constructed according to the image data of visual detector detection；
In the case where environmental form is dark surrounds, map is constructed using the range data of laser detector detection.
11. device according to claim 10, wherein the calibration module is configured as according to switching moment nobody
The continuity of steer pose, will be under the map unification to the same coordinate system that constructed before and after switching.
12. device according to claim 11, wherein the calibration module is configured as:
Obtain most final position of the unmanned equipment described in the map structuring mode that switching moment stops under global coordinate system
Appearance, coordinate origin is in the position when global coordinate system is started using the unmanned equipment；
Obtain initial bit of the unmanned equipment under local coordinate system described in the map structuring mode that switching moment opens
Appearance；
According to the final pose and the initial pose, the coordinate that the initial pose is transformed to the final pose is determined
It is transformation matrix；
It will be under the map that constructed after switching unification to the global coordinate system according to the coordinate system transformation matrix.
13. device according to claim 12, wherein it is described according to the final pose and the initial pose, it determines
Include: by the coordinate system transformation matrix that the initial pose is transformed to the final pose
According to formula
Tglobal=Tconversion·Tlocal
Determine transformation matrix of coordinates Tconversion, wherein TglobalFor pose of the unmanned equipment under global coordinate system,
TlocalFor pose of the unmanned equipment under local coordinate system；Transformation matrix of coordinates TconversionIncluding spin matrix
RconversionWith translation vector tconversion, according to formula
cglobal=Rconversion·clocal+tconversion
Determine spin matrix RconversionWith translation vector tconversion, cglobalIt is the unmanned equipment in global coordinate system
Under coordinate, clocalFor coordinate of the unmanned equipment under local coordinate system.
14. device described in 1,12 or 13 according to claim 1, wherein the calibration module be also configured to when it is described nobody
When steer enters outdoor bright light environments:
The coordinate of unmanned equipment according to the map structuring mode stopped in switching moment determines the visual detection
The scale factor of the image data of device detection；
According to the map under the scale factor and described image data building local coordinate system.
15. according to right want 10 described in device, wherein
If the environmental form when map structuring module is additionally configured to unmanned equipment starting is outdoor bright ring
Border then determines the scale factor of described image data according to the detection data of range sensor, according to the scale factor and institute
State the map under image data building local coordinate system.
16. according to right want 10 or 15 described in device, wherein
The calibration module is also configured to
If environmental form when the unmanned equipment starts is outdoor bright light environments, part is constructed according to image data and is sat
Map under mark system；
When environment changing occurs, the coordinate of unmanned equipment according to the map structuring mode opened in switching moment
Determine the scale factor of described image data；
According to the scale for the map that scale factor amendment is constructed according to image data, the map under global coordinate system is generated,
And it will be under the map unification to the global coordinate system that constructed before and after switching.
17. a kind of map structuring device, comprising:
Memory；And
It is coupled to the processor of the memory, the processor is configured to based on the instruction execution for being stored in the memory
Method as claimed in any one of claims 1 to 8.
18. a kind of computer readable storage medium, is stored thereon with computer program instructions, real when which is executed by processor
The step of method described in existing claim 1 to 8 any one.
19. a kind of unmanned equipment, comprising:
Environmental sensor is configured as detecting environment locating for the unmanned equipment；
At least two in laser detector, depth camera or visual detector, it is configured as providing detection number for building map
According to；
With,
Map structuring device described in claim 9~17 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810408970.3A CN110444102B (en) | 2018-05-02 | 2018-05-02 | Map construction method and device and unmanned equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810408970.3A CN110444102B (en) | 2018-05-02 | 2018-05-02 | Map construction method and device and unmanned equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110444102A true CN110444102A (en) | 2019-11-12 |
| CN110444102B CN110444102B (en) | 2021-10-01 |
Family
ID=68427612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810408970.3A Active CN110444102B (en) | 2018-05-02 | 2018-05-02 | Map construction method and device and unmanned equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110444102B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110986920A (en) * | 2019-12-26 | 2020-04-10 | 武汉万集信息技术有限公司 | Positioning navigation method, device, equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103786806A (en) * | 2014-01-20 | 2014-05-14 | 北京航空航天大学 | Multifunctional leg-and-wheel combination robot and multi-movement-mode intelligent switching method thereof |
| CN103900583A (en) * | 2012-12-25 | 2014-07-02 | 联想(北京)有限公司 | Device and method used for real-time positioning and map building |
| US20140309924A1 (en) * | 2013-04-16 | 2014-10-16 | Apple Inc. | Seamless transition from outdoor to indoor mapping |
| US20160179095A1 (en) * | 2013-08-08 | 2016-06-23 | Israel Aerospace Industries Ltd. | Classification of environment elements |
| CN107160395A (en) * | 2017-06-07 | 2017-09-15 | 中国人民解放军装甲兵工程学院 | Map constructing method and robot control system |
| CN107392547A (en) * | 2017-08-08 | 2017-11-24 | 北京京东尚科信息技术有限公司 | Goods handling method, apparatus, system and computer-readable recording medium |
| CN107421465A (en) * | 2017-08-18 | 2017-12-01 | 大连理工大学 | A kind of binocular vision joining method based on laser tracker |
| CN107917712A (en) * | 2017-11-16 | 2018-04-17 | 苏州艾吉威机器人有限公司 | A kind of synchronous superposition method and apparatus |
-
2018
- 2018-05-02 CN CN201810408970.3A patent/CN110444102B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103900583A (en) * | 2012-12-25 | 2014-07-02 | 联想(北京)有限公司 | Device and method used for real-time positioning and map building |
| US20140309924A1 (en) * | 2013-04-16 | 2014-10-16 | Apple Inc. | Seamless transition from outdoor to indoor mapping |
| US20160179095A1 (en) * | 2013-08-08 | 2016-06-23 | Israel Aerospace Industries Ltd. | Classification of environment elements |
| CN103786806A (en) * | 2014-01-20 | 2014-05-14 | 北京航空航天大学 | Multifunctional leg-and-wheel combination robot and multi-movement-mode intelligent switching method thereof |
| CN107160395A (en) * | 2017-06-07 | 2017-09-15 | 中国人民解放军装甲兵工程学院 | Map constructing method and robot control system |
| CN107392547A (en) * | 2017-08-08 | 2017-11-24 | 北京京东尚科信息技术有限公司 | Goods handling method, apparatus, system and computer-readable recording medium |
| CN107421465A (en) * | 2017-08-18 | 2017-12-01 | 大连理工大学 | A kind of binocular vision joining method based on laser tracker |
| CN107917712A (en) * | 2017-11-16 | 2018-04-17 | 苏州艾吉威机器人有限公司 | A kind of synchronous superposition method and apparatus |
Non-Patent Citations (7)
| Title |
|---|
| C. JIANG, D. P. PAUDEL, Y. FOUGEROLLE, D. FOFI，C. DEMONCEAUX: "Static-Map and Dynamic Object Reconstruction in Outdoor Scenes Using 3-D Motion Segmentation", 《IEEE ROBOTICS AND AUTOMATION LETTERS》 * |
| D. SHI, Y. WU, X. MO, R. WANG，J. WEI: "Activity Recognition Based on the Dynamic Coordinate Transformation of Inertial Sensor Data", 《2016 INTL IEEE CONFERENCES ON UBIQUITOUS INTELLIGENCE & COMPUTING》 * |
| RAÚL MUR-ARTAL，J. M. M. MONTIEL，JUAN D. TARDÓS: "ORB-SLAM: A Versatile and Accurate Monocular", 《 IEEE TRANSACTIONS ON ROBOTICS》 * |
| ZHEHANG TONG，DIANXI SHI，SHAOWU YANG: "SceneSLAM: A SLAM Framework Combined with Scene Detection", 《2017 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO)》 * |
| 刘浩敏，章国锋，鲍虎军: "基于单目视觉的同时定位与地图构建方法综述", 《计算机辅助设计与图形学学报》 * |
| 周绍磊,吴修振,刘刚,张嵘,徐海刚: "一种单目视觉ORB-SLAM/INS组合导航方法", 《中国惯性技术学报》 * |
| 李学龙，史建华，董永生，陶大程: "场景图像分类技术综述", 《中国科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110986920A (en) * | 2019-12-26 | 2020-04-10 | 武汉万集信息技术有限公司 | Positioning navigation method, device, equipment and storage medium |
| CN110986920B (en) * | 2019-12-26 | 2021-06-22 | 武汉万集信息技术有限公司 | Positioning navigation method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110444102B (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11127203B2 (en) | Leveraging crowdsourced data for localization and mapping within an environment | |
| Scaramuzza et al. | A toolbox for easily calibrating omnidirectional cameras | |
| JP4984650B2 (en) | Mobile device and self-position estimation method of mobile device | |
| CN109544636A (en) | A kind of quick monocular vision odometer navigation locating method of fusion feature point method and direct method | |
| Scaramuzza | Omnidirectional vision: from calibration to root motion estimation | |
| CN112785702A (en) | SLAM method based on tight coupling of 2D laser radar and binocular camera | |
| Dorfmüller | Robust tracking for augmented reality using retroreflective markers | |
| CN108352056A (en) | System and method for correcting wrong depth information | |
| Borrmann et al. | The project thermalmapper–thermal 3d mapping of indoor environments for saving energy | |
| CN111156998A (en) | Mobile robot positioning method based on RGB-D camera and IMU information fusion | |
| WO2020000395A1 (en) | Systems and methods for robust self-relocalization in pre-built visual map | |
| CN110375739A (en) | A kind of mobile terminal vision fusion and positioning method, system and electronic equipment | |
| CN112734765A (en) | Mobile robot positioning method, system and medium based on example segmentation and multi-sensor fusion | |
| KR102075844B1 (en) | Localization system merging results of multi-modal sensor based positioning and method thereof | |
| CN112258567A (en) | Visual positioning method and device for object grabbing point, storage medium and electronic equipment | |
| Kyriakoulis et al. | Color-based monocular visuoinertial 3-D pose estimation of a volant robot | |
| JP2017117386A (en) | Self-motion estimation system, control method and program of self-motion estimation system | |
| Aliakbarpour et al. | An efficient algorithm for extrinsic calibration between a 3d laser range finder and a stereo camera for surveillance | |
| CN108332752A (en) | The method and device of robot indoor positioning | |
| CN111161337A (en) | Accompanying robot synchronous positioning and composition method in dynamic environment | |
| CN112015187B (en) | Semantic map construction method and system for intelligent mobile robot | |
| Hochdorfer et al. | 6 DoF SLAM using a ToF camera: The challenge of a continuously growing number of landmarks | |
| Hu et al. | An indoor positioning framework based on panoramic visual odometry for visually impaired people | |
| Núnez et al. | Data Fusion Calibration for a 3D Laser Range Finder and a Camera using Inertial Data. | |
| CN110675455A (en) | Self-calibration method and system for car body all-around camera based on natural scene |
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: 20210302 Address after: Room a1905, 19 / F, building 2, No. 18, Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Beijing Jingdong Qianshi Technology Co.,Ltd. Address before: 101, 1st floor, building 2, yard 20, Suzhou street, Haidian District, Beijing 100080 Applicant before: Beijing Jingbangda Trading Co.,Ltd. Effective date of registration: 20210302 Address after: 101, 1st floor, building 2, yard 20, Suzhou street, Haidian District, Beijing 100080 Applicant after: Beijing Jingbangda Trading Co.,Ltd. Address before: 100195 Beijing Haidian Xingshikou Road 65 West Cedar Creative Garden 4 District 11 Building East 1-4 Floor West 1-4 Floor Applicant before: BEIJING JINGDONG SHANGKE INFORMATION TECHNOLOGY Co.,Ltd. Applicant before: BEIJING JINGDONG CENTURY TRADING Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |